The Mediterranean Basins: Tertiary Extension within the Alpine Orogen
Geological Society Special Publications Series Editors A. J. FLEET
R. E. HOLDSWORTH A. C. MORTON M. S. STOKER
It is recommended that reference to all or part of this book should be made in one of the following ways.
DURAND, B., JOLIVET, L., HORVATH, E d~; SERANNE,M. (eds) 1999. The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156. DOGLIONI, C., GUEGUEN, E., HARABAGLIA,P. ~f~MONGELLI,E 1999. On the origin of west-directed subduction zones and applications to the western Mediterranean. In: DURAND, B., JOLIVET, L., HORVATH, E & SERANNE,M. (eds) 1999. The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 541-561.
G E O L O G I C A L S O C I E T Y S P E C I A L P U B L I C A T I O N NO. 156
The Mediterranean Basins: Tertiary Extension within the Alpine Orogen EDITED
BY
BERNARD DURAND Institut Franqais du Petrole, Rueil Malmaison, France LAURENT JOLIVET Universit6 Pierre et Marie Curie, Paris, France FRANK HORVATH Lor~ind E6tv6s University, Budapest, Hungary and MICHEL St~RANNE Universit6 Montpellier 2, France
1999 Published by The Geological Society London
THE G E O L O G I C A L SOCIETY
The Society was founded in 1807 as The Geological Society of London and is the oldest geological society in the world. It received its Royal Charter in 1825 for the purpose of 'investigating the mineral structure of the Earth'. The Society is Britain's national society for geology with a membership of around 8500. It has countrywide coverage and approximately 1500 members reside overseas. The Society is responsible for all aspects of the geological sciences including professional matters. The Society has its own publishing house, which produces the Society's international journals, books and maps, and which acts as the European distributor for publications of the American Association of Petroleum Geologists, SEPM and the Geological Society of America. Fellowship is open to those holding a recognized honours degree in geology or cognate subject and who have at least two years' relevant postgraduate experience, or who have not less than six years' relevant experience in geology or a cognate subject. A Fellow who has not less than five years' relevant postgraduate experience in the practice of geology may apply for validation and, subject to approval, may be able to use the designatory letters C Geol (Chartered Geologist). Further information about the Society is available from the Membership Manager, The Geological Society, Burlington House, Piccadilly, London W l V 0JU, UK. The Society is a Registered Charity, No. 210161. Published by The Geological Society from: The Geological Society Publishing House Unit 7, Brassmill Enterprise Centre Brassmill Lane Bath BA1 3JN UK (Orders: Tel. 01225 445046 Fax 01225 442836)
Distributors
USA A A P G Bookstore PO Box 979 Tulsa OK 74101-0979 USA (Orders: Tel. (918) 584-2555 Fax (918) 560-2652)
First published 1999
Australia The publishers make no representation, express or implied, with regard to the accuracy of the information contained in this book and cannot accept any legal responsibility for any errors or omissions that may be made. 9 The Geological Society 1999. All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No paragraph of this publication may be reproduced, copied or transmitted save with the provisions of the Copyright Licensing Agency, 90 Tottenham Court Road, London WIP 9HE. Users registered with the Copyright Clearance Center, 27 Congress Street, Salem, MA 01970, USA: the item-fee code for this publication is 0305-8719/99/$15.00.
Australian Mineral Foundation 63 Conyngham Street Glenside South Australia 5065 Australia (Orders: Tel. (08) 379-0444 Fax (08) 379--4634)
India Affiliated East-West Press PVT Ltd G-1/16 Ansari Road New Delhi 110 002 India (Orders: Tel. (11) 327-9113 Fax (11) 326-0538)
Japan British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. ISBN 1-86239-033-9 Typeset by Type Study, Scarborough, UK Printed in Great Britain by Cambridge University Press, Cambridge, UK.
Kanda Book Trading Co. Cityhouse Tama 204 Tsurumaki 1-3-10 Tama-Shi Tokyo 206q3034 Japan (Orders: Tel. (0423) 57-7650 Fax (0423) 57-7651)
Contents
Foreword
VII
JOLIVET, L., FRIZON DE LAMOTTE,D., HORVATH,E, MASCLE,A. & SI~RANNE,M. The Mediterranean basins: Tertiary extension within the Alpine Orogen - an introduction Western Mediterranean
St2RANNE,M. The Gulf of Lion continental margin (NW Mediterranean) revisited by IBS: an overview
15
CHAMOT-ROOKE, N., GAULIER, J.-M. & JESTIN, E Constraints on Moho depth and crustal thickness in the Liguro-Provenqal basin from a 3D gravity inversion: geodynamic implications
37
VERGIe,S, J. & SABAT,E Constraints on the Neogene Mediterranean kinematic
63
evolution along a 1000 km transect from Iberia to Africa BENEDICTO, A., St~GURET,M. & LABAUME,P. Interaction between faulting, drainage and sedimentation in extensional hanging-wall syncline basins: example of the Oligocene Matelles basin (Gulf of Lion rifted margin, SE France)
81
ZECK, H. P. Alpine plate kinematics in the western Mediterranean: a westward-directed subduction regime followed by slab roll-back and slab detachment
109
MASCLE, A. & VIALLY, R. The petroleum systems of the Southeast Basin and Gulf of Lion (France)
121
WILSON,M. & BIANCHINI, G. Tertiary-Quaternary magmatism within the Mediterranean
141
and surrounding regions
MAUFFRET,A. & CONTRUCCI,I. Crustal structure of the North Tyrrhenian Sea: first
169
result of the multichannel seismic LISA cruise Pannonian Basin
HORVATH, E & TARI, G. IBS Pannonian Basin project: a review of the main results and their bearings on hydrocarbon exploration
195
TARI, G., DOVt~NYI,P., DUNKL, I., HORVATH,E, LENKEY,L., STEFANESCU,M., SZAFIAN,P. & TOTH, T. Lithospheric structure of the Pannonian Basin derived from seismic, gravity and geothermal data
215
GYORFI, I., CSONTOS,L. & NAGYMAROSY,A. Early Tertiary structural evolution of the border zone between the Pannonian and Transylvanian Basins
251
GERNER, P., BADA, G., DOVt~NYI,P., MULLER, B., ONCESCU,M. C., CLOETINGH, S. & HORWX~TH,E Recent tectonic stress and crustal deformation in and around the Pannonian Basin: data and models FODOR, L., CSONTOS,L., BADA, G., GYORFI, I. & BENKOVICS,L. Tertiary tectonic evolution of the Pannonian Basin system and neighbouring orogens: a new synthesis of palaeostress data
269 295
vi
CONTENTS
JUHASZ, E., PHILLIPS,L., MOLLER, P., RICKETTS, B., T(3TH-MAKK,/~., LANTOS,M. KovAcs, L. 6. Late Neogene sedimentary facies and sequences in the Pannonian Basin, Hungary
335
SACCHI, M., HORV.~TH, F. & MAGYARI,O. Role of unconformity-bounded units in stratigraphy of continental record: a case study from the Late Miocene of western Pannonian Basin, Hungary
357
VAN BALEN, R. T., LENKEY, L. HORVATH, F. & CLOETINGH,S. A. E L. Two-dimensional modelling of stratigraphy and compaction-driven fluid flow in the Pannonian Basin
391
Eastern Mediterranean
HATZFELD,D. The present-day tectonics of the Aegean as deduced from seismicity
415
JOLIVET,L. & PATRIAT,M. Ductile extension and the formation of the Aegean Sea
427
LIPS, A. L. W., WIJBRANS,J. R. & WHITE,S. H. New insights from 4~ laserprobe dating of white mica fabrics from the Pelion Massif, Pelagonian Zone, Internal Hellenides, Greece: implications for the timing of metamorphic episodes and tectonic events in the Aegean region
457
OKAY,A. I. & TOYsOz, O. Tethyan sutures of northern Turkey
475
General
ZIEGLER,P. A. & ROURE,E Petroleum systems of Alpine-Mediterranean fold belts
517
and basins DOGLIONI, C., GUEGUEN, E., HARABAGLIA,P. • MONGELLI,F. On the origin of west-directed subduction zones and applications to the western Mediterranean
541
FOREWORD This book derives from the Integrated Basin Studies Project (IBS), which ran during the years 1992-1995 with the support of the European Commission DGXII. The papers produced by the IBS group have been complemented by eight papers arising from the conference 'Mediterranean Basins: Tertiary Extension within the Alpine Orogen', held in Cergy-Pontoise, France, 11-13 December 1996. The title of this conference has also been retained as the title of the book. The papers included here collectively cover the majority of the Mediterranean Tertiary extensional basins, with the exception of those in the SW of the region. The book is the second in a series of three that are devoted to studies carried out in European Basins as part of the IBS project. The first volume, Cenozoic Foreland Basins of Western Europe, has already been published by the Geological Society (A. Mascle et al. 1998). A third volume concerned with the Norwegian rifted margin is currently in preparation. The main concept of the IBS project was to develop methods and techniques of modelling in which the main physical phenomena responsible for formation, development and the infilling of sedimentary basins or subbasins are linked together and related to large-scale, deep processes such as the convective movements of the mantle. Critical aspects of such models were as follows. The capacity to realistically link and describe the interactions of deep deformations to the near surface deformations that determine the geometrical evolution of basins and sub-basins. This has to be possible in both extensional and compressional tectonic contexts. The capacity to couple such tectonic models to simple but realistic models of linked erosion and sedimentation processes. The capacity to incorporate a model of fine-grained sediment compaction so as to include the effect of sediment loading and water escape on sediment deformation. The final objective was to obtain practical outputs and deliverables such as accounts of the architecture of basin or sub-basin fills (down to the reservoir scale), their evolution through geological times, fluid pressure regimes and stress histories and informations concerning the thermal evolution. Such methods and techniques are intended to result in better designed reservoir geological models and therefore to contribute to the improvement of field development planning and exploitation. In order to achieve these goals, the IBS teams used the following methodology. (1)
(2)
(3)
Using prototypes proposed mainly by groups from Vrije University in Amsterdam (S. Cloetingh and coworkers), models have been developed based on the existing knowledge concerning the rheology of crust and sediments and from published and unpublished data already available from various thoroughly studied basins in different tectonic settings. At the same time, a group, under the leadership of a team from Newcastle University (A. Aplin and colleagues) re-examined the compaction behaviour of fine-grained sediments and how it may be modelled using theoretical, experimental and observational approaches. A synthesis of this work has been published recently as a special issue of Marine and Petroleum Geology (A. Aplin & G. Vasseur (eds) 1998, Marine and Petroleum Geology, vol. 15, No. 2) A small number of European basins, set in their appropriate tectonic contexts, have been utilized as natural laboratories in which to document the interaction of tectonic and sedimentation processes using extensive syntheses of seismic well and data obtained in the field. This work resulted in a considerable updating of geological knowledge on the chosen basins and for this reason the IBS teams decided to publish these studies in the form of the present series of books.
These basins were chosen as follows. Tertiary rifted basins within the Alpine orogen: the Gulf of Lion in France and the Pannonian basin in Hungary were studied respectively under the leadership of the University of Montpellier (M. S6ranne and colleagues) and of the E6tv6s-Lorand laboratory in Budapest (F. Horvfith and colleagues).
viii
FOREWORD
Foreland basins: the south Pyrenean basins, the Guadalquivir basin in Spain, the molasse basin in Germany and the Barr~me syncline in France were studied respectively by teams from the University of Barcelona (M. Marzo and colleagues), from the Institut de Ciencias de la Terra, Barcelona (M. Fernandez and colleagues), from the University of Ttibingen (H. P. Luterbacher and colleagues), and ETH, Zurich (M. Ford and colleagues), associated under the leadership of the Servei Geol6gic de Catalunya (C. Puigdef~bregas and colleagues). The Norwegian margins: the northern Viking Graben, the Mere basin, the Voring basin and the mid-Norwegian Margin in Norway, studied by teams from the Norwegian universities and Norwegian oil companies under the leadership of Norsk Hydro (A. Nottvedt and colleagues). The cooperation with industry increased during the project and finally, 21 oil companies helped in a significant way. Some, like the Norwegian oil companies were directly involved. In particular, Norsk-Hydro had the leadership of module 3 (Dynamics of the Norwegian Margin). Others contributed less directly by giving documents or helping in the interpretation. The IBS teams acknowledge this help and thank these companies which are listed below (in alphabetical order): Amoco, BEB, BP, Coparex, DEE, EEP, Esso, INA Naftaplin, Norsk-Hydro, MOL, Mobil, OMV, Preussag, Repsol, RWE-DEA, Saga, Shell, Statoil, Total, VVNP, Wintershall. Overall, more than 200 researchers belonging to 38 institutions and 15 countries (eight E U countries, six non-EU European countries and the USA) have participated in the IBS project. The IBS teams have collaborated and participated with groups from the International Lithosphere Programme (Origin of Sedimentary Basins), the network E B R O (European Basin Research Organisation), and the 'Human Capital and Mobility' programme of the European Commission. Through the IBS project, the D G XII of the European Commission has clearly demonstrated its capacity to create a European research space. This capacity was enhanced by the access given to the IBS Program to a European program with Hungary, which resulted in the IBS program on the Pannonian basin. In addition, the cooperation of the Research Council of Norway (NFR) allowed the IBS-DMN program on the Dynamics of the Norwegian Margin to be launched. The Bundesamt for Bildung und Wissenschaft of Switzerland, also helped to launch the IBS-ETH cooperation. We also wish to acknowledge here the role played by these institutions and we thank them for their financial support. Many thanks are also due to DGXII experts and executives and particularly to J. C. Imarisio and J. M. Bemtgen for having helped to make the IBS project realistic and effective. The IBS teams are also indebted to M. Rougeaux, Managing Director of Groupement d'Etudes et de Recherches en Technologie des Hydrocarbures (GERTH) for his help in the management of the project. The European research space which has been created consists in academic teams who voluntarily worked on geological problems of interest for the oil industry, and who now have a solid capacity in this field. Many of these teams are now associated in the Eurobasin School, where a number of European Universities and Research Institutions cooperate, under the auspices of Academia Europea.
Mediterranean Basins: Tertiary Extension within the Alpine Orogen An international workshop on Mediterranean basins was held at the University of Cergy-Pontoise, France, on December 11-13 in 1996. It was co-organized by the University of Cergy-Pontoise and IFP. Funds were provided by the University, the Conseil G6n6ral du Val d'Oise and the Syndicat de l'Agglom6ration Nouvelle de Cergy-Pontoise. Forty-five oral communications were presented on all the major basins from the Alboran Sea to the Pannonian Basin, as well as on surrounding mountain belts. Recently acquired data, in particular those obtained within the framework of the IBS project, were presented and regional syntheses proposed. The coexistence in space and time of growing mountain belts and actively extending basins poses a number of yet unsolved questions in terms of mechanics. This problem is particularly crucial in the Mediterranean region where all Cenozoic basins opened in the internal zones of mountain belts. The Tyrrhenian Sea opened in the backarc region of the Apennines, the Aegean Sea in the backarc domain of the Hellenides and Hellenic arc, the Pannonian Basin behind the Carpathians and the Alboran Sea between the Betics and the Rift In some examples such as the Tyrrhenian Sea and the Aegean Sea, extension is still ongoing while peripheral compression and convergence are active. The Alboran and Pannonian basin are now in a stage of compression. Several models have been proposed to explain this coexistence of compression and extension: slab retreat during subduction process, detachment of a deep lithospheric root under the internal zones leading to radial extension and peripheral compression and slab detachment. The conference
FOREWORD
ix
acted as a forum for interactions between geologists and geophysicists in the study of the complex dynamic problem posed by the Mediterranean region. This volume presents a wealth of new data on various topics centered around the Mediterranean region from the deep mantle structure to the detailed geometry of sedimentary basins. B. D U R A N D Project Leader of IBS L. J O L I V E T Chairman of the Conference on Mediterranean Basins: Tertiary Extension within the Alpine Orogen
References APLIN,A. C. & VASSEUR,G. (eds) 1998. Geological Compaction of Fine Grained Sediments. Marine and Petroleum Geology, 15. CLOETINGH,S., DURAND,B. & PUIGDEFABREGAS,C. (eds) 1995. Integrated Basin Studies. Marine and Petroleum Geology, 12. DURAND, B. & MASCLE,A. 1996. Interest for the European Oil Industry of the Results Obtained by the Integrated Basin Studies JOULE Project no: CT92-120 In: 'The Strategic Importance of Oil and Gas Technology'. Proceedings of the 5th European Union Hydrocarbons Symposium, Edinburgh, 26-28 November 1996, 2, 1151-1167. MASCLE,A., PUIGDEFABREGAS,C., LUTERBACHER,H. P. • FERNANDEZ,M. (eds) 1998. Cenozoic Foreland Basins of Western Europe. Geological Society, London, Special Publications, 134.
The editors thank warmly the following persons for their participation in the review process: D. Avigad, J. M. Azafi6n, B. Biju-Duval, C. Bois, B. Bonin, D. Bonijoly, J. R. Borgomano, R. Caby, B. Colletta, J. M. Daniel, B. de Voogd, J. F. Dewey, R. W. England, C. Faccenna, D. Frizon, J. M. Gaulier, P. Gautier, P. Guennoc, I. Gy0rfy, F. Horv~th, T. Jacquin, A. Jambon, L. Jolivet, M. R. Leeder, I. Lerche, A. Lips, L. Longergan, A. Mascle, G. Matavelli, A. Mauffret, G. Nesen, J. Platt, A. Poisson, A. Richard, E. Roca, L. E. Ricou, F. Roure, M. S6ranne, G. Tari, P. Tremoli6res, J. Verg6s, R. Vially, A. B. Watts, J. Wheeler, T. White, H. Zeck. The following companies are thanked for their generous support of colour printing in the volume: BP, CGG, Elf-EP, Gaz de France, Institut Fran~ais du P6trole, Norsk Hydro and Total-Fina.
FOREWORD This book derives from the Integrated Basin Studies Project (IBS), which ran during the years 1992-1995 with the support of the European Commission DGXII. The papers produced by the IBS group have been complemented by eight papers arising from the conference 'Mediterranean Basins: Tertiary Extension within the Alpine Orogen', held in Cergy-Pontoise, France, 11-13 December 1996. The title of this conference has also been retained as the title of the book. The papers included here collectively cover the majority of the Mediterranean Tertiary extensional basins, with the exception of those in the SW of the region. The book is the second in a series of three that are devoted to studies carried out in European Basins as part of the IBS project. The first volume, Cenozoic Foreland Basins of Western Europe, has already been published by the Geological Society (A. Mascle et al. 1998). A third volume concerned with the Norwegian rifted margin is currently in preparation. The main concept of the IBS project was to develop methods and techniques of modelling in which the main physical phenomena responsible for formation, development and the infilling of sedimentary basins or subbasins are linked together and related to large-scale, deep processes such as the convective movements of the mantle. Critical aspects of such models were as follows. The capacity to realistically link and describe the interactions of deep deformations to the near surface deformations that determine the geometrical evolution of basins and sub-basins. This has to be possible in both extensional and compressional tectonic contexts. The capacity to couple such tectonic models to simple but realistic models of linked erosion and sedimentation processes. The capacity to incorporate a model of fine-grained sediment compaction so as to include the effect of sediment loading and water escape on sediment deformation. The final objective was to obtain practical outputs and deliverables such as accounts of the architecture of basin or sub-basin fills (down to the reservoir scale), their evolution through geological times, fluid pressure regimes and stress histories and informations concerning the thermal evolution. Such methods and techniques are intended to result in better designed reservoir geological models and therefore to contribute to the improvement of field development planning and exploitation. In order to achieve these goals, the IBS teams used the following methodology. (1)
(2)
(3)
Using prototypes proposed mainly by groups from Vrije University in Amsterdam (S. Cloetingh and coworkers), models have been developed based on the existing knowledge concerning the rheology of crust and sediments and from published and unpublished data already available from various thoroughly studied basins in different tectonic settings. At the same time, a group, under the leadership of a team from Newcastle University (A. Aplin and colleagues) re-examined the compaction behaviour of fine-grained sediments and how it may be modelled using theoretical, experimental and observational approaches. A synthesis of this work has been published recently as a special issue of Marine and Petroleum Geology (A. Aplin & G. Vasseur (eds) 1998, Marine and Petroleum Geology, vol. 15, No. 2) A small number of European basins, set in their appropriate tectonic contexts, have been utilized as natural laboratories in which to document the interaction of tectonic and sedimentation processes using extensive syntheses of seismic well and data obtained in the field. This work resulted in a considerable updating of geological knowledge on the chosen basins and for this reason the IBS teams decided to publish these studies in the form of the present series of books.
These basins were chosen as follows. Tertiary rifted basins within the Alpine orogen: the Gulf of Lion in France and the Pannonian basin in Hungary were studied respectively under the leadership of the University of Montpellier (M. S6ranne and colleagues) and of the E6tv6s-Lorand laboratory in Budapest (F. Horvfith and colleagues).
viii
FOREWORD
Foreland basins: the south Pyrenean basins, the Guadalquivir basin in Spain, the molasse basin in Germany and the Barr~me syncline in France were studied respectively by teams from the University of Barcelona (M. Marzo and colleagues), from the Institut de Ciencias de la Terra, Barcelona (M. Fernandez and colleagues), from the University of Ttibingen (H. P. Luterbacher and colleagues), and ETH, Zurich (M. Ford and colleagues), associated under the leadership of the Servei Geol6gic de Catalunya (C. Puigdef~bregas and colleagues). The Norwegian margins: the northern Viking Graben, the Mere basin, the Voring basin and the mid-Norwegian Margin in Norway, studied by teams from the Norwegian universities and Norwegian oil companies under the leadership of Norsk Hydro (A. Nottvedt and colleagues). The cooperation with industry increased during the project and finally, 21 oil companies helped in a significant way. Some, like the Norwegian oil companies were directly involved. In particular, Norsk-Hydro had the leadership of module 3 (Dynamics of the Norwegian Margin). Others contributed less directly by giving documents or helping in the interpretation. The IBS teams acknowledge this help and thank these companies which are listed below (in alphabetical order): Amoco, BEB, BP, Coparex, DEE, EEP, Esso, INA Naftaplin, Norsk-Hydro, MOL, Mobil, OMV, Preussag, Repsol, RWE-DEA, Saga, Shell, Statoil, Total, VVNP, Wintershall. Overall, more than 200 researchers belonging to 38 institutions and 15 countries (eight E U countries, six non-EU European countries and the USA) have participated in the IBS project. The IBS teams have collaborated and participated with groups from the International Lithosphere Programme (Origin of Sedimentary Basins), the network E B R O (European Basin Research Organisation), and the 'Human Capital and Mobility' programme of the European Commission. Through the IBS project, the D G XII of the European Commission has clearly demonstrated its capacity to create a European research space. This capacity was enhanced by the access given to the IBS Program to a European program with Hungary, which resulted in the IBS program on the Pannonian basin. In addition, the cooperation of the Research Council of Norway (NFR) allowed the IBS-DMN program on the Dynamics of the Norwegian Margin to be launched. The Bundesamt for Bildung und Wissenschaft of Switzerland, also helped to launch the IBS-ETH cooperation. We also wish to acknowledge here the role played by these institutions and we thank them for their financial support. Many thanks are also due to DGXII experts and executives and particularly to J. C. Imarisio and J. M. Bemtgen for having helped to make the IBS project realistic and effective. The IBS teams are also indebted to M. Rougeaux, Managing Director of Groupement d'Etudes et de Recherches en Technologie des Hydrocarbures (GERTH) for his help in the management of the project. The European research space which has been created consists in academic teams who voluntarily worked on geological problems of interest for the oil industry, and who now have a solid capacity in this field. Many of these teams are now associated in the Eurobasin School, where a number of European Universities and Research Institutions cooperate, under the auspices of Academia Europea.
Mediterranean Basins: Tertiary Extension within the Alpine Orogen An international workshop on Mediterranean basins was held at the University of Cergy-Pontoise, France, on December 11-13 in 1996. It was co-organized by the University of Cergy-Pontoise and IFP. Funds were provided by the University, the Conseil G6n6ral du Val d'Oise and the Syndicat de l'Agglom6ration Nouvelle de Cergy-Pontoise. Forty-five oral communications were presented on all the major basins from the Alboran Sea to the Pannonian Basin, as well as on surrounding mountain belts. Recently acquired data, in particular those obtained within the framework of the IBS project, were presented and regional syntheses proposed. The coexistence in space and time of growing mountain belts and actively extending basins poses a number of yet unsolved questions in terms of mechanics. This problem is particularly crucial in the Mediterranean region where all Cenozoic basins opened in the internal zones of mountain belts. The Tyrrhenian Sea opened in the backarc region of the Apennines, the Aegean Sea in the backarc domain of the Hellenides and Hellenic arc, the Pannonian Basin behind the Carpathians and the Alboran Sea between the Betics and the Rift In some examples such as the Tyrrhenian Sea and the Aegean Sea, extension is still ongoing while peripheral compression and convergence are active. The Alboran and Pannonian basin are now in a stage of compression. Several models have been proposed to explain this coexistence of compression and extension: slab retreat during subduction process, detachment of a deep lithospheric root under the internal zones leading to radial extension and peripheral compression and slab detachment. The conference
FOREWORD
ix
acted as a forum for interactions between geologists and geophysicists in the study of the complex dynamic problem posed by the Mediterranean region. This volume presents a wealth of new data on various topics centered around the Mediterranean region from the deep mantle structure to the detailed geometry of sedimentary basins. B. D U R A N D Project Leader of IBS L. J O L I V E T Chairman of the Conference on Mediterranean Basins: Tertiary Extension within the Alpine Orogen
References APLIN,A. C. & VASSEUR,G. (eds) 1998. Geological Compaction of Fine Grained Sediments. Marine and Petroleum Geology, 15. CLOETINGH,S., DURAND,B. & PUIGDEFABREGAS,C. (eds) 1995. Integrated Basin Studies. Marine and Petroleum Geology, 12. DURAND, B. & MASCLE,A. 1996. Interest for the European Oil Industry of the Results Obtained by the Integrated Basin Studies JOULE Project no: CT92-120 In: 'The Strategic Importance of Oil and Gas Technology'. Proceedings of the 5th European Union Hydrocarbons Symposium, Edinburgh, 26-28 November 1996, 2, 1151-1167. MASCLE,A., PUIGDEFABREGAS,C., LUTERBACHER,H. P. • FERNANDEZ,M. (eds) 1998. Cenozoic Foreland Basins of Western Europe. Geological Society, London, Special Publications, 134.
The Mediterranean Basins: Tertiary Extension within the Alpine O r o g e n - an introduction J O L I V E T , L. 1'2, D. F R I Z O N D E L A M O T T E 1, A. M A S C L E 3, M. S I ~ R A N N E 4
1DOpartement des Sciences de la Terre, E S A 7072 CNRS, Universitd de Cergy-Pontoise, Avenue du Parc, 8, le Campus, 95011 Cergy-Pontoise cedex, France 2present address: Laboratoire de Tectonique, E S A 7072 CNRS, UniversitO Pierre et Marie Curie, T 26-0 El, case 129, 4 Place Jussieu, 75252 Paris cedex 05, France 3IFP School, 228-232 avenue Napoleon Bonaparte, 92506 Rueil-Malmaison, France 4ISTEEM, Universitd Montpellier II, 34095 Montpellier cedex, France Abstract: The recent evolution of ideas on the Mediterranean region has been triggered by very active data acquisition over the last 15 years. Seismic tomography provides an unique view of mantle heterogeneities, space geodesy leads to precise determinations of the present strain and velocity fields, the combination of structural geology, radiometric dating and metamorphic petrology allows the description of P-T-t-D paths of exhumed metamorphic rocks, and exploration geophysics, onshore and offshore, gives a detailed view of the crustal geometry. Extension started in the Gulf of Lion and propagated eastwards and southwestwards to form the Liguro-Proven~ai basin, Tyrrhenian Sea and the Alboran Sea. It also started, at much the same time, in the Panonnian basin as well as in the Aegean back-arc region. Thus a seminal event occurred some 30 Ma ago that produced a sharp change from overall compression to back-arc extension. Although gravitational forces due to the collapse of a thick crust have affected most basins, it is now almost certain that this event ultimately originated in the mantle, either by slab detachement, slab rollback or both processes acting in sequence.
The reason why, suddenly, some 30 Ma ago, several extensional basins started to form within the overall compressional domain sandwiched between the African and European plates (Fig. 1), is a still open question. This problem was addressed by the IBS (Integrated Basins Studies) program and was the main topic of an international symposium, held at the Universit6 de Cergy-Pontoise (France) in December 1996 and co-organized by the Universit6 de CergyPontoise and by Institut Fran~ais du P6trole (IFP) (Fig. 2). In the past ten years or so, an entirely new view of the kinematic evolution and deformation pattern of the Mediterranean region has emerged, particularly with regards to these processes that may be responsible for the extension during basin formation. Early ideas proposed in the seventies involved mostly crustal dynamics (e.g. Tapponnier 1977), were superceded by m o r e recent models involving postorogenic gravitational collapse (Platt & Vissers 1989; Platt 1993; Gautier & Brun 1994a; Jolivet et al. 1994a), and mantle dynamics (Doglioni 1991; Wortel & Spakman 1992; Spakman et al. 1993; Platt & England 1994). Some of these ideas concerning dynamics were already in the literature
much earlier ( B e r c k h e m e r 1977; Le Pichon 1981), but it is only quite recently that the data sets have emerged to critically test model predictions. At the same time, several detailed 2D reconstructions at various scales were published based on plate kinematic data and geological information (Dercourt et al. 1986; Dewey et al. 1989; Dercourt et al. 1993; Ricou 1994). Such reconstructions have been used as tests of the 3D structure of the mantle obtained from tomography (de Jonge et al. 1993). Recent findings come from new tomographic techniques that provide a detailed view of the mantle heterogeneities, from space geodesy that leads to precise determinations of the present strain and velocity fields, from the combination of structural geology, radiometric dating and m e t a m o r p h i c petrology which allows the description of P-T-t-D paths of exhumed metamorphic rocks, and from exploration geophysics, onshore and offshore.
Lithospheric and crustal structure Seismic t o m o g r a p h y and seismic anisotropy recently changed our understanding of the Mediterranean region. A 3D view of the mantle
JOLWET,L., FRlZON DE LAMOTTE,D. MASCLE,A. & SI~RANNE,M. 1999. The Mediterranean Basins: Tertiary Extension within the Alpine Orogen - an introduction. In: DURAND,B., JOLIVET,L., HORVATH,F. & St~RANNE, M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156,1-14.
2
L. JOLIVET
ETAL.
"--~ 4..a
0
E . ~, .-...
g~z< 4....'
~.~
o~
.
9, ~
.~
~ ~ o=o o
~o
~
~
~-~ o
~a. ~
~-~
r
~:~ ,..~
~::~ . ~
INTRODUCTION
oZ
oz
oZ
3
oZ
oZ
oZ
o
o
~
/ o
~
9
o
o
I.~
9
-' . ,,,,~ 4..,a
.9
f
~
o
"G =
t~
~
8
r
o
oz
~
oZ
oZ
oz
Z
o
'q
4
L. JOLIVET E T A L .
is now available for the whole region with some remaining uncertainties. High seismic velocity anomalies extend seismic slabs at depths down to 1200 km. As shown on Fig. 1, deep slabs are seen everywhere below the actively extending domains though they are not continuous (de Jonge et al. 1993, 1994). Only a small dense body is seen east of the Pannonian Basin. Most slabs imaged by this technique suggested slab break off (de Jonge etal. 1994). Only the Aegean slab appears continuous down to 1000 km or more. More detailed analysis were performed in the Apennine region (Amato et al. 1993, 1998; Chiarraba & Amato 1996; Mele et al. 1997) with sometimes different results. They suggest the presence of a continuous high velocity anomaly in the upper mantle below the northern Apennines, contrasting with the absence of such anomaly below the central Apennines. The presence of the vertical cold slab below the northern Apennines is consistent with the occurrence of deep earthquakes at mantle depth at this latitude. A low velocity anomaly is observed at lower crustal depth below the most elevated part of the central Apennines (Chiarraba & Amato 1996). This suggests high temperature in the lower crust right above the cold slab recognized before where deep earthquakes occur. A similar stratification has been documented along a N-S transect across the Alboran Sea with an intermediate aseismic domain separating two seismic zones in the upper crust and upper mantle (Seber et al. 1996). The distribution of seismicity in the upper crust shows important variations from the frontal zones of subduction zones to back-arc regions. The seismic-aseismic transition is deeper close to the slab than in the back-arc domain. This has been observed across the northern Apennines and Tyrrhenian Sea as well as across the Hellenic arc (Amato et al. 1993; Hatzfeld et al. 1993). Two seismic anisotropy profiles were obtained across the northern and central Apennines (Margheriti et al. 1996; Amato et al. 1998). They reveal a progressive change in the detailed structure of the lithosphere from the foreland domain to the backarc region. The fast direction is parallel to the NW-trending belt in the frontal domain and is progressively rotated toward a more E - W strike in the back-arc region, from Tuscany to Elba island and Corsica. This E-W direction is parallel to the strike of stretching lineations in the exhumed metamorphic rocks along the same transect (Fig. 1). Lithospheric thickness below the Pannonian Basin reflects an Early to Mid-Miocene
extensional event rather than the present-day dynamics which has been compressional since the Pliocene (Tari et aL this volume). Tertiary oceanic crust is present in the LiguroProvencal Basin (Burrus 1984) and southern Tyrrhenian Sea (Kastens et al. 1988). Crustal thickness in the Liguro-Provenqal Basin is only about 5 km on average which is thinner than typical oceanic crust. This suggests a mantle temperature lower than normal and a slow spreading rate (Chamot-Rooke et al. this volume). Wide-angle refraction and the ECORS reflection profiles, as well as inversion of gravity data (De Voogd et al. 1991; Pascal et al. 1993) show the progressive attenuation of crustal thickness across the Gulf of Lion margin. If moderate extension, mostly thin-skinned, is seen in the upper margin, a classical geometry of tilted blocks and intracrustal detachments as well as ductile thinning of the lower crust are observed in the offshore domain. These data provide a detailed image of the lithosphere and the crust. High-temperature anomalies are often present in the lower crust and upper mantle below the actively extending regions suggesting a sharp change in the thermal regime after the formation of a crustal scale thrust wedge. The extension seen at the surface is parallel to the fast direction of seismic waves in the deeper portions of the lithosphere suggesting that the whole lithosphere is deformed in a similar way during extension and that the field observations at the surface are significant at lithospheric scale.
Present-day kinematic framework Global kinematic data (De Mets et al. 1990) show a slow convergence between Africa and Eurasia in a NNW-SSE direction. The rate of convergence increases from less than 5 mm a-1 in the west to approximately 1 cm a -I in the east. Space geodesy provides kinematic data within the Mediterranean deformed region. GPS and SLR measurement in the Aegean show the almost rigid extrusion of the Anatolian block delimited to the north by the North Anatolian Fault, to the west by a series a grabens (such as the Corinth Gulf) and to the south by the Hellenic subduction (Le Pichon et al. 1994; Kahle et al. 1995, 1996; Reilinger et al. 1995; Noomen et al. 1996). This pattern is very similar to the earlier model proposed by McKenzie (1978). The Anatolian block rotates counterclockwise relative to Eurasia about a pole located north of Sinai. The instantaneous rate of westward expulsion of Anatolia is much higher than the rate of convergence between Eurasia and Africa so that
INTRODUCTION the velocity of subduction in the Hellenic trench is mostly controlled by the Anatolia-Africa relative motion which amounts to more than 3 cm a -1. N-S extension in the Gulf of Corinth results from the accomodation of the same motion with respect to stable Eurasia. More than 1.5 cm a -1 of extension are calculated across the Gulf from the regional data set, and approximately the same amount is measured directly across the Gulf (Rigo 1994). Velocities are much slower in the western Mediterranean. Measurement around the Tyrrhenian Sea reveals a slow N-S shortening in the whole region except in the SE (southern Apennines) where a strong E - W extension is measured (Ward 1994). Studies of active faults, focal mechanisms and breakout analysis show stress and strain patterns in the Aegean and Tyrrhenian regions (Armijo et al. 1992, 1996; Frepoli & Amato 1997; Hatzfeld et al. 1990, 1993; Jackson 1994; Montone et al. 1997; Hatzfeid this volume). Radial compression is active at the periphery of extensional basins. Compression is active along the Hellenic trench and its strike changes from E N E - W S W in the west to NNE-SSW in the east (Hatzfeld et al. 1993, 1997; Hatzfeld this volume). Radial compression is also observed around the northern Apennines (Frepoli & Amato 1997). A radial pattern of the compressional stress axis is also recorded in the Pannonian Basin (Gerner et a L this volume). E - W extension is predominant in the northern Apennines west of the crest line. Most of the southern Apennines are under an extensional regime. N-S extension predominates in the Aegean Sea, and is mostly active along the western margin of the Anatolian block as in the Corinth Gulf, whereas arc-parallel extension is recorded in Crete and Peloponnese (Hatzfeld et al. 1993, 1997). The present-day kinematics of the Aegean region is strongly influenced by the westward extrusion of the Anatolian block and the most active extension is recorded in the Gulf of Corinth (Le Pichon et al. 1994). In this example extrusion tectonics strongly shapes the extensional region (McKenzie 1972). Elsewhere the present-day kinematic pattern shows radial extension in the internal zones and compression in the external zones.
Major structures and kinematic evolution Apart from the North Anatolian Fault, which results from the extrusion of the Anatolian block out of the Arabia-Africa collision zone (McKenzie 1972, 1978) and the Mid-Hungarian Shear Zone (Gyfrfi et aL this volume, van Balen
5
et aL this volume), there are no large-scale strike-slip faults in the Mediterranean domain during the Tertiary. Some minor strike-slip component is recorded locally along some normal faults in the Apennines. The Kefallonia fault transfers the convergence from the front of the Mediterranean ridge to the Ionian islands (Le Pichon et al. 1997). A strike-slip component is also postulated along some NE-trending faults in the eastern Betic Cordillera and in the eastern Rif (Frizon de Lamotte et al. 1991; Leblanc & Olivier 1984). Contractional structures are often highly reworked by late orogenic extension. A description of the pre-extension structure and tectonic history is provided by Okay & Tiiysiiz (this volume) from an example in Western Turkey. Radial compression is the rule around major basins and mountain belts with the notable exception of the Atlas and the Pyr6n6es. All belts show an outward migration of the foreland domain (Malinverno & Ryan 1986; Patacca & Scandone 1989; D'Offizi et al. 1994; Jolivet et al. 1994b, 1998; Linzer 1996; Lonergan & White 1997; Fodor et aL this volume; Gy6rfi et al. this
volume; Jolivet and Patriat this volume; Verges & S~bat this volume; Zeck this volume). Lithospheric extension is predominantly E - W in the Western Mediterranean, N-S in the Eastern Mediterranean and E - W in the Pannonian basin. The first large-scale structure to form is the West European Rift System which runs from the east Alboran basin to the Rhine Graben (Brunet al. 1992, S~ranne this volume). This Oligocene structure appears to have formed independently from the backarc basins which developed afterward (S~ranne this volume). It is oblique to the alpine thrust front which is clearly cut in Provence in the eastern Alboran region (Doglioni et al. 1997). The Liguro-Proven~al basin and the Valencia Trough developed mainly on normal thickness continental crust (Mauffret et al. 1995). However its northern part (Gulf of Lion and Ligurian Sea) formed on a continental crust which had been previously thickened during the formation of the Pyr6n6es, Languedoc and Provence mountain chains and during the formation of the Alps (Corsica). The width of the margins varies greatly and the pre-extension crustal thickness might explain these variations (S~ranne this volume). The Liguro-Proven~al basin can be considered to display the characteristics of a normal oceanic basin (Gueguen 1995). Large-scale palaeomagnetic rotations are documented in and around all basins. They were first discovered in the Aegean region (Laj et al.
6
L. JOLIVET E T A L .
1982; Kissel & Laj 1988) and have been subsequently found in the Southern Apennines (Sagnotti 1992; Scheepers et al. 1993; Mattei et al. 1995), in the Pannonian Basin (Gy6rti et a L this volume; van Balen et ai. this volume) and the Betic Cordillera (Allerton et al. 1993; Allerton 1994). If a complex pattern of rotations is seen in the Alboran domain, the Aegean, Tyrrhenian and Pannonian Basins show rather simple outward rotations forming the present-day arcuate geometry. The Alboran, Tyrrhenian and Aegean Seas and the Pannonian Basin formed as postorogenic basins superimposed on compressional structures (Dewey 1988; Faccenna et al. 1997; Mauffret & Contrucci, this volume; Tari et ai. this volume). Extension is distributed within large domains and is associated with the exhumation of high temperature metamorphic core complexes (Lister et al. 1984; Dinter & Royden 1993; Sokoutis et al., 1993; Gautier & Brun 1994b; Jolivet et al. 1998; Platt & England 1994; Jolivet & Patriat this volume). The direction of extension recorded in these deep-seated core complexes is parallel to the present-day extension vector. As a whole, vertical-axis rotations have not perturbed significantly the fossil strain field during extension, it should be noted, however, that the direction of extension seen in Alpine Corsica should be rotated back to a more NW-SE direction before the opening of the Liguro-Provenqal basin (Montigny et al. 1981; Vigliotti & Kent 1990). The observed parallelism of ancient and modern extension directions is quite easy to understand in the northern Tyrrhenian margin where palaeomagnetic data show that no rotation has occurred (Mattei et al. 1996). It is more puzzling in the Aegean where clockwise rotations are documented onland and in several islands of the Cyclades archipelago (Kissel & Laj 1988). Some of the NE-trending lineations of the Cyclades have probably been rotated by 10-20 ~ clockwise after their exhumation (Avigad et al. 1998; Morris & Anderson 1996). This raises the question of the thickness of rotating blocks, whether the whole crust is affected by the rotations or only its upper part (Allerton 1993, 1994). During the most important stage of extension (Miocene) N-S or NE-SW extension is observed in the whole Aegean Sea from Crete to the northern Cyclades, while E - W extension predominates in the Tyrrhenian and Alboran seas (Jolivet et al. 1994b, 1998; Vissers et al. 1995). Extension is achieved by large-scale, steeply dipping, normal faults, which usually localize the largest earthquakes. These faults control the topography of the Aegean region in the Peloponese and
southeastern continental Greece (Jackson & White 1989; Armijo et al. 1992, 1996; Jackson 1994) and in the internal Apennines until the maximum of topography (D'Agostino et al. 1998). The most important topographic gradient of Corsica is a large, east-dipping, normal fault (Daniel et al. 1996). The internal domain of the Apennines, the northern Tyrrhenian sea, the Alboran sea, the eastern Betics and the Cyclade archipelago all display a basin and range topography controled by normal faults. Extension is also achieved along flat-lying normal faults and extensional shear zones. The sense of shear is constant over large regions (100-200 km wide). The hanging wall is always displaced toward the external zones in the northern Tyrrhenian Sea and internal Apennines (Jolivet et al. 1998) as well as in the BeticRif orogen (Vissers et al. 1995). It is instead displaced toward the internal zones in the Aegean Sea (Gautier & Brun 1994a; Jolivet & Patriat this volume). A continuum is observed from ductile to brittle extension. The most spectacular examples occur in the Aegean Sea where shallow, north-dipping ductile shear zones active in the Cyclades during the Miocene are relayed by more recent north-dipping, steep normal faults which root into shallow, north-dipping d6collements within the brittle-ductile transition zone (Rigo et al. 1996; Jolivet & Patriat this volume). Extension reactivates earlier thrusts and basins can develop above ramp and flat decollements such as seen, for example, along the rifted margin of the Gulf of Lion (Benedicto et al. this volume). A continuum is also observed from the deformation in the frontal zones near the subduction slab where high-pressure and low-temperature core complexes are exhumed below detachments (Crete) (Fassoulas et al. 1994; Jolivet et al. 1996, Jolivet & Patriat this volume) and in the back-arc region where high-temperature and low-pressure core complexes come to the surface below detachements with the same geometry (Cyclades) (Buick 1991; Gautier et al. 1993). This continuum is associated with a migration of extension and compression toward the external regions (Malinverno & Ryan 1986; Lonergan & White 1997; Jolivet et al. 1998; Jolivet & Patriat this volume) and by a coeval migration of magmatism, at least in the Tyrrhenian and Aegean regions (Serri et al. 1993) and the Pannonian Basin (Linzer 1996). The migration of deformation regimes and styles, and from the Miocene to the present are significant observations that should be included in lithospheric-scale models.
INTRODUCTION
Rifting history Inception of extension
7
(GyOrfi et aL this volume; van Balen et al. this volume). They were deposited during fast
Subduction is thought to be the major driving mechanism for compression and extension in the Mediterranean and its history can be controlled by the related magmatism (Wageman et al. 1970; Fytikas et al. 1984; Serri et al. 1993;
outward rotations of the Carpathian arc, counterclockwise in the north, clockwise in the south. Back-arc extension and eastward migration of the thrust front is then recorded during an eastward slab retreat and progressive locking of the northern border of the basin.
Sacchi et aL this volume; Wilson & Bianchiui this volume). The history of rifting is summar-
35 Ma. Oceanic lithosphere (Fig. 3a) subducts
ized in four figures spanning the recent evolution from the Oligocene to the Present (Fig. 3). They are modified from Dercourt et al. (1993, 1986). Slab rollback seemingly starts to produce back-arc extension around 30 Ma in the whole Mediterranean region. In the Western Mediterranean the opening of the Liguro-Provenqal, Alboran and Tyrrhenian basins can be described as a continuum starting some 30 Ma ago in the Gulf of Lion. The Late Oligocene and Early Miocene extensional structures in Alpine Corsica (Jolivet et al. 1998) formed in the time gap that exists between the Liguro-Provenqal and Tyrrhenian rifting episodes. Extension in Corsica started approximately 30 Ma ago and lasted until the Mid-Miocene at least. ChamotRooke et aL (this volume) also argue for a continuum of extension based on a new compilation of palaeomagnetic data and crustal structure. S~ranne (this volume) in his overview on the Gulf of Lion and Mascle & Vially (this volume) conclude that the back-arc rifting episode should be differentiated from the earlier formation of the West European Rift which started earlier. Extension in the Alboran Sea started in the Aquitanian (23-20 Ma) (Comas et al. 1992, 1996; Lonergan & White 1997) as a propagation toward the southwest of the back-arc basin initiated in the Gulf of Lion.
Zeck (this volume), Verges & S~bat (this volume) and Doglioni et al. (this volume) describe kinematic models which take into account this progressive extensional history. The timing of extension in the Aegean Sea is less well constrained. The first marine basins were formed in the Cyclades in the Aquitanian and the oldest radiometric dates in high-temperature metamorphic core complexes are around 25 Ma (Altherr et al. 1982; Gautier & Brun 1994a; Wijbrans & McDougall 1988). Extension should have started early enough to allow the exhumation of metamorphic rocks and a significant decrease of crustal thickness. The migration of subduction-related volcanism suggests a rollback from 30 Ma to the present (Fytikas et al. 1984; Jolivet et al. 1998). Early syn-rift deposits in the Pannonian Basin date back to the Early Miocene or Late Oligocene
below the Aegean domain and the Western Mediterranean region. Tomographic data suggest that the Hellenic subduction was initiated at least 40 Ma ago (Spakman et al. 1993) which significantly differs from classical interpretations which suggest that subduction started only in the Neogene. The subducting oceanic lithosphere is of Mesozoic age, probably Cretaceous (Truffert-Luxey 1992) and can thus subduct easily below the southern margin of Eurasia. Slab rollback starts soon after the initiation of subduction. The first extensional event is the formation of the West European Rift
(S~ranne this volume). 20 Ma. In the early Miocene (Fig. 3b), extension
is widespread in the A e g e a n Sea and the Western Mediterranean. Oceanic crust is forming in the Liguro-Provenqal basin. Intracontinental extension shapes the Aegean Sea and the west Alboran Basin. The first marine basins form in Corsica after the collapse of the Alpine belt and the formation of the Tyrrhenian Sea has started. Syn-rift sediments are deposited in the Pannonian Basin and the Carpathian arc rotates outward. Crustal collapse toward active subduction zones appears to control the geometry of extension. 10 M a to Present. The Tyrrhenian Sea (Fig. 3c,
d) is actively rifting between the Apennines and the Corsica-Sardinia block. Oceanic crust eventually forms in the southeastern part in the Pliocene. Extension in the Aegean progressively localizes along the western limit of the Anatolian block while the North Anatolian Fault propagates toward the west and the trench. The extrusion of the Anatolian block toward the low-stress boundary of the Hellenic subduction controls the kinematic and deformation regimes in the Aegean region. Tectonic inversion occurs in the Pannonian Basin (van Balen et al. this volume) progressively from 10 to 5 Ma (Juhasz et al. this volume; Sacchi et aL this volume). The recent compression explains the peripheral upilft and central subsidence of the Pannonian Basin (van Balen et all. this
volume).
8
L. JOLIVET E T A L .
Fig. 3. Tectonic scenario from the Oligocene to the present modified from Dercourt et al. (1993, 1986). Light shade represents the submerged continental crust, and darker shade Mesozoic oceanic crust. Cenozoic oceanic crust is shown in black.
The dynamics of extension Three mechanisms are generally proposed to explain the observed extension in the Mediterranean region. We do not include here models which relate the formation of extensional basins to strike-slip faulting because although such features may influence the recent evolution of the Aegean Sea (Armijo et al. 1996), they cannot account for the major part of extension which predates the formation of the North Anatolian Fault (Gautier et al. 1998). Collapse o f an overthickened crust. The widespread distribution of extension in the Aegean and Alboran Sea, together with the presence of metamorphic core complexes with high-temperature metamorphic overprint is consistent with a weak continental crust (Platt & Vissers 1989; Platt 1993; Gautier & Brun 1994a; Jolivet et al. 1994a; Vissers et al. 1995). The progressive heating of a thick crust following thickening will alter its rheology and lower its overall resistance leading to distributed extension similar to that
observed in the Basin and Range Province (Wernicke 1992). Similar behaviour can also be observed in the northern Tyrrhenian Sea where lower crustal melts and high temperature metamorphic core complexes have also been described (Jolivet et al. 1994b). In this case the extensional domain is narrower at any given time period compared to the sudden general collapse of the Aegean or the Betic-Rif orogen. The three basins show a progressive evolution from a cold toward a warm geotherm associated to crustal collapse and extension. The deep crust below the Apennines is apparently also warmer than expected (Chiarraba & A m a t o 1996) and the whole belt seems now to be collapsing. Crustal collapse is likely to be one important cause of extension. It is believed, however, that it is other processes such as slab rollback, slab detachment, or convective removal that are ultimately responsible for extension and which increase the heat flow and the potentiality of the continental crust to spread laterally.
INTRODUCTION Slab rollback, slab detachment and convective removal Distinguishing between slab rollback
and convective removal (detachment of the dense lithospheric root (Fleitout & Froidevaux 1982; Houseman et al. 1981)) is difficult in terms of thermal and mechanical consequences in the crust. Both processes will lead to the replacement of cold lithospheric material by hot asthenosphere below the mountain belt. Similar consequences in terms of crustal resistance, magmatism and extension are to be expected. The only real difference is the outward migration of magmatism and extension in the case of slab rollback. Convective removal is thought by some authors to be the major process responsible for the formation of the Alboran Sea (Platt & England 1994; Vissers et al. 1995; Seber et al. 1996), although slab rollback has also been proposed to explain the westward translation of the Alboran domain (Frizon de Lamotte et al. 1991; Royden 1993; Lonergan & White 1997). In the case of the northern Tyrrhenian Sea, slab rollback is obviously an active and important process, based on the observed migration of magmatism, extension and frontal compression processes (Jolivet et al. 1998; Malinverno & Ryan 1986; Patacca et al. 1990; Serri et al. 1993). In the case of the Aegean Sea a migration of volcanism is also seen from north to south (Fytikas et al. 1984) but the extensional event seems more widespread. The crustal collapse component may be more important in this region. Numerical and analogue experiments have shown how slab rollback and gravitational forces cooperate to control the extensional process (Faccenna et al. 1996; Giunchi et al. 1994). The slab rollback component can only increase with time, except when slab detachement occurs, while forces due to gravitational collapse can only decrease through time. Slab detachement may have some consequences that are easier to recognize such as a possible isostatic rebound of the upper and lower plates, although this has been challenged by numerical modelling studies (Giunchi et al. 1994, 1996). Slab breakoff has otherwise consequences quite similar to those of convective removal and slab retreat in terms of thermal evolution and magmatism (von Blanckenburg & Davies 1995). Another consequence is the along-strike migration of the foreland basin and the progressive concentration of the slab-pull force on the still attached region during progressive tear. This has been applied in the case of the Tyrrhenian Sea (Wortel & Spakman 1992). Multiple slab detachment events can also be envisaged for the evolution of the western Mediterranean, each trigerring a pulse of extension (Carminati et al. 1998).
9
Thus, across-strike migration of magmatic and extensional events suggest slab rollback or progressive delamination of the lithospheric mantle (and part of the lower crust), whilst along strike migration of depocenters may be used to infer slab detachement (van der Meulen et al. 1998). In any case, there may be several ways of testing deep processes by looking at the surface geology. Sedimentary basins and petroleum exploration.
Thick and quite diverse sedimentary basins are present in the Mediterranean area. The onshore basins have been explored for hydrocarbons over a long time period and some are petroleum provinces of quite significant economic importance (Panonian basin, the Apennines, Northern Africa). The offshore area has been little explored and, apart from some successes in the southern Adriatic Sea and Nile Delta, most of the Mediterranean basins and margins can be regarded as underexplored. Water depths have been a serious limitation in the past but the polyphase geological history and structural complexity have prevented a proper evaluation of their petroleum potential. The papers in this volume provide the necessary geodynamic and structural framework for the future assessment of these basins (Ziegler & Roure this volume). Broadly speaking, the Mediterranean basins fall into four categories. (1) The basins belonging to the African foreland to the south (Macgregor et al. 1998) have been only moderately affected by Tertiary compressional and extensional events. They include Palaeozoic basins of the northern African craton infilled with thick continental deposits with marine influences developing to the north. These basins are presently the focus of an extensive and successfull hydrocarbon exploration in the Algerian Sahara, following a first phase of field discoveries and developments in the 1960s. The Palaeozoic basins are overlain by a Mesozoic cover sequence that was laid down on the southern continental margin of the Tethys ocean. These basins are well developed onshore in northern Africa from Algeria to Egypt. Active exploration of Mesozoic sequences is currently active in Tunisia, Libya and Egypt, with spectacular success in the Western Desert of Egypt. The Tethyan Mesozoic basins are also well preserved in some offshore segments on the northern African margin such as the Gulf of Gabes, the Pelagian Sea, the Gulf of Sirte (with Tethyan oceanic crust possibly being preserved in the deepest parts of the Ionian Sea), below the
10
L. JOLIVET E T A L .
Nile delta and in the Adriatic Sea. Shallow water domains have been explored with only a limited amount of success (Gulf of Gabes). Little attention has focused on the deeper offshore area apart from the southern Adriatic Sea where the Aquila oil field has been developped in water depth in excess of 500 m. (2) Offshore accretionary wedges are developing and propagating southward as the last remaining fragments of the Tethyan oceanic crust are subducted northward below the southern edge of the Alpine orogen. These wedges form the Ionian ridge, the Mediterranean ridge and the Cyprus arc. They typically develop over a d6collement hosted in Messinian salt at their southern external edge, but deeper decollement are expected to occur below their northern inner edge. These wedges are poorly imaged by the seismic reflection data due to structural complexities and, locally, to mud volcanism. As a result they have been little explored and petroleum plays have still to be defined and tested. (3) Tertiary extension within the Alpine orogen has led to the opening of young oceanic basins (Western Mediterranean, Tyrrhenian Sea) or to the subsidence of basins floored by actively thinning continental crust (Aegean Sea and Pannonian Basin). These basins are the main focus of this volume and structural characteristics have been discussed above. Late stage alpine compressions have led locally to tectonic inversion sometimes with quite significant thrusting, as documented in the southern Alboran Sea (Chalouan et al. 1997). Petroleum exploration has only been successfull in the Pannonian Basin (Horvfith & Tari this volume), the other basins being probably devoid of source rocks of regional extent, although some local petroleum systems have been recognized (southern France, Masde & ViaUy this volume), or even large oil fields discovered and developped (Valencia Gulf). (4) Deltas and deep sea fans have developped in Neogene times in different structural settings: the Ebro and Rhone delta at the edge of the Western Mediterranean Neogene oceanic basin, and the Nile Delta superimposed over the Mesozoic Tethyan margin. Only the Nile Delta has proven so far to be an important petroleum province, with hydrocarbons originating from Oligocene source rocks and trapped within turbidites of Late Miocene to Pliocene age.
Conclusion Although some uncertainties remain concerning the nature of the seminal event that produced a sharp change from overall compression to backarc extension and compression some 30 Ma ago in the Mediterranean region, it is now certain
that this event originates in the mantle, either by slab detachement or slab rollback or both acting in sequence. Gravitational collapse of a thick crust may have influenced the development of most basins, and strike-slip faults, such as the North Anatolian Fault have also affected basin evolution recently. If the shape and kinematics of individual basin-mountain belt pairs is controlled by the local geometry of plate boundaries and the local history of crustal thickening, the late Oligocene event results from a change at a larger scale. Thus the 30 Ma event requires a change in mechanical or kinematic boundary conditions on the scale of the Mediterranean region. The subduction regime has changed somehow at this time. It is furthermore possible that one (or several) event(s) of slab detachment trigerred extension after the late Oligocene major reorganization.
References ALLERTON, S. 1994. Vertical-axis rotation associated
with folding and thrusting: an example from the eastern subbetic zone of Southern Spain. Geology, 22, 1039-1042. --,
LONERGAN,L., PLATT, J. P., PLATZMAN,E. S. &
MCCLELLAND, E. 1993. Paleomagnetic rotations in the eastern Betic cordillera, southern Spain. Earth and Planetary Science Letters, 119, 225-241. ALTHERR, R., KREUZER, H., WENDT, I., LENZ, H., WAGNER, G. A., KELLER,J., HARRE, W. & HOHN-
DORF,A. 1982. A Late Oligocene/Early Miocene high temperature belt in the anti-cycladic crystalline complex (SE Pelagonian, Greece). Geologische Jahrbuch, 23, 97-164. AMATO, A., ALESSANDRINI,B., CIMINI, G., FREPOLI,A. & SELVAGGI,G. 1993. Active and remanant sub-
ducted slabs beneath Italy: evidence from seismic tomography and seismicity. Annali di Geofisica, 36, 201-214. --,
MARGHERITI, L., ARAZZA, R., BASILI, A., CHIARABBA, C., CIACCO, M. G., CIMINI, G. B., D1 BONA, M., LUCENTE, E P., NOSTRO, C. & SEL-
VAGGI, G. 1998. Passive seismology and deep structure in central Italy. In: Geodynamics of the Lithosphere and the Earth's Mantle Pageoph. Special volume, in press. ARMIJO, R., LYON-CAEN,H. & PAPANIKOLAOU,D. 1992. East-west extension and Holocene normal fault scarps in the Hellenic arc. Geology, 20, 491-494. - - - , MEYER, B., KING, G. C. P., RIGO, A. & PAPANASTASSIOU, D. 1996. Quaternary evolution of the Corinth Rift and its implications for the Late Cenozoic evolution of the Aegean. Geophysical Journal International, 126, 11-53. AVlGAD,D., BAER,G. & HEIMANN,A. 1998. Block rotations and continental extension in the Central Aegean Sea: paleomagnetic and structural evidence from Tinos and Mykonos. Earth and Planetary Science Letters, in press. BENEDICTO, A., SI~GURET, M. & LABAUME, P. 1999.
Interaction between faulting, drainage and
INTRODUCTION sedimentation in extensional hangingwall syncline basins: example of the Oligocene Matelles basin (Gulf of Lions rifted margin, SE France). This volume. BERCKHEMER, H. 1977. Some aspects of the evolution of marginal seas deduced from observations in the Aegean region. In: (eds) International Symposium on Structural History of the Mediterranean Basins. Technip, Paris, 303-314. BRUN, J. P., GUTSCHER, M. A. & TEAMS, D.-E. 1992. Deep crustal structure of the Rhine Graben from D E K O R P - E C O R S seismic reflection data: a summary. Tectonophysics, 208, 139-147. BUICK, I. S. 1991. Mylonite fabric development on Naxos, Greece. Journal of Structural Geology, 13, 643-655. BURRUS, J. 1984. Contribution to a geodynamic synthesis of the Provenqal basin (north-western Mediterranean). Marine Geology, 55, 247-269. CARMINATI,E.,WORTEL, M. J. R., SPAKMAN,W. & SABADINI, R. 1998. The role of slab detachment processes in the opening of the western-central Mediterranean basins: some geological and geophysical evidence. Earth and Planetary Science Letters, 160, 651-665. CHALOUAN,A., SAJI, R., MICHARD,A. & BALL5",A. W. 1997. Neogene tectonic evolution of the Southwestern Alboran Basin as inferred from seismic data off Morocco. A A P G Bulletin, 81, 1161-1184. CHAMOT-ROOKE, N., GAULIER,J.-M. & JESTIN,E 1999. Constraints on Moho depth and crustal thickness in the Liguro-Proven~al basin from a 3D gravity inversion: geodynamic implications. This volume. CHIARRABA, C. ~ AMATO, A. 1996. Crustal velocity structure of the Apennines (Italy) from P-wave travel time tomography. Annali di Geofisica, 39,
1133-1148. COMAS, M. C., GARCfA-DUE/qAS,W. & JURADO, M. J. 1992. Neogene tectonic evolution of the Alboran Sea from MCS data. GeoMarine Letters, 12, 157-164. COMAS, M. C., ZANH, R. & KLAUS A. ETAL. 1996. Leg 161 - Western Mediterranean. Initial Reports of the Ocean Drilling Program, 161. D'AGOSTINO, N., CHAMOT-ROOKE,N., FUNICIELLO,R., JOLIVET,L. & SPERANZA,F. 1998. The role of preexisting thrust faults and topography on the styles of extension in the Gran Sasso range (Central Italy). Tectonophysics, 292, 229-254. D'OFFIZI, S., MINELLI, G. & PIALLI, G. 1994. Foredeeps and thrust systems in the northern Apennines. Bolletino di Geofisica Teorica ed Applicata, 36, 141-144. DANIEL, J. M., JOLIVET, L., GOFFI~, B. & POINSSOT, C. 1996. Crustal-scale strain partitionning: footwall deformation below the Alpine Corsica OligoMiocene detachment. Journal of Structural Geology, 18, 41-59. DE JONGE, M. R.,WORTEL, M. J. R. & SPAKMAN,W. 1993. From tectonic reconstruction to upper mantle model: an application to the Alpine-Mediterranean region. Tectonophysics, 223, 53-65. --, -& -1994. Regional scale tectonic evolution and the seimic velocity structure of the lithosphere and upper mantle: the Mediterranean
11
region. Journal of Geophysical Research, 99, 12091-12108. DE METS, C., GORDON, R. G., ARGUS, D. E & STEIN, S. 1990. Current plate motions. Geophysical Journal International, 101, 425-478. DE VOOGD, B., NICOL1CH,R., OLIVET,J. L. ETAL. 1991. First deep seismic reflection transect from the Gulf of Lions to Sardinia (ECORS-CROP profiles in Western Mediterranean). In: MEISSNER,R., BROWN, L., DURBAUM,H. J., FUCHS, K. & SEIFER, E (eds) Continental lithosphere: deep seismic reflections. American Geophysical Union, Geodynamic Series, 22, 265-274. DERCOURT,J., RJCOU,L. E. & VRIELINCK,B. 1993. Atlas Tethys Palaeo environmental Maps. Paris, Gauthier-Villars. --, ZONENSHAIN, L. E, RICOU, L. E. ET AL. 1986. Geological evolution of the Tethys belt from the Atlantic to the Pamir since the Lias. Tectonophysics, 123, 241-315. DEWEY, J. F. 1988. Extensional collapse of orogens. Tectonics', 7, 1123-1139. - - , HELMAN, M. L., TORCO, E., HUTTON, D. H. W. & KNOTT, S. D. 1989. Kinematics of the Western Mediterranean. In: COWARD,M. E, DIETRICH, D. & PARK, R. G. (eds) Alpine Tectonics. Geological Society, London, Special Publications, 45, 265-283. DINTER, D. A. & ROYDEN, L. 1993. Late Cenozoic extension in northeastern Greece: Strymon valley detachment system and Rhodope metamorphic core complex. Geology, 21, 45-48. DOGLIONI, C. 1991. A proposal for the kinematic modelling of W-dipping subductions-possible applications of the Tyrrhenian-Apennines system. Terra Nova, 3, 423-434. --, GUEGUEN, E., HARABAGLIA,P. & MONGELLI, F. 1999. On the origin of W-directed subduction zones and applications to the Western Mediterranean. This volume. --, SABAT, E ~; FERNANDEZ, M. 1997. The Western Mediterranean extensional basins and the Alpine orogen. Terra Nova, 9, 109-112. FACCENNA, C., DAVY, P., BRUN, J. P., FUNICIELLO, R., GIARDINI, D., MATTEI, M. & NALPAS,Y. 1996. The dynamic of backarc basins: an experimental approach to the opening of the Tyrrhenian Sea. Geophysical Journal International, 126, 781-795. , MATTEI, M., FUNICIELLO,R. & JOLIVET,L. 1997. Styles of back-arc extension in the Central Mediterranean. Terra Nova, 9, 126-130. FASSOULAS, C., KILIAS, A. & MOUNTRAKIS, D. 1994. Postnappe stacking extension and exhumation of high-pressure/low-temperature rocks in the island of Crete, Greece. Tectonics, 13, 127-138. FLEITOUT, L. & FROIDEVAUX, C. 1982. Tectonics and topography for a lithosphere containing density heterogeneities. Tectonics, 1, 21-56. FODOR, L., CSONTOS, L., BADA, G., GYORFI, Y. d~; BENKOVICS,L. 1999. Tertiary tectonic evolution of the Pannonian basin system and neighbouring orogens: a new synthesis of palaeostress data. This volume. FREPOLI, A. & AMATO, A. 1997. Contemporaneous extension and compression in the Northern
12
L. JOLIVET E T A L .
Apennines from earthquakes plane solutions. Geophysical Journal International, 129, 368-388. FRIZON DE LAMOTTE, D., ANDRIEUX, J. & GuI~zou, J. C. 1991. CinGmatique des chevauchements NGogGnes dans l'arc bGtico-Rifains, discussion sur les modGles gGodynamiques. Bulletin de la Socidt~ Gdologique de France, 4, 611-626. , POISSON, A., AUBOURG, C. & TEMIZ, H. 1995. Chevauchements post-tortoniens vers l'ouest puis vers le sud au coeur de l'angle d'Isparta (Taurus, Turquie). ConsGquences gGodynamiques. Bulletin de la Socidtd Gdologique de France, 166, 59-67. FYTIKAS, M., INNOCENTI, E, MANETTI, P., MAZZUOLI, R., PECCERILLO,A. & VILLARI,L. 1984. Tertiary to Quaternary evolution of volcanism in the Aegean region. In: ROBERTSON,A. H. E & DIXON, J. E. (eds) Geological Evolution of the Eastern Mediteranean. Geological Society, London, Special Publications, 17, 687-699. GAUTIER, E & BRUN, J. P. 1994a. Crustal-scale geometry and kinematics of late-orogenic extension in the central Aegean (Cyclades and Evvia island). Tectonophysics, 238, 399-424. - & -1994b. Ductile crust exhumation and extensional detachments in the central Aegean (Cycladcs and E w i a islands). Geodinamica Acta, 7, 57-85. - & JOLIVET,L. 1993. Structure and kinematics of upper Cenozoic extensional detachement on Naxos and Paros (Cyclades Islands, Greece). Tectonics, 12, 1180-1194. - - , MORICEAU,R., SOKOUTIS,D., MARTINOD,J. JOLIVET, L. 1999. Aegean extension: timing, kinematics and cause. Tectonophysics, in press. GERNER, P., BADA, G., DOVI~NYI, P., MI3LLER, B., ONCESCU, M. C., CLOETINGH, S. & HORVATH, E 1999. Recent tectonic stress and crustal deformation in and around the Pannonian basin: data and models. This volume. GIUNCHI, C., GASPERINI, P., SABADINI, R. & D'AGOSTtNO, G. 1994. The role of subduction on the horizontal motions in the Tyrrhenian basin: a numerical model. Geophysical Research Letters, 21,529-532. --, SABADINI, R., BOSCHI, E. ~r GASPERINI, P. 1996. Dynamic models of subduction: geophysical and geological evidence in the Tyrrhenian Sea. Geophysical Journal International, 126, 555-578. GUEGUEN, E. 1995. Le bassin Liguro-Provenqal: un vdritable ocdan. Exemple de segmentation des marges et de hiatus cindmatiques. Implications sur le processus d'amincissement crustal. Thesis, Universit6 de Bretagne Occidentale. GYORFI, I., CSONTOS, L. & NAGYMAROSY, A. 1999. Early Tertiary structural evolution of the border zone between the Pannonian and Transylvanian basin. This volume. HATZFELD, D. 1999. The present-day tectonics of the Aegean as deduced from seismicity. , BESNARD, M., MAKROPOULOS,K. ~% HATZIDIMITRIOU, P. 1993. Microearthquake seismicity and fault-plane solutions in the southern Aegean and its geodynamic implications. Geophysical Journal International, 115, 799-818. , MARTINOD,J., BASTET,G. & GAUTIER,P. 1997. An
analog experiment for the Aegean to descibe the contribution of gravitational potential energy. Journal of Geophysical Research. 102, 649-659. - - , PEDOTTI, G., HATZIDIMITRIOU,P. ~; MAKROPOULOS, K. 1990. The strain pattern in the western Hellenic arc deduced from a microearthquake survey. Geophysical Journal International, 101, 181-202. HORVATH, E & TARI, G. 1999. IBS Pannonian Basin Project: a review of its main results and their bearing on hydrocarbon exploration. This volume. HOUSEMAN, G. A., MCKENZIE, D. E & MOLNAR, P. 1981. Convective instability of a thickened boundary layer and its relevance for the thermal evolution of the continental crust. Journal of Geophysical Research, 86, 6115-6132. JACKSON, J. 1994. Active tectonics of the Aegean region. Annual Review of Earth and Planetary Science, 22, 239-271. & WHITE, N. J. 1989. Normal fauting in the upper continental crust: observations from regions of active extension. Journal of Structural Geology, 11, 15-36. JOLIVET, L. & PATRIAT,M. 1999. Ductile extension and the formation of the Aegean Sea. This volume. , BRUN, J. P., GAUTIER, P., LALLEMANT, S. & PATRIAT,M. 1994a. 3-D kinematics of extension in the Aegean from the Early Miocene to the Present, insight from the ductile crust. Bulletin de la SocidtO Gdologique de France, 165, 195-209. - - , DANIEL, J. M., TRUFFERT, C. d~. GOFFI~,B. 1994b. Exhumation of deep crustal metamorphic rocks and crustal extension in back-arc regions. Lithos, 33, 3-30. - - , FACCENNA,C., GOFFI~,B., MATTEI,M., ROSSETTI, E, BRUNET,C., STORTI,E, FUNICIELLO,R., CADET, J. P. & PARRA,T. 1998. Mid-crustal shear zones in post-orogenic extension: the northern Tyrrhenian Sea case. Journal of Geophysical Research, 103, 12 123-12 160. --, GOFFI~, B., MONII~, P., TRUFFERT-LUXEY, C., PATmAT, M. & BONNEAU, M. 1996. Miocene detachment in Crete and exhumation P-T-t paths of high pressure metamorphic rocks. Tectonics, 15,
1129-1153.
JUHASZ, E., PHIIJJPS, L., M()I,LER, P., RICKETTS, B., TOTH-MAKK, g,., LANTOS, M. & KovAcs, L. O. 1999. Late Neogene sedimentary facies and sequences in the Pannonian basin, Hungary. This volume. KAHLE, H. G., MULLER, M. V., GEIGFR, A., DANUSER, G., MUELLER, S., VEIS, G., BUJJRIS, H. ~ PARADIsis, D. 1995. The strain field in NW Greece and the Ionian Islands: results inferred from GPS measurements. Tectonophysics, 249, 41-52. , & VElS, G. 1996. Trajectories of crustal deformation of western Greece from GPS observations 1989-1994. Geophysical Research Letters, 23, 677-680. KASTENS, K., MASCLE, J., AUROUX, C., BONATTI, E., BROGLIA, C., CHANNELL, J., CURZI, P., EMEIS, K. C., GLAGON, G., HASEGAWA, S., HIEKE, W., MASCLE, G., F McCoY, E, MCKENZlE, J., MENDELSON, J., MOLLER, C., RI~HAULT, J. P., ROBERTSON, A., SARTORI, R., SPROVIER, R. &
INTRODUCTION TORIt, M. 1988. O D P Leg 107 in the Tyrrhenian Sea: insight into passive margin and back-arc basin evolution. Geological Society of America Bulletin, 100, 1140-1156. KISSEL, C. & LAJ, C. 1988. The Tertiary geodynamic evolution of the Aegean arc: a paleomagnetic reconstruction. Tectonophysics, 146, 183-201. LAJ, C., JAMET, M., SOREL, D. & VALENTE, J. P. 1982. First paleomagnetic results from Mio-Pliocene series of the Hellenic sedimentary arc. Tectonophysics, 86, 45-67. LE PICHON, X. 1981. Land-locked oceanic basins and continental collision, the eastern Mediterranean as a case example. In: HsO, K. J. (ed.) Mountain Building Processes. Academic Press, London, 201-211. - - , CHAMOT-ROOKE,N., LALLEMANT,S. L., NOOMEN, R. & VEIS, G. 1994. Geodetic determination of the kinematics of Central Greece with respect to Europe: implications for eastern Mediterranean tectonics. Journal of Geophysical Research, 100, 12675-12690. - - , LALLEMANT,S., CHAMOT-ROOKE,N., LEMEUR,D. & PASCAL, G. 1999. The Mediterranean ridge backstop: a consequence of the Hellenic nappes seaward migration. Marine Geology, in press. LEBLANC, D. & OLIVIER, R 1984. Role of strike-slip faults in the Betic-Rifian orogeny. Tectonophysics, 101, 345-355. LINZER, H. G. 1996. Kinematics of retreating subduction along the Carpathian arc, Romania. Geology, 24, 167-170. LISTER, G. S., BANGA, G. & FEENSTRA,A. 1984. Metamorphic core complexes of cordilleran type in the Cyclades, Aegean Sea, Greece. Geology, 12, 221-225. LONERGAN, L. & WHITE, N. 1997. Origin of the BeticRif mountain belt. Tectonics, 16, 504-522. MACGREGOR, D. S., MOODY, R. J. & CLARK-LOWES,D. D. (eds) 1998. Petroleum Geology of North Africa. Geological Society, London, Special Publications, 132. MALINVERNO, A. & RYAN, W. 1986. Extension in the Tyrrhenian sea and shortening in the Apennines as result of arc migration driven by sinking of the lithosphere. Tectonics, 5, 227-245. MARGHERITI, L., NOSTRO, C., COCCO, M. & AMATO,A. 1996. Seismic anisotropy beneath the Northern Apennines (Italy) and its tectonic implications. Geophysical Research Letters, 23, 2721-2724. MASCLE,A. & VIALLY,R. 1999. The petroleum system of the Southeast Basin and Gulf of Lions (France). This volume. MATTEI, M., FUNICIELLO,R. & KISSEL, C. 1995. Paleomagnetic and structural evidence for Neogene block rotations in the Central Apennines, Italy. Journal of Geophysical Research, 100, 17863-17883. MATTEI, M., K1SSEL, C., SAGNOTTI,L., FUNICIELLO,R. & FACCENNA, C. 1996. Lack of Late Miocene to Present rotation of the northern Tyrrhenian margin (Italy): a constraint on geodynamic evolution. In: MORRIS,A. & TARLING,D. H. (eds)
Palaeomagnetism and Tectonics of the Mediterranean Region. Geological Society, London, Special Publications, 105, 141-146.
13
MAUFFRET,A. & CONTRUCCI,I. 1999. Crustal structure of the North Tyrrhenian Sea: first results of the multichannel seismic LISA cruise. This volume. , PASCAL, G., MAILLARD, A. & GORINI, C. 1995. Tectonics and deep structure of the north-western Mediterranean basin. Marine and Petroleum Geology, 12, 645-666. MCKENZIE, D. 1972. Active tectonics in the Mediterranean region. Geophysical Journal of the Royal Astronomical Society, 30, 109-185. - 1978. Active tectonics of the Alpine-Himalayan belt: the Aegean Sea and surrounding regions.
Geophysical Journal of the Royal Astronomical Society, 55, 217-254. MELE, G., ROVELLI, A., SEBER, D. & BARAZANGI, M. 1997. Shear wave attenuation in the lithosphere beneath Italy and surrounding regions: tectonic implications. Journal of Geophysical Research, 102, 11863-11875. MONTIGNY, R., EDEL, J. B. & THU1ZAT,R. 1981. OligoMiocene rotation of Sardinia: K-Ar ages and paleomagnetism data of Tertiary volcanics. Earth and Planetary Science Letters, 54, 261-271. MONTONE, E, AMATO,A., FREPOLI,A., MARIUCCI,M. T. & CESARO,M. 1997. Crustal stress regime in Italy. Annali di Geofisica, 4 0 , 741-757. MORmS, A. & ANDERSON, A. 1996. First paleaomagnetic results from the Cyclaic Massif, Greece, and their implications for Miocene extension directions and tectonic models in the Aegean. Earth and Planetary Science Letters, 142, 397-408. NOOMEN, R., SPRINGER, T. A., AMBROSIUS, B. A. T., HERZBERGER, K., KUIJPER, D. C., METS, G. J., OVERGAAUW, B. & WAKKER, K. E 1996. Crustal deformations in the Mediterranean are computed from SLR and GPS observations. Journal of Geodynamics, 21, 73-96. OKAY,A. & TOYSIDZ,0.1999. Tethyan sutures of northern Turkey. This volume. PASCAL, G. P., MAUFFRET,A. & PATRIAT,P. 1993. The ocean-continent boundary in the Gulf of Lion from analysis of expanding spread profiles and gravity modelling. Geophysical Journal International, 113, 701-726. PATACCA, E., SARTORI, R. & SCANDONE, P. 1990. Tyrrhenian basin and Apenninic arcs: kinematic relations since late Tortonian times. Memorie della Societd Geologica Italiana, 45, 425-451. - & SCANDONE, S. 1989. Post-Tortonian mountain building in the Apennines. The role of the passive sinking of a relic lithospheric slab. In: BORIANI, A., BONAFEDE, M., PICCARDO, G. B., VAI, G. B. (eds) The lithosphere in Italy. Advances in science research. Accademia Nazionale dei Lincei, Rome, 157-176. PLATY,J. P. 1993. Exhumation of high-pressure rocks: a review of concept and processes. Terra Nova, 5, 119-133. -& ENGLAND, P. 1994. Convective removal of lithosphere beneath moutain belt: thermal and mechanical consequences. American Journal of Science, 294, 307-336. & VISSERS,R. L. M. 1989. Extensional collapse of thickened continental lithosphere: A working hypothesis for the Alboran Sea and Gibraltar arc. Geology, 17, 540-543.
14
L. J O L I V E T E T A L .
REILINGER, R., MCCLUSKY, S., KAHLE, H. O., MULLE, M. V., STRAUB, C., KASTENS, K., GILBERT, L., HURST, K., VEIS, M., PARADIS1S,D. & BARKA, A. 1995. GPS evidence for westward continuation of dextral strike-slip motion from the North Anatolian Fault Zone across the North Aegean and North-Central Greece. Abstract. American Geophysical Union (AGU). EOS, Supplement, Nov. 7, F620-F621. RIcou, L. E. 1994. Tethys reconstructed: plates, continental fragments and their boundaries since 260 Ma from Central America to south-eastern Asie. Geodinamica Acta, 7, 169-218. RIGO, A. 1994. Etude sismotectonique et gOoddsique du golfe de Corinthe. Thesis, Universit6 Denis Diderot, Paris. --, LYON-CAEN, H., ARMIJO, R., DESCHAMPS, A., HATZFELD, D., MAKROPOULOS, K., PAPADIMITRIOU,P. & KASSARAS,I. 1996. A microseismicity study in the western part of the Gulf of Corinth (Greece): implications for large-scale normal faulting mechanisms. Geophysical Journal International, 126, 663-688. ROYDEN, L. H. 1993. Evolution of retreating subduction boundaries formed during continental collision. Tectonics, 12, 629-638. SACCHI,M., HORVATH,E & MAGYARI,O. 1999. Role of unconformity-bounded units in stratigraphy of continental record: a case study from the Late Miocene of western Pannonian basin, Hungary. This volume. SAGNOrrI, L. 1992. Paleomagnetic evidence for a Plio-Pleistocene counterclockwise rotation of the Sant'Arcangelo basin, Southern Italy. Geophysical Research Letters, 19, 135-138. SCHEEPERS, P. J. J., LANGEREIS, C. G. & HILGEN, E J. 1993. Counter-clockwise rotations in the southern Apennines during the Pleistocene: paleomagnetic evidences from the Matera area. Tectonophysics, 225, 379-410. SEBER, D., BARAZANGI, M., IBENBRAHIM, A. & DEMNATI, A. 1996. Geophysical evidence for lithospheric delamination beneath the Alboran Sea and Rif-Betic mountains. Nature, 379, 785-790. SI~RANNE, M. 1999. The Gulf of Lions continental margin (NW Mediterranean) revisited by IBS: an overview. SERRI, G., INNOCENTI, E & MANETTI, P. 1993. Geochemical and petrological evidence of the subduction of delaminated Adriatic continental lithosphere in the genesis of the NeogeneQuaternary magmatsim of central Italy. Tectonophysics, 223, 117-147. SOKOUTIS, D., BRUN, J. P., DRIESSCHE, J. V. D. & PAVLIDES, S. 1993. A major Oligo-Miocene detachment in southern Rhodope controlling north Aegean extension. Journal of the Geological Society, London, 150, 243-246. SPAKMAN,W., VAN DER LEE, S. & VAN DER HILST, R. 1993. Travel-time tomography of the EuropeanMediterranean mantle. Physics of the Earth and Planetary Interiors, 79, 3-74. TAPPONNIER, P. 1977. Evolution du syst6me alpin en M6diterran6e: poinconnement et 6crasement
rigide-plastique. Bulletin de la Socidt~ Geologique de France, 7 (19), 437-460. TARI, G., DOVI~NYI, P., HORV,~TH, E, DUNKL, I., LENKEY, L., STEFANESCU,M., SZAFIAN,P. & TOTH, T. 1999. Lithospheric structure of the Pannonian basin derived from seismic, gravity and geothermal data. This' volume. TRUFFERT-LUXEY,C. 1992. De la compression de la ride mOditerrandenne d l'extension en Mer Egde: gdodynamique de la MdditerranOe orientale. Thbse, Universit6 Pierre et Marie Curie. VAN BALEN, R. T., LENKEY, L., HORVA,TH, E & CLOETINGH, S. A. P. L. 1999. Towdimensional modeling of stratigraphy and compaction driven fluid flow in the Pannonian basin. This volume. VANDER MEULEN,M. J., MEULENKAMP,J. E. & WORTEL, M. J. R. 1998. Lateral shifts of Apenninic foredeep depocentres reflecting detachment of subducted lithosphere. Earth and Planetary Science Letters, 154, 201-218. VERGES, J. & SABAT,F. 1999. Contraints on the western Mediterranean kinematic evolution along a 1000 km transect, from Iberia to Africa. This volume. VIGLIOrrl, L. & KENT, D. V. 1990. Paleomagnetic results of Tertiary sediments from Corsica: evidence for post-Eocene rotation. Physics of the Earth and Planetary Interiors, 62, 97-108. VISSERS, R. L. M., PLATr, J. P. & VAN DER WAL, D. 1995. Late orogenic extension of the Betic Cordillera and the Alboran domain: a lithospheric view. Tectonics, 14, 786-803. VON BLANCKENBURG,E & Huw DAVIES,J. 1995. Slab breakoff: a model for syncollisional magmatism and tectonics in the Alps. Tectonics, 14, 120-132. WAGEMAN, J. M., HILDE, T. W. C. & EMERY, K. O. 1970. Structural framework of the of the East China Sea and Yellow Sea. American Association of Petroleum Geologists Bulletin, 54, 1611-1643. WARD, S. N. 1994. Constrains on the seismotectonics of the central Mediterranean from Very Long Baseline Interferometry. Geophysical Journal International, 117, 441-452. WERNICKE, B. 1992. Cenozoic extensional tectonics of the U.S. cordillera. In: BURCHFIEL,B. C., LIPMAN, P W. & ZOBACK, M. L. (eds) The Cordilleran Orogen: Conterminous U.S. Geological Society of America, Geology of North America, G-3, 553-581. WIJBRANS, J. R. & McDOUGALL, I. 1988. Metamorphic evolution of the Attic Cycladic Metamorphic Belt on Naxos (Cyclades, Greece) utilizing 4~ age spectrum measurements. Journal of Metamorphic Geology, 6, 571-594. WILSON, M. & BIANCHINI, G. 1999. Tertiary-Quaternary magmatism within the Mediterranean and surrounding regions. This" volume. WORTEL, M. J. R. & SPAKMAN,W. 1992. Structure and dynamic of subducted lithosphere in the Mediterranean. Procedings of the Koninks Nederlandse Akadamie voor Wetenschappen, 95, 325-347. ZECK, H. P. 1999. Alpine kinematic evolution in the W Mediterranean: a westward directed subduction regime followed by slab roll-back and slab detachment. This volume.
The Gulf of Lion continental margin (NW Mediterranean) revisited by IBS: an overview MICHEL
St~RANNE
GOophysique Tectonique SOdimentologie, Universit( Montpellier 2, cc. 060, 34095 Montpellier cedex 05, France (e-mail."
[email protected], univ-montp2.fr) Abstract: The Gulf of Lion margin is one of the Tertiary extensional basins of the western
Mediterranean that opened during convergence of Africa and Europe. This Oligocene-Aquitanian rifted margin and associated Burdigalian oceanic basin have been used as case study for stretching models of 'Atlantic-type' margins. However, when the Integrated Basin Study (IBS) project was initiated, several outstanding questions remained about the present structure and the geodynamic setting of the margin within the Western Mediterranean. IBS-Gulf of Lion research was based on the existing onshore and offshore, industrial and academic data, which were heterogeneous and unevenly distributed. Compilation of the stratigraphic correlations on a regional scale allowed precise calculation of the timing of rifting, and clarification of the relationships with Alpine and Mediterranean geodynamics. Reprocessing of the existing ECORS deep seismic reflection profiles shed new light on the extensional structure and mechanisms of extension of the continental margin. Structural and sedimentological studies onshore led to the definition of new tectonostratigraphic models for extensional basins. Results of structural analyses showed a partitioning of the extensional deformation processes across the continental margin. 3D gravity modelling of the margin and basin area led to the production of a new map of the Moho depth by inversion, and testing several hypotheses for the origin of the present day subsidence. Although the Gulf of Lion margin displays structural and stratigraphic features similar to 'Atlantic-type' margins, its structure and evolution corresponds to that of a rifted margin of a large continent formed during the opening of a marginal basin. Integration of the new results of IBS-Gulf of Lion within the geodynamic evolution of the western Mediterranean suggests that the Oligocene rifting of the Gulf of Lion represents the initial stage of a succession of rifting events and back-arc basin formation, due to continuously retreating subduction during convergence of Africa and Europe.
'Atlantic-type' divergent passive margins are a prime site for hydrocarbon accumulation, because of optimum geological conditions generated during their evolution (e.g. Edwards & Santogrossi 1989). As new models of basin formation and basin-fill architecture were developed, it appeared that 'Atlantic-type' models did not account for the evolution of many extensional margins. For example, a number of extensional sedimentary basins are formed during or shortly after m o u n t a i n building (e.g. S6ranne & Malavieille 1994; Cloetingh et al. 1995); in SE Asia and in the Mediterranean/Alpine region, continental extension leads to oceanic crust accretion, whilst the bordering lithospheric plates converge at rate of several centimetres per year (e.g. D e w e y et al. 1989; Rangin et al. 1990). The geodynamics of such extensional basins imply specific geological conditions controlling oil accumulation, which need to be understood in order to produce successful exploration models. One of the modules of the Integrated Basin Studies program addressed the problems of the
formulation of extensional basins in convergent settings. T h e Gulf of Lion in the Western Mediterranean was chosen as a natural laboratory where the structure and evolution of a fully developed rifted continental margin was formed during convergence of the African and Eurasian plates. The Pannonian Basin is the other natural laboratory chosen by IBS where continental extension occurred within the Alps (Horvfith & Tari this volume). The Gulf of Lion (Fig. 1) is located between the Valencia Trough to the west and the Provenqal margin of the Ligurian Sea to the east. These basins represent the rifted continental margin of SW Eurasia, whose conjugate margins have drifted southward or southeastwards, above the northwest-dipping subduction of the African and related plates (Apulia). This contribution aims to give an overview of the structure and development of the Gulf of Lion within the tectonic f r a m e w o r k of the Western Mediterranean, in the light of the new results reached by IBS. The tectonics of the Gulf of Lion are then compared with examples of
SERANNE,M. 1999. The Gulf of Lion continental margin (NW Mediterranean) revisited by IBS: An overview.
In: DURAND,B., JOLIVET,L., HoRvATN, E & SERANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 15-36.
16
M. SI~RANNE
Fig. 1. Present-day seismic-tectonic map of the Mediterranean-Alpine region (modified from Reba'f et al. 1992), showing the location of the Oligocene-Recent extensional basins formed during N-S convergence of Africa and Eurasia. classic 'Atlantic-type' margins and of back-arc basins.
The western Mediterranean setting The present-day structure of the Mediterranean region (Fig. 1) results from the overall convergence of Africa and Eurasia which involved several local and successive rifting and collision episodes during the Tertiary (Le Pichon et al. 1971; Auzende e t al. 1973; Tapponnier 1977; Biju-Duval et al. 1978; R6hault et al. 1984; Dercourt el at. 1985; Dewey et al. 1989). In the eastern Mediterranean, part of the convergence is still accommodated by northward subduction of Mesozoic Tethyan oceanic crust. On the other hand, the western Mediterranean is floored by oceanic crust formed during Neogene time as a
consequence of rifting and drifting of small (100 km wide) continental blocks. Formation of such extensional basins occurred simultaneously with thrusting and mountain building in the surrounding areas, during continued northward motion of Africa with respect to Eurasia. The exact kinematics and chronology of the events resulting in the present-day western Mediterranean is still intensely discussed; the number of unsolved questions increases as one goes back further in time (for example see the contrasting E o c e n e reconstructions by Biju-Duval e t al. 1978; Boillot et al. 1984; Dercourt et al. 1985; Dewey et al. 1989; Bois 1993). However there is a consensus on the main features and on the succession of tectonic stages which makes it possible to sketch out the Tertiary evolution of the western Mediterranean (Fig. 2).
Fig. 2. Sketch of the geodynamic evolution of the NW Mediterranean since Eocene. Compiled from different sources, including Bellon (1976); Coulon (1977); Girod & Girod (1977); Wildi (1983); Boillot et al. (1984); R6hault et al. (1984); Bergerat (1985); Dercourt et al. (1985); Bouillin (1986); Dewey et al. (1989); Sartori (1990); Bartrina et al. (1992); Mart/et al. (1992); Reba/et al. (1992); Roca & Guimer~t (1992); Ziegler (1992); Keller et al. (1994); Thompson (1994); Tricart et al. (1994); Mauffret et al. (1995); S6ranne et al. (1995); Olivet (1996); Saadallah & Caby (1996).
G U L F OF L I O N O V E R V I E W
17
18
M. SI2RANNE
Until late Eocene time (Fig. 2a, b), the Iberian plate was associated with several continental blocks characterized by Hercynian basement: the Balearics, Corsica, Sardinia, Calabria and the Kabylies blocks. To the SE, Iberia and the Hercynian blocks were bordered by a NWdipping subduction, south of which extended the African plate and derived continental blocks such as Apulia (Dercourt et al. 1985). To the north, Iberia and the Hercynian blocks were bounded by the North Pyrenean Fault Zone, along which they had been translated eastwards during the late Cretaceous opening of the Bay of Biscay (Olivet 1996). The northward motion of Iberia (and associated blocks) resulted in collision with the European plate, and the formation of the Pyrenean thrust and fold belt (Olivet 1996). The Languedoc-Provenqal segment of the Pyrenees results from the collision of Sardinia-Corsica with Provence. The kinematics east of this zone and the linkage with the Alps-Apennines are not clear, mostly due to the complex 3D geometry of the continental subductions converging in the 'Ligurian knot' (Laubscher et al. 1992). During latest Eocene (Priabonian, Fig. 2b), shortening proceeded between Iberia and Europe while E-W extension initiated intra-continental rift basins in western Europe (e.g. Rhine Graben, Bergerat 1985). The intermediate zone of the future Gulf of Lion, was affected by left-lateral strike-slip tectonics, involving formation of oblique extensional basins in releasing bends. From Rupelian time, the Hercynian blocks began to drift apart in several stages. During the first stage, in late Rupelian (Fig. 2c), continental rifting occurred between southern France and Corsica-Sardinia-Calabria, while the Balearics and Kabylies, located SW of the North Balearic Transfer Zone (Mauffret et al. 1995) or 'Accident Paul Fallot' (Durand-Delga & Fontbot6 1980) remained stable. In the Valencia Trough, rifting was delayed until latest Oligocene-Aquitanian times (Fig. 2d) (Batrina et al. 1992). Subduction-related calc-alkaline volcanism in Provence, Sardinia and the Valencia Trough (Bellon 1976; Coulon 1977; Girod & Girod 1977) indicate NW-dipping subduction of the Apulian plate beneath Calabria-SardiniaCorsica during Chattian to Burdigalian. The subduction system consisted of an accretionary prism located in Calabria (e.g. Thompson 1994), and a volcanic arc extending from Sardinia to Provence through the western margin of Corsica (Bellon 1976; Girod & Girod 1977). Intra-continental rifting in the Gulf of Lion-Ligurian basins was therefore in a backarc setting (R6hault et al. 1984). On this
transect, the onset of oceanic accretion is dated at latest Aquitanian-early Burdigalian (Fig. 2e) (Burrus 1984)and it is marked in the stratigraphic record by a break-up unconformity (Gorini et al. 1993). Oceanic accretion in the basin allowed the anticlockwise rotation of Corsica-Sardinia-Calabria. The pole and amount of rotation are still under discussion (see review in Vially & Tr6moli6res 1996). Palaeomagnetic data suggest a rotation of some 30 ~ which would have taken place between 21 and 19 Ma (Montigny et al. 1981; Burrus 1984). However, there is a growing body of evidence for a longer period of rotation and oceanic accretion lasting until latest Burdigalian-early Langhian (Fig. 2f) (Vigliotti & Kent 1990; Vigliotti & Langenheim 1995; Chamot-Rooke et al. this volume). Continental stretching on the margins and accretion of some 200 km of oceanic crust is accommodated by equivalent southeastward retreat of the subduction zone with respect to Europe, while between 35.5 Ma and 19.5 Ma Africa (at the longitude of Algiers) had a northwards motion of 160 km (Dewey et al. 1989). Southwest of the North Balearic Transform Zone, subduction occurred south of the Kabylies blocks. Calc-alkaline volcanism started in the Valencia Trough in the Aquitanian (Fig. 2d), i.e. with the same delay as the onset of extension with respect to the Gulf of Lion and lasted throughout the Burdigalian (Fig. 2e) (Mart/et al. 1992; Verg6s & Sabat this volume). Continental break-up occurred between the Balearics and the Kabylies during Langhian (Bartrina et al. 1992), leaving the Valencia Trough as an aborted intra-continental rift. The Kabylies blocks drifted southwards and were thrust over northern Africa (Wildi 1983; Bouillin 1986). The southward retreat of the subduction is confirmed by migration of the calcalkaline volcanic arc from the Valencia Trough to the Kabylies around Langhian time. Late Miocene to Recent magmatism in the Valencia Trough (Fig. 2f-h) is characterized by intraplate alkaline magmatism (Mart/et al. 1992). The second stage of evolution started during the mid-Miocene. After the end of oceanic accretion in the Gulf of Lion and Ligurian basins, and cessation of the rotation of Corsica-Sardinia, rifting occurred between the latter and Calabria (Fig. 2g, h). Continental extension gave way to oceanic accretion in the Tyrrhenian during late Miocene, both extension and axis of accretion migrated southeastward during Messinian to Pliocene (Kastens et al. 1988; Sartori 1990). By Tortonian, calc-alkaline volcanism in Sardinia had ceased and was
GULF OF LION OVERVIEW replaced by intraplate alkaline volcanism (Fig. 2g) (Bellon 1976; Coulon 1977). Subduction was still active, but the calc-alkaline volcanic arc had jumped to the Tyrrhenian Sea. The Calabrian accretionary wedge collided with the continental Apulian plate during this interval. In the northern Tyrrhenian, rifting did not reach continental break-up. Extension migrated eastward, so did compression in the accretionary wedge represented by the Apennines (Sartori 1990). Outward migration of the subduction system or subduction retreat is recognized throughout the Mediterranean (Jolivet et al. 1994) and it is explained by sinking of the dense lithospheric footwall of the subduction in the asthenosphere (Malinverno & Ryan 1986). Doglioni (1991) has suggested that such process is enhanced by westward-dipping subduction, due to the generalized eastward flow of the mantle with respect to lithospheric plates.
19
were located on structural highs, thus 'missing' the syn-rift sedimentary record (Cravatte et aL 1974); the refraction experiment covered the Ligurian Sea and stopped along the eastern side of the Gulf of Lion margin (Le Douaran et al. 1984); ECORS deep seismic profile, located along these ESPs, left unexplored the deepest rift basins and the corresponding deep crustal structure; besides, the Italian equivalent CROP profile off Sardinia is not across the exact conjugate margin (de Voogd et aL 1991). As a result, the structure of the Gulf of Lion margin remained a poorly understood segment of the NW Mediterranean compared to neighbouring Valencia Trough and Ligurian Sea, as illustrated by the blanks in structural maps of Mauffret et al. (1995).
A g e o f rifting: W e s t E u r o p e a n
rifts v e r s u s
back-arc basin
The recent acquisitions: IBS-Gulf of Lion The objectives of IBS-Gulf of Lion were to: (1) establish the 2D and 3D tectonics and kinematics of extension, (2) define tectono-sedimentary models of extensional basins, (3)investigate thermal and mechanical processes controlling lithospheric extension, in such complex settings. Along with other independent projects (e.g. Gorini 1993; Guennoc et al. 1994; Mauffret et al. 1995) one of the aims of the IBS-Gulf of Lion project, was to re-evaluate, develop and synthesize existing data. No data acquisition was planned except selective field observations. Instead, it was decided: (i) to compile and homogenize available data (bathymetry, topography, gravimetry, heat flow) in the Gulf of Lion and associated Alg6ro-Proven~al oceanic basin; (ii) to use recent synthetic structural maps based on commercial seismic data to be published by BRGM, (iii) to re-process the ECORS-Gulf of Lion deep seismic profile shot in 1988 which was still largely unexploited, (iv) to update the structural study of the onshore part of the margin, integrating the newly released seismic surveys. In spite of many studies undertaken during the 1970s and 80s, including thousands of kilometres of industrial and academic seismic profiling, industrial drillings, which turned the Gulf of Lion into a natural laboratory where subsidence (e.g. Steckler & Watts 1980), and basin models (e.g. Burrus & Audebert 1990) were tested, the formation and evolution of this continental margin was still largely unknown at the beginning of the 1990s. Many of the operations undertaken fell short of the expectations of the community. For example, all exploratory wells
The Gulf of Lion is seen either as a southern extension of the Tertiary West European Rift system (e.g. Bergerat 1985; Ziegler 1992) or as one of the back-arc basins developed above a NW-dipping subduction in the western Mediterranean (e.g. Auzende et al. 1973; Biju-Duval et al. 1978; R6hault et al. 1984; Mauffret et al. 1995). One way to distinguish between the two models is to document precisely the chronology of events. Stratigraphic correlations of the different onshore basins of the southern margin of Eurasia from Valle Penedes to Northern Vat half-grabens (Fig. 3) show remarkable correlation and contemporaneous unconformities related to otherwise identified geodynamic events. The West European Rift system initiated in the Rhine Graben (Fig. 4) during late Eocene time and extension migrated northward and southward (Ziegler 1992). The mid-Eocene tectonics of this area is characterized by left-lateral motion along the NNW-striking bordering faults, consistent with N-S 'Pyrenean' contraction (Bergerat 1985; Larroque 1987). East-west extension started during Priabonian time in the Rhine Graben (Larroque 1987; Maurin 1995), the Limagne (Morange et al. 1971; Burg et al. 1982; Bl6s et at,. 1989;Busson 1992), and the Bresse and Valence grabens (Debrand-Passard & Courbouleix 1984; Busson 1992). The earliest syn-rift sediments of the Camargue basin, only known from industrial boreholes, are not dated. However, these evaporites and shales have been correlated with the evaporite series of the Valence, Bresse and Rhine basins spanning the Priabonian-early Rupelian periods (DebrandPassard & Courbouleix 1984; Busson 1992). In
20
M. SI~RANNE
GULF OF LION OVERVIEW
Fig. 3. (a) Tectonostratigraphic correlations of major onshore rift basins of the northern margin of the Alg6ro-Provenqal basin, compared with the Sardinia conjugate margin, and the West European Rift system. Location of the basins is indicated in Fig. 4. Time scale from Cande & Kent (1992). Data compiled from different sources, including Morange et al. (1971); Cherchi & Montadert (1982); DebrandPassard & Courbouleix (1984); Valette (1991); Bartrina et al. (1992); Biondi et al. (1992); Busson (1992); Roca & Guimerh (1992); Ziegler (1992); Maerten & S6ranne (1995); Benedicto (1996). (b) Key. the Manosque-Forcalquier and A16s basins onset of syntectonic sedimentation along NNEand NE-trending faults respectively, occurred in late Priabonian time, following Pyrenean syntectonic sedimentation (Bartonian) (DebrandPassard & Courbouleix 1984; Bergerat 1985; Biondi et al. 1992). Finally, in the Campidano basin of Sardinia, which represents the Gulf of Lion conjugate margin, the earliest sediments date from the Priabonian (Cherchi & Montadert 1982). When Sardinia is replaced in a pre-rift position, Campidano basin-bounding faults strike NNE, similarly to the rifts on the European margin. E - W extensional stress along NEto NNE-trending faults involves oblique extension with a component of left-lateral strike-slip. Late Rupelian continental syntectonic sediments were deposited, after a hiatus, above an angular unconformity in small continental basins and in earlier basins formed during Eocene (Debrand-Passard & Courbouleix 1984) (Fig. 3). The late Rupelian unconformity can be correlated from southern France to Sardinia; however it is not documented in SW France and Catalunya, where syntectonic sedimentation related to extension started in late Chattian-early Aquitanian times (DebrandPassard & Courbouleix 1984; Bartrina et al. 1992). As a result of N W - S E extension during late Rupelian, northeast-trending fault-bounded
21
basins were formed, and earlier Priabonian basins were reactivated (Fig. 4). From these correlations it can be gathered that the West European Rift system was initiated in Priabonian time, during a phase of E - W extension. The large basins (Al6s, ManosqueForealquier, and probably Camargue) in Southern France were formed during this time by oblique extension. A second phase of rifting due to N W - S E extension started in late Rupelian in the Gulf of Lion area (from southern France to Sardinia). This later phase of extension is contemporaneous with intraplate alkaline volcanism in Languedoc and onset of calc-alkaline subduction-related volcanism in Sardinia (Bellon & Brousse 1977). It is concluded that rifting of the Gulf of Lion margin occurred in late Rupelian time, in relation to back-arc extension, as a geodynamical event distinct from the West European Rift. Spatial overlap of the two events in southern France resulted in the reactivation of Priabonian basins. S t r u c t u r e o f the m a r g i n a n d m o d e s o f crustal s t r e t c h i n g
The Gulf of Lion margin extends SE of the C6vennes Fault (Fig. 5). Extensional structures are predominantly oriented NE, and overprint the Pyrenean thrust and fold belt. The interaction of inherited Pyrenean structures with extension has been analysed by S6ranne et al. (1995). The margin is segmented by several transfer fault zones sub-parallel to extension. The most representative segment which is analysed in the following sections is located between the Arl6sienne and S6toise transfer zones (Fig. 5). Moho depth data (wide-angle refraction and the ECORS deep seismic reflection profiles) are mostly restricted to a dip section across the Gulf of Lion. The lateral crustal variability introduced by the along-strike segmentation of the margin could only be addressed by a 3D approach. Gravity data can be used to obtain reliable geometry of the Moho. Compilations and homogenization of gravity data over the Liguro-Proven~al basin resulted in the preparation of maps, presented in a companion paper (Chamot-Rooke et al. this volume). A totally new Moho depth map was derived by inversion of 3D gravity data. This map was successfully tested against the few available Moho depth measurements. A section across the continental margin (Fig. 6) allows to distinguish: (1) the poorly deformed upper-margin, located mostly onshore, (2)the continental shelf, (3) the slope and (4) the basin floored by intermediate or oceanic crust.
22
M. SF,R A N N E
Fig. 4. Structural map of Priabonian to Aquitanian syn-rift basins from the Rhine Graben to the Valencia Trough. The Gulf of Lion margin is located at the junction of the West European Rift system and the rifts of the NW Mediterranean. Although these basins are often associated in the literature, they differ by the age of rifting onset (Priabonian on one hand, and late Rupelian to Aquitanian on the other hand), structural trend (N-S and NE-SW), and orientation of extensional stress (E-W, and NW-SE). The two systems overlap in the onshore part of the Gulf of Lion, where the Camargue, Albs and Manosque-Forcalquier basins were initiated during Priabonian under E - W extensional stress, and were reactivated during Oligocene-Aquitanian times under NW-SE extensional stress. Compiled from R6hault et al. (1984); Bergerat (1985); Larroque (1987); Bartrina et al. (1992); S6ranne et al. (1995); Benedicto (1996). Bar, Barcelona; Mtp, Montpellier; Mar, Marseille.
GULF OF LION OVERVIEW
23
Fig. 5. Structural map of the Gulf of Lion margin. The rift basins are controlled by NE-trending extensional faults overprinting the E-W Pyrenean thrusts in Provence and Languedoc. Thin-skinned extension (light shading) characterizes the onshore part of the margin, whereas the shelf and the slope area are deformed by basement-involved tectonics (dark shading). The margin is segmented by transfer zones parallel to extension. The ECORS seismic profile and ESP's corresponding to the section Fig. 6 are indicated. Ma, Marseille; Ni, Nimes; Mo, Montpellier; Na, Narbonne; Pc, Perpignan. NPFZ, North Pyrenean Fault Zone; MFB, Manosque-Forcalquier Basin; CB, Camargue Basin; AB, Albs Basin; LMB, Les Matelles Basin; HB, H6rault Basin. Modified after Sdranne et al. (1995).
Upper margin. Landward, extensional deformation is bounded by the C6vennes Fault (Fig. 6). The Mesozoic-Eocene cover is affected by NE-striking, SE-facing normal faults bounding Oligocene or Early Miocene basins. The Jurassic and Neocomian limestones cropping out in Languedoc display n u m e r o u s micro-faults which have yielded extensional stress tensors with a m i n i m u m stress axis oriented N120 ~ (Arthaud et al. 1977), consistent with the basinformation event. Interpretation of industrial seismic reflection profiles indicates that the basin-bounding faults detach above the Palaeozoic basement, in Triassic evaporites and shales (Maerten & S6ranne 1995; Benedicto 1996). The underlying continental crust is therefore not affected by extension in this part of the margin. However, the crust is thinning towards the margin from 30 km beneath the south Massif Central (Sapin & Hirn 1974) to 20 km close to
the shoreline (de Voogd et al. 1991; ChamotR o o k e et al. this volume), suggesting either lower-crustal thinning during Oligocene rifting, or crustal stretching and thinning inherited from the Mesozoic rifting. Immediately east of the considered section, there is a good correlation between the isopaches of Mesozoic series (Debrand-Passard & Courbouleix 1984) and the Moho (Chamot-Rooke et al. this volume) which favours the second alternative. The onshore Gulf of Lion margin is therefore characterized by thin-skinned extensional tectonics of the prerift sedimentary cover, which accommodated only several kilometres extension. Continental shelf. Structurally, the continental shelf extends mainly offshore, SE of the basement-ramp in which thin-skinned extension is transferred. On the section presented (Fig. 6) this ramp is a reactivated E - W - o r i e n t e d
24
M. St~RANNE
Fig. 6. Section of the Gulf of Lion showing the distribution of tectonic style across the margin. Based on field observations and industrial seismic onshore data, and on the ECORS seismic profiled reprocessed during IBS project (Pascal et al. pers. comm.). Moho data from Pascal et al. (1993) and Chamot-Rooke et al. (this volume). Vertical = horizontal scale.
P y r e n n e a n thrust; however the NE-trending N~mes Fault (Fig. 5) corresponds to this ramp for most of the width of the studied area. Formation of the onshore syn-rift Camargue basin was controlled by the N~mes Fault. This basin comprises two half-grabens: the Vistrenque and the Petit Rh6ne grabens, and presents the thickest (>4000 m) and most complete syn-rift succession of the margin (see below). Seismic reflection profiles interpretation of the area undertaken during the course of IBS project has shown that the basinbounding NJmes Fault was a ramp dipping 25-30 ~ SE, into the upper crust; upwards, the fault cuts through the Mesozoic cover at a higher angle (Fig. 6) (Benedicto et al. 1996). Offshore, the structures of the continental shelf are known from the E C O R S deep seismic profile, and from industrial seismic reflection (Gorini 1993; Gorini et al. 1994). Beneath the continental shelf the Moho is 25 to 20 km deep. The major extensional faults dipping 25-30 ~ can be traced in the upper crust and merge with the 5 km thick reflective lower crust (S6ranne et al. 1995) (Fig. 6). They strike parallel to the N~mes Fault (Gorini et al. 1993) and bound thin (22-17 Ma), and the earliest nappe-sealing sedimentary rocks vary regionally in age from c. 22 to 17 Ma (Table 1). Structural, stratigraphic and chronological work in Alpine Corsica, Apennines and Calabria suggest oldest metamorphic ages of c. 80-60 Ma ( K - A r age of c. 90 Ma, Maluski 1977; Rb-Sr isochron age of 105 _+20 Ma, Cohen et al. 1981; and somewhat younger stratigraphic constraints, Egal 1992; De Roever 1972). The older ages should perhaps now be regarded with caution (H. Maluski 1996 pers. comm.). More recent 4~ work suggests ages ranging from c. 65 to 35 Ma (Moni6 et al. 1996). A recent 35 Ma single-zircon age for the HP metamorphism in the western Alps (Gebauer et al. 1997) may be relevant to note. Late E - W - and S W - N E directed extensional tectonics range from c. 27 Ma to recent (e.g., Jolivet et al. 1991; Kastens et al. 1988), oldest nappe sealing rocks from c. 22 to 10 Ma (Egal 1992; Carmignani et al. 1995). In the Kabylies, early Alpine metamorphic ages of c. 80 Ma were reported for high-grade core complexes (4~ muscovite, biotite and feldspar, Moni6 et al. 1988). Younger ages of c. 25 Ma (4~ and Rb-Sr muscovite and biotite, Moni6 et al. 1988) were obtained from regional mylonite zones with regional southdirected tectonic transport (cf. Mahdjoub & Merle 1990), which continued until Recent time (Mauffret et al. 1987). Ages of oldest nappe sealing sedimentary rocks were given as c. 23-21 Ma (Saadallah & Caby, 1996). The analogous structural, stratigraphic and geochronological development of these Alpine complexes, although c. 1000 km apart, suggests that they are part of the same regional
113
geodynamic development. The data indicate an early stage characterized by subduction activity starting in earliest Tertiary time, perhaps latest Cretaceous, and a later orogenic stage characterized by thin-skinned, extensional tectonics which took place from c. 30 Ma to Recent time. The oldest sedimentary rocks sealing these extensional allochthons have ages between c. 25 and 10 Ma.
Reconstruction of the Alpine collisional belt in the western Mediterranean The original configuration of the Alpine subduction zone system may be reconstructed by considering on the one hand the regional pattern of tectonic transport directions and on the other hand the distribution of late-stage extensional basins in the western Mediterranean region. The southern termination of the Alpine subduction zone in the western Mediterranean (here called the Betic-Ligurian subduction system) is given by the sinking slab under the Betic-Rif orogen (Figs 2, 3 and 4b,c). Its abrupt southwestern termination represents the transcurrent contact towards the African plate. The equally abrupt nature of its northeastern termination suggests that the Betic-Ligurian subduction system was formed not as a continuous, smooth belt, but rather had a composite character, segments with active subduction being separated by transform fault zones. Such development of the subduction system may have been controlled mainly by factors such as the shape of the Iberian continental plate and the variable nature of the subducting/colliding Tethyan realm lithosphere. An outline of its preferred configuration is given in Fig. 4b and c; constraints are discussed below. Two approximately N W - S E transform fault zones are proposed (Fig. 4b): one north of Alicante, the other, following an earlier suggestion by Cohen (1980), south of the pre-drift position of Sardinia. It is further proposed that subsequent Miocene extensional regimes which broke up the subduction system were patterned upon these fundamental discontinuities and may be described in part by two spreading poles (Dewey et al. 1989), one in the Genoa area for the Corsica-Sardinia block, and one southwest of Alicante for the Balearic Islands. The length of the Sardinia-Menorca transform fault zone is suggested to be c. 250 km (Fig. 4b). This estimate is based on reconstruction of the CorsicaSardinia rotation (sinistra150~50 ~ c. 400 km for south Sardinia; c. 23-18 Ma, Dewey et al. 1989; 21-16 Ma, Todesco & Vigliotti 1993) and the much later Tyrrhenian extension. The length of the Alicante transform fault zone is not very well
114
H.P. ZECK
defined; it may range from very small to up to c. 100 km. The Balearic subduction zone segment, located between the two proposed transform
fault zones (Fig. 4b, c), comprised the Kabylian collisional prism. The segment was translated towards the SE from its suggested pre-drift position along the present Iberian coast (Dewey et
ALPINE PLATE KINEMATICS, W MEDITERRANEAN al. 1989; Fig. 4b) through opening of the Gulf of Valencia (c. 23-19 Ma, Dewey et al. 1989, Banda & Santenach 1992; c. (25-)15-10 Ma, Torres et al. 1993). Additional extension south of the Balearic Islands (c. 22-13 Ma, Dewey et al. 1989) would have translated it further southeastward (Fig. 4c). This translation is suggested to have been by slab roll-back which requires old subductable oceanic lithosphere southeast of the Balearic subduction zone segment and concomitant spreading of back-arc character between the Kabylian collision prism and the present Balearic Isles (for details on the slab roll-back mechanism, see Elsasser 1971 and Le Pichon & Angelier 1981). Concurrent N-S AfricanIberian convergence (Dewey et al. 1989; Srivastava et al. 1990) might have been accommodated in part along this retreating slab system and in part along a parallel subduction zone somewhat further south (Fig. 4c; see below). This kinematic evolution resulted in the Kabylian core complex being translated from its collisional position close to the slab trench (Fig. 4b) towards the African plate where it was emplaced onto the African foreland (Fig. 4c), such in agreement with consistent top-to-south, Miocene and younger regional tectonic transport directions observed in the Kabylies (see above). Seismic tomography information on the northern termination of the African plate is not conclusive (cf. Mueller 1989) and therefore it is difficult to decide whether late-stage slab breakoff which would have supported uplift of the Kabylian core complex and extensional emplacement upon the African foreland, similar to the model proposed for the Betic-Rif, has been part of the process. Neither is the crucial cooling information, which was an important argument supporting the Betic-Rif model, available for the Kabylies. There is some tomographic support, though, for a steeply N-dipping slab north of the African plate (Mueller 1989), and this might represent the 20-30 ~ dextrally rotated, peeled back oceanic lithosphere of the Balearic segment (Fig. 4b,c). The model is
115
supported by older Nll0-140~ stretching lineations (c. N85-115~ after drift correction) reported by Moni6 et al. (1988) and Saadallah & Caby (1996) for Kabylian metamorphic core complexes. These directions are identical to syncollisional directions of tectonic transport in core complexes of the Betic Cordilleras (Vauchez & Nicolas 1991; Zeck 1996a) and thus conform to the overall E - W convergent character of the Betic-Ligurian subduction system implied by the regional model here presented (Fig. 4b). The Corsica-Sardinia segment of the Betic-Ligurian subduction system also went through a slab roll-back translation after subduction (Fig. 4b, c). This involved coeval backarc type extension west of the Corsica-Sardinia block and formation of the (in part) oceanic Provencal-Algerian basin (Fig. 4c; cf. ChamotRooke et al. 1996). In analogy with the model outlined above for the Balearic-Kabylies segment, this roll-back operation requires old subductable oceanic lithosphere east of the west-dipping subduction zone indicated in Fig. 4b. Results of earlier work which concluded that Alpine Corsica consists of an ophiolite containing nappe sequence which for its major part was formed in a west-dipping subduction zone, mainly by underthrusting from the east (Egal 1992; Jolivet et al. 1994; Carmignani et al. 1995) are in good agreement with this model. Final allochthon emplacement was by west-directed thin-skinned tectonics (Warburton 1986; Carmignani et al. 1995). Complementary eastward sense of tectonic transport in the northern Apennines and Elba suggests regional rock uplift over a sinking slab in the area between Corsica and Apennines and agrees well with earlier work in this particular region (e.g., Jolivet et al. 1994; Carmignani et al. 1995; Keller & Coward 1996; and references therein) and conforms to the Betic-Rif model outlined above. Seismic tomography information (Wortel & Spakman 1992) indicates that the detached, lithospheric slab at present is located c. 100 km
Fig. 4. Outline of the kinematic evolution of the western Mediterranean (modified after Zeck 1996b). (a) Location of metamorphic core complexes, modified after Coward and Dietrich (1989), showing major geographic entities. (b) Final stage of subduction with Iberian plate colliding with continental Betic-Ligurian lithosphere after Mesozoic oceanic Tethyan lithosphere has been subducted westwards under eastward drifting Iberia. The subduction system did not form a smooth, continuous belt: three segments of active subduction being separated by two transform fault zones. (c) Miocene stage of extensional regimes with slab roll-back and local slab detachment involving opening of the Gulf of Valencia, the Provenqal-Algerian basin and Tyrrhenian basin, but prior to the latest stage extension in the southern Tyrrhenian Sea basin (Vavilov and Marsili oceanic basins in dark shading) which translated the slab further east (hatched double arrows) towards its present position (Fig. 1, stage 4) and assisted in the final emplacement of the Calabria core complex, overriding the E-W-trending subduction system (hinge line in bold) accomodating major N-S Europe-Africa convergence (cf. Figs 5, 6). Black arrows indicate late stage extensional tectonic transport directions.
116
H.P. ZECK
more to the east, directly below t h e Apennine chain (Fig. 1, stage 4), that is below the Evergent cover complexes, suggesting that the detached slab underwent some late stage movement towards the east relative to the overlying lithosphere. This suggestion is supported by a recent fault kinematic study by Keller & Coward (1996) indicating an eastward migration of the centre of extension across the Tyrrhenian Sea. Displacement of the southern part of the Corsica-Sardinia subduction zone segment has been much larger than for its northern part (Fig. 4c) and therefore a scenario similar to that given above for the Balearic-Kabylian segment is suggested with additional extension taking place between the Sardinian block and the Calabrian collisional wedge which was located close to the slab trench (Fig. 4c). This resulted in slab rollback towards its present position under the Apennines and initial emplacement of Neogene, E-vergent allochthons in the Apennine chain.
The situation in the southern part of the Corsica-Sardinia segment is complicated by considerable, very fast latest stage extension (Figs 4c, 5, 6) with neoformation of oceanic crust in the southern part of the Tyrrhenian Sea (c. 250 km long, NW-SE elongated area with Vavilov and Marsili basins, c. 7 Ma to Present, younging southeastwards, Kastens et al. 1988). This latestage, predominantly E - W - and N W - S E directed extension has advanced final emplacement of E-(SE-)vergent allochthons in the southern Apennines and brought the Calabrian core complex into its present position, possibly, as in the Betic Rif, involving extrusion tectonics (Figs 4c, 6). This model for the Calabrian region explains a number of local geological features which otherwise appear enigmatic. Consistent top N E - w a r d tectonic transport directions in collisional tectonic melanges in Calabria and southern Apennines (Monaco &
Fig. 5. Sicily-Calabria area with seismic epicentres of events >50 km depth during the period 1988-1993; data from Selvaggi & Chiarabba (1995). Marsili oceanic crust floored basin in striped signature. In bold: hinge line of subduction system which steeply N-dipping Tethyan oceanic lithosphere is overridden by European (crustal) lithosphere (cf. Figs 4c, 6).
ALPINE PLATE KINEMATICS, W MEDITERRANEAN
NW
SE MARSILI
~ CALABRIA
o
,,
-k
r
117
area, the subduction zone has much lower level of >50 km depth seismicity (Figs 5, 6). The present lithospheric slab configuration in the Calabria region according to current interpretations (e.g., Channel & Mareschal 1989; Wortel & Spakman 1992; Robertson & Grasso 1995; Catalano et al. 1995) would comprise a continuous subduction zone (Fig. 1, stage 4) which runs southwards along the Apennines, curves westward around Calabria and continues north of the African plate. The model suggested here implies an alternative lithospheric slab configuration in which the southern Apennines, Calabria and Sicily are dominated by thrust units derived from the Betic-Ligurian collisional realm and emplaced in a regionally extensional setting controlled by eastward slab roll-back and E - W and N W - S E extension (Fig. 4c). The subduction system running north of the African plate and eastward over Sicily and Calabria (Figs 4c, 5, 6) in this model is a separate entity, continuing eastward to connect with the subduction zone dipping northward under Crete.
r 100
km
Fig. 6. Cross section showing seismicity outlining steeply N-dipping subducted Tethyan oceanic lithospheric slab; for location see Fig. 5; diamond on top frame indicates intersection point with the subduction zone hinge line indicated in Figs 4c and 5.
Tortorici 1995) are at variance with a N-S-convergent subduction regime, which is currently claimed (e.g., Dewey et al. 1989), but are in good agreement with the top-to-the-E tectonic regime of the Betic-Ligurian subduction zone, if the 45 ~ post-collisional sinistral rotation is taken into account (Fig. 4c). A Cretaceous age for H P metamorphism in core complexes in Calabria (De Roever 1972) is difficult to combine with Late Oligocene or even later collision as currently suggested (Monaco & Tortorici 1995), but agrees well with the collision timing in the Betic-Ligurian subduction system (see above; Table 1). A high level o f seismicity in a restricted area N W o f Calabria is explained by N W - S E convergence under the influence of extension centered in Vavilov and Marsili basins. Earthquake hypocentres are concentrated within the denser lithospheric slab. Outside this restricted
Conclusions An outline for the tectonic evolution of the Alpine orogen in the western Mediterranean is presented which has the westward subduction of Tethyan oceanic lithosphere under eastward drifting Iberia as its driving force. The resulting SW-NE-striking Betic-Ligurian subduction zone system had a segmented character: three sectors with subduction activity being separated by two transform fault zones. The Tethyan realm lithosphere which was subducted under, and ultimately collided with, the Iberia plate had a laterally variable character. In the Betic segment, the southern one, the Betic-Rif collisional wedge, formed after Tethyan oceanic lithosphere had been subducted, was not subsequently affected by a slab roll-back development. The subducted, and detached, Betic-Rif lithospheric slab is at present located under the Betic-Rif orogen (Figs 2, 3, 4). The intermediate, BalearicKabylian, segment underwent slab roll-back after collision which indicates that the Kabylian collisional wedge represents a small continental fragment located between Tethyan oceanic lithosphere which was subducted westward before collision and Tethyan oceanic lithosphere which was peeled back southeastward after collision. The northern, Corsica-Sardinia, segment likewise underwent considerable slab roll-back after subduction, concomitant with back-arc type extension both in the Provenqal-Algerian basin west of Sardinia and
118
H.P. ZECK
b e t w e e n eastern Sardinia and the Calabrian collisional wedge which was located close to the slab trench (Figs 4b, c). This indicates that also the Calabrian collisional prism represents a small continental fragment within the Tethyan realm. It thus seems that the rather irregular p a t t e r n of extensional basins and roll-back t r e n c h e s which characterizes the w e s t e r n M e d i t e r r a n e a n reflects the distribution pattern of continental and oceanic lithosphere in the Tethyan realm prior to Alpine tectogenesis (cf. A l v a r e z 1976). The ultimate s o u t h e a s t w a r d e m p l a c e m e n t of the Calabrian core complex took place under the influence of the latest stage, c. 7-0 Ma extension involving generation of oceanic crust in Vavilov and Marsili basins. The tectonic evolution model presented here implies that N-S convergence played a minor role in western M e d i t e r r a n e a n tectonics, whereas in the eastern Mediterranean its influence was much more important due to the sinistral hinged m o v e m e n t of the African plate with respect to Eurasia (Dewey et al. 1989; Mueller 1989; Srivastava et al. 1990). This led to developm e n t of the well defined E - W - t r e n d i n g subduction z o n e system dipping n o r t h w a r d u n d e r Crete. It is suggested here that this subduction zone continues westward over Calabria and Sicily and further westward north of the African plate, loosing importance on the way. The tectonic complexities in the SicilyCalabria region (e.g., Dewey et al. 1989; Channel & Marechal 1989; Robertson & Grasso 1995; Catalano et al. 1995; Monaco & Tortorici 1995) may be explained by its location in a zone where two different tectonic regimes meet, one controlled by the E-W-convergent Betic-Ligurian subduction zone system and its off-spin of E(SE)-vergent allochthons following slab roll-back translations, the other controlled by northward subduction of Tethyan oceanic lithosphere under the influence of northward drift of Africa with respect to Europe. Supported by the Danish Research Council (SNF) and Carlsberg Foundation, and in earlier stages by NATO. I thank H. Maluski (Montpellier) for hospitality, A. Saadallah (Z&S, Stavanger) for introduction in North African geology, L. Jolivet (Paris) and A. Mascle (IFP, Rueil Mahnaison) for organizing the 1996 Mediterranean geology meeting in Cergy, and R. Caby (Montpellier) and J. Verg6s (Barcelona) for inspiring and constructive reviews.
References AGUADO, R., FEINBERG, H., DURAND-DEI,GA, M., MART[N-ALGARRA,A., ESTERAS, M. & DIDON,J. 1990. Nuevos datos sobre la edad de formaciones
miocenas transgresivas sobre las Zonas Internas b6ticas: la formaci6n de San Pedro de Alc~ntera (Prov. de M~ilaga). Revista Sociedad Geolologia Espa~a, 3, 79-85. ALVAREZ,W. 1976. A former continuation of the Alps. Bulletin Geolological Society of America, 87, 891-896. ANDRIESSEN,P. A. M. & ZECK, H. P. 1996. Fission track constraints on timing of Alpine nappe emplacement and rates of cooling and exhumation, Torrox area, Betic Cordilleras, S Spain. Chemical Geology (Isotope Geosciences), 131, 199-206. ARGAND, E. 1922. La tectonique de l'Asie. In: 8th International Geolological Congress, 171-372. AZA/q0N, J. M. & GOFFE, B. 1997. Ferro- and magnesiocarpholite assemblages as record of high-P, Iow-T metamorphism in the Central Alpujarrides, Betic Cordillera (SE Spain). European Journal of Mineralogy, 9, 1035-1051. BALANYA,J. C., AZAIq0N,J. M., SANCHEZ-GOMEZ,M. & GAr~C|A-DUESAS, V. 1993. Pervasive ductile extension, isothermal decompression and thinning of the Jubrique unit in the Paleogene (Alpujarride Complex, W Betics, Spain). Comptes Rendus de l'Acad~mie des Sciences Paris, 316, s6r. II, 1595-1601. BANDA, E. • SANTANACH,P. 1992. The Valencia trough (western Mediterranean): an overview. Tectonophysics, 208, 183-202. BLANCO,M. J. & SPAKMAN,W. 1993. The P-wave velocity structure of the mantle below the Iberian Peninsula: evidence for subducted lithosphere below southern Spain. Tectonophysics, 221, 13-34. BOUYBAOUENE, M. L., GOFFE, B. & MICHARD, A. 1995. HP-LP metamorphism in the Sebtides nappes, northern Rif, Morocco. Geogaceta, 17, 117-119. CARMIGNANI,L., DECANDIA,E A., DISPERA'FI,L., FANTOZZI, P. L., LAZZAROTTO, A., LIOTFA, D. & OGGIANO, G. 1995. Relationships between the Tertiary structural evolution of the SardiniaCorsica-Provenqal Domain and the Northern Apennines. Terra Nova, 7, 128-137. CATALANO,R., INFUSO,S. & SULH, A. 1995. Tectonic history of the submerged Maghrebian Chain from the Southern Tyrrhenian Sea to the Pelagian Foreland. Terra Nova, 7, 179-188. CHAMOT-ROOKE, N., MAILI.ARD,A., GAUI.IER, J. M. & PASCAl., G. 1996. Crustal structure of the LiguroProvenqal basin from gravity modelling: geodynamic implications. In: The Mediterranean Basins: Tertiary extension within the Alpine orogen. Workshop 11-13.12.1996, Cergy-Pontoise, Paris, 35-36. CHANNELL, J. E. T. & MARESCHAI,,J. C. 1989. Delamination and asymmetric lithospheric thickening in the development of the Tyrrhenian rift. In: COWARD, M. P., PARK, R. G. & DIETRICH,D. (eds) Alpine Tectonics. Geological Society, London, Special Publications, 45, 285-302. COHEN,C. R. ! 980. Plate tectonic model for the OligoMiocene evolution of the western Mediterranean. Tectonophysics, 68, 283-311.
A L P I N E PLATE KINEMATICS, W M E D I T E R R A N E A N , SCHWEICKERT,R. A. & ODOM, A. L. 1981. Age of emplacement of the Schistes Lustr6s nappe, Alpine Corsica. Tectonophysics, 73, 267-283. COWARD, M. & DmTmCH, D. 1989. Alpine tectonics an overview. In: COWARD, M. P., PARK, R. G. & DIETRICH, D. (eds) Alpine Tectonics, Geological Society, London, Special Publications, 45, 1-29. DE JONG, K. 1991. Tectono-metamorphic studies and radiometric dating in the Betic Cordilleras (SE Spain) - with implications for the dynamics of extension and compression in the western Mediterranean area. Thesis, Amsterdam Christian University. DE ROEVER, E. W. E 1972. Lawsonite-albite-facies metamorphism near Fuscaldo, Calabria (S Italy), its geological significance and petrological aspects. Proefschrift Amsterdam University. DEWEY, J. E 1988. Extensional collapse of orogens. Tectonics, 7, 1123-1139. - - , HELMAN, M. L., TURCO, E., HuTroN, D. H. W. & KNOTT, S. D. 1989. Kinematics of the western Mediterranean. In: COWARD,M. R, PARK, R. G. & DIETRICH, D. (eds) Alpine Tectonics. Geological Society, London, Special Publications, 45, 265-283. EGAL, E. 1992. Structures and tectonic evolution of the external zone of Alpine Corsica. Journal of Structural Geology, 14, 1215-1228. ELSASSER,W. M. 1971. Seafloor spreading as thermal convection. Journal of Geophysical Research, 76, 1101-1112. FLINCH,J. E, BALLY,A. W. & WU, S. 1996. Emplacement of a passive-margin evaporitic allochthon in the Betic Cordillera of Spain. Geology, 24, 67-70. GEBAUER, D., SCHERTL, H. R, BRIX, M. & SCHREYER, W. 1997. 35 Ma old ultrahigh-pressure metamorphism and evidence for very rapid exhumation in the Dora Maira massif, western Alps. Lithos, 41, 5-24. HARLAND,W. B., ARMSTRONG,R. L., Cox, A. V., CRAIG, L. E., SMITH,A. G. & SMITH,D.G. 1990. A geologic time scale 1989. Cambridge Univ. Press, Cambridge. JOLIVET, L., DANIEL, J. M. & FOURNIER, M. 1991. Geometry and kinematics of extension in Alpine Corsica. Earth and Planetary Science Letters, 104, 278-291. - - , TRUFFERT,C. & GOFFE, B. 1994. Exhumation of deep crustal metamorphic rocks and crustal extension in arc and back-arc regions. Lithos, 33, 3-30. KASTENS ETAL. 1988. ODP Leg 107 in the Tyrrhenian Sea: insights into passive margin and back-arc basin evolution. Geological Society of America Bulletin, 100, 1140-1156. KELLER, J. V. A. & COWARD,M. P. 1996. The structure and evolution of the N Tyrrhenian Sea. Geolological Magazine, 133, 1-16. LE PICHON,X. & ANGELIER, J. 1979. The Hellenic arc and trench system: a key to the evolution of the eastern Mediterranean area. Tectonophysics, 60, 1-42. MAHDJOUB, Y. & MERLE, 0. 1990. Cindmatique des ddformations tertiaires dans le massif de Petite
119
Kabylie (Alg6rie orientale). Bulletin de la Soci~tO GOologique de France, 6, 629-634. MALUSKI, H. 1977. Application de la mOthode 4~ 39Ar aux min~raux des" roches cristallines perturbOes par des ~v~nements thermiques et tectoniques en Corse. Thesis Univ. Montpellier. MAUFFRET, A., EL-ROBRINI, M. & GENNESSEAUX, M. 1987. Indice de la compression r6cente en mer M6diterrande: un basin losangique sur la marge nord-alg6rienne. Bulletin de la SociOtd Gdologique de France, 3, 1195-1206. MONACO, C. & TORTORICI, L. 1995. Tectonic role of ophiolite-bearing terranes in the development of the Southern Apennines orogenic belt. Terra Nova, 7, 153-160. MONIfl, P., JOLIVET, L., BRUNET, C., TORRES-ROLDAN, R. L., CABY, R., GOFFE, B. & DUBOIS, R. 1996. Cooling paths of metamorphic rocks in the western Mediterranean region and tectonic implications. In: The Mediterranean Basins: Tertiary extension within the Alpine orogen. Workshop, Cergy-Pontoise, Paris, 16-17. - - , MALUSKI,H., SAADALLAH,A. & CABY, R. 1988. New 39Ar-4~ ages of Hercynian and Alpine thermotectonic events in Grande Kabylie (Algeria). Tectonophysics, 152, 53-69. --, TORRES-ROLDAN, R. L. & GARCiA-CASCO, A. 1994. Cooling and exhumation of the Western Betic Cordilleras, 4~ thermochronological constraints on a collapsed terrane. Tectonophysics, 238, 353-379. MUELLER, S. 1989. Deep-reaching geodynamic processes in the Alps. In: COWARD,M. R, PARK, R. G. & DIETRICH, D. (eds) Alpine Tectonics. Geological Society, London, Special Publications, 45, 303-328. NIJHUIS, M. 1964. Plurifacial Alpine metamorphism in the SE Sierra de los Filabres S, of Lubrfn, SE Spain. Proefschrift, Amsterdam University. PLATT,J. P. & VISSERS,R. L. M. 1989. Extensional collapse of thickened continental lithosphere: a working hypothesis for the Alboran Sea and Gibraltar arc. Geology, 17, 540-543. ROBERTSON,A. H. E & GRASSO,M. 1995. Overview of the Late Tertiary-Recent tectonic and palaeoenvironmental development of the Mediterranean region. Terra Nova, 7, 114-127. SAADALLAH,A. & CABY,R. 1996. Alpine extensional detachment tectonics in the Grande Kabylie metamorphic core complex of the Maghrebides (northern Algeria). Tectonophysics, 267, 257-273. SELVAGGI,G. • CHIARABBA,C. 1995. Seismicity and Pwave velocity image of the southern Tyrrhenian subduction zone. Geophysical Journal International, 121, 818-826. SERRANO, E 1990. Presencia de Serravalliense marino en la cuenca de Nijar (Cordillera B6tica, Espafia). Geogaceta, 7, 95-97. SRIVASTAVA,S. P., ROEST, W. R., KOVACS,G., OAKLEY, G., Lt~VESQUE,S.,VERHOEF,J. & MACNAB,R. 1990. Motion of Iberia since the Late Jurassic: results from detailed aeromagnetic measurements in the Newfoundland Basin. Tectonophysics, 184, 229-260.
120
H.P. Z E C K
THOMPSON, A. B., SCHULMAN, K. & JEZEK, J. 1997. Extrusion tectonics and elevation of lower crustal metamorphic rocks in convergent orogens. Geology, 25, 491-494. TODESCO, M. & VIOLIOTTI,L. 1993. When did Sardinia rotate? Statistical evaluation of the paleomagnetic data. Annali di Geofisica, 36, 119-134. TORRES, J., BOIS, C. & BURRUS,J. 1993. Initiation and evolution of the Valencia Trough (western Mediterranean): constraints from deep seismic profiling and subsidence analysis. Tectonophysics, 228, 57-80. TUBiA, J. M. 1994. The Ronda peridotites (Los Reales nappe): an example of the relationship between lithosphere thickening by oblique tectonics and late extensional deformation within the Betic Cordillera (Spain). Tectonophysics, 238, 381-398. & IBARGUCHI, J. I. 1991. Eclogites of the Oj6n nappe: a record of subduction in the Alpuj~irride complex (Betic Cordilleras, S Spain). Journal of the Geological Society, London, 148, 801-804. VAN DER WAL, D. & VISSERS, R. L. M. 1993. Uplift and emplacement of upper mantle rocks in the western Mediterranean. Geology, 21, 1119-1122. VAUCHEZ, A. & NrCOLAS,A. 1991. Mountain building: strike-parallel motion and mantle anisotropy. Tectonophysics, 185, 183-201. VISSERS, R. L. M., PLATr, J. P. & VAN DER WAL, D. 1995. Late orogenic extension of the Betic Cordillera and the Alboran domain: a lithospheric view. Tectonics, 14, 786-803. WARBURTON, J. 1986. The ophiolite-bearing Schistes Lustr6s nappe in Alpine Corsica: a model for the emplacement of ophiolites that have suffered HP/LT metamorphism. Memoirs of the Geological Society of America, 164, 313-331. WORTEL, M. J. R. d(z SPAKMAN,W. 1992. Structure and -
-
dynamics of subducted lithosphere in the Mediterranean region. Verhandelingen der
Koninklijke Nederlandse Akademie van Wetenschappen, Amsterdam, 95, 325-347. ZECK, H. E 1996a. Betic-Rif orogeny: subduction of Mesozoic Tethys lithosphere under E-ward drifting Iberia, slab detachment shortly before 22 Ma, and subsequent uplift and extensional tectonics. Tectonophysics, 254, 1-16. 1996b. Alpine kinematic evolution in the W Mediterranean - mosaic of local extensional tectonic regimes following upon regional W-ward directed subduction of Betic-Ligurian lithosphere. In: The Mediterranean Basins: Tertiary extension within the Alpine orogen. Workshop, Cergy-Pontoise, Paris, 34. 1997. Mantle peridotites outlining the Gibraltar Arc - centrifugal extensional allochthons derived from the earlier Alpine, westward subducted nappe pile. Tectonophysics, 281, 195-207. , ALBAT, E, HANSEN, B. T., TORRES-ROLDAN, R. L., GARCIA-CASCO, A. & MART[N-ALGARRA, A. 1989. A 21 _+2 Ma age for the termination of the ductile Alpine deformation in the internal zone of the Betic Cordilleras, S Spain. Tectonophysics, 169, 215-220. , MONI[!, R,VILLA, I. & HANSEN,B.T. 1990. Mantle diapirism in the W-Mediterranean and high rates of regional uplift, denudation and cooling. In: -
-
Proceedings of the Symposium on Diapirism, 2. Tehran, Iran Geological Survey, 403-422. & -1992a. Very high rates of cooling and uplift in the Alpine belt of the Betic Cordilleras, southern Spain. Geology, 20, 79-82. -- & -1992b. Reply to K. de Jong (1992), Comment on "Very high rates of cooling... " Geology, 20, 1053-1054.
The petroleum systems of the Southeast Basin and Gulf of Lion (France) ALAIN MASCLE 1& ROLAND
VIALLY 2
IlFP School, 228 avenue Napol(on Bonaparte, 92 852 Rueil Malmaison cedex, France 2Institut Franfais du P(trole, 1-4 avenue de Bois Pr6au, 92 500 Rueil Malmaison, France
Abstract: The Southeast Basin is the thickest onshore French sedimentary basin where up to 10 km of Mesozoic-Cenozoic sediments can locally be found. This basin is surrounded to the east and to the south by two segments of the Alpine Thrust Belt, the Western Alps and the Pyrenees-Provence respectively, and to the west by recently uplifted elements of the Palaeozoic Basement (Massif Central). The development of the basin was related to several stages of subsidence between late Carboniferous and late Cretaceous times. Partial tectonic inversion took place during two Alpine compressive events in early Tertiary and late Tertiary times. They were separated by an intervening stretching event of Oligocene age which further south led to the opening of the western Mediterranean oceanic basin in Burdigalian times and, as a result, to the formation of the Gulf of Lion passive continental margin. In Neogene times the Palaeozoic basement of the Massif Central was uplifted to approximately 2000 m as the result of an ascending athenospheric plume. A large oil seep near the city of Gabian has been exploited since the beginning of the seventeenth century. Most of the exploration undertaken from 1945 (onshore) and 1965 (offshore) to the present time has, however, been disappointing as no significant oil or gas accumulations have been discovered, despite drilling of about 150 wells. A recent re-assessment of the potential remaining prospectivity of the Southeast Basin and Gulf of Lion basin has been undertaken by IFP. This study has benefited from scientific researches developed within the Integrated Basin Studies programme. This paper focuses on the source rocks and the petroleum system aspects. The review of all potential source rocks indicates that, from a qualitative and quantitative point of view, the best source rocks are located within three specific stratigraphic intervals (Stephanian-Autunian, late Lias and late EoceneOligocene). This analytical work allowed the reconstruction of the history of the different petroleum systems from the Jurassic to the present day. Because of the severe tectonic disturbances that these areas have experienced in Tertiary times, we can conclude that the best potential for economical discoveries are within the Gulf of Lion and some sub-basins of the Southeast Basin where subsidence has been active in Neogene times, or, in other words, where the processes of hydrocarbon generation, expulsion and migration can still be active today.
Following the detailed structural analysis undertaken at different crustal levels in the Southeast Basin and Gulf of Lion (Fig. 1) within the IBS project, a re-assessment of petroleum plays in these two areas is proposed. This should convince operators that opportunities still exist for significant discoveries in these two basins despite the previous disappointing exploration campaigns. This re-assessment is based upon the identification of all potential source rocks that have b e e n deposited at different stages of the basin's development. The reconstruction of their burial and uplift history allows modelling of the timing of hydrocarbon generation and expulsion. Correlations b e t w e e n p r o d u c e d hydrocarbons and the source rocks are based upon geochemical analyses. A more complete definition of petroleum systems requires, however, additional geological data (or h y p o t h e s e s ) including those related to tectonic and thermal
histories, distribution of potential reservoirs and migration paths (Perrodon 1992; Demaison & Huizinga 1991). These data are more difficult to discuss at a regional scale as they are mostly closely related to the local geology. They will thus be shown in different geological settings that illustrate a variety of unexplored potential plays still present in the Southeast Basin and the Gulf of Lion.
Regional Framework The Palaeozoic basement was consolidated in Carboniferous time during the last stages of Variscan thrusting and magmatism (Fig. 2). This basement crops out in the Massif Central (Ledru et al. 1994), the massif des Maures (Crevola & Pupin 1994) and in the axial zone of the Pyrenees (Majeste-Menjoulas & D e b a t 1994). Basin d e v e l o p m e n t started in Stephanian and/or
MASCLE,A. & VIALLY,R. 1999. The petroleum systems of the Southeast Basin and Gulf of Lion (France). In: DURAND,B., JOLIVET,L., HORVATH,F. & St~RANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156,121-140.
122
A. M A S C L E & R. V I A L L Y
PETROLEUM SYSTEMS, SE FRANCE
123
9
r
124
A. MASCLE & R. VIALLY
Permian times as a result of the post-collisional collapse of the Variscan thrust belt (Malavielle et al. 1990; Echtler & Malavieille 1990; Costa & Rey 1995). These basins were approximatively 2-4 km thick but of limited lateral extend. However, late Permian-early Triassic erosions probably removed large volumes of sediments thus explaining why their actual distribution and initial thicknesses are poorly known. The basin subsidence and morphology were controlled by the previous Variscan structures. The older ones (Stephanian) contain the inprint of the latest Variscan compressive events (Genna & Debriette
MA -. 0
STRATIGRAPHY
GEODYNAMIC EVENTS
TECTONIC PHASES
OROGENY
Alpine phase
II~
~
L,t*
9
.~
M., :w..,...M~,. .......... p~,...
23-- ~
PALAEO-STRESS SOURCE-ROCK FIELD DISTRIBUTION |
l
i : WEST-EUROPEAN I Olig....ne . RIFTING
~[
1994) as opposed to the distensive stress regime established in Permian times (Mascle 1989; Legrand et al. 1994). The late CarboniferousAutunian deposits are contemporaneous with HT/LP metamorphism, anatexis and granite intrusions (Chenevoy et al. 1995), and to metallogenesis processes well documented in the Massif Central (Bril et al. 1994). The Triassic-Jurassic history of the basin represents the development of a passive margin related to the Tethyan rifting (Trias to Malta) and to the subsequent opening of the 'LiguroPidmontais' ocean in Callovian times. The
Oligocene exten.
I, III
+
Mhh
Paleocene
PYRENEAN OROGENY
65 Late
Pyrenean phase
4
Bay of Bi,'~'ay riftingand ,,pening
Early
~ ~
/\
9
III
9
II
9
II
"~
~
135 --
Maim
[~ ~=~ ~i ~ 205-
~1
Dogger Li'*s
i
TETHYAN AND ATLANTIC RIFT AND OPENING -
+ E - Mid Jurassic rifting /
~
I1
,r
Late
Middle Early
245
295-- ~
9
[~
360
_
,.2
LATE VARISCAN
Autunian
COLLAPSE
Stephanian
Permian exten. I, III III
Saalian phase
~
Westphalian
Na,..aa,,
5": "':;';"+ :5 :.%.+ 1000 m). Most of our knowledge of the volcanism in this sector is based on the subaerial record of activity on the islands of Pantelleria and Linosa (Fig. 5). Stratigraphic evidence, supported by K - A t dating, indicates that volcanic activity commenced in the late Miocene (c. 10 Ma) and has continued to the present day. The last observed manifestation of activity was in 1891 when a small volcanic island emerged and was subsequently eroded. The most primitive volcanic rocks on Pantelleria are transitional-mildly alkaline basalts and hawaiites. More silica-rich magma types (trachytes and peralkaline rhyolites) are widespread and are usually interpreted as the differentiation products of parental basic magmas. Linosa is predominantly composed of alkali-basalts with subordinate hawaiites. Beccaluva et al. (1981) and Calanchi et al. (1989) studied samples of dredged submarine volcanic rocks from the area, noting the occurrence of a wider range of basic magma compositions from tholeiites to nepheline basanites. There are many similarities with the volcanism of the Iblean area of Sicily to the NE, although the latter is characterised by predominantly mafic magma compositions. Aeolian Archipelago
The Aeolian Archipelago consists of seven volcanic islands plus a number of seamounts (located in the western and northeastern parts of the archipelago) which define a horse-shoe shaped structure. A wide spectrum of volcanic rocks of orogenic affinity (island arc tholeiite, calcalkaline, high-K calcalkaline and shoshonitic) have been recorded (Beccaluva et al. 1985; Ellam et al. 1989; Francalanci et aL 1993; Galassi 1995a, b). In contrast with the volcanism of central Italy, that of the Aeolian Archipelago
151
is predominantly calcalkaline to high-K calcalkaline, ranging in composition from basalt to rhyolite. In several volcanoes a transition from calcalkaline to highly potassic shoshonitic and leucite tephrite magmas has been observed. Such potassium-rich alkaline magma compositions are, however, generally subordinate in volume. Volcanism is clearly related to subduction of a slab of Ionian Sea oceanic crust towards the NW along the Calabrian arc (Spakman 1990; Giardini & Velona 1991; Francalanci & Manetti 1994). Magmatic activity initiated in the Pleistocene and becomes progressively younger moving counterclockwise along the archipelago towards the northeast (Beccaluva et al. 1985). The islands of Stromboli and Vulcano are volcanically active. The oldest calcalkaline magmatism, based on a dredge sample from the westernmost seamount, commenced at c. 1.3 Ma whilst the oldest recorded shoshonitic activity occurred at 0.85-0.64 Ma. Central-south Italy
Tertiary-Quaternary potassic magmatism occurred in central-south Italy in a postcollisional extensional tectonic setting. The magmatism of this area (Fig. 5) has traditionally been divided into two provinces: the Tuscan magmatic province in the north and the Roman magmatic province (including Campania and Latium) further south. More recently, some authors (e.g. Serri 1990; Beccaluva et al. 1991) have proposed a further subdivision of the Roman Province into a N W Campania-Latium sub-province and a Central Campania subprovince, based on significant differences in the nature of the erupted magmas. Many authors (e.g. Peccerillo et al. 1987; Conticelli & Peccerillo 1992; Serri et al. 1993) have suggested that the onset of magmatism in the area occurred at c. 14 Ma, based upon a small outcrop of ultrapotassic igneous rock at Sisco in North Corsica. This remains a subject for debate, however, since the Sisco magmatism is both temporally and spatially separated from the magmatism of central-south Italy. As reported by Serri et al. (1993) the first recorded occurrence of magmatic activity in the area (c. 7 Ma) is in t h e Tuscan archipelago in northern part of the Tyrrhenian Sea and along the Tuscan coast (Capraia, Elba, Giglio and Montecristo islands, Tolfa and S.Vincenzo, with activity between 7 and 3.5 Ma,), with subsequent migration of the volcanism towards the east and southeast (Roccastrada, Mt Amiata, Radicofani, Mt Cimini, where activity is much younger, ranging from 2.3 to 0.18 Ma).
152
M. WILSON & G. BIANCHINI
In the Roman Province the oldest magmatism occurs in the island of Ponza (4.4-3.9 Ma), whilst on the adjacent mainland the oldest volcanic products (c. 2 Ma) have been located in boreholes in the westernmost part of the Volturno plain (Beccaluva et al. 1991). In the Latium-Roccamonfina area volcanic activity ended between 0.05 and 0.1 Ma, but continues to the present day further south in Campania around the Gulf of Naples (Ischia, the Phlegrean Fields and Mt Vesuvius). The Pleistocene (0.8-0.58 Ma) volcanic activity of Mount Vulture represents the southeasternmost extent of the province. This is the only volcano located in the eastern side of the Apenninic orogenic belt (De Fino et al. 1986) and, consistent with its unique geodynamic setting, the volcanic products have distinctive geochemical characteristics. Within the Roman Province, most of the volcanic rocks belong to potassic and highly potassic alkaline and calcalkaline magma series (Appleton 1972; Serri et aL 1993). Among the most potassium-rich rocks are exotic lamproitic and kamafugitic magma types, similar to those which occur in the Ugandan segment of the East African rift (Peccerillo et al. 1987; Conticelli & Peccerillo 1992). In addition, rare carbonatite magmas, closely associated with rocks of kamafugitic affinity, have been recorded in the easternmost part of the area (Stoppa & Cundari 1996). Subalkaline intermediate-acid volcanic and plutonic rocks, mainly of high-K calcalkaline affinity, occur in Tuscany and in the Tuscan archipelago. The more acid magmas are usually interpreted as hybrids between crustal and mantle derived melts, although a pure crustal, anatectic origin has been proposed for rhyolites from S. Vincenzo, Roccastrada, and Cerveteri (Pinarelli 1987; Ferrara et al. 1989) and for some acid plutonic rocks exposed in the Vercelli seamount and on the islands of Montecristo, Elba and Giglio. The geodynamic setting of the volcanism of south-central Italy has been hotly debated in the past. Its association with extensional tectonic structures (grabens) has led some authors to assume that the magmas are of anorogenic affinity. The recognition of typical calcalkaline magmas associated with the more prevalent Krich alkaline magmas (Beccaluva et al. 1991), however, clearly demonstrates an orogenic affinity, related to subduction. Serri et al. (1993) have recognized a temporal evolution within the province from regional compression to extension (rifting) and the formation of sedimentary basins, followed by uplift and the onset of vol-
canism. Eastward migration of the focus of the extensional tectonic activity and of the location of the magmatism, while compression was still occurring in more external portions of the orogenic belt, suggests that the geodynamic setting of the magmatism is most probably postcollisional related to subducted slab roll-back, possibly associated with tearing of the slab and the formation of a slab window. N o r t h I t a l y a n d s u r r o u n d i n g areas: I n s u b r i c - P e r i a d r i a t i c line a n d the V e n e t o v o l c a n i c district
There is no direct evidence for Eoalpine (prelate Eocene) orogenic magmatic activity associated with Neotethys subduction apart from the presence of andesitic clasts in the Tavayanne formation of the Haute Savoie, France (Delaloye & Sawatzky 1975). Magmatism within the Alpine orogenic belt occurred between 42 and 25 Ma, with a climax between 33 and 29 Ma, postdating the main compressional phases. This included the emplacement of granitoid intrusions (e.g. Adamello) as well as of extensive basic-acid dyke swarms cutting Austroalpine, Southalpine and, rarely, Penninic units. Most of the dykes are located along the Insubric-Periadriatic tectonic lineament (Fig. 5). Geochemical studies of the basic dykes indicate a spectrum of magma compositions ranging from calcalkaline and high-K calcalkaline to shoshonitic and ultrapotassic types (Beccaluva et al. 1979; 1983; Venturelli et al. 1984b; von Blanckenburg & Davies 1995). The most primitive magma compositions appear to be mantle-derived melts, minimally affected by crustal contamination or high-level differentiation processes. Their geochemical characteristics indicate a subductionrelated affinity (von Blanckenburg & Davies 1995) and it has been suggested that they were derived from a lithospheric mantle source intensely metasomatized by subduction zone fluids/melts. Partial melting of the lithospheric mantle may have been triggered by slab breakoff. It is possible that the widespread occurrence of shoshonitic/ultrapotassic magmas in this region could reflect the subduction of continental crustal material. Beccaluva et al. (1983) recognized a spatial zonation in the chemical composition of the dykes with an increase in the K20 content from the SE (where the magmas are typically of calcalkaline affinity) towards the NW (shoshonitic/ultrapotassic magma types). On this basis they proposed the existence of an Oligocene subduction zone dipping towards the NW, in the opposite sense to the generally
TERTIARY-QUATERNARY MAGMATISM accepted southward directed Eoalpine subduction of the European plate. If correct, the OligoMiocene magmatism of this sector could represent the westward continuation of the Sardinian Oligo-Miocene orogenic (subductionrelated) magmatic arc. Late Eocene-early Miocene magmatism of calcalkaline affinity also occurs in Croatia and Slovenia in the easternmost part of the Periadriatic zone close to the southwestern part of the Pannonian Basin (Pamic 1993; Altherr et al. 1995). Miocene-Pliocene basalts, andesites and pyroclastics have been reported from the Adriatic coast of Croatia (Marjanac pers. comm. 1996), although these have not been characterized geochemically and therefore their tectonic affinity is unknown. Localized occurrences of Palaeocene-Oligocene extension-related (anorogenic) alkali basalts, basanites and subordinate transitional basalts occur in the Veneto (Lessini Mountains, Southern Trentino area, Euganei Hills, Berici Hills, Asiago Plateau and Marostica Hills) region of northern Italy (Siena & Coltorti 1989; De Vecchi & Sedea 1995; Milani 1996). More differentiated magmas occur only in the Euganei Hills. The age of the magmatism is based primarily upon stratigraphical constraints and may extend into the Miocene.
Aegean area Fytikas et al. (1984) recognized two main phases of volcanic activity in the Aegean area (Fig. 7).
153
The first developed in the northernmost sector from Oligocene to mid-Miocene times, mainly consisting of intermediate magmas of calcalkaline to shoshonitic affinity. There appears to have been a continuous migration of the focus of volcanic activity towards the south, accompanied by a variation in the chemical composition; the younger volcanic products becoming more K-rich. An exception to this trend, however, are the islands of Skiros and Evia (15-13 Ma) where more normal calcalkaline magmas were erupted (Pe-Piper & Piper 1994). After a hiatus in activity during the midMiocene to Pliocene a second volcanic cycle initiated in the early Pliocene, along a restricted zone in the southern part of the Aegean erupting a typical calcalkaline association with both basic and more evolved magmas (Briqueu et al. 1986; Pe-Piper & Piper 1989; Robert et al. 1992; Pe-Piper 1994). This magmatic arc is considered to be the expression of active subduction towards the north of the oceanic crust of the Herodotus abyssal plain (a remnant of the Mesozoic Tethys; Robertson & Grasso 1995).
Macedonia Late Miocene to early Pleistocene volcanism occurred in Macedonia characterized by high-K calcalkaline to shoshonitic products, including both basic and more evolved magmas (Kolios et al. 1980; Karamata et al. 1994).
Oligocene Miocene Pliocene Quaternary Fig. 7. Distribution of TertiaryQuaternary magmatism of different ages in the Aegean. Data sources are given in the text. Abbreviations: AL, Alexandroupolis; AT, Athens; AF, Afyon; BO, Bodrum; CH, Chios; CR, Crete; DO, Doirani; ED, Edessa; EV, Evia; EZ, Ezine; IP, Isparta; IS, Istanbul; IZ, Izmir; KO, Kos; KU, Knla; LE, Lesbos; LI, Limnos; PA, Patmos; SA, Samothrathrakj; SM, Samos; ST, Santorini; TH, Thessaloniki; VO, Volos.
154
M. WILSON & G. BIANCHINI
Western Turkey and the Levant The Neotectonic evolution of Turkey (Fig. 7) has been dominated by the collision of the African and Arabian plates with the Eurasian plate along the Hellenic arc to the west and the Bitlis-Zagros suture to the east (Seng6r & Yilmaz 1981; Seng6r et at. 1985; Dewey et al. 1986; Yilmaz 1993; Oral et al. 1995). This resulted in extensive mafic to felsic volcanism in eastern, central and western Anatolia, as well as along the North Anatolian and East Anatolian strike-slip fault zones (Yilmaz 1990; Pearce et al. 1990; Notsu et al. 1995; Wilson et al. 1997). The Western Anatolian volcanic province (Fig. 7) is located at the eastern end of the Aegean volcanic arc, which results from the northward subduction of the African plate beneath the Aegean. Calcalkaline volcanic activity commenced in the late Oligocene-early Miocene followed by alkali basaltic volcanism from late Miocene to Recent times. This change in the style of the volcanism has been attributed by some authors to a change in the regional stress field from N-S compression to N-S extension (Yilmaz 1990; Gt~leq 1991). Seyitoglu & Scott (1992), however, consider that the transition to N-S extensional tectonics actually commenced much earlier in the latest Oligocene-early Miocene. Seyitoglu et al. (1997) have demonstrated that within the youngest volcanic sequence there is a change from potassic magmatism in the Miocene to more sodic (anorogenic) alkaline magmatism in the Quaternary. The geochemical characteristics of the potassic magmas are inferred to reflect the presence of an inherited subduction-modified component in their mantle source. Widespread, extension-related alkali basaltic volcanism of Late Miocene-Pliocene to Recent age also occurs throughout the Middle East in Israel, Syria and eastward within the Zagros collision zone (Giannerini et al. 1988; Garfunkel 1989; Sawaf et al. 1993; Seng6r et al. 1993; Alavi 1994).
Geodynamic setting of magmatism in the central Mediterranean region It is generally accepted that the main orogenic phase of the Alpine chain was associated with southward subduction of the European lithosphere beneath the African plate. As noted previously, however, there is little evidence for Eoalpine subduction-related magmatic activity. Eocene-Oligocene igneous activity associated with the Insubric-Peradriatic lineament has been interpreted by some authors to reflect the
post-collisional detachment of the subducted slab (e.g. von Blanckenburg & Davies 1995). If correct, the geochemical characteristics of the magmas suggest that there may be a correlation between the generation of highly potassic magmas and slab detachment which deserves further investigation. It seems reasonable to assume that after the main Eoalpine compressional phase several oceanic strands (former parts of the Mesozoic Tethys) remained in the Mediterranean area which were subsequently consumed by subduction. Focussing on the circum-Tyrrhenian region (Fig. 5), it is possible to recognize two distinct subduction-related magmatic phases (Beccaluva et aL 1987, 1994; Galassi 1995b) which can be attributed to the subduction of these remnants of Tethyan oceanic lithosphere. Oligo-Miocene
cycle
Two distinct subduction systems seem to have been operative at this time. In the western Mediterranean a subduction system directed towards the northwest (Fig. 8) is required to explain the oldest orogenic volcanism in Provence (34-20 Ma), Sardinia (32-13 Ma), North Africa (Kabylies; c. 18-20 Ma) and southern Spain (c. 18 Ma). Calcalkaline igneous rocks dated at c. 20 Ma have also been recorded from the Alboran Sea and Valencia Trough (DSDP holes 122, 123; K - A r ages confirmed by fission track dating). The easternmost propagation of this subduction system could be represented by the Oligocene magmatism along the Periadriatic-Insubric lineament in the Alpine domain, providing an alternative hypothesis to the slabbreakoff model of von Blanckenburg & Davies (1995). During this phase the Sardinia-Corsica microcontinent and the Kabilies blocks would have been joined to the European continent as the Liguro-Provenqal-Balearic Basin had not yet opened. A second Oligo-Miocene subduction system must have dipped towards the north in the eastern part of the Mediterranean in the Aegean area (Fytikas et al. 1994), progressively consuming the northernmost oceanic strand present in this sector. The southernmost oceanic strand present in the eastern Mediterranean area was not involved during this stage (Robertson & Grasso 1995). Lonergan & White (1997) have recently suggested that the collision of the Kabylies block with the North African margin by about 18 Ma effectively divided the previously continuous Oligo-Miocene subduction zone into two segments (Fig. 8). They attribute the Miocene calcalkaline magmatism of the Western
TERTIARY-QUATERNARY MAGMATISM
155
Fig. 8. Tectonic reconstructions illustrating the Neogene evolution of the Western Mediterranean at c. 30 and 18 Ma. Modified after Lonergan & White (1997). By 18 Ma Sardinia and Corsica had rotated counterclockwise and the Balearic Islands clockwise as a consequence of the opening of the the Liguro-Provenqal and Valencia basins. The Kabylies block had collided with the North African margin splitting the formerly continuous subduction system into two branches.
Mediterranean to a short, arcuate, eastward dipping subduction zone. This view is somewhat controversial as most authors consider that the subduction zone polarity was towards the north at this stage (e.g. Doglioni et aL 1997).
N e o g e n e - Q u a t e r n a r y cycle (15-0 Ma) By 16-13 Ma orogenic magmatic activity along the two Oligo-Miocene subduction systems seems to have ceased. It is possible that in some sectors (e.g. North A p p e n n i n e area) no more oceanic crust was left to 'feed' the arc magmatism, with the result that these subduction zones became partially locked by continental collision (e.g. Mantovani et al. 1997). To accommodate the continuing convergence between Europe and Africa subduction is inferred to h a v e migrated, by a process of slab roll-back, towards the southeast in the Western Mediterranean and to the south in the Eastern Mediterranean. In these areas subductable oceanic lithosphere was still present (Ionian Sea lithosphere in the west and the southernmost oceanic strand of
Tethys in the Eastern Mediterranean). Beccaluva et al. (1987) proposed that there was a single subduction system in the Central Mediterranean which migrated progressively, if discontinuously, from its Oligocene configuration to the present one. In the Central Mediterranean region a new orogenic magmatic cycle began in the uppermost Tortonian (c. 7 Ma). The products of this cycle are widespread in Central Italy (Tuscany, Latium, C a m p a n i a and Umbria) and in the Aeolian archipelago. Active subduction is currently only occurring along the Calabrian arc, where subduction of Ionian Sea oceanic lithosphere is associated with the Pleistocene-Recent orogenic volcanism of the Aeolian Archipelago, and further east in the south Aegean, where subduction of the oceanic lithosphere of the Herodotus abyssal plain has been associated with orogenic volcanism from Pliocene to historical times. The relative positions of the subduction systems of this younger orogenic phase are clearly revealed by seismic tomography (e.g. Spakman et aL 1993).
156
M. WILSON & G. BIANCHINI
Relation between subduction and extensional and cornpressional tectonics A combination of compressional tectonics and simultaneous extension (in the more internal areas) seems to be a common feature of the subduction systems of both the Oligo-Miocene and Neogene-Quaternary orogenic phases within the Mediterranean region. For example, extension occurs in the internal zone of the South Aegean arc while subduction and a compressive regime (associated with the formation of the socalled Mediterranean ridge) is active in the external part of the arc; extension occurs in the internal zone of the Apenninic belt while the more external zones further east are under compression; Oligo-Miocene extension occurred in South France and Sardinia simultaneous with subduction. SengOr (1993) noted that such behaviour is not unique to the Mediterranean area, reporting similar cases of coeval and codirectional shortening and extension in active orogenic zones (e.g. Western Pacific island arcs). Lonergan & White (1997) argue that this is an inevitable consequence of slab roll-back.
Sea-floor spreading Sea-floor spreading in the Balearic Basin (c. 20-15 Ma) and the Tyrrhenian Sea (1, whereas the anorogenic magmas have a KzO/Na20 ratio 6-7 wt% MgO should be considered.
M. W I L S O N & G. B I A N C H I N I
158 8
,
i
9
i
zx
i
9
orogenic
t'q
9
A
t"q
anorogenic
t'q
r
A~~
A zxzx~
oooo ,
,
~,~rW'*
40
I>
,
I
50
0
70
60
80
40
50
SiO 2 9
Aeolian Archipelago
[]
Ernici (Latium)
O
Roccamonfina
A
Vulsini(Latium)
@
Sardinia 32-13 Ma
6O
70
SiO 2 I~
Pantelleria
@
Etna
@
Iblean tholeiites
V
Iblean transitional
[]
Iblean alkaline
4,
Iblean ankaramites
Sardinia Plio-Pleistocene subalkaline series
t>
Sardinia Plio-Pleistocene alkali basalts
Fig. 9. V a r i a t i o n of K 2 0 / N a 2 0 (weight r a t i o ) versus wt % SiO2 for T e r t i a r y - Q u a t e r n a r y volcanic suites f r o m the C e n t r a l M e d i t e r r a n e a n region. D a t a sources: see c a p t i o n to Fi~. 12.
1000
i
i
I
I
I
I
I
I
I
I
I
I
I
I
000
I
i
i
i
i
i
l
i
i
i
i
i
i
t
i
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Ernici
~ 100
lO0
~ lO
lO
?.
] m
1
i
i
t
RbBaTh
i
i
i
i
i
i
t
K N b L a C e Sr N d P
i000 ...............
i
t
t
[
I
t
R b B a T h K N b L a C e Sr N d P Z r S m E u T i Y b
Z r S m E u Ti Y b
OROGENIC BASIC MAGMAS Vulsini
100
10001................ A e o l i a n
~
100
" , " ~
10
lfl I I I I i I f I R b B a T h K N b L a C e Sr N d P Z r S m E u Ti Y b
10
Islands
,~
..... " O . . . O
\'b
""
"...... ":'-"
1 . . . . . i i , i ~ , , i i i I RbBaTh K NbLaCe SrNd P ZrSmEu TiYb
Fig. 10. P r i m i t i v e m a n t l e - n o r m a l i s e d t r a c e e l e m e n t v a r i a t i o n d i a g r a m s for p r i m i t i v e o r o g e n i c basic volcanic r o c k s f r o m the C e n t r a l M e d i t e r r a n e a n region. D a t a sources: see c a p t i o n to Fig. 12.
TERTIARY-QUATERNARY MAGMATISM 1000
1000 . . . . . . . . . . . . . . . 9-~
Iblean lOO
Plateau
I
.,..~
~ 100
subalkaline
lo
lO E
~ 9 a.
1
RbBaTh KNbLaCeSrNd P ZrSmEuTiYb
1000
I
!{' IbiS'anPlateau~ alkaline I
I
I
I
I
I
I
I
I
I
1
I
@
100
~100
r
I
I
I
I
10
~
........9 ........ P a n t e l l e r i a
1RbBaTh K NbLaCe SrNd P ZrSmEuTi Yb I
I
I
I
I
I
I
I
I
I
I
I
I
9
I
St. Helena
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
---o .... S a r d i n i a . . . . . . 9 ........
9 ..,---~O9
Z.
I
I
ANOROGENICBASIC MAGMAS ~1000
o
-~
(
RbBaTh K NbLaCe SrNd P ZrSmEuTiYb
---~--- E t n a
.,..,
159
"-%
. . . . . . .....
..
10
.,.~ .,--4
1
.
.
.
.
7"7
RbBaTh KNbLaCe SrNd P ZrSmEu TiYb
Fig. 11. Primitive mantle-normalized trace element variation diagrams for anorogenic basic volcanic rocks from the Central Mediterranean region. Data sources: see caption to Fig. 12.
Even for the more primitive magmas, interpretation of the trace element signature can be ambiguous. For example, negative Nb anomalies can indicate either shallow-level crustal contamination or introduction of subducted sediment into the mantle source of the magmas. Plots of Th/Yb versus Ta/Yb (Fig. 12) and Th/Zr versus Nb/Zr (Fig. 13) for primitive mafic magma compositions, clearly discriminate basalts of orogenic from those of anorogenic affinity. The mafic samples (tephrites-phonotephrites) from Mt Vulture plot in an intermediate position between the orogenic and anorogenic fields in both diagrams, consistent with the unique tectonic setting of this volcano.
S r - N d - P b isotopes Diagrams showing the variation of 143Nd/144Nd versus STSr/S6Sr (Fig. 14) and 2~176 versus 2~176 (Fig. 15) clearly distinguish the anorogenic from the subduction-related magmatic suites. These isotope diagrams include all the available data in the literature for both primitive mafic and more differentiated magmatic rocks. Data sources are given in the figure captions. Acid volcanics from the Tuscan province are most likely to have highly radiogenic Sr
isotope compositions as a consequence of high degrees of crustal contamination. The radiogenic Sr isotope compositions of the orogenic volcanic suites from the Aeolian Archipelago, Sardinia, Provence, Campania and Latium most probably reflects a combination of mantle source enrichment by aqueous fluids or silicate melts released from the subducted oceanic crust (which m a y inherit a continental crustal isotopic signature as a consequence of subduction of continentally derived sediments), combined with shallow-level crustal contamination. Anorogenic alkaline magmas from Sicily (Etna and Iblean area) and the Sicily Channel (Pantelleria) have Nd-Sr isotope compositions identical to those of primitive mafic alkaline magmas from central and western Europe (e.g Massif Central, Wilson & Downes 1991; Wilson et al. 1995), inferred to originate by partial melting of a distinct mantle source component within the European upper mantle, the European Asthenospheric Reservoir or E A R , which may be plume-related. The subalkaline (tholeiitic) basalts from the Iblean area of Sicily have N d - S r - P b isotope compositions transitional between those of the E A R and the depleted mantle source of M O R B (DM). The MORBlike basalts from the Tyrrhenian Sea have higher
160
M. WILSON & G. BIANCHINI
Fig. 12. Variation of ThfYb versus Ta/Yb for primitive mafic igneous rocks from the Central Mediterranean region. Data sources: Iblean area, Sicily, Bianchini (1995); Pantelleria, Esperan~a & Crisci (1995); Sardinia, Plio-Pleistocene anorogenic lavas, Rutter (1987; alkaline) and Dostal et al. (1982b; subalkaline); Etna, 1991-1993 eruption Armienti et al. (1994); Treuil & Joron (1994); Mount Vulture, De Fino et al. (1996); Campania, Beccaluva et al. (1991); Aeolian Archipelago, Ellam et al. (1989); Galassi (1995a, b); Vulsini (Latium), Coltorti et al. (1991); Rogers et al. (1985); Ernici (Latium), Civetta et al. (1981): South Tuscany-North Latium high-Mg# samples, Conticelli & Peccerillo (1992).
Fig. 13. Variation of Th/Zr versus Nb/Zr for primitive mafic igneous rocks from the Central Mediterranean region. Data sources: see caption to Fig. 12.
TERTIARY-QUATERNARY MAGMATISM
161
Fig. 14. Variation of 143Nd/144Ndversus 8VSr/86Sr(age corrected) for orogenic and anorogenic magma series from the Central Mediterranean region. Data sources: Iblean area, Beccaluva et al. (1998); Tonarini et al. (1996); Bianchini (1995). Pantelleria, Esperan~a & Crisci (1995). Etna, D'Orazio (1994). Sardinia, Rutter (1987; alkaline); Cioni et al. (1982; subalkaline). Aeolian Archipelago, Ellam et aL (1989); Francalanci et aL (1993); Galassi (1995a, b). Sardinia, Oligocene-Miocene orogenic lavas, Galassi (1995a, b). Provence, Galassi (1995a, b). Mount Vulture, Vollmer, (1976); Hawkesworth & Vollmer (1979). Campania, Vollmer (1976); Hawkesworth & Vollmer (1979); Civetta et al. (1991a, b); Galassi (1995a, b). Latium, Vollmer (1976); Hawkesworth & Vollmer (1979); Rogers et al. (1985). Tuscany, Vollmer (1976); Hawkesworth & Vollmer (1979); Peccerillo et al. (1987).
87Sr/86Sr and 2~176 and lower 143Nd/144Nd ratios than typical depleted mantle, consistent with the involvement of a subduction-related fluid flux in their petrogenesis. The subalkaline Plio-Pleistocene basalts of Mt Arci in Sardinia have Sr isotope compositions intermediate between those of the Plio-Pleistocene alkali basalts and the subduction-related OligoMiocene andesites, consistent with the involvement of a subduction-modified mantle source component in their petrogenesis.
Discussion: Magma source regions and mantle dynamics The generation of Tertiary-Quaternary basaltic magmas within the Mediterranean domain was most probably triggered by adiabatic decompression partial melting of the asthenopheric upper mantle, the solidus of which was locally lowered by the infiltration of slab-derived fluids
above adjacent subduction zones. Throughout the region, both orogenic and anorogenic magmas appear to share a common asthenospheric mantle source component (EAR) which could be plume-related. Locally (e.g. central Spain, French Massif Central), asthenospherederived anorogenic basaltic magmas appear to mix with potassium-rich partial melts of enriched domains within the mantle lithosphere, which modifies their trace element and Sr-Nd-Pb isotope signature (Wilson &Downes 1991; Cebrigl & L6pez-Ruiz 1995). The widespread pollution of the shallow upper mantle by a geochemically distinct mantle plume component (EAR) need not necessarily be a Tertiary phenomenon. Available data (Patterson 1996) suggest that it may have occurred during the Mesozoic. The convective instabilities (mantle diapirs) which appear to trigger the magmatism in many areas have been imaged by seismic tomography (e.g. Granet e t al. 1995). Their scale-length
162
M. WILSON & G. BIANCHINI
Fig. 15. Variation of 2~176
versus 2~176 for orogenic and anorogenic magma series from the Central Mediterranean region. Data sources: see captions to Figs 12 & 14.
suggest that they most probably originate from t h e r m a l b o u n d a r y layers within the upper mantle (400 or 650 km discontinuities). Convective de-stabilization of the upper mantle may have been initiated by the Alpine collision and the consequent global reorganization of plate motions. The distinctive geochemical and isotopic characteristics of the orogenic magmas can be related to the 'pollution' of the shallow asthenospheric mantle by subduction zone fluids/melts, which in some regions (e.g. T u s c a n y - C a m p a n i a - L a t i u m ) carry a particularly strong crustal signature. This may locally reflect subduction of continental lithosphere during the collision of continental micro-plates. The distribution of highly potassic magmas within the Mediterranean region may provide an important indicator of those locations at which collision of continental micro-plates triggered slab break-off. We would like to thank L. Beccaluva for his encouragement and support for this project. Constructive
comments from B. Bonin, L. Jolivet and an anonymous reviewer helped to clarify our ideas.
References ALAVI,M. 1994. Tectonics of the Zagros orogenic belt of Iran: new data and interpretations. Tectonophysics, 229, 211-238. ALTHERR,R., LUGOViC,B., MEYER, H.-P. & MAYER,V. 1995. Early Miocene post-collisional calc-alkaline magmatism along the easternmost segment of the Periadriatic fault system (Slovenia and Croatia). Mineralogy and Petrology, 54, 225-247. APARICO,A., MITJAVILA,J. M., ARANA,g. • gILA, I. M. 1991. La edad del volcanism de las islas Columbrete Grande Alboran (Mediterraneo Occidentale). Boletim Geologica y Minero, 102, 562-570. APPLETON, J. D. 1972. Petrogenesis of potassium rich lavas from the Roccamonfina volcano, Roman Region, Italy. Journal of Petrology, 13, 425-456. ARMIENTI, P., CLOCCHIATFI,R., D'ORAZIO, M., INNOCENTI, F., PETRINI, R., POMPILIO,M., TONARINI,S. & Vn~LAm, L. 1994. The long standing 1991-1993 Mount Etna eruption: petrography and geochemistry of lavas. Acta Vulcanologica, 4, 15-28.
TERTIARY-QUATERNARY MAGMATISM BECCALUVA,L., BIGIOGGERO,B., CHIESA,S., COLOMBO, A., FANTI,G., GATTO, G. O., GREGNANIN,A., MONTRASIO,A., PICCmlLLO, E. M. & TUNESI,A. 1983. Post collisional orogenic dyke magmatism in the Alps. Memorie della Societd Geologica Italiana, 26, 341-359. , BONATTI, E., DuPuY, C., FERRARA, G., INNOCENTI, E, LUCCHINI,E, MACERA, P., PETRINI, R., Rossl, R L., SERRI, G., SEYLER, M. & SIENA, E 1990. Geochemistry and mineralogy of volcanic rocks from ODP sites 650, 651,655, and 654 in the Tyrrhenian sea. Proceedings of the ODP: Scientific Results, 107, 49-74. , BROTZU, P., MACCIOTTA, G., MORBIDELLI, L., SERRI, G. & TRAVERSA, G. 1987. Cainozoic tectonomagmatic evolution and inferred mantle in the Sardo-Tyrrhenian area. The lithosphere in Italy. Advances in Earth Science Research. Accadamia Nazionale Lincei, 80, 229-248. , COLANTONI,P., DI GIROLAMO, P. • SAVELLI.C. 1981. Upper-Miocene submarine volcanism in the Strait of Sicily (Banco senza nome). Bulletin Volcanologique, 44, 573-581. --, COLTORTI, M., GALASSI, B., MACCIOTTA, G. & SIENA, E 1994. The Cainozoic calcalkaline magmatism of the western Mediterranean and its geodynamic significance. Bollettino Geofisica Teoretica ed Applicata, 36, 294-308. - - , DERIU, M., MACCIOTrA, G., SAVELLI. C., VENTURELLI G. 1977. Geochronology and magmatic character of the Pliocene-Pleistocene volcanism in Sardinia (Italy). Bulletin Volcanologique, 40-3, 1-16. - - , DI GIROLAlVIO,P. & SERRI, G. 1991. Petrogenesis and tectonic settings of the Roman volcanic province. Lithos, 26, 191-221. --, GABBIANELLI,G., LUCCHINI, E, ROSSI, P. L. & SAVELL1,C. 1985. Petrology and K/Ar ages of volcanics dredged from the Eolian seamounts: implication for geodynamic evolution of the southern Tyrrhenian basin. Earth and Planetary Science Letters, 74, 187-208. - - , GATTO,G.D., GREGNANIN,A., PICCIRILLO,E.M. & SCOLAm,A. 1979. Geochemistry and petrology of dyke magmatism in the Alto Adige (Eastern Alps) and its geodynamic implications. Neues Jahrbuch fiir Geologic und Palaiiontology Monatshefte, 6, 321-339. --, SIENA, E, COLTORTI, M., Dt GRANDE, A., LO GIUDICE, A., MACCIOTrA, G., TASSINARI, R. & VACCARO, C. 1998. Nephelinitic to tholeiitic magma generation in a transtensional tectonic setting: an integrated petrogenetic model for the Iblean volcanism, Sicily. Journal of Petrology, 39, 1547-1577. BELANTEUR, 0, BELLON, H., MAURY, R. C., OUABADI, A., COUTELLE,A., SEMROUD,B., MEGARTSI,M. & FOURCADE, S. 1995. Le magmatisme miocbne de l'Est Alg6rois gdologie, g6ochimie et g6ochronologic 40K-40Ar. Comptes Rendus de l'Acaddmie des Sciences, Paris, 321, serie II a, 489-496. BELLON, H. 1981. Chronologic radiom6trique (K-Ar) des manifestations magmatiques autour de la
163
M6diterran6e occidentale entre 33-1 MA. F.C. In: WEZEL, E C. (ed.) Sedimentary basin of Mediterranean margins, CNR Italian Project of Oceanography, Tectonoprint Bologna, 341-360. BERGMAN, S. 1987. Lamproites and other potassiumrich igneous rocks: a review of their occurrence, mineralogy and geochemistry. In: FITTON,J. G. & UPTON, B. G. J. (eds) Alkaline Igneous Rocks. Geological Society, London, Special Publications, 30, 103-190. BERRAHMA, M., DELALOYE, M., FAUREMURET,A. & RACHDI, H. 1993. Preliminary and geochronological data on alkaline volcanism of Jbel-Saghro, Anti-Atlas (Morocco). Journal of African Earth Sciences, 17, 333-341. B1ANCHINI, G. 1995. Petrogenesi delle vulcaniti sottomarine dell'area Iblea. PhD thesis, Univ. Ferrara, Italy. BRIQUEU, L., JAVOY,M. LANCELOT,J. R. & TATSUMOTO, M. 1986. Isotope geochemistry, of recent magmatism in the Aegean arc: Sr, Nd, Hf, and O isotopic ratios in the lavas of Milos and Santorini - geodynamic implications. Earth and Planetary Science Letters, 80, 41-54. CAHEN, L., SNELLING,N. J., DELHAL, J. & VAIL, J. R. 1984. The Geochronology and Evolution of Africa. Clarendon Press. CALANCHI,N., COLANTONI,P., RossI, P .L., SAITTA,M. & SERRI, G. 1989. The Strait of Sicily continental rift system: physiography and petrochemistry of the submarine volcanic centres. Marine Geology, 87, 55-83. CEBmA, J. M. & LOeEz-Ruiz, J. 1995. Alkali basalts and leucitites in an extensional intracontinental plate setting: The Late Cenozoic Calatrava Volcanic Province (Central Spain). Lithos, 35, 27-46. C1ONI, R., CLOCCHIATTI, R., DI PAOLA, G. M., SANTACROCE,R. & TONARINI,M. 1982. Miocene calcalkaline heritage in the Pliocene volcanism of Monte Arci (Sardinia Italy). Journal of Volcanology and Geothermal Research, 14, 133-167. CIVETTA,L., CARLUCCIO,E., INNOCENTI,E, SBRANA,A. & TADDEUCCI, G. 1991a. Magma chamber evolution under the Phlegrean Fields during the last 10 Ka: trace element and isotope data. European Journal of Mineralogy, 3, 415-428. --, GALATI, R. d~c SANTACROCE,R. 1991b. Magma 9 mixing and convective compositional layering within the Vesuvius magma chamber. Bulletin Volcanologique, 53, 287-300. --, INNOCENTI, E, MANETrI, P., PECCERILLO,A. & POLI, G. 1981. Geochemical characteristics of potassic volcanics from Mts. Ernici (Southern Latium-Italy). Contribributions to Mineralogy and Petrology, 78, 37-47. CLOCCHIATI'I, R., JORON, J. ~; TREUIL, M. 1988. The role of selective alkali contamination in the evolution of recent historic lavas of Mount Etna. Journal of Volcanology and Geothermal Research, 34, 241-249. CONEY, C. R. 1980. Plate tectonic model for the OligoMiocene evolution of the Western Mediterranean. Tectonophysics, 68, 283-311. COLTORTI,M., DI BATTISTINI,G., NAPPI, G., RENZULLI,
164
M. WILSON & G. BIANCHINI
A. & ZEDA, O. 1991. Structural setting and magmatic evolution of Montefiascone volcanic complex, Vulsini district, Central Italy. Journal of Volcanology of Geothermal Research, 46, 99-124. COMAS, M. C., GARC[A-DUElqAS,V. JURADO,M. J. 1992. Neogene tectonic evolution of the Alboran Sea from MCS data. Geomarine Letters, 12, 157-164. CONDOMINES, M., TANGUY, J. C., KIEFFER, G. & ALLEGRE, C .J. 1982. Magmatic evolution of a volcano studied by 23~ disequilibrium and trace elements systematics: the Etna case. Geochimica et Cosmochimica Acta, 46,1397-1416. CONTICELLI, S. & PECCERILLO,A. 1992. Petrology and geochemistry of potassic and ultrapotassic volcanism in central Italy: petrogenesis and inferences on the evolution of the mantle sources. Lithos, 28, 221-240. D'ORAZIO, M. 1994. Natura ed evoluzione delle vulcaniti dell'Etna e loro relazioni con il magmatismo Ibleo. Doctoral thesis, Universith di Pisa, Italy. DE FINO, M., LA VOLPE, L., PECCERILLO,A., PICCARRETTA, G. &POLI, G. 1986. Petrogenesis of Monte Vulture volcano (Italy): inferences from mineral chemistry, major and trace element data. Contributions to Mineralogy and Petrology, 92, 135-145. DE VECCHI, G. & SEDEA, R. 1995. The Paleogene basalts of the Veneto Region. Memorie della Societd Geologica Italiana, 47, 253-274. DELALOYE, M. & SAWATZKY, G. 1975. G6ochronom6trie des 616ments volcaniques du flysch Helv6tique du synclinal de Th6ne (Haute Savoie, France). Archives du SociOt~ GenOve, 28, 95-99. DEVERCHERE, J., BELLON, H., BESLIER, M-O. ET AL. 1996. Geodynamics of the Ligurian Basins and margins. In: Workshop Abstracts: The Mediterranean Basins: Tertiary extension within the Alpine Orogen. Cergy-Pontoise, Paris, France, December 1996. DEWEY, J. E 1988. Extensional collapse of orogens. Tectonics, 7, 1123-1139. - - , HEMPTON, M. R., KIDD, W. S. E, SAROGLU, E & SENGOR, A. M. C. 1986. Shortenening of continental lithosphere: the neotectonics of Eastern Anatolia-a young collision zone. In: COWARD,M. E & RIES, A. C. (eds) Collision Tectonics. Geological Society, London, Special Publications, 19, 3-36. DOBLAS, M. & OYARZUN, R. 1989. Neogene extensional collapse in the Western Mediterranean (Betic-Rif orogenic belt): Implications for the generation of the Gibraltar arc and magmatic activity. Geology, 17, 430-433. DOCHERTY, C. & BANDA, E. 1995. Evidence for the eastward migration of the Alboran Sea based on regional subsidence analysis: a case for basin formation by delamination of the subcrustal lithosphere? Tectonics, 14, 804-818. DOGLIONI, C., GUEGUEN, E, SABAT,F & FERNANDEZ, M. 1997. The Western Mediterranean extensional basins and the Alpine orogen. Terra Nova, 9, 109-112. DOSTAL, J., COULON, C. & DUPUY, C, 1982a. Cainozoic andesitic rocks of Sardinia (Italy). In: THORPE R. S. (ed.) Andesites, Wiley and Sons Ltd, 353-370.
, DuPuY, C., & VENTURELLI,G. 1982b. Geochemistry of volcanic rocks from the Monte Arci (West Sardinia, Italy). Chemical Geology, 35, 247-264. DOWNES, H. 1987. Tertiary and Quaternary volcanism in the Massif Central, France. In: F~TTON,J. G. & UPTON, B. G. J. (eds) Alkaline Igneous Rocks. Geological Society, London, Special Publications, 30, 517-530. ELLAM, R. M., HAWKESWORTH, C. J, MENZIES, M. A. & ROGERS N. W. 1989. The volcanism of Southern Italy: Role of subduction and the relationship between potassic and sodic alkaline magmatism. Journal of Geophysical Research, 9 4 , 4589-4601. ELLOUZ, N. 1984. Etude de la subsidence de la Tunisie atlasique, orientale et de lamer POlagienne. Th6se 3e cycle, Paris 6. ESPERANfA, S. & CRISCI, G. M. 1995. The Island of Pantelleria: a case for development of DMMHIMU isotopic composition in a long lived extensional setting. Earth and Planetary Science Letters, 136, 167-182. FACCENNA,C., MATTEI, M., FUNICIELLO,R. & JOLIVET, L. 1997. Styles of back-arc extension in the Central Mediterranean. Terra Nova, 9, 126-130. FERRARA, G., PETRINI, R., SERRI, G. & TONARINI, S. 1989. Petrology and isotope-geochemistry of San Vincenzo rhyolites (Tuscany, Italy). Bulletin Volcanologique, 51, 379-388. FERAUD G. 1990. 39Ar/4~ analysis on basaltic lavas series of Vavilov basin, Tyrrhenian sea (Ocean Drilling Program, Leg 107, Holes 655B and 651A). In: Proceedings of the O D P Sci. Results, 107, 93-97. FRANCALANCI, L. & MANETTI, P. 1994. Geodynamic models of the southern Tyrrhenian region: constraints from the petrology and geochemistry of the Aeolian volcanic rocks. Bollenino Geofisica Teoretica Applicata, 36, 283-292. - - , T A Y L O R , S. R., MCCULLOCH,M.T. & WOODHEAD, J. D. 1993. Geochemical and isotopic variations in the calc-alkaline rocks of Aeolian arc, southern Tyrrhenian Sea, Italy: constraints on magma genesis. Contributions to Mineralogy and Petrology, 113, 300-313. FYTIKAS, M., INNOCENTI, E, MANETTI, P., MAZZUOLI, R., PECCEmLLO,A. & VmLAm, L. 1984. Tertiary to Quaternary evolution of volcanism in the Aegean region. In: RO~ERTSON, A. H. E & D~XON, J. G. (eds) Geological Evolution of the Eastern Mediterranean. Geological Society, London, Special Publications, 17, 687-699. GALASSI, B. 1995a. Evoluzione magmatologica delle associazioni orogeniche cenozoiche del Mediterraneo Centro-Occidentale. Phd thesis, Universitia di Ferrara, Italy. - 1995b. Magmatological evolution of the Cainozoic orogenic associations of the Central-Western Mediterranean. Plinius - European Journal of Mineralogy Italian Supplement, 13, 86-91. GARFUNKEL, Z. 1989. Tectonic setting of Phanerozoic magmatism in Israel. Israeli Journal of Earth Sciences, 38, 51-74.
TERTIARY-QUATERNARY MAGMATISM GENESSEAUX,M. & REHAULT,J. P. 1975. La marge continentale corse. Bulletin de la Soci~tO GOologique de France, 17, 505-518. GIANNERINI, G., CAMPREDON,R. ~:; FERAUD, G. 1988. Gen~se des rifts intracontinentaux par propagation polygonale de transformantes continentales: exemple du triplet golfe d'Aden, mer Rouge, Afrique de l'Est. Comptes Rendus de l'Acaddmie des Sciences, Paris, 306, Sdrie II, 1507-1514. GIANNETTI, B. & ELLAM, R. 1994. The primitive lavas of Roccamonfina volcano, Roman region, Italy: new constraints on melting processes and source mineralogy. Contributions to Mineralogy and Petrology, 116, 21-31. GIARDINI, D & VELONA,M. 1991. The deep seismicity of the Tyrrhenian Sea. Terra Nova, 3, 57-64. GILLOT, R Y., KIEFFER, G. & ROMANO, R. 1994. The evolution of Mount Etna in the light of potassiumargon dating. Acta Volcanologica, 5, 81-87. GOMEZ, E, BARAZANGI, M. & BENSAID, M. 1996. Active tectonism in the intracontinental Middle Atlas Mountains of Morocco: synchronous crustal shortening and extension. Journal of the Geological Society, London, 153, 389-402. GRANET,M.,WILSON,M. & ACHAUER,U. 1995. Imaging a mantle plume beneath the French Massif Central. Earth and Planetary Science Letters, 136, 281-296 GUEGUEN, E., DOGLIONI, C. & FERNANDEZ,M. 1997. Lithospheric boudinage in the Western Mediterranean back-arc basin. Terra Nova, 9, 184-187. GUIRAUD, R. & MAURIN, J. C. 1991. Le rifting en Afrique au Cr6tac6 inf6rieur: synthbse structurale, mise en 6vidence de deux 6tapes dans la genese des bassins, relations avec les ouvertures oc6aniques p6ri-africaines. Bulletin de la SociOt~ G~ologique de France, 162, 811-823. GUIRAUD, R., BINKS,R. M., FAIRHEAD,J. D. ~%WILSON, M. 1992. Chronology and geodynamic setting of Cretaceous-Cenozoic rifting in West and Central Africa. Tectonophysics, 213, 227-234. G12LE~, N. 1991. Crust-Mantle interaction in Western Turkey: Implications from Sr and Nd geochemistry of Tertiary and Quaternary volcanics. Geological Magazine, 128, 417-435. HAMMUDA, O. S., VAN HINTE, J. E. & NEDERBRAGT,S. 1992. Geohistory Analysis Mapping in Central and Southern Tarabulus Basin, Northwestern Offshore of Libya. In: SALEM,M. J. (ed.) The geology of Libya. Elsevier, Amsterdam, 1657-1680. HAWKESWORTH,C. J. & VOLLMER,R. 1979. Crustal contamination versus enriched mantle: 143Nd/144Nd and 87Sr/a6Sr evidence from the Italian volcanics. Contributions to Mineralogy and Petrology, 69, 151-165. HERNANDEZ, J., LAROUZIERE DE, E D., BOLZE, J. BORDEr, R 1987. Le magmatisme n6ogbne b6ticorifain et le couloir de d6crochement transAlboran. Bulletin de la SociOtO GOologique de France, III, 2, 257-267. - - & LEPVRIER,C. 1979. Le volcanisme calco-alcalin miocene de la r6gion d'Alger (Alg6rie): p6trologie et signification g6odynamique. Bulletin de la Socidt6 GOologique de France, XXI, 73-86.
165
KARAMATA, S., PAVICEVIC, K., KORIKOVSKIJ, S. K., BORONIHIN,V. A. & AMTHAUER, G. 1994. Petrology and mineralogy of Neogene volcanic rocks from the Allchar area, the FY Republic of Macedonia. Neues Jahrbuch fiir Mineralogie Abhandlung, 167, 317-328. KASTENS, K., MASCLE, J., ET AL. 1990. The geological evolution of the Tyrrhenian Sea: an introduction to the scientific results of ODP Leg 107. In: KASTENS,K., MASCLE,J. ET AL. Proceedings of the ODP, Scientific Results, 107, 3-26. KOLIOS,N., INNOCENTI,E, MANETTI,P., PECCERILLO,A. & GIULIANI, O. 1980. The Pliocene volcanism of the Voras Mts (Central Macedonia, Greece). Bulletin Volcanologique, 43, 553-568. LONERGAN, L. & WHITE, N. 1997. Origin of the BeticRif mountain belt. Tectonics, 16, 504-522. LOUNI-HACINI, A., BELLON, H., MAURY, R.C., MEGARTSI,M., COULON,C., SEMROUD,B., GOTTEN, J. & COUTELLE,A. 1995. Datation 40K-40Ar de la transition du volcanisme calco-alcalin au volcanisme alcalin en Oranie au Miocbne supdrieur. Comptes Rendus de l'AcadOmie des Sciences, Paris, 321, serie II a, 975-982. LOPEZ-RuIz, J., CEBRIA,J. M., DOBLAS, M., OYARZUN, R., HoYOS, M. & MARTIN,C. 1993. The late Cenozoic alkaline volcanism of the Central Iberian Peninsula (Calatrava Volcanic Province, Spain): Intra-plate volcanism related to extensional tectonics. Journal of the Geological Society, London, 150, 915-922. MACCIOTTA, G., VENTURELLI, G. d~; BECCALUVA, L. 1978. Geochemistry of mafic Cainozoic volcanic rocks from Sardinia (Western Mediterranean). Bulletin Volcanologique, 41, 56-78. MANTOVANI,E., ALBARELLO,D., TAMBURELLI,C., BABBUCCI, D. & VITI, M. 1997. Plate convergence, crustal delamination, extrusion tectonics and minimization of shortening work as a main controlling factor of the recent Mediterranean deformation pattern. In: MANTOVANI,E. (ed.) Geodynamics of the Mediterranean Region and its implications. Annali di Geofisica, 40, 611-643. MERGOIL, J. & BOIVIN,P. 1993. Le Velay. Son volcanisme et les formations associ~es. Notice de la carte 1/100 000. Geologie de la France. Orl6ans, BRGM. 3. MICHARD, A. 1976. ElOments de GOologie Marocaine. Notes et Mdmoires du Service G~ologique Maroc, 252. MILANI, X. 1996. Petrogenesis and metallogenic aspects of Tertiary volcanics of the Euganean Hills magmatic complex (Venetian Province). Plinius, European Journal of Mineralogy Italian Supplement, 15, 129-132. MONIE, E, MONTIGNY, R. & MALUSKI, H. 1992. Age burdigalien de la tectonique ductile extensive dans le massif de l'Edough (Kabylie, Alg6rie). Donn6es radiom6triques 39Ar-4~ Bulletin de la Socidt~ G~ologique de France, 163, 571-584. MONTANINI, A. & VILLA, I. M. 1993. 39Ar-4~ chronostratigraphy of Monte Arci volcanic complex (Western Sardinia, Italy). Acta Vulcanologica, 3, 229-233.
166
M. WILSON & G. B I A N C H I N I
MOl~Er~, J.-M. & CABANIS,B. 1993. Mise en 6vidence d'une association magmatique dans le volcanisme plio-quaternaire du Moyen-Atlas marocain. Comptes Rendus de l'Acad~mie des Sciences, Paris, 316, S6rie II, 537-362. MORRA, V., SECCHI, F .A.G., MELLUSO, L. & FRANCIOSl, L. 1997. High-Mg subduction related Tertiary basalts in Sardinia, Italy. Lithos, 40, 69-91. NIXON, P. E., THIRLWALL,M. F., BUCKLEY,E & DAVIES, C. J. 1984. Spanish and Western Australian lamproites: aspects of whole rock geochemistry. In: KORNPROBST,J. (ed.) Kimberlites I: Kimberlites and Related Rocks. Elsevier, Amsterdam, 285-296. NOTSU, K., FUJITANI,T., UI, T., MATSUDA,J. & ERCAN, T. 1995. Geochemical features of collision-related volcanic rocks in central and eastern Anatolia, Turkey. Journal of Volcanology and Geothermal Research, 64, 171-192. ORAL, M. B., RELINGER, R. E.,TOKSOZ, M. N., KING, R. W., BARKA, A. A., KINIK, I. & LENK, 0. 1995. Global positioning system offers evidence of plate motions in eastern Mediterranean. EOS Transactions of the AGU, 76, 1-3. OTTAVIANI-SPELLA,M.-M, GIRAD, M. & CHEILLETZ,A. 1996. Les ignimbrites burdigaliennes du Sud de la Corse. Comptes Rendus de l'AcadOmie des Sciences, Paris', 323, serie II a, 771-778. PAM1C, J. 1993. Eoalpine to Neoalpine magmatic and metamorphic processes in the northwestern Vardar Zone, the Easternmost Periadriatic zone and the southwestern Pannonian Basin. Tectonophysics, 226, 503-518. PATTERSON, R. 1996. Geochemical characteristics and geodynamic setting of Cenozoic mafic magmatism in western and central Europe. PhD thesis, University of Leeds, UK. PEARCE, J. A., BENDER, J. E, DELONG, S. E., KIDD,W. S. E, Low, R J., GONER, Y., SAROGLU,E, YILMAZ,Y., MOOI~BATH,S. & MYTCHELL,J. G. 1990. Genesis of collision volcanism in Eastern Anatolia, Turkey. Journal of Volcanology and Geothermal Research, 44, 189-229. PECCERILLO, A., CONTICELLI, S. & MANETTI, E 1987. Petrological characteristics and the genesis of the recent magmatism of South Tuscany and North Latium. Period. Mineral., 56, 167-183. PE-PIPER, G. 1994. Lead isotopic composition of Neogene volcanic rocks from the Aegean extensional area. Chemical Geology, 118, 27-41. & PWER, D. J. V. 1989. Spatial and temporal variation in Late Cenozoic back-arc volcanic rocks, Aegean Sea region. Tectonophysics, 169, 113-134. & -1994. Miocene magnesian andesites and dacites, Evia, Greece: adakites associated with subducting slab detachment and extension. Lithos, 31, 125-140. PICCOLI, G. 1970. Outlines of volcanism in northern Tripolitania (Libya). Bollettino della Societd Geologica Italiana, 89, 449-461. PINARELLI,L. 1987. Considerazioni geochimiche-petroIogiche sul vulcanismo acido della Provincia Magmatica Toscana. Univ. Firenze, Italy. PLATT, J. P. & V1SSERS,R. L. M. 1989. Extensional collapse of thickened continental lithosphere: A -
-
-
-
working hypothesis for the Alboran Sea and the Gibraltar Arc. Geology, 17, 540-543. - - , SOTO, J. I., COMAS, M. C. & LEG 161 SHIPBOARD SCIENTISTS 1996. Decompression and high temperature, low-pressure metamorphism in the exhumed floor of an extensional basin, Alboran Sea, western Mediterranean. Geology, 24, 447-45O. RACHDI, H., BERRAHMA, M., DELALOYE, M., FAUREMURET,A. & DAHMANI M. 1997. Tertiary volcanism in Rekkame (Morocco) - Petrology, geochemistry and geochronology. Journal of African Earth Sciences, 24, 259-269. REHAULT,J. P., BOILLOT,G. & MAUFFRET,A. 1984. The Western Mediterranean basin geological evolution. Marine Geology, 55,447-477. ROBERT, U., FODEN, J. & VARNE, R. 1992. The Dodecanese Province, SE Aegean: a model for tectonic control on potassic magmatism. Lithos, 28, 241-260. ROBERTSON,A. H. E & GRASSO,M. 1995. Overview of the Late Tertiary-Recent tectonic and paleoenviromental development of the Mediterranean region. Terra Nova, 7, 114-127. ROGERS, N. W., HAWKESWORTH,C. J,, PARKER, R. J. & MARSH, J. S. 1985. The Geochemistry of potassic lavas from Vulsini, Central Italy and implications for mantle enrichment processes beneath the Roman region. Contributions to Mineralogy and Petrology, 90, 224-257. ROMANO,R. 1983. Succession of the volcanic activity in the Etnean area. Memorie della Societd Geolologica Italiana, 23, 27-28. RtJrrER, M. J. 1987. Evidence for crustal assimilation by turbulently convecting, mafic alkaline magmas: geochemistry of mantle xenoliths-bearing lavas from Northern Sardinia. Journal of Volcanology and Geothermal Research, 32, 343-354. SAVELI,I, C. 1988. Late Oligocene to Recent episodes of magmatism in and around the Tyrrhenian Sea: implications for the process of opening in a young inter-arc basin of intra-arc-orogenic (Mediterranean) type. Tectonophysics, 146, 163-181. - - - , BECCALUVA, L., DERIU, M., MACCIOTTA, G. & MACCIONI, L. 1979. K/Ar geochronology and evolution of the Tertiary "calc-alkalic" volcanism of Sardinia (Italy). Journal of Volcanology and Geothermal Research, 5, 257-269. dh; GASPAROTTO,G. 1994. Calc-alkaline magmatism and rifting of the deep water volcano of Marsili (Aeolian back-arc, Tyrrhenian sea). Marine Geology, 119, 137-157. SAWAF,T., AL-SAAD, D., GEBRAN, A., BARAZANGI,M., BEST, J. A. & CHAIMOV,T. A. 1993. Stratigraphy and structure of eastern Syria across the Euphrates depression. Tectonophysics, 220, 267-281. SENGOR, A. M. C. 1993. Some current problems on the Tectonic evolution of the Mediterranean during the Cainozoic. In: B o s c m E., MANTOVANIE. 8,: MORELLI A. (eds) Recent volution and seismicity of the Mediterranean Region. Nato ASI Series C, D Reidel, Dordrecht-Boston, 402, 403-422. & YILMAZ, Y. (1981) Tethyan evolution of -
-
T E R T I A R Y - Q U A T E R N A R Y MAGMATISM Turkey: a plate tectonic approach. Tectono-
physics, 75, 181-241. , CIN, A., ROWLEY, D. B. & NIE, S. Y. 1993. Spacetime patterns of magmatism along the Tethysides: a preliminary study. Journal of Geology, 101, 51-84. , GOROR, N. & SAROGLU,E 1985. Strike-slip faulting and related basin formation in zones of tectonic escape: Turkey as a case study. In: B1DDLE, K. T. & CRISTIE, B. N. (eds) Strike-slip defor-
mation,
basin formation and sedimentation.
Society of Economic Paleontologists and Mineralogists. Special Publications, 37, 227-64. SERRI, G. 1990. Neogene-Quaternary magmatism of the Tyrrhenian region: characterization of the magma sources and geodynamic implications. Memorie della Societti Geologica Italiana, 41, 219-242. - - , INNOCENTI,E & MANETTI,P. 1993. Geochemical and petrological evidence of the subduction of delaminated Adriatic continental lithosphere in the genesis of the Neogene-Quaternary magmatism of central Italy. Tectonophysics, 223, 117-147. SEYITOGLU,G. & SCOTT,B. C. 1992. Late Cenozoic volcanic evolution of the northeastern Aegean region. Journal of Volcanology and Geothermal Research, 54, 157-176. - - , ANDERSON,D., NOWELL,G. & SCOTT,B. 1997. The evolution from Miocene Potassic to Quaternary sodic magmatism in western Turkey: implication for enrichment processes in the lithospheric mantle. Journal of Volcanology and Geothermal Research, 79, 127-148. SIENA, E ~ COLTORTI,M. 1989. Lithospheric mantle evolution: Evidences from ultramafic xenoliths in the Lessinian volcanics (northern Italy). Chemical Geology, 77, 47-364. SPADINI, G., BERTOTTI, G. & CLOETINGH, S. 1995. Tectono-stratigraphic modelling of the Sardinian margin of the Tyrrhenian Sea. Tectonophysics, 252, 269-284. SPAKMAN, W. 1990. Images of the upper mantle of central Europe and the Mediterranean. Terra Nova, 2, 542-553. , VAN DER LEE, S. AND VAN DER HILST,R.D. (1993) Travel-time tomography of the EuropeanMediterranean mantle down to 1400 km. Physics of the Earth and Planetary Interiors, 79, 3-74. STOPPA,E & CUNDAR[,A. 1995. A new Italian carbonatite occurrence at Cupaelto (Rieti) and its genetic significance. Contributions to Mineralogy and Petrology, 122, 275-288. TONARINI, S., D'ORAZIO, M., ARMIENTI,P., INNOCENTI, E, SCRIBANO V. 1996. Geochemical features of eastern Sicily lithosphere as probed by Hyblean xenoliths and lavas. European Journal of Mineralogy, 8, 1153-1173. TORRES-ROLDAN, R. L., POLI, G. & PECCERILLO, A. 1986. An early Miocene arc-tholeiitic magmatic dyke event from the Alboran Sea: Evidence for pre-collisional subduction and back arc crustal extension in the westernmost Mediterranean. Geologisches Rundschau, 71, 219-234.
167
TREUIL, M. & JORON, J.-L. 1994. Etude g6ochimique des 616ments en traces dans les laves 6mises au cours de l'6ruption 1991-1993 de l'Etna. Mise en evidence des contributions de la source, de la fusion partielle, de la differenciation et des modalitds de tranfert des magmas. Acta Volcanologica, 4, 29-45. TRICART, P., TORELLI, L., ARGNANI, A., REKHISS, E & ZITELLINI,N. 1994. Extensional collapse related to compressional uplift in the Alpine Chain off northern Tunisia (Central Mediterranean). Tectonophysics, 238, 317-329. VAN DER MEER, E & CLOETINGH, S. 1993. Intraplate stresses and the subsidence history of the Sirte basin (Libya). Tectonophysics, 226, 37-58. VENTURELLI, G., CA PEDRI, S., DI BATTISTINI,G., CRAWFORD, A., KOGARKO, L. N. & CELESTINI,S. 1984a. The ultrapotassic rocks from southeastern Spain. Lithos, 17, 37-54. - - , THORPE, R. S., DAL PIAZ, G. V., DEL MORO, A. & POTTS, E J. 1984b. Petrogenesis of calc-alkaline, shoshonitic and associated ultrapotassic Oligocene volcanic rocks from the Northwestern Alps, Italy. Contributions to Mineralogy and Petrology, 86, 209-220. VIGLIOTTI, L.& LANGENHEIM, V. E. 1995. When did Sardinia stop rotating? New paleomagnetic results. Terra Nova, 7, 424-435. VISSERS,R. L. M., PLATT,J. P. & VANDER WAAL, D. 1995. Late orogenic extension of the Betic Cordillera and Alboran Domain: A lithospheric view. Tectonics, 14, 786-803. VOLLMER, R. 1976. Rb-Sr and U - T h - P b systematics of alkaline rocks: the alkaline rocks from Italy. Geochimica et Cosmochimica Acta, 40, 283-295. VON BLANCKENBURG, E & DAVIES, J. H. 1995. Slab breakoff: a model for syncollisional magmatism and tectonics in the Alps. Tectonics, 14, 120-131. WEZEL, E C. 1976. Widespread manifestations of Oligocene-Lower Miocene volcanism around the Western Mediterranean. In: BIJU-DUvAL, X. & MONTADERT,X. (eds) International symposium on
the structural history of the Mediterranean basin. Split, Yugoslavia. Editions Technip, Paris, 287-302. WILSON, M. 1989. Igneous Petrogenesis. International Thompson, London. & DOWNES, H. 1991. Tertiary-Quaternary extension related alkaline magmatism in Western and Central Europe. Journal of Petrology, 32, 811-849. & -1992. Mafic alkaline magmatism in the European Cenozoic rift system. Tectonophysics, 173-182. & GUIRAUD, R. 1998. Late Permian to Recent magmatic activity on the African-Arabian margin of Tethys. In: MACGREGOR. D. CLARK-LOWES,D. D. & MOODY, R. (eds) Petroleum Geology of North Africa. Geological Society, London, Special Publications, 132, 231-263. - - - , DOWNES, H. & CEBRIA, J.-M. 1995. Contrasting fractionation trends in coexisting continental alkaline magma series; Cantal, Massif Central, France. Journal of Petrology, 36, 1729-53.
-
-
-
-
-
-
2 0 8 ,
168
M. WILSON & G. BIANCHINI
-,TANKU'I,A. &GOLE~ N. 1997. Tertiary volcanism of the Galatia Province, Central Anatolia, Turkey. Lithos, 42, 105-121. YILMAZ,Y. 1990. Comparison of young volcanic associations of western and eastern Anatolia formed under a compressional regime: a review.
1992. Very high rates of cooling and uplift in the Alpine belt of the Betic Cordilleras, southern Spain. Geology, 20, 79-82. ZmGLER, R A. 1992. European Cenozoic rift system. In: ZIEGLER, P. A. (ed.) Geodynamics of Rifting,
Volume I. Case History Studies on Rifts: Europe and Asia. Tectonophysics, 208, 91-111.
Journal of Volcanology and Geothermal Research, 44, 69-87. 1993. New evidence and model for the evolution of the southeast Anatolian orogen. Geological Society of America Bulletin, 195, 251-271. ZECK, H. R, MONIE, P., VILLA, I. M. & HANSEN, B. T.
-
-
--,
CLOETINGH, S. & VAN WEES, J.-D. 1995. Dynamics of intra-plate compressional deformation: the Alpine foreland and other examples. Tectonophysics, 252, 7-59.
Crustal structure of the North Tyrrhenian Sea: first result of the multichannel seismic LISA cruise A. M A U F F R E T
& I. C O N T R U C C I
D @ a r t e m e n t de Gdotectonique, Case 129, Universitd Pierre et M a r i e Curie, 4 Place Jussieu, 75252, Paris C e d e x 05, France
Abstract: During a deep-penetration multichannel seismic cruise, the North Tyrrhenian basin was investigated. This basin has a shallow Moho along the Italian coast. Two kinds of dipping reflections were identified in the crust. The westward-dipping reflections are interpreted as Apennines or/and Alpine thrusts whereas the eastward-dipping reflections may be the seismic expression of extensional detachments. The deep part of the 10 km thick Corsica Basin is well imaged on the continental shelf. This basin was formed before or during the opening of the Balearic-Ligurian basin and is related to the collapse of the Alpine Corsica belt. We emphasize the active role of plutonic bodies that rise from a 3 km depth during the mid-Pliocene. The thinning of the North Tyrrhenian crust could be explained by a delamination of the Adriatic lithosphere that retreats towards the east.
The formation of extensional basins in convergent setting is a major problem of Earth sciences (Faccena et al. 1996). In the past years formation of marginal basins behind subduction zones was the best explanation for extensional tectonics. Indentation theory was an alternate hypothesis to explain extension in the collided plate. In recent years several studies highlighted the collapse of former orogens to explain the extension. The Mediterranean Sea is an ideal place to study the formation of extensional basins that were fully developed in the Western Mediterranean basins (Balearic and Ligurian Seas) with early Miocene emplacement of oceanic crust. The North Tyrrhenian Basin is underlain by continental basement and its evolution is very recent (late Miocene to Pliocene). Consequently in this region we can study the young extensional processes that are inactive in the others basins of the Western Mediterranean Sea.
Tectonic framework The African plate has converged towards the E u r o p e a n plate since the Late Cretaceous (Olivet 1987). The Adriatic Promontory (Fig. 1) represents a n o r t h e r n prolongation of the African plate (Alvarez 1991). This promontory collided with E u r o p e during the Cretaceous-Eocene with the closure of the LiguroPiemont Ocean (Stampfli & Marchant 1997) and formation of the Alps. During this event the Corsican Alps were overthrust onto western Hercynian Corsica (Mattauer et al. 1981). Recent dates, obtained (Brunet et al. 1997) by the 39Ar/4~ method, suggest a Late Cretaceous
(65 Ma) to late Eocene (37 Ma) age for the emplacement of Corsican nappes. Soon after this compressional episode a large extension occurred in the European plate with formation of Oligocene grabens (Rhine, Limagnes and Bresse Basins; Fig.l). The subsequent formation of oceanic crust in the Balearic and Ligurian Seas and the coeval rotation of Corsica-Sardinia block occurred (Cravatte et al. 1974; Montigny et al. 1981; Rehault et al. 1984) from the mid-Aquitanian to Burdigalian (22-19 Ma). During the late Oligocene (33-29 Ma and 25 Ma; Brunet et al. 1997) Alpine Corsica underwent an extensional event (Jolivet et al. 1990, 1991, 1994; Daniel & J o l i v e t 1996) while compression migrated towards the east, in the inner Apennines Belt (27 Ma; Carmigniani & Kligfield 1990), and then in the present-day front of deformation along the Adriatic coast of Italy (Fig. 1). During the same time span extension also migrated from the North Tyrrhenian Sea to the central part of Apennines Belt (Elter et al. 1975; D'Offizi et al. 1994).
Geological and geophysical setting of the North Tyrrhenian Basin This region is limited (Figs 2 and 3) by Corsica to the west and the Tuscany coast of Italy to the east, the north Ligurian Sea to the north and the South Tyrrhenian sea to the south. Four physiographic provinces can be distinguished (Fig. 2): the Corsica Basin with moderate depth and smooth topography, a central ridge where Capraia, Elba, Pianosa, and M o n t e Cristo
MAUFFRET,A. dCz;CONTRUCCI,I. 1999. Crustal structure of the North Tyrrhenian Sea: first result of the muitichannel seismic LISA cruise. In: DURAND,B., JOUVE1,L., HORVATH,F. & SERANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 169-193.
170
A. MAUFFRET & I. CONTRUCCI European Plate Iberian Plate African Plate
Molassic Basin
Adriatic Plate Austro-Alpine appes
Po Basin
i " I
"-,..J ~
v--,c_,L.gJ ~-,
Apennines Brianconnais and Alpine Corsica
I Neogene basin Oceanic crust
v
0 I
500 km I
I
I
I
i
Fig. 1. Tectonic framework of the North Tyrrhenian Basin. From Stampfli & Marchant (1997) modified.
Islands are located, the Tuscany shelf and the north-south-trending Etruschi and Cialdi seamounts that formed the transition with the deeper South Tyrrhenian Basin. The Corsica basin has an onshore extension (Aleria Basin) flanked by Alpine Corsica (Fig. 3). The central high is divided into two ridges: Pianosa sedimentary ridge and Capraia-Monte Cristo Ridge constituted of recent granitic intrusions (Elba, Monte Cristo) and volcanic extrusion (Capraia). Similarly, the Giglio Island intrusion is a prominent feature of the Tuscany Shelf that is characterized by a horst and graben structure (Bartole et al. 1991; Bartole 1995). Monte Etruschi and Cialdi are also recent extensional features (Zitellini et al. 1986). The main offshore structures of the North Tyrrhenian Sea have been investigated by an industrial multichannel seismic survey described by Bartole et al. (1991) and Bartole (1995). Two exploratory wells located on the Capraia ridge (Martina and Mimosa, Fig. 3) help to calibrate
the seismic sections. However the results of coring and dredging surveys (Aleria 1980) show that this ridge has a composite and irregular basement (Upper Cretaceous and Eocene; Fig. 3). Moreover, Triassic rocks crop out in Africa islet (Figs 2 and 3). Therefore, the results of the exploratory wells projected on a distant seismic line (Fig. 4a) may not be accurate. The line drawing L 122 (Fig. 4a; Bartole et al. 1991) illustrates the main feature of the North Tyrrhenian Sea. The Ligurian nappes are characterized by eastern vergent thrusts. Martina well bottomed into Palaeocene-middle Eocene layers of Alpine affinity that are overlain unconformably by Oligocene to early Miocene pelites. This layer, 1 km thick in the Marina and Mimosa wells, is not deformed by compression, but affected by the Tortonian to early Pliocene extension. However, the seismic sequence attributed to the Oligocene (Bartole et al. 1991) may be younger if the correlation between the well and the seismic profile is wrong. The Ligurian nappes have been
CRUSTAL STRUCTURE OF THE NORTH TYRRHENIAN SEA
171
Fig. 2. Track map of the LISA cruise. Bathymetric contour: 0.2 km interval. The position of the seismic profiles shown are indicated by the number of the figures. emplaced in an accretionary prism (Principe & Treeves 1984) that has been overprinted by the late Oligocene collision between Corsica and Adria continental crust (Keller & Pialli 1990;
D'Offizi et al. 1994; Keller & Coward, 1996). In this context the Oligocene-early Miocene unit is interpreted as piggy back deposits if the correlation between the line 122 and the Marina well
172
A. MAUFFRET & I. CONTRUCC1
Fig. 3. Structural map of the North Tyrrhenian Basin. Extension of the synrifts sediments are from Bartole et (1991); Bartole (1995). Results of sampling are from Aleria (1980) and Bartole et al. (1991).
al.
Fig. 4. (a) Line drawing of the industrial seismic L 122 redrawn from Bartole et al. (1991). (b) Transect crossing the North Tyrrhenian Basin constrained by refraction (Hirn & Sapin 1976; Letz et al. 1977a,b; Egger et al. 1988; Egger 1992) and gravity (Carrozzo & Nicolich 1977) data. The depth of the Corsica Basin is only 5 km deep in the northern part of the Corsica Basin; it is much deeper in the southern part. Note the thin crust beneath the Tuscany Margin. The refractor indicating a deep European Moho (Letz et al. 1977a,b) could be a multiple (Ponziani et al. 1995).
CRUSTAL STRUCTURE OF THE NORTH TYRRHENIAN SEA
173
174
A. MAUFFRET & I. CONTRUCCI
is correct (Bartole et al. 1991). In an alternate hypothesis this unit may represent the base of the sedimentary fill deposited in the Corsica fore-arc basin (Rehault 1981). The acoustic basement off Elba Island is supposed to be formed by the Tuscan metamorphic rocks that outcrop in the island (Fig. 4a; Bartole et al. 1991; Keller & Coward 1996), but this basement may be also representative of the 6.2-5.1 Ma old (Saup6 et al. 1982)granitic intrusion of Porto Azzuro located on the eastern coast of Elba Island. The refraction ( H i r n & Sapin 1976; Letz et al. 1977a,b; Egger et al. 1988; Egger 1992) and gravity studies (Carrozzo & Nicolich 1977) show the transition from the 32 km thick crust of Corsica (Fig. 4b) to the 22 km thick thinned crust of Tuscany. The base of the Corsica basin is 5.2
km deep (Carrozzo & Nicolich 1977) that corresponds to 3.6 s two-way travel time (TWTT). However, this transect crosses the northern part of the Corsica basin whereas the deepest part of the basin is located southwards (see later). In the former refraction experiments a 50-70 km deep 'European' Moho was supposed beneath the 'Italian' Moho (Letz et al. 1977a,b) but a recent study (Ponziani et al. 1995) show that the supposed refracted deep arrival was in fact a multiple artifact. The penetration of the industrial multichannel seismic (MCS) lines in the north Tyrrhenian Sea are no greater than 4 s TTWT (Fig. 4a). On the other hand the refraction results (Fig. 4b) penetrated deeply, but do not show the fine structure. The Lisa MCS lines fill the gap between the two types of seismic data.
Fig. 5. Eastern part of the LISA 7 seismic profile showing dipping reflections with Apennine vergence. A 7 s TWTT deep reflector is correlated with the Moho identified on the refraction line shown in Fig. 4b. The horizontal reflection at 7.5 s is an artifact generated by the migration processing. Position of the seismic profile is indicated in Fig. 2.
CRUSTAL STRUCTURE OF THE NORTH TYRRHENIAN SEA
LISA Cruise A multichannel seismic cruise (LISA, Fig. 2) was carried out in 1995 on the RV Nadir in the Western Mediterranean Sea. We used a 2.4 km streamer, 96 channels, towed 20 m below sea level. The same depth was adopted for the immersion of 10 GI guns with a total volume of 1140 cubic inches. These guns were tuned in single bubble mode (Avedik et aL 1993). This special mode and the deep immersion of streamer and guns generate low frequencies and allow deep penetration of the crust, although the volume of guns is relatively small. We presented the results of on board processing: velocity analysis, normal move out correction and stack. LISA seismic data need a deconvolution and filtering in the F-K domain. Multiple and reverberation is an acute problem, but we were lucky that the primary reflections are not parallel to the sea floor and can be distinguished, particularly in the Corsica Basin, through the multiples.
Main results of the LISA cruise Structures with Apenninic
vergence
T h e east vergent A p e n n i n e s thrusts are identified by p r o m i n e n t reflections that dip towards the west (Bartole et al. 1991; B a r t o l e 1995).
175
T h e s e reflections w e r e not visible b e l o w 4 s T W T T in the industrial MCS lines (Fig. 4a). T h e Lisa 7 profile (Fig. 5) shows w e s t w a r d - d i p p i n g reflectors as d e e p as 7 s TWTT. A t this d e p t h an horizontal reflector can be c o r r e l a t e d to the M o h o (a d e p t h of 22 km, Fig. 4b, c o r r e s p o n d s to 7 s T W T T ) . T h e s e d i m e n t a r y c o v e r of t h e Capraia Basin is separated into two parts (Fig. 5) by a p r o m i n e n t unconformity, probably Messinian in age. T h e strong reflection, which is linked to this u n c o n f o r m i t y , is characteristic of the Messinian level s h o w e d in the Corsica basin (Fig. 6), although Bartole et al. (1991) attributed a m i d - P l i o c e n e age to this unconformity. T h e u p p e r part of the s e d i m e n t a r y fill is horizontal w h e r e a s the lower part is affected by extensional m o v e m e n t s . T h e base of the s e d i m e n t a r y fill could be O l i g o c e n e (Bartole et al. 1991). We n o t e d that the Oligocene, if t h e assumption of Bartole et al. (1991) is correct, is not involved in the thrusts, which can be d a t e d as late E o c e n e . In this case the f o r m a t i o n of thrusts is coeval to the A l p i n e Corsica c o m p r e s s i o n with already an A p e n n i n i c vergence. S o m e n o r m a l faults s e e m to b e b r a n c h e d a n d r e a c t i v a t e d t h e f o r m e r thrusts (200-250 shot poins, Fig. 5).
Fig. 6. LISA 7 seismic profile. The northern Corsica Basin is relatively shallow (3.5 s TWq-T). The Messinian reflector is prominent and cut by a normal fault. Position of the seismic profile is indicated in Fig. 2.
176
A. MAUFFRET & I. CONTRUCCI
Fig. 7. Western part of LISA 8 seismic profile. The northern Corsica Basin is relatively shallow (3 s TWTT). The Messinian reflector is upturned on the western flank of the Pianosa Ridge. Burdigalian sediments have been sampled by coring (Viaris de Lesegno 1978) on the Pianosa Ridge near the seismic profile. 3 km of uplift are calculated if the erosion on the continental shelf of Elba Island is taken into account. The chaotic configuration of the seismic reflectors below the Messinian unconformity indicates that erosion on Elba Island was already active during the late Messinian. Position of the seismic profile is indicated in Fig. 2.
Corsica Basin (Figs 6 to 11) In the onshore extension of the Corsica Basin (Aleria Basin) the marine lower Pliocene overlies the Messinian erosional unconformity. Beneath this unconformity lies the upper Tortonian which rest u n c o n f o r m a b l y upon the L a n g h i a n (middle Miocene). The Miocene layers has a 20 ~ dip towards the east (OrzagSperber & Pilot 1976). In the Corsica basin the P l i o c e n e - Q u a t e r n a r y sedimentary unit is limited at its base by a prominent reflector that is Messinian in age
(Viaris de Lesegno 1978; Viaris de Lesegno et al. 1978). The lower part of the Pliocene-Quaternary unit forms a wedge that thins towards the east (Figs 6 and 7). This sedimentary wedge shape suggests that the main source of sediments is Corsica (Viaris de Lesegno, 1978; Viaris de Lesegno et al. 1978), although we cannot exclude a recent tilting of Corsica in the n o r t h e r n m o s t seismic profile (Fig. 6). The Messinian reflector is affected by normal faulting and uplifted near Corsica in the northern part of the Corsica Basin (Fig. 6). The normal fault is also evident in the early Pliocene. The
CRUSTAL STRUCTURE OF THE NORTH TYRRHENIAN SEA
177
Fig. 8. Western part of LiSA 10 seismic profile. The Corsica Basin is 5-6 s TWTT thick (horizontal arrows show the real reflectors, dipping arrows show the multiples) Pianosa Ridge is an uplifted part of the Corsica Basin. The Messinian reflector and a Miocene horizon are 1.1 km and 2.4 km uplifted respectively. Position of the seismic profile is indicated in Fig. 2.
Messinian reflector is flat and deep (more than 2 s TWTT) in the middle part of the basin (Figs 7 and 8). It is shallower and cut by several canyons in the southern part of the study area (Figs 10 and 11). The acoustic basement is moderately deep in the northern part of the Corsica Basin (3.5 and 3 s T W T T Figs 6 and 7 respectively). This depth corresponds to the 5.2 km depth (3.6 s TWTT) observed on the refraction and gravity results (Fig. 4b; Carrozzo & Nicolich 1977). This basement is very deep in the middle part of the Corsica Basin (about 5.5 s TWTT; Fig. 8; Finetti & Morelli 1973). For the first time the deepest part of the Corsica Basin is imaged on the Corsica continental shelf (Fig. 9). The basal unit (between 5 and 6 s TWTT) is tilted and the middle unit (between 3.5 and 5 s
TWTT, Fig. 9) shows a fan-shaped configuration characteristic of a synrift sequence. This huge graben is limited by the Solenzara listric fault (Fig. 3) and by other normal faults evident in the easternmost Alpine Corsica (Daniel et al. 1996). South of this fault the acoustic basement is shallow (3-2 s TWTT, Figs 10 and 11). The onshore extension (Aleria Basin) of the Corsica Basin is 4 km thick (Bayer et al. 1976) and the acoustic basement is more than 10 km deep in the offshore part of this basin. The thickness of the Corsica Basin is shallower (5.2 km) in the transect presented in Fig. 4b, which crosses the northern part of the basin. The Bouguer gravity map on land (Bayer et al. 1976) and the free air gravity map at sea (Sandwell et al. 1995) show (Fig. 12) the shape of the Corsica basin and in
178
A. MAUFFRET & I. CONTRUCCI
Fig. 9. LISA 11 seismic profile. This profile shows the listric Solenzara fault (Fig. 3) that limits the Corsica Basin. The acoustic basement is 6 s TWTT deep. The synrift formation has a fan shaped configuration. Position of the seismic profile is indicated in Fig. 2.
particular its northern and southern termination (+50 milligal contour). The southern positive gravity anomaly is related to Hercynian Corsica whereas the northern one is linked with Alpine Corsica.
Capraia-Monte Cristo and Pianosa ridges (Figs 6, 7, 8, 10, 13 and 14) The e m p l a c e m e n t of magmatic bodies of Capraia, M o n t e Capanne (Elba), Giglio and Monte Cristo have been dated at 6.9-3.5 Ma,
C R U S T A L S T R U C T U R E OF T H E N O R T H T Y R R H E N I A N S E A
179
Fig. 10, Western part of LISA 12 seismic profile. The Corsica Basin has a shallow basement. Note the Monte Cristo Ridge that is correlated with a prominent magnetic anomaly. Position of the seismic profile is indicated in Fig. 2.
Fig. 11. Western part of L I S A 13 seismic profile. This profile shows the eastern extension of the Hercynian Corsica (see Fig. 12). Position of the seismic profile is indicated in Fig. 2.
180
A. MAUFFRET & I. CONTRUCCI
., /0
/~
~'.50../
25 /
f~O'~ l 'z
~
25__---~
"
i
]
,
,
-50
25
.,
-- 7---.~.:.v.~.~ .
41"N
1 C~
rN
Fig. 12. Gravity map of Corsica (Bayer et al. 1976) and North Tyrrhenian Basin (Sandwell et al. 1995). Bouguer anomalies on land and free air anomalies at sea. Alpine Corsica is characterized by positive anomalies and Aleria Basin (on shore extension of the Corsica Basin, see Fig. 3) by negative anomalies. The Corsica Basin is correlated to negative gravity anomalies (0 to -25 mGal) whereas the western extension of the Hercynian Corsica shows a positive anomaly (50 mGal).
CRUSTAL STRUCTURE OF THE NORTH TYRRHENIAN SEA 5 Ma and 7 Ma respectively (Serri et al. 1993 for a review). This ridge is delineated by a prominent magnetic anomaly (up to 500 nT near Monte Cristo Island). A relationship between magmatism and magnetism is likely, although the magnetic positive anomaly has been attributed to an ophiolitic body (Cassano 1991). However, Monte Capanne is not related to a magnetic anomaly that can be explained by the low magnetic susceptibility of the granite that forms this intrusion (Boullin et al. 1993). In Pianosa Island (Colantoni & Borsetti 1974) the middle Pliocene-Quaternary series overlies directly the lower Miocene (Langhian, 15 Ma). On the continental shelf of this island Burdigalian marls have been sampled by coring (Viaris de Lesegno 1978). The Miocene layers of the island and on the Pianosa continental shelf dip 20 ~ towards the west. In the exploratory well Martina (Fig. 3), located south of Pianosa Island the upper Pliocene overlies directly 1 km thick Oligocene series unaffected by compressional tectonics, if the distant correlation (Fig. 4a) between the seismic profile L 122 and the Martina well is correct. In the Martina well this thick Oligocene sequence rests on Palaeocene-middle Eocene turbiditic layer and ophiolitic breccias that is deformed by east vergent thrusts (Bartole et aL 1991). Triassic rocks crop out in the Africa Islet (Figs 1 and 2). Eocene, Upper Cretaceous Palaeocene (Helmintoides flysch) rocks were sampled (Aleria 1980) in the southern part of Pianosa Ridge (near Monte Cristo Island, Fig. 3). From these results it is evident that the Corsica Basin is underlain by Alpine Corsica formations (Helmintoides flysch, Shistes Lustr6s and ophiolites) and that the Miocene and Oligocene sedimentary layers of the Pianosa Ridge are equivalent, but uplifted, to the deep part of the Corsica Basin. In the Corsica Basin the Messinian reflector is flat but it shows a i km tilt (Figs 6, 7 and 8) along the western flank of Pianosa Ridge. The Lower Pliocene is restricted to the basin (sedimentary wedge, Figs 6 and 7); therefore, we conclude that the main episode of uplift occurred during the early Pliocene. However, Messinian uplift is also recorded by the accumulation of erosional detritus below the Messinian unconformity (Fig. 7). The Pianosa continental shelf is an erosional surface (Fig. 8) and 1 km of uplift is a minimum value; therefore, 3 km can be estimated if we take into account this erosion (Fig. 7). A greater value (10 km) may be proposed if we assume that the Oligocene layers drilled on the Pianosa Ridge Martina and Mimosa wells (Fig. 3) were initially located in the deepest part of the Corsica Basin. However, we do not know the
181
initial topography of the Pianosa Ridge area that was probably the hanging wall of the Corsica Basin and consequently shallower than the bottom of the basin. Moreover the basement that underlies the basin must be backstripped (work in progress) to obtain a correct evaluation. On the eastern flank of the Monte Cristo Ridge tilting is also evident (Figs 13 and 14). The mid-Pliocene tilting can be evaluated to 0.9 km (Fig. 13). A 1 km thick Oligocene layer was drilled in Mimosa well (Bartole et al. 1991), but folded Alpine units crop out at the western end of the seismic profile presented in Fig. 14 although this profile is very close to Mimosa well. This seismic profile illustrates the difficulty in tying the results of the exploratory wells to seismic stratigraphy. In this profile the minimum tilting is evaluated to 1 km. We note that the normal faults are west facing (Figs 13 and 14) and these faults may be antithetic relative to a deep detachment. This detachment is not observed because prominent multiples obscure the seismic profiles. 1 to 3 km of mid-Pliocene uplift are recorded by the tilting of the sedimentary layers and acoustic basement on the two flanks of the Elba-Monte Cristo Ridge. Monte Capanne (Elba) was emplaced 6-7 Ma ago (late Tortonian) at a depth of 2.5 km, then a very rapid 2 Ma old uplift is recorded by fission tracks in apatites (Bouillin et al. 1994). Monte Capanne is 1 km high and Elba Island underwent a 3.5 km total uplift during the mid-Pliocene. These results are in complete agreement with our observations. Eurite (tourmaline-rich aplite) pebbles has been found in the early Pliocene (Tongiorgi & Tongiorgi 1964) and latest Messinian (Marinelli et al. 1993) basins of Tuscany. These pebbles originated from the Monte Capanne intrusion of Elba Island, which was an extension of the Tuscany continental landmass at this time, Marinelli et al. (1993) concluded that he Monte Capanne was 2.5 km high during the Messinian. The presence of erosional products beneath the Pliocene-Miocene unconformity (Fig. 7) suggests, indeed, that the Monte Capanne was already high during the latest Messinian. The tectonic denudation of the Monte Capanne is commonly attributed to low-angle simple shear fault dipping to the east (Keller & Pialli 1990; Keller & Coward 1996; Daniel & Jolivet 1995) associated with the Tyrrhenian Sea extension. The emplacement of the plutonic body is passive in this hypothesis. However, we suggest, in agreement with Zitellini et al. (1986), that the mid-Pliocene tilting on the two flanks of the Monte Capanne-Monte Cristo Ridge is caused by the rise of the intrusion. Therefore, we favour
182
A. MAUFFRET
& I. C O N T R U C C I
. ,.,,,, .,,,,a
9
~z
9
9
o ~3
+,,.,
o3
r
r~
.~
,,,~
C R U S T A L S T R U C T U R E OF T H E N O R T H T Y R R H E N I A N SEA
183
Fig. 14. Eastern part of LISA 10 seismic profile. We observe again a prominent tilting along the eastern flank of the Monte Cristo Ridge. Alpine Formation outcrops on the ridge (see Fig. 2) but the Mimosa well sampled 1 km thick Oligocene series. These Oligocene series may be included in the compressional deformed layers or undeformed (Fig. 4a) but absent in the area crossed by the seismic profile. This profile illustrates the difficulty to tie the results of the wells and the seismic stratigraphy. Position of the seismic profile is indicated in Fig. 2.
184
A. MAUFFRET & I. CONTRUCCI
an active role for the magmatic body. Ductile extensional shear zone localized at the pluton boundary accompanied the exhumation at depth of the Elba intrusives (Daniel & Jolivet 1995). This history does not contradict the fact that the pluton in its last stages of ascent, already in its solid-state, rises up by isostatic forces and the brittle extension of its cover. A part of the uplift may be due to elastic rebound and extensional unloading of the footwall during the tectonic denudation. Offshore Elba several eastwarddipping reflectors are observed in the upper part of the crust (3 s TWTT, Fig. 15) but also in the lower part (6-7 s TWTT, Fig. 15). A well-layered lower crust shows a prominent eastward dip whereas the reflectors corresponding to the Moho (7 s TWTT, Fig. 15) are horizontal. The presence of the magmatic body may favour the localization of detachment faults at crustal scale, probably in a ductile regime. Tuscany M a r g i n a n d extensional tectonics (Figs 14, 15 a n d 16) We crossed two main basins: one between Monte Cristo and Giglio islands that we name Monte Glio Basin (Fig. 14) and the Punta Ala Basin (Figs 15 and 16). A shallow unconformity is related to the end of the mid-Pliocene extensional event (Bartole et al. 1991). A Tortonian extension is sealed by the Messinian reflector. The Monte Glio Basin has a relatively simple geometry (Fig. 14), but the mid-Pliocene uplift of the Monte Cristo Ridge tilted the basin (see above). The Punta Ala Basin is a graben where the fan-shaped synrift layers thin alternatively through time towards the east and west (Fig. 15). This basin is underlain by a shallow horizontal detachment (3 and 2 s TWTT Figs 15 and and 16 respectively) also found by Bartole (1995). S o u t h e r n p a r t o f the study area (Fig. 17) Two highs and adjacent basins that trends north-south form the main structures of this area. A mid-Pliocene rifting event (Zitellini et al. 1986) was defined in this area. Calcshists and ophiolites dredged on the two flanks of the Monte Cialdi (Aleria 1980). These samples demonstrated the extension towards the south of the Alpine formations. We found in the western part of this area (Fig. 3) several reflectors dipping towards the west (Fig. 11). These dipping reflections, also noted by Zitellini et al. (1986), could be related to Alpine thrusts with eastwards vergence.
Discussion The tectonic evolution of the North Tyrrhenian Sea is not fully understood in terms of time and space. The collision between Europe and Adria occurred during the late Cretaceous-Eocene. In Corsica the last compressional event is 37 Ma old (late Eocene) (Brunet et al. 1997). During or before the early Oligocene the oceanic Liguride Domain was closed and subducted beneath Corsica (Principe & Treeves 1984; Abbate et aL 1994), whereas the Alpine Corsica overthrust the Hercynian crust. Seismic profile (Fig. 5) north of Elba Island shows that late Eocene thrusts, but with an eastwards vergence, cut the entire crust. The Ligurides Units of the Northern Apennines can be considered as an accretionary prism at a crustal scale (Principe & Treeves 1984; Abbate et al. 1994) related to westwards subduction. The Adria continental crust is involved in the collisional prism (Figs 18 and 19). In other studies (Alvarez 1991; Keller & Coward 1996), the Liguride oceanic crust is not completely consumed after the Eocene epoch. A space problem is related, indeed, to the subsequent opening of the Balearic-Ligurian Basin (Carmignani et al. 1994, 1995). In this basin the rifting starts during the Oligocene (30 Ma) and the onset of oceanic crust is clearly dated (Gorini 1994; Gorini et al. 1993, 1994; Mauffret et al. 1995) by the middle Aquitanian (22 Ma) break-up unconformity. The drifting of the Corsica-Sardinia block occurred (Montigny et al. 1981; Rehault et al. 1984) from the mid-Aquitanian to the Burdigalian (22-19 Ma). The first compression in the modern Apennines is 27 Ma old (Carmigniani & Kligfield 1990). This compression cannot be related to the drifting in the Western Mediterranean Sea (Keller & Coward 1996), but is probably controlled by the subduction of the Adria continental lithosphere. After the emplacement of Alpine nappes in Corsica, Eastern Corsica underwent an extensional event (33-29 Ma and 25 Ma; Brunet et al. 1997) along east-dipping detachments faults (Jolivet etal. 1990, 1991,1994; Daniel etal. 1996). The formation of the Corsica Basin is related to this extensional event. The 5 km thick synrifl formation (Fig. 9) is at least Burdigalian (21-16 Ma) old but the sampling by industrial wells of a thick Oligocene layer beneath Pianosa Ridge suggests that the basal infilling of the Corsica Basin could be Oligocene. Therefore, this basin may have been formed at the same time as the grabens of the European margin before the rotation of the Corsica-Sardinia block. The Adriatic Promontory may be an indenter that collided with the European plate during the Eocene
CRUSTAL STRUCTURE OF THE NORTH TYRRHENIAN SEA
185
Fig. 15. Eastern part of LISA 8 seismic profile. Note several detachments and the dipping of the layered lower crust. A reflection between 7 and 8 s TWTT corresponds to the 22 km deep Moho of the Tuscany margin (Letz et al. 1977a, b; Fig. 4b). The Punta Ala Basin has a complex geometry. Position of the seismic profile is indicated in Fig. 2. (Stampfli & Marchant 1997; Fig. 18a). During the late Oligocene-early Miocene Sardinia and may be Corsica u n d e r w e n t a transpressive
motion with a left-lateral reactivation of the N E - S W H e r c y n i a n faults (Carmignani e t al. 1994, 1995). The Corsica Basin, that shows a
186
A. M A U F F R E T & I. C O N T R U C C I
Fig. 16. LISA 9 seismic profile. Note the shallow detachment beneath the Formiche di Giglio Basin. Observe a deep reflection (white arrow) corresponding to the shallow Moho beneath the Tuscany margin. Position of the seismic profile is indicated in Fig. 2.
CRUSTAL STRUCTURE OF THE NORTH TYRRHENIAN SEA
187
Fig. 17. Eastern part of LISA 12 seismic profile. Monte Etruschi and Cialdi are two horst formed during the middle Pliocene extension (Zitellini et al. 1986). Rocks of Alpine affinity have been dredged (arrows) along the Monte Cialdi scarps (Aleria 1980).
rhomboedric shape (Fig. 3), could have been controlled by these strike-slip faults or preferably by the conjugate N W - S E right-lateral faults (Fig. 18b) before and/or during the rotation of the Corsica-Sardinia block. During the formation of the deep Corsica Basin by extension, the crust was stretched but we do not know if the thinning is local, beneath the basin, or regional because the North Tyrrhenian Basin has subsequently been affected by recent rifting episodes. During and after its formation the Corsica Basin was not affected by the Apennines compression; consequently, at that time, the backstop of the Apennines nappes is located eastwards of this basin probably along the Capraia-Monte-Cristo Ridge. During Tortonian time (10-6 Ma) the whole Tyrrhenian Sea underwent a rifting event (Kastens et al. 1988; Sartori 1990). In the South Tyrrhenian Sea this extensional tectonics preceded the emplacement of oceanic crust whereas
the North Tyrrhenian Sea has yet a continental crust but intruded by plutonic and volcanic bodies. These intrusions were located in the deep and intermediate levels of the crust. The present geometry of the North TyrrhenJan Basin results from a middle Pliocene extension. This extension was coeval to a 3 km ascent of the plutonic bodies that upturned the sedimentary cover of the Corsica Basin (Fig. 19). 0.6 to 1 km Pliocene uplift is also observed on land above the geothermal fields (Mongelli et al. 1991; Marinelli et al. 1993). A regional high heat flow (100 W m -2) indicates that the lithosphere is thin in the study area (Mongelli & Zitto 1991). Very high heat flow anomalies are superimposed on the regional field. These anomalies are related to geothermal fields like Larderello. However, the extensional unloading of the western Elba footwall may also concur to the tectonic uplift (Fig. 19e, right side). The extensional tectonics of the North Tyrrhenian is
188
A. M A U F F R E T & I. C O N T R U C C I
A. Late Eocene
B. Late Oligocene
CRUSTAL STRUCTURE OF THE NORTH TYRRHENIAN SEA characterized by small basins that trend n o r t h - s o u t h separated by transfer faults (Bartole et al. 1991; Bartole 1995). The emplacement of Porto Azzurro granite (eastern Elba) and Larderello intrusions (Camelli et al. 1994) could be related to the Piombino-Faenza transfer fault (Fig. 20) that intersects the N-S to NNW-SSE grabens. A magnetic susceptibility study of Monte Capanne (Boullin et al. 1993) suggests that this pluton was emplaced in a pullapart environment along E - W to N E - S W leftlateral strike-slip faults although an E - W extensional regime is proposed by Daniel & Jolivet (1995). Monte Cristo Island can also be related to the Grosseto-Pienza transfer fault. We observe that Monte Capanne and Monte Cristo correspond to an orientation change of the Capraia-Monte Cristo Ridge and the trend of the C a p r a i a - M o n t e Cristo Ridge and the trend of the ridge south of Monte Cristo Island is compatible with a left-lateral strike-slip fault (Fig. 20) with a small component of extension. These transfer faults and the change of trend of the extensional basins from N W - S E in the north to N-S in the south suggest a pole of rotation located in the north. Although no tectonic rotation since the Messinian is recorded on land by palaeomagnetic studies (Mattei et al. 1996) a Tortonian rotation is probable in the North Tyrrhenian Basin. Several studies (Channel & Mareschal 1989; Serri et al. 1993) suggest that the thinning of the North Tyrrhenian lithosphere is related to a d e l a m i n a t i o n of the A d r i a lithosphere that retreats towards the east. High heat flow, the active role of the plutons and space problems are in favour of this hypothesis but we will wait the results of the land recording of the shots fired during the LISA cruise to propose an interpretation of the deep structure.
Summary and conclusions The LISA seismic cruise investigated the upper sedimentary layers and the deep crust. The results of this cruise are complementary to the industrial seismic lines, where the resolution is
189
good but the penetration too shallow, and the refraction data, which show the deep structure but fail to give a good resolution. The Tertiary history of the North Tyrrhenian Basin began with the Alpine collision in Corsica. Probably the Liguride oceanic crust has been consumed at this time but several authors disagree with this complete subduction of the oceanic crust. After the formation of Alpine nappes in Corsica, this mountain collapsed along an east-dipping extensional detachment. The Corsica Basin is related to this extension, which is probably coeval with the rifting of the Western Mediterranean Basin. The rhombohedric shape of the 10 km deep Corsica Basin suggests a strike-slip component related to late Oligocene-early Miocene transpressional tectonics in the Corsica-Sardinia block. The Corsica basin is preserved from the A p e n n i n e compression that began during the Oligocene, and is characterized in the North Tyrrhenian Basin by crustal thrusts. We have no evidence of tectonic activity in the N o r t h T y r r h e n i a n Basin from the early to midMiocene. A n intense normal faulting event is recorded from the Tortonian to the midPliocene in the grabens of the Tuscany margin and in the southern part of the study area. The late Miocene rifting is contemporaneous with the emplacement of plutonic bodies at intermediate and deep levels in the crust. During the mid-Pliocene extensional event these plutonic bodies rose to the surface and induced a rapid uplift of the sedimentary layers. During this event deep detachments were formed and the lithosphere was t h i n n e d by an hypothetical delamination process. We thank the officers and crew of RV N a d i r for assistance in the LISA project. The authors are grateful to the scientists particularly C. Truffert, and J. Begot who processed on board the LISA seismic data. This study benefited from the Diplome d'Etudes Approfondies presented by S. Poignard. ELF-SNEA(P), CFPTOTAL and IFP kindly provided a seismic line and allow us to publish this seismic profile. We thank C. Faccena and J. M. Daniel for their constructive reviews and helpful suggestions. This work is supported by the IDYL-INSU program. URA 1759 Contribution.
Fig. 18. (a) Late Eocene reconstruction. Position of Corsica-Sardinia block from (Olivet 1987; Gueguen et aL 1993; Mauffret et al. 1995). General tectonic framework from Stampfli & Marchant (1996). A. C., Alpine Corsica. (b) Oligocene-early Miocene reconstruction. Position of Corsica-Sardinia block from (Olivet 1987; Gueguen et aL 1993; Mauffret et aL 1995). General tectonic framework from Stampfli & Marchant (1996). The Corsica Basin, adjacent to the Alpine Corsica, may have been initiated before the rotation of Corsica by normal faulting along the former thrusts of Alpine Corsica (Jolivet et al. 1990, 1991, 1994; Daniel et al. 1996). The Corsica pull-apart basin may be related to the northwards motion of the Sardinia-Corsica block (Carmignani et al. 1994, 1995).
190
A. M A U F F R E T & I. C O N T R U C C I
Fig. 19. (a to c) Sketch of the North Tyrrhenian tectonic evolution. The delamination model is inspired from Serri et al. (1993). (d) Uplifting of the sedimentary layers fill of the Corsica Basin by the ascent of an intrusive body. (e) Left: formation of the Corsica Basin during the Oligocene-Early Miocene. (e) Right: the rise of the intrusive body is coeval to an extensional unloading of the Elba footwall. The horst and graben structures of the Tuscany margin form the hanging wall.
CRUSTAL S T R U C T U R E OF T H E N O R T H T Y R R H E N I A N SEA
191
Llrderelio
Porto Azzurro ~.~
4"
~,
......
~ ....
,
~
~.__:-~.::::::
"-'-
i-:-U.-.'::'-.'-'.-----_ ...4 ............ ~,d,c,~o
',
~
"
~ 'i
. . . . . . . . . .
r
___3
..........
t
sf
~--~2~ ,!
/
Fig. 20. Transfer faults in the North Tyrrhenian Basin (from Bartole 1995). Pull-apart basin interpretation of the Monte Capanne emplacement is from Bouillin et aL (1993). Larderello geothermal field and Porto Azzurro granite are in strike along the Piombino-Faenza transfer fault. Monte Cristo Island is located on the southwestern extension of the Grosseto-Pienza transfer fault. The ridge south of Monte Cristo Island may correspond to left-lateral strike-slip fault with a component of extension.
References
ABBATE,E., BORTOLOTTI,V., PASSERINI,P., PRINCIPI,a. & TREVES, B. 1994. Oceanisation processes and sedimentary evolution of the Northern Apennine ophiolitic suite: a discussion. Memorie della Societd Geologica Italiana, 48, 117-136. ALERIA, A. 1980. Alpine metamorphic rocks and Late Cretaceous rocks and Late Cretaceous-Eocene flysch sediments in the Northern Tyrrhenian Sea: tectonic implications. Centro Grafico, Universith di Parma. ALVAREZ,W. 1991. Tectonic evolution of the CorsicaApennines-Alps studied by the method of Successive Approximations. Tectonics, 10, 936-947. AVEDIK, E, RENARD, g., ALLENOU, J. P. • MORVAN,B. 1993. "Single Bubble" air-gun array for deep exploration. Geophysics, 58, 366-382.
BARTOLE, R. 1995. The North Tyrrhenian-Northern Apennines post-collisional system:constraints for a geodynamic model. Terra Nova, 7, 7-30. - - . , TORELLI,L., MATTEI, G., PEIS, D. & BRANCOLINI, G. 1991. Assetto stratigrafico-structurale del Tirreno settentrionale: stato dell'arte. In: Studi Geologici Camerti, Volume speciale 1991/1, AGIP and CNR, 115-140. BAYER, M., BAYER, R. & LESQUER,A. 1976. Quelques remarques sur la structure g6ologique de la Corse d'apr~s la gravim6trie et le magn6tisme. Bulletin de la SociOt~ GOologique de France, 18,1189-1194. BOULLIN, J. E, BOUCHEZ, J. L., LESHNASSE, E & PgCHER, A. 1993. Granite emplacement in an extensional setting: an AMS study of the magmatic structures of Monte Capanne (Elba, Italy). Earth and Planetary Science Letters, 118, 263-279.
192
A. M A U F F R E T & I. C O N T R U C C I
, POUPEAU, G. & SABIL, N. 1994. Etude thermochronologique de la ddnudation du pluton du Monte Capanne (ile d'Elbe, Italie). Bulletin de la SociOtd GOologique de France, 165, 19-25. BRUNET, C., MONIt~,P. 8`: JOLIVET,L. 1997. Geodynamic evolution of Alpine Corsica based on new Ar/Ar data. Terra Nova, 9, 493. CAMELI, G. M., DINI, I. 8`: LIOTTA, D. 1994. Upper crustal structure of the Larderello geothermal field as a feature of post-collisional extensional tectonics (Southern Tuscany, Italy). Tectonophysics, 413-423. CARMIGNIANI,L . 8`: KLIGFIELD,R. 1990. Crustal extension in the Northern Apennines: the transition from compression to extension in the Alpi Apuanes core complex. Tectonics, 9, 1275-1303. - - , BARCA,S., D1SPERATI,L., FANTOZZI,P., FUNEDDA, A., OGGIANO, G. 8`: PASCI, S. 1994. Tertiary compression and extension in the Sardinian basement. Bollettino di Geofisica Teorica ed Applicata, 36, 45-62. - - , DECANDIA, E A., DISPERATI,L., FANTOZZI,P. L., LAZZAROTTO,A., LIOTTA,D. 8`: OGGIANO, G. 1995. Relationships between the Tertiary structural evolution of the Sardinia-Corsica-Provenqal domain and the Northern Apennines. Terra Nova, 7, 128-137. CARROZZO, M. T. 8`: NICOLICH, R. 1977. Quantitative interpretation of gravity and magnetic data. Bollettino di Geofisica Teorica e Applicata, 19, 236-248. CASSANO, E. 1991. Dati magnetici lungo il profilo CROP 03. In: Studi Geologici Carnerti, Volume speciale 1991/1. AGIP and CNR 4%53. CHANNEL, J. E. Y. 8`: MARESCHAL,J. C. 1989. Delamination and asymmetric lithospheric thickening in the development of the Tyrrhenian rift. In: COWARD, M. E, DmTRICH, D. & PARK, R. G. (eds) Alpine Tectonics. Geological Society, London, Special Publications, 45, 285-302. COLANTONI, R & BORSETTI,A. M. 1974. Geologia e stratigrafia dell'isola Pianosa (Archipelago toscano-Mar tireno). Giornale di Geologia, 29, 287-302. CRAVATI'E, J., DUFAURE, P., PRIM, M. & ROUAIX, S. 1974. Les sondages du Golfe du Lion: stratigraphie et sddimentologie. Notes et Memoire de la CFP, 2, 209-274. DANIEL, J. M. 8,: JOLIVET, L. 1995. Detachment fault and pluton emplacement: Elba Island (TyrrhenJan Sea). Bulletin de la SociOtO G~ologique de France, 166, 341-354. - - , GOFFE, B. & POINSSOq; C. 1996. Crustalscale strain partitioning: footwall deformation blow the Alpine Oligo-Miocene detachment of Corsica. Journal of Structural Geology, 18, 41-59. D'OFFIZI, S., MINELLI, G. & PIALLI, G. 1994. Foredeeps and thrust systems in the Northern Apennines. Bollettino di Geofisica Teorica ed Applicata, 36 91-102. EGGER, A.P. 1992. Lithospheric structure along a transect from the northern Apennines to Tunisia derived from seismic refraction data. Th6se, Universit6 de Lausanne.
--,
DEMARTIN, M., ANSORGE, J., BANDA, E. & MAIESTRELLO,M. 1988. The gross structure of the crust under Corsica and Sardinia. Tectonophysics, 150, 363-389. ELSTER, P., GIGLIA, G., TONGIORGI, M. & TREVISAN,L. 1975. Tensional and compressional areas in the recent (Tortonian to present)evolution of the Northern Apennines. Bollettino Geofisica Teorica ed Applicata, 17, 3-19. FACCENA, C., DAVY, P., BRUN, J. R, FUNICELLO, R., GIARDIM, D., MATTEI, M. & NALPAS,T. 1996. The Dynamics of back-arc extension: an experimental approach to the opening of the Tyrrhenian Sea. Geophysical Journal International, 126, 781-795. FINETrI, L. & MORELLI, C. 1973. Geophysical exploration of the Mediterranean Sea. Bollettino Geofisica Teorica ed Applicata, 15, 263-341. GORINI, C. 1994. GOodynamique d'une marge passive: le Golfe du Lion (MOditerranOe occidentale). Thase de l'Univ. Toulouse, 256 p. --, LE MARREC, A. & MAUFFRET, A. 1993. Contribution to the structural and sedimentary history of the Gulf of Lions, (Western Mediterranean), from the ECORS profiles, industrial seisnaic profiles and well data. Bulletin de la SociOtO GOologique de France, 164, 353-363. , MAUFFRET,A., GUENNOC, E & LE MARREC, A. 1994. Structure of the Gulf of Lions (Northwestern Mediterranean sea): a review. In: MASCLE,A. (ed.) Hydrocarbon and Petroleum Geology of France. European Association of Petroleum Geologists, Special Publications, 4, 223-243. GUEGUEN, E., OLIVET,J. L. & REHAULT,J. R 1993. Kinematics model in the Western Mediterranean: new constraints. Terra Abstracts', 5, 167. HmN, A. & SAPIN,M. 1976. La crofite terrestre sous la Corse: donn6es sismiques. Bulletin de la SociOtO G(ologique de France, 18, 1195-1199. JOLIVET, L., DANIEL, J. M. & FOURNIER, M. 1991. Geometry and kinematics of extension in Alpine Corsica. Earth and Planetary Science Letters, 104, 278-291. - - , TRUFFERT,C. & GOFFI~, B. 1994. Exhumation of deep crustal metamorphic rocks and crustal extension in back-arc regions. Lithos, 33, 30. - - , DUBOlS, R., FOURN1ER,M., GOFFI~, B., MICHARD, A. & JOURDAN, C. 1990. Ductile extension in Alpine Corsica. Geology, 18, 1007-1010. KASTENS, K., MASCLE,J. t~r AL. 1988. ODP Leg 107 in the Tyrrhenian Sea: insights into passive margin and black-arc evolution. Geological Society of America Bulletin, 100, 1140-1156. KELLER, J. V. A. & COWARD, M. P. 1996. The structure and evolution of the Northern Tyrrhenian Sea. Geological Magazine, 133, 1-16. - 8`: PIALLI,G. 1990. Tectonics of the island of Elba: a reappraisal. Bolletino della Societd Geologica Italiana, 109, 413-425. LETZ, H., REICHERT,C. & WIGGER, P. 1977a. Results of two refraction lines in the Northern Apennines. Bolletino di Geofisica Teorica e Applicata, 19, 225-232. - - , - - , - - & GIESE, P. 1977b. Seismic refraction
CRUSTAL S T R U C T U R E OF T H E N O R T H T Y R R H E N I A N SEA measurements in the Ligurian Sea and in Northern Apennines. In: CLOSS, H. & ROEDER, D. (eds) Alps, Apennines and Hellenides. Stuttgart, 413-430. MARINELLI, G., BARBERI,E & CIONI, R. 1993. Sollevamento neogenice e intrusion acide della Toscana e del Lazio Settentrionale. Memorie della Societd Geologica Italiana, 49, 279-288. MATTAUER, M., FAURE, M. & MALAVIELLE,J. 1981. Transverse lineation and large scale structure related to alpine obduction in Corsica. Journal of Structural Geology, 3, 401-409. MATTEI, M., KISSEL, C. & FUNICELLO,R. 1996. No tectonic rotation of the Tuscan Tyrrhenian margin (Italy) since late Messinian. Journal of Geophysical Research, 101, 2835-2845. MAUFFRET,A., PASCAL,G., MAILLARD,A. & GORINI, C. 1995. Structure of the deep Northwestern Mediterranean basin. Journal of Petroleum Geology, 12, 645-666. MONGELLI, E & ZITO, G. 1991. Flusso di calore helle regiona toscana. In: Studi Geologici Camerti, Volume speciale 1991/1. AGIP and CNR, 91-98. , PUXEDDU, M., SQUARCI, P., TAFFI, L. & ZITO, G. 1991. I1 flusso di calore e l'anomalia geotermica dell'area Tosco-Laziale: implicazioni profonde. In: Studi Geologici Camerti, Volume speciale 1991/1. A G I P and CNR, 399-402. MONTIGNY, R., EDEL, J. B. & THUIZAT,R. 1981. OligoMiocene rotation of Sardinia: K - A r ages and palaeomagnetism data of tertiary volcanics. Earth and Planetary Science Letters, 54, 261-271. OLIVET,J. L. 1987. L'origine du bassin nord-occidental de la M6diterran6e du point de vue de la cin6matique des plaques. In: BURRUS,J. & OLIVET, J. L. (eds) Profils ECORS. Golfe du Lion: rapport d'implantation. IFP, Paris, 35 941-1, 10-49. OZRAG-SPERBER,E & PILOT,M. D. 1976. Grands traits du N6ogbne en Corse. Bulletin de la Socidt~ G~ologique de France, 18, 1183-1187. PONZIANI, E, DE FRANCO,R., MINELLI, G., BIELLA, G., FEDERICO, C. &; PIALLI, G. 1995. Crustal shortening and duplication of the Moho in the northern Apennines: a view from seismic refraction data. Tectonophysics, 252, 391-418. PRINCIPE, G. & TREEVES, B. 1984. I1 systema CorsoApennino come prisma d'accrezione. Riflessi sul problema generale del limite Alpi Apennino, I1
193
systema Corso-Apennino come prism d'accrezione. Riflessi sul problema generale del limite Alpi Apennino, 28, 549-576. REHAULT, J. P. 1981. Evolution tectonique et sddimentaire du bassin Ligure (MOditerran~e occidentale). Th6se d'Etat de l'Univ. P. et M. Curie, Paris. - - , BOILLOT, G. & MAUFFRET,A. 1984. The western Mediterranean Basin, geological evolution. Marine Geology, 55, 447-477. SANDWELL,D. Y., YALE, M. M. & SMITH,W. H. E 1995. Gravity anomaly profiles from ERS-1, Topex and Geosat Altimetry. LOS, Transactions of the AGU, 76, 89. SARTORI,R. 1990. The main results of O D P Leg 107 in the frame of Neogene to recent geology of Perithyrrhenian areas: In: KASTENS, K. MASCLE, J. (eds) Proceeding of Ocean Drilling Program, Scientific Results, 107, 715-730. SAUPt~,E, MARIGNAC, C., MOINE, B., SONET, J. • ZIMMERMANN, J. L. 1982. Datation par les m6thodes K/Ar et Rb/Sr de quelques roches de la partie orientale de l'~le d'Elbe province de Livourne, Italie. Bulletin de Min~ralogie, 105, 236-245. SERRI, G., INNOCENTI, E & MANETTI, P. 1993. Geochemical and petrological evidence of the subduction of delaminated Adriatic continental lithosphere in the genesis of the NeogeneQuaternary magmatism of central Italy. Tectonophysics, 223, 117-147. STAMPFLI, G. M. & MARCHANT, E H. 1997. Geodynamic evolution of the Tethyan margins of the western Alps, In: PFIFFNER, O. A., ET AL. (eds) Deep structure of Switzerland. Results from NPF 20, in press. TONGIORGI, E. & TONGIORGI, M. 1964. Age of the Miocene-Pliocene limit in Italy. Nature, 201, 365-367. VIARIS DE LESEGNO, L. 1978. Etude structurale de la met Tyrrh~nienne septentrionale. Th6se 3 ~ cycle, Paris, 6, 1-170. - - , GENNESSEAUX,M. & REHAULT,J. P. 1978. La tectonique n6og6ne et les s6ries s6dimentaires dans le bassin nord-tyrrh6nien. Bulletin de la Soci~t~ G~ologique de France, 20, 29-42. ZITELLINI,N., TRINCARDI,E, MARANI,M. & FABBRI,A. 1986. Neogene tectonics of the Northern Tyrrhenian Sea. Giornale di Geologia, 48, 25-40.
IBS Pannonian Basin project: a review of the main results and their bearings on hydrocarbon exploration FRANK
HORVATH
1 & GABOR
TARI 2
1Geophysical Department, EOtvOs University, H-1083, Budapest, Ludovika tOr 2. 2AMOCO Production Company, 501 WestLake Park Bld, Houston, Texas, USA
Abstract: The IBS Pannonian Basin project presents a good example of fruitful joint activity between Hungarian and other European scientists, and beneficial co-operation of academia and the petroleum industry. This allowed us to achieve significant progress in the understanding of the structure, tectonic evolution, stratigraphic features and hydrocarbon generation in the Pannonian Basin. The Pannonian region has been an integral part of the Alpine belt, and it reveals the complexity of orogenic evolution. Continental to oceanic rifting, followed by convergence, subduction and continental collision shaped the Palaeozoic-Mesozoic substrata of the region. Subsequently, two periods of basin formation (Late Cretaceous and Palaeogene) occurred, most probably in compressional setting. From the earliest Miocene large scale lateral displacement and block rotation took place in the internal domain of the orogen, together with the formation of the Pannonian Basin. This has been characterized by lithospheric extension, however, interrupted by compressional events. The modern Pannonian Basin is in an initial phase of positive structural inversion. All of these tectonic events had significant impacts on the formation and the economic value of the various petroleum systems in the area. Located completely within the E u r o p e a n Alpine belt, the Pannonian Basin has been traditionally an area of intensive geological research, and a classical test site of models to explain areas of subsidence within active orogens. As in many other regions of the world, a significant part of the geological knowledge has come from data acquired during h y d r o c a r b o n exploration. Deliberate search for hydrocarbons has been going on for more than 80 years in this area. The Pannonian basin is now a mature exploration area, and known hydrocarbon reserves together with production in H u n g a r y has undergone a slow but steady decrease during the past decade (Fig. 1). One way to keep exploration prosperous and effective is to make further progress in understanding the structure and evolution of the basin and its orogenic substrata. The aim of this introductory paper is to review the main scientific results achieved during the IBS project and, in the light of this progress, redefine the petroleum systems of the basin. First, a summary is presented about the general geological setting and tectonostratigraphic units. This is followed by a listing of some of the classical geological problems in the Pannonian Basin, which have been addressed and partially solved during the past three years. More detailed reports about recent results in different subjects can be found in subsequent papers in this volume. In the final part of the paper three petroleum systems are outlined. First, potential traps in the
Neogene basin fill, which can be charged by Miocene source rocks are reviewed. Then, reservoirs in the underlying Palaeogene rocks which can contain h y d r o c a r b o n s derived from an Oligocene source rock are reported. Last, but not least, prospect possibilities are discussed on the basis of the distribution of potential traps and source rocks in the structurally complex Mesozoic/Palaeozoic substrata of the basin.
Geological setting and main tectonostratigraphic units The P a n n o n i a n Basin is located in eastern central Europe, and situated inside the European Alpine belt (Fig. 2). At the western margin of the Pannonian basin, the Eastern Alps apparently bifurcate and continue to the SE (Dinarides) and the N E (Carpathians). While the Dinarides constitute a remarkably linear mountain belt, the Western, Eastern and Southern C a r p a t h i a n s form an almost complete loop before continuing into the Balkans. Thus, the Alpine chain is about 300, 1000 and 400 km wide respectively in the Alps, Dinarides-Pannonian b a s i n - C a r p a t h i a n s , and in the H e l l e n i d e s Balkans transects. The pronounced widening in the Pannonian sector is mainly the result of the Neogene extensional basin formation in this area. The surrounding Eastern Alps, Carpathians and Dinarides indeed project below the
HORVATH,F. 8r TARI,G. 1999. IBS Pannonian Basin project: a review of the main results and their bearings on hydrocarbon exploration. In: DURAND,B., JOLIVET,L., HORVATH,E & SI~RANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 195-213.
196
F. HORVATH & G. TARI
PANNONIAN BASIN OVERVIEW
197
Fig. 2. Main geological and geographic units of the Pannonian region and surrounding Alpine-Carpathian-Dinaric mountains. Pannonian Basin, which is superimposed on these A l p i n e fold-thrust belts. While these mountains are characterized by relatively high average elevation (Carpathians: 1500 m; Dinarides: 1000 m; Eastern and Southern Alps: 2000 m), the Pannonian Basin is a lowland with an average height of 150 m above sea level. In this lowland, which is about 400 km wide from N to S and 800 km long from E to W, isolated mountains (or inselberg) emerge from the plain with elevations up to 1000 m. These ranges subdivide the P a n n o n i a n Basin into a number of subbasins. The Vienna Basin is not part of the Pannonian Basin strictly speaking. It is located between the Eastern Alps and the West Carpathians (Fig. 2). The Danube Basin is bounded by the Eastern
Alps and the West Carpathians to the west and north, respectively, while it is bordered by the Transdanubian Range to the south. The Hungarian (southern) part of this basin is called the Little Hungarian Plain. Transdanubia is that part of Hungary which is located to the south and west of the river Danube. It is interesting to know that two thousand years ago this western part of the basin was a R o m a n province for about four centuries, and called Pannonia. Historically, this R o m a n name was extended to apply to the whole basin. The southern edge of Transdanubia is given by the Drava Trough, which is an elongate and curvilinear basin, just like the Sava trough further to the South. The Zala Basin constitutes the southwestern corner of Transdanubia and passes to the Styrian and
Fig. 1. Diagrams showing the history of the cumulative amount of hydrocarbon reserves (a), annual production (b), and a few data to illustrate that Hungary is a mature exploration area (K6kai 1994; Hungarian Geological Survey 1998).
198
E HORVATH & G. TARI
Mura Basins in Austria and Slovenia, respectively. The Great Hungarian Plain is the largest sub-basin of the Pannonian Basin and occupies the eastern portion of the area. The Transylvanian Basin located between the Apuseni Mts and the Eastern and Southern Carpathians (Fig. 2) does not belong in a strict sense to the Pannonian Basin. The separation of the Vienna and Transylvanian basins from the Pannonian Basin can be considered as mainly geographical. There is, however, a more profound geological reason for it: the Vienna and Transylvanian Basins are different in tectonic origin from the Pannonian Basin. The Pannonian Basin is the result of whole-scale lithospheric extension (Tari et al. this volume). In contrast, the Vienna Basin formed by thin-skinned extension, while the Transylvanian Basin is not of extensional origin (Royden 1988). Accordingly, when the name Pannonian Basin is used, this excludes the Vienna and Transylvanian Basins. However, the term intra-Carpathian basins or area refers to all the basins and ranges inside the Carpathian loop. The present-day assembly of the superimposed basins and the underlying orogenic structures suggest a complex origin of the intraCarpathian area. Four major periods of tectonic evolution can be distinguished. (i) Late Permian-Early Cretaceous. Two distinct episodes of continental break-up occurred during this period in the future Alpine-Mediterranean domain of Pangaea. While the Triassic rifting aborted in most of the area, the Jurassic extension led to the formation of the Tethys ocean, flanked by two rifted continental margins: the African-Adriatic (on the south) and the European (on the north). After the rifting, the continental margins and the intervening ocean were controlled by thermal subsidence until the end of Early Cretaceous. (ii) Late Cretaceous-Palaeocene. It was a period of first major compressions in the Alpine system, when many of the oceanic troughs and passive margins disappeared, due to convergence of the continental margins and subduction of the Tethys ocean. During this period, three subperiods of compressional events can be recognized: the Austrian phase (Aptian-Albian), the PreGosau phase (Cenomanian-Turonian) and the
Laramian phase (Maastrichtian-Danian). These events played a decisive role in shaping the structure of the pre-Tertiary strata in the intraCarpathian area. In addition, it was an important period of basin formation, which took place in the Senonian, after the second and before the third compressional event. Accordingly, the Senonian basin fill always represents a seal on the thrusts and folds developed during the Austrian and pre-Gosau phases. (iii) Eocene-Early Miocene. This period represents the second major interval of collision and compression in the Alps, which mostly affected the more external parts. In the internal part, a set of basins developed. The two most remarkable basins are the Palaeogene 'epicontinental' basin and the Szolnok-Maramures 'flysch' basin (Fig. 3). At the end of this period (latest Oligocene-Early Miocene) large-scale lateral displacement (continental extrusion) and/or rotation of internal blocks occurred, which disintegrated the former Alpine fold-thrust belt, and also strongly dismembered the Palaeogene basins. Disintegration includes juxtaposition in the intra-Carpathian area of two Alpine terranes of different early Mesozoic palaeographic position (Yilmaz et al. 1996): the North Pannonian terrane (African-Adriatic continental margin) and the South Pannonian terrane (European continental margin). The boundary of the two juxtaposed terranes is actually a wide zone of intensive early Miocene deformation, called the mid-Hungarian shear zone (Fig. 3). (iv) Mid-Miocene-Recent. Continuing convergence between Europe and Africa has formed further fold-thrust belts in the outermost domains. In the rearranged internal domain the Mid-Miocene was the period when widespread continental rifting initiated the formation of the intra-Carpathian basins. It was followed by the postrift thermal subsidence; however, compressional events causing local basin inversion and fault reactivation also occurred. These broad periods are close to those defined by Trtimpy (1973) in the classic Swiss sector of the Alps. Although there are slight differences in the timing, one can refer to his terms for the major stages outlined above. These are the Early Alpine, the Eoalpine, the Mesoalpine and the Neoalpine stages, respectively. In the intra-Carpathian area, the main Alpine
Fig. 3. Distribution of Palaeogene basins in the intra-Carpathian area. Keys: C, Central Carpathian flysch basin; Sz, Szolnok flysch Basin; Tc, Maramures flysch Basin; K, Krappfeld Basin; S, Slovenian Basin; H, Hungarian Basin; Ts, Transylvanian Basin.
PANNONIAN BASIN OVERVIEW
199
200
F. HORVATH & G. TARI
evolutionary stages led to the formation of four major tectonostratigraphic units. A simplified chronostratigraphic diagram of the area illustrates these major units (Fig. 4). The lower tectonostratigraphic unit is a composite of two different terranes made up of Mesozoic and Palaeozoic successions, which developed during the Early Alpine stage. These are the North Pannonian terrane (often called Alcapa block) and the South Pannonian terrane (often called Tisza-Dacia block). Structurally, this is the most complex unit, as it was affected also by all later phases of deformations. The overlying tectonostratigraphic unit is composed of the Senonian basins, which developed during the Eoalpine stage. They are now preserved in the Little and Great Hungarian Plains and in Transylvania. In these parts of the intra-Carpathian area they represent the immediate post-tectonic cover of the nappes developed in the lower tectonostratigraphic unit. The next tectonostratigraphic unit above comprises the Palaeogene epicontinental basin and the flysch basin formed during the Mesoalpine stage (Fig. 3). The uppermost tectonostratigraphic unit
"
=--1 "~;:;in-
comprises the Neogene Pannonian and Vienna basins which opened and filled during the Neoalpine stage. It is interesting to note that the Transylvanian Basin developed during both the Neoalpine and Mesoalpine stages.
Major problems of the Alpine evolution of the Pannonian basin and progress during the IBS project The Miocene to Quaternary basin fill of the uppermost tectonostratigraphic unit covers almost the entire area and locally is very thick (8 km). Neogene volcanics crop out at the basin margins, but they are also abundant in the subsurface. The middle and lower units are mostly unexposed. There are only isolated blocks where they crop out, particularly in the Transdanubian and North Hungarian Ranges, and Transylvania (Fig. 3). This clearly shows the importance of subsurface geology. Many geological problems in the Pannonian Basin, debated for almost a century, can be resolved only by systematic evaluation of surface and subsurface geological information, combined with acquisition of new data applying modern
P a n n o n i a
z-1
b a s i n
~
"~Transylvanlar~
J
basin
z
~ ~z _
w z
~
r
Trans.
.
~
m ..a
,
z
North Pannonian
(Alcapa
)!
,
'
South Pannonian
(Tisz
")tlll
~ ~
4 km and >6 km depth). Note the coincidence of a large positive Bouguer anomaly with the deep B6k6s basin, contrary to the expectations. See text for detailed explanation.
232
G. TARI E T A L .
nappe structure of the pre-Tertiary basement may partly be responsible for the observed discrepancies (see Szafifin et al. in press), but a complete explanation is most probably hidden in the dynamic processes of the basin's formation.
Lithospheric structure of the Pannonian basin based on geothermal data The Pannonian basin is characterized by high heat flow (Fig. 16). For a description of heat flow determinations the reader is referred to Appendix B. The average near-surface heat flow is about 90 mW m -2, in contrast with a characteristic value of about 60 mW m -~ in the Carpathians and 40 mW m -2 on the Ukrainian shield. The high heat flow in the Eastern Alps is based on a few measurements explained by rapid uplift and erosion (Oxburgh & England 1980). Towards the Dinarides the heat flow is decreasing rapidly, while the Outer Dinarides represent an extremely cold region. The Adriatic Sea is characterized by low, although variable heat flow values. From the Inner Dinarides towards the Pannonian basin, particularly along the Sava and Drava troughs, geothermal highs occur. A large positive heat flow anomaly can be found at the southern part of the Pannonian basin around Belgrade, which continues to the SE along the Vardar zone. The exact areal extent of this anomaly is doubtful because of sporadic data. The Inner Carpathian calcalkaline volcanic regions are also characterized by high heat flow. Simple calculations show that the extra heat cannot be derived from the cooling of magmatic bodies intruded below the subsurface during Miocene and early Pliocene (Horv~th et al. 1986). Such local effect may be expected only in the Eastern Carpathians, where the volcanic activity occurred from late Pliocene to early Pleistocene (Szab6 et al. 1992). The high heat flow is probably the result of elevated temperatures in the deeper crust. The heat flow shown in Fig. 16 also reflects the disturbances caused by groundwater circulation. The outcropping Mesozoic carbonate complex in the Transdanubian Central Range is fractured and karstified and the infiltrating cold meteoric water reduces the heat flow to 40-50 mW m -2. Heat balance calculations (D6v6nyi et al. 1983) show that the background heat flux beneath the Transdanubian Central Range without water circulation is 90 mW m -2. Similar convective systems are active in NE Hungary in karstic areas (Bakk, Aggtelek-Gemer mountains) and in the Outer Dinarides.
The large-scale groundwater flow system, acting in the porous Neogene and Quaternary sediments driven by the difference of hydraulic heads between the recharge and discharge areas, does not alter the regional heat flow significantly. It is because large part of the circulation system is shallow, contained mostly in permeable Quaternary sediments (Erd61yi 1985). The groundwater flow system in the deeper and hotter sediments has an insufficiently low flux, especially vertical flux, to significantly modify the heat flow. The amplitude of the disturbance on the regional heat flow caused by the largescale groundwater system is estimated to be _+10% of the measured heat flow. The rapid Neogene and Quaternary sedimentation decreased the surface heat flow, because sediments have not attained thermal equilibrium yet, and are still colder in the deep subbasins than would be in equilibrium state. The effect of sedimentation on the temperature and heat flow was calculated by solving the heat diffusion equation in vertical direction using finite-difference method (Appendix C). The compaction of sediments was taken into account. The sedimentation history was simplified in the calculations. The Pannonian basin was filled up by a large delta system prograded from NW, N and NE directions into the basin (Vakarcs et al. 1995). We assumed four depositional units: (1) synrift sediments, (2) Lower Pannonian, (3) Upper Pannonian and (4) Quaternary sediments. The Lower Pannonian sediments roughly correspond to prodelta and delta slope sediments and the Upper Pannonian sediments to delta plain sediments. The ages of the Lower and Upper Pannonian units were obtained from the timing of the delta shelf edge progradation given by Vakarcs et al. (1995). All sediments deposited before delta sediments were regarded as synrift sediments. The NW part of the Pannonian basin (Little Hungarian Plain-Danube basin) has been filled up already 10-12 Ma, while the SE part (Mak6 trough-B6k6s basin) only 6 Ma. As a result, in the NW part of the basin the present day heat flow is reduced only by 10%, but in the SE part of the basin the present day heat flow is about 30% less than it would be without sedimentation. The error of the correction, due to the different possibilities for thermal parameters, is 10% of the corrected heat flow (Appendix B), which is within the error of heat flow determinations and estimations. After correction for the Neogene and Quaternary sedimentation the average heat flow in the Pannonian basin increases to 100 mW m -2 and the heat flow
LITHOSPHERIC STRUCTURE OF PANNONIAN BASIN
233
Fig. 17. Heat flow map of the Pannonian-Carpathian region corrected for Neogene and Quaternary sedimentation, Lenkey (in prep). For detailed explanation see text and also Appendix C.
pattern also changes (Fig. 17). Before the correction the deep subbasins were characterized by local heat flow minimum, after correction their heat flow reaches or becomes higher than the average heat flow value of the Pannonian basin. The Pannonian basin is characterized by thin lithosphere and crust (Fig. 7). In contrast, the mountain arc around the basin system has a deep root with thick crust and lithosphere. Comparing these features to the heat flow distribution the overall correlation is good. It was Stegena et al. (1975), who first suggested that the thin crust and high heat flow are related to the formation of the Pannonian basin. Royden et al. (1983b) show that the heat flow and the subsidence in the Pannonian basin may be explained by nonuniform thinning of the lithosphere. To model the high heat flow and high postrift subsidence rate they assume that the mantle part of the lithosphere was almost completely thinned. We used the non-uniform extensional model of Royden & Keen (1980) to derive crustal and mantle thinning factors along two regional sections (Figs 18 and 19), which are prolongations of sections A and B in Fig. 2. The sections were divided into series of boxes of 5 km width
representing lithospheric columns. According to the nonuniform extensional model, the crustal and mantle thinning factors determine the subsidence and heat flow any time after rifting (Appendix D). The model was used in a simplified form and the thinning factors for each box were calculated from the present day mean basement depth (total subsidence until now) and present day mean heat flow of the box, assuming instantaneous rifting 17 Ma. Parameters of the model are given in Table 2. Before calculating the thinning factors, the heat flow was corrected for sedimentation and was filtered by a low-pass filter to remove heat flow anomalies, which had wavelength less than 100 km. The average lithospheric thickness in the Pannonian basin is 60 km (Fig. 7), therefore we assumed that heat flow variations caused by changes in the lithospheric thickness had the same or even larger wavelength. Therefore we r e m o v e d the smaller wavelength anomalies from the heat flow before modelling (Figs 18a and 19a). We assumed that the error of the corrected and filtered heat flow was _+15%, which primarily derives from the error of heat flow determinations and estimations. Minimum and maximum thinning factors were calculated from
234
G. TARI E T A L .
Fig. 18. Thermal model along section AA' (for location see Fig. 15) based on crustal and mantle thinning factors. (a) Heat flow, crosses: heat flow taken from the heat flow map (Fig. 15), circles: heat flow corrected for Neogene and Quaternary sedimentation, taken from map in Fig. 17, solid line: low-pass filtered corrected heat flow, used as input to calculate thinning factors. The error of the corrected and filtered heat flow, which primarily derives from the uncertainty of heat flow determinations and heat flow estimations, is estimated to be +15% and shown by hatching. (b) Thinning factors derived from the present day corrected and filtered heat flow and basement depth using the nonuniform extensional model of Royden & Keen (1980). The error of the thinning factors due to the error of the heat flow is shown by hatching. (c) Present day temperature distribution in the lithosphere calculated using the nonuniform extensional model with the medium thinning factors, shown by solid lines in (b). Thick line in (c) denotes the bottom of the lithosphere taken from Fig. 7. Note the strong correlation amongst the elevated heat flow, high mantle thinning factors and observed thin lithosphere.
the lower and upper error limits of heat flow, respectively (Figs 18b and 19b). The mantle thinning factor strongly depends on the present day heat flow and relatively small change in the heat flow (+15%) causes large variation in the mantle thinning factor (2-8). In spite of the large error, we obtained that similarly to other subsidence and heat flow modelling studies (Royden et al. 1983b; R o y d e n & D6v6nyi 1988; Lankreijer et al. 1995; Sachsenhofer et al. 1997), the mantle thinning factors
are 1.5-6 times higher than crustal thinning factors. The present day temperature distribution in the lithosphere was calculated using the extensional model with thinning factors corresponding to m e d i u m heat flow. The bottom of the lithosphere, taken from Fig. 7, is also shown in Figs 18c and 19c by thick solid line. Comparing the surface heat flow, mantle thinning factors and temperature distribution in the lithosphere to the thickness of the lithosphere, it is evident
LITHOSPHERIC STRUCTURE OF PANNONIAN BASIN
235
Table 2. Summary o f parameters used in the thermal model Parameter Thermal conductivity of upper crust Thermal conductivity of lower crust Thermal conductivity of mantle Heat generation in the upper crust Heat generation in the lower crust Heat generation in the mantle Thickness of the upper crust before rifting Thickness of the lower crust before rifting Thickness of the lithosphere before rifting Temperature of the asthenosphere Volumetric heat capacity Volumetric heat expansion coefficient Duration of rifting
Value 3 W in 1 ~ 2.3 W m a ~ 4 W m -1 ~ 2 • 10 .6 W in -3 0.5 • 10 -6 W m -3 0.01 • 10 .6 W m -3 15 km 20 km 120 km 1333~ 3.85 • 106 J m -3 ~ 1 3.1 • 10 -5 ~ 17.5 - 13.5 Ma
Fig, 19, Thermal model along section BB'. For detailed explanation see Fig. 18. (a) Observed and corrected heat flow. (b) Thinning factors derived from heat flow and basement depth using nonuniform extensional model. (e) Modelled present-day temperature in the lithosphere calculated from the extensional model using the thinning factors shown in (b). Note the strong correlation amongst the elevated heat flow, high mantle thinning factors and observed thin lithosphere.
236
G. TARI E T A L .
that the high heat flow in the Pannonian basin is caused by the thinning of the lithosphere. The area of thin lithosphere very well correlates with the increased temperature in the lithosphere and elevated surface heat flow. According to the definition of the mantle thinning factor (Appendix D), the knowledge of the present day thickness of the lithosphere allows to give an independent estimation for the mantle thinning factor:
[3-
a l i - Yci
al - Yc
where all and Yci are the initial thicknesses of the lithosphere and crust, respectively al and Yc are the present day thicknesses of the lithosphere and crust, respectively. Assuming 120 km initial lithospheric thickness, 35-50 km initial crustal thickness, and 60 km and 25 km present-day m e a n lithospheric and crustal thicknesses, respectively, the resulting mantle thinning factor ranges between 2 and 2.43. Mantle thinning factors derived from modelling are higher, because the base of the lithosphere is controlled by the thermal state of the mantle and its shape and depth are modified by the thermal decay during postrift cooling. Therefore the mantle thinning factor derived from the present day lithospheric thickness is a minimum estimation. In the nonuniform extensional model the heat transfer occurs only in vertical direction. The decay of thermal anomaly, caused by the thinning of the lithosphere is faster due to lateral heat conduction. The thickness of the nondeformed lithosphere might have been around 120 km, which is reduced to 60 km. Thus lateral heat transport is significant in the peripheral areas of the basin, in a distance of about 100 km from the edges. In these areas the mantle thinning factors were likely underestimated. In the central areas the one-dimensional approach is acceptable. Thinning factors were calculated from the present day basement depth and heat flow, thus in the areas of outcrops (Mecsek Mts and Apuseni Mrs in Fig. 19) no thinning factors were derived and no temperature was calculated (Fig. 19). It is evident from the heat flow, the extrapolation of mantle thinning factors and temperature in the mantle that the lithospheric mantle was thinned beneath the Mecsek Mts, located inside the basin, and by lesser extent beneath the Apuseni Mts, being in peripheral position. In calculating the thinning factors standard crustal and lithospheric thicknesses were assumed (Table 2). However, the thinning factors strongly depend on the initial conditions. The effect of variation of initial crustal thickness
on the thinning factors was investigated at the Algy6 high, where metamorphic core complexes were found (Fig. 19c). Increased initial crustal thickness has two i m p o r t a n t effects: it results in elevated land surface before rifting due to isostatic compensation, and elevated initial surface heat flow due to the higher concentration of radioactive elements in the crust compared to the mantle (Fig. 20). Initial elevation was calculated assuming local isostatic compensation at the base of the crust, implying crustal and mantle densities given in Table 2. Calculating the initial surface heat flow steady-state was assumed (Appendix A). Thinning factors were derived the same way as for the lithospheric boxes along the sections, except that the present-day basement depth and heat flow, 2301 m and 121 _+ 15% mW m -2, respectively, were fixed according to the observed (corrected) values at the specific location. The initial crustal thickness, elevation and heat flow were varied. With increasing crustal thickness the crustal thinning factor increases, because starting from higher initial elevation larger subsidence is required to attain to the present day basement
Fig. 20. Variation of thinning factors in the function of increasing crustal thickness. Over 50 km of initial crustal thickness the model fails, because the initial heat flow and the crustal thinning factor become too high, thus the predicted heat flow gets higher than the observed (and corrected) heat flow. Mantle thinning factor estimated from the present day thickness of the lithosphere (see equation in text) suggests that mantle thinning factor was higher than two. It is attained when initial crustal thickness is less than 50 km.
LITHOSPHERIC STRUCTURE OF PANNONIAN BASIN depth (Fig. 20). The decrease of the mantle thinning factor has two reasons: (1) the initial heat flow is higher, thus less mantle thinning is necessary to obtain the present day heat flow, (2) heat flow depends on both the crustal and mantle thinning factors (Appendix D), and due to higher crustal thinning less mantle thinning results in the same present day heat flow. For initial crustal thickness more than 50 km, the initial heat flow and the crustal thinning factor become so high that crustal thinning alone, without mantle thinning, results in higher present day heat flow than observed, thus the model fails. The same failure occurs for initial crustal thickness more than 45 km, if the lower error limit of the present day corrected heat flow is used. The present day lithospheric thickness (compare Figs 7 & 21) places strong constraints on the minimum mantle thinning factor ([3 > 2). According to Fig. 20, mantle thinning factor larger than two occurs, when the initial crustal thickness is less than 50 km. Therefore, modeling shows that initial crustal thickness higher than 45-50 km, and initial heat flow higher than 75 mW m -2 were not likely.
237
Magmatism Note that, besides the voluminous subductionrelated calc-alkaline Mid-Late Miocene volcanics of the Carpathian region (e.g. Szab6 et al. 1992), there are volcanics of the same age (i.e. >10 Ma; P6cskay et al. 1995) with a seemingly different origin. For example, Badenian/Sarmatian volcanic rocks (i.e. trachytes, shoshonites) in the Styrian and Danube basins (Fig. 1) have a generally more alkaline character than their Carpathian counterparts (Ebner & Sachenshofer 1995; Harangi et al. 1995a). The formation of these enigmatic volcanics is better understood in the context of syn-rift extension (Tari & Horvfith 1995). The rise and decompression of lower crustal material underneath areas characterized by extreme extension (core-complex type) may induce melting of lower and middle crustal rocks. Since systematic trace element studies are not available from these alkaline volcanics, this hypothesis remains to be tested. In contrast, the Late Miocene/Pliocene (i.e. 1 Ma cycle (Fig. 12c) was observed. From measuring the average distance between the extreme points of each curve, rough estimates were obtained for the amplitudes of the oscillations of the layer thickness: c. 0.5 m for the >1.0 Ma, c. 0.3 m for
LATE NEOGENE SEDIMENTS, PANNONIAN BASIN
345
Kas.2 Maim
5oi
I
I SB1
i JPLIOCENE ~AGGRADATIONAL U N ~ SB2
//////PPER AGGRADATIONAL UNIT
r
Legend:
2~
77~ Fig. 10. The upper aggradational set of genetic units with elementary cycles from borehole Kaskantyti-2. together with the Pliocene sequence. For legend see Fig. 8. Additional legend: 1, bedded; laminated silt; 2, large-scale cross-bedding; 3, lost section; 4, shell fragments; 5, bioturbation; 6, cycle number.
the c. 370 ka, c. 1 m for the c. 71 ka, and c. 1.5 m for the c. 19 ka cycle. From this, the relative importance of the cyclic events on sedimentation can be estimated (Juhfisz et al. 1997).
Inferences for basin subsidence Thermal subsidence, beginning about 13 Ma ago, followed an earlier phase of rifting (Royden
1988). Pronounced differential subsidence also resulted in several sub-basins, some containing more than 5000 m of sediment. A l t h o u g h chronostratigraphic control is poor for the deeper portions of the sub-basins, deposition probably spanned a similar period to that represented by the four boreholes studied here, attesting to very high subsidence and sedimentation rates in the different parts of the basin.
346
E. JUHASZ E T A L .
----
5 -
4
~aoa] (Tp-i)
-
i
..=
coal
~3._h /
\
/
2 -
~,~
J /
9 \ \ N ,~
~ 'D ~D"D " Q' mD.b q,
"
1
0 1600
/
9 o 9149
o
~
-
~.
\
/ / /
/ /
"
I
I
I
I
I
i
I
1400
1200
1000
800
600
400
200
depth in m
5
-
4
-
! !
0!
(Szh-II) o o "*%
~3-
9 # S #
~**
o0
1
| I
I I
t
%%
J I I I
J
-
I 1000
I 800
I
I
600 depth
400
I 200
I 0
in m
Fig. 11. Thickness trend of the lithological units (sand, silt, clay and coal), revealed by numerical cycle analysis, in borehole Tp-1 and Szh-II. (a) The sand and coal curves and also the silt and clay curves move together and show the same pattern through the borehole. The degree of fitted polynomials is 5. There is a one fourth wavelength shift between the two sets of curves. (b) Sand and coal thickness trends in borehole Szh-II, represented by ten-degree polynomials. For more explanation see text.
The subsidence analyses shown here therefore r e p r e s e n t only the 'shallower' parts of the Pannonian Basin. In subsidence analysis, back-stripping a column of sediment separates the isostatic effects of sediment and water load, from the effects of tectonic subsidence (Steckler & Watts 1978). The data used to calculate tectonic subsidence, according to the techniques outlined by Sclater & Christie (1980), and Bond & Kominz (1984) are outlined below. The software used in these analyses was developed by the Basin Analysis G r o u p at the Free University of A m s t e r d a m (S. Cloetingh & R. Stephenson, pers. comm.).
(1) Stratigraphic time-depth data for the Pannonian Basin are derived from logged drill core and m a g n e t o s t r a s t i g r a p h y (discussed in this paper), and a few K - A r dates on felsic volcanic rocks and tufts. Stratigraphic-depth intervals, for which bracketing age dates are known, and the proportions of main lithological constituents were determined directly from continuous core in the deepest well (summarized in Table 2). The chronostratographic position of unconformities and their corresponding sequence boundaries are shown in Fig. 6. (2) Porosity-depth data, used to correct for changes in compaction and cementation, are not directly available in the boreholes: the standard
LATE NEOGENE SEDIMENTS, PANNONIAN BASIN 6
6th order', average period=19 ka
~
347
~
E
"~
4
a
depth in m
0
i
I
700
I
I
600
i
I
500
i
I
400
I
200
9.44 5 -
I
300
9 . 3 4 a g e in M a
5th order: average period=71 ka
E~ 4 ._q 3-
b
2m
1
I
--
depth in m
0
I
1200 4
-
9.54
I
1000
I
i
I
800
I
I
I
600
I
I
400
200
T
i
0
9.31age in Ma
4th order: average period=370 ka 3rd order
3- 4 0 0 ka
~
1 -
( T ~ l -1 ? 800
[
1
depth into
i
i
i
I
i
i
700
600
500
400
300
200
7. 74
7.0
age
m Ma
Fig. 12. Cyclicities of different order shown by smoothed sand or silt thicknesses. (a) c. 19 ka cycles in borehole Szh-II, the curve was gained by a nine point running average; (b) c. 71 ka cycles revealed by fitting a nine-degree polynomial in borehole Szh-II; (c) c. 370 ky and >1 My cycles represented by a nine-degree and a three-degree polynomials, respectively, in borehole Tp-1. The age control is given from magnetostratigraphic measurements.
exponential p o r o s i t y - d e p t h relations and material parameters of Sclater & Christie (1980) were used. (3) The problem of determining palaeobathymetry, perhaps the most difficult parameter to establish in any basin analysis, is exacerbated in the Pannonian Basin by a highly endemic molluscan fauna. In the analysis we have used conservative estimates of palaeodepths based on facies analyses a n d
extensive palaeoecological studies (MiJller & Magyar 1992a,b). Palaeodepths range from zero where subaerial exposure is indicated (fluvial deposits, palaeosols), 0-10 m for shoreface, delta plain and lagunal-estuarine deposits, up to 50 m for shelf-like and outer delta plain deposits, to a maximum of 200 m for deeper basin and delta slope facies. Backstripping of four of the boreholes (Fig. 13) indicates two patterns of subsidence, again
E. J U H A S Z ETAL.
348
Table 2. Summary of data used in the back-stripping analyses D e p t h to base of unit (m)
Age (Ma)
Sand
Silt
Clay/ Shale
Carbonate
Water depth
Kaskantygt-2 50 MPU 151 SBI 241 SB2 300 410 490 580 700 830 910 1160 SB3 1185
0.1 1 1.9 3.2 3.8 5.5 5.9 6.5 6.7 7.4 8.2 8.9 9.2 11 12.8 14
0.34
0.5
0.16
0
0
0.8
0.1
0.1
0
0
0.5
0.3
0.2
0
0
0.15 0.43 0.17 0.08 0.21 0.13 0.08 0.08
0.85 0.57 0.83 0.8 0.7 0.7 0.62 0.3
0 0 0 0.12 0.09 0.17 0.3 0.47
0 0 0 0 0 0 0 0.15
0
0
0
1
20
0.1 1 1.9 3.2 3.8 5.3 8.8 9.8 12.8
0.86
0.14
0
0
0
0.9
0.07
0.03
0
0
0.47
0.53
0
0
0
0.65 0.01
0.35 0.79
0 0.2
0 0
200 10
0.1 1 1.9 3.2 3.8 6.2 6.4 6.7 7.4 7.9 8.2 8.9 10
0.34
0.5
0.16
0
0
0.71
0.06
0.23
0
0
0.42
0.22
0.36
0
0
0.45 0.62 0.53 0.25 0.48 0.38 0.13
0.28 0.23 0.4 0.61 0.47 0.52 0.77
0.27 0.15 0.07 0.14 0.05 0.1 0.1
0 0 0 0 0 0 0
0 5 5 20 20 50 200
0.1 1 1.9 3.2 3.8 5.2 8.8 9.6 10
0.29
0.53
0.18
0
0
0.44
0.23
0.33
0
0
0.35
0.58
0.07
0
0
0.21 0.1 0.11
0.69 0.1 0.1
0.07 0.8 0.7
0.03 0 0.09
10 20 50 100 100 100 100 200
Jdnoshalma-1 92 MPU 165 SB1 218 SB2 335 538 SB3
Tiszapalkonya-1 50 MPU 128 SB 1 266 SB2 400 510 870 1060 1220 1570 1987
Bglcsalmas-1 50 MPU 145 SB 1 190 SB2 280 495 532
200 200 200
D e p t h correspond to the base of each time unit and measure from the top of each well. Ages also correspond to the base of each unit and unconformity (sequence boundaries SB1, SB2 and SB3, and the middle Pleistocene unconformity MPU, see Fig. 15). The proportion of principal lithologies (sandstone, siltstone, shale, carbonate) are averaged over the thickness of each unit. Water depths (metres) are below sea level.
LATE NEOGENE SEDIMENTS, PANNONIAN BASIN illustrating the differential nature of the basin dynamics. All four wells indicate rapid subsidence between 10 and 9 Ma, corresponding to the basal transgressive, and subsequent deep water deposits of the succeeding highstand. H o w e v e r there are significant differences between the Bficsalmfis-l-Jfinoshalma-1 boreholes, and the Kaskantyti-2-Tiszapalkonya-1 boreholes (note the scale difference in Fig. 13). The actual amount of accommodation space created in the latter two boreholes during this time period (tracked by the basement curve), is almost double that of the Bficsalmfis-1 Jfinoshalma-1 holes. Furthermore, the significant uplift recorded in both the tectonic and basement subsidence curves in the Bficsalmfis-lJfinoshalma-1 boreholes between about 6 and 4 Ma, is not observed in the other two boreholes: in the latter the basement and tectonic curves are markedly divergent, suggesting an additional component of accommodation space to that formed by thermal-isostatic subsidence. After about 4 Ma the pattern of subsidence in the Bficsalm~is-l-Jfinoshalma-1 wells is similar to that in the Kaskantyti-2 and Tiszapalkonya-1 wells. The disparities among the subsidence curves and the sea-level curve further illustrate the
peculiar nature of Pannonian Basin dynamics and the corresponding creation or loss of stratigraphic accommodation space (Royden 1988; Juhfisz 1991). Early subsidence during the Pannonian (s.l.) was interrupted in the Bficsalmfisl-J~inoshalma-1 successions because of differential tectonic uplift and reduction in sediment influx, perhaps related to strike slip movement on fault bounding subbasins (Horvfith 1984). During the same time interval, basement subsidence and high sedimentation rates continued the Kaskantyfi-2-Tiszapalkonya-1 successions. After 4 Ma, thermo-isostatic subsidence seems to have dominated in all four successions. In all four wells, the 6-4 Ma interval corresponds to the SB2 (subaerial) unconformity. Clearly, the dominant factors that determined the style and geometry of Pannonian Basin fill during the Late N e o g e n e were tectonism (including thermo-isostatic subsidence) and sediment influx.
Discussion Subsidence and infilling of the Pannonian Basin occurred mainly during the Late Miocene and Pliocene. The studied boreholes, with full core 0 . _
KASKANTYU-2 g:O.2 v
o 0.6
349
E v
-
o
-
TISZAPALKONYA ~._~
u \
"r~cto~lc +--+
~
I #
+--'
73 m 1.0-
I
1
I
I
I
t
-~
~ "x_~Basementl -I J.- :I ,1 ~I
10
5
time
Base
1.8 10
0
i
i
i
i
i
i
9
8
7
6
8
4
Ma
'1
3
i,
-~----~.
2
1
0
t i m e Ma.
O
I
E v
~
....
JANOSHALMA-1
E 0.2
O r-9
r-
73 "~
_Q -.j
"12
0.4
-~04
o~
I 10
9
8
7
6
5
time Ma.
4
3
2
l
I
I
t 10
I
I
t
[ 5
time Ma.
~
}
I
l 0
Fig. 13. Back-stripped subsidence curves for the Kaskantyti-2, Tiszapalkonya-1, Bficsalm~is-1and Jfinoshalma-1 boreholes. Note the scale differences in the graphs. The tectonic curve tracks the basin subsidence with the isostatic effects of sediment and water loads removed.
350
E. JUHASZ ETAL.
recovery, are sufficiently scattered over the area of the basin to depict a general picture of its evolution, except for the evolution of the eastern margin and the deepest southeastern subbasins. Sequence stratigraphical remarks are summarized on Fig. 6. On the figure the generalized depositional environments are pictured, as function of time, for each section. The curves illustrate the shoreline-shifts, which give information for the relative lake level changes. The major subaerial unconformities and the sequences between them are presented as well. For comparison, the eustatic sea level curve is added (Haq et al. 1987).
Unconformities Magnetostratigraphic data were applied to some of the borehole sections. At SB3 the basal Pannonian sedimentary rocks have a magnetostratigraphic signature that corresponds to an age of 11.8 and 9.7 Ma; the unconformity represents therefore a significant hiatus. The Berhida section is exceptional in this respect, since the sedimentation seems to be continuous between the Sarmatian and the Pannonian. The three unconformities (SB1, SB2, SB3), identified in the basin fill represent significant erosion and a minimum 0.5 Ma and maximum 7.5 Ma time gap. They are interpreted as regional or simple (SB3), and super or composite unconformities (SB2 and SB1). In terms of sequence stratigraphy they can bound third and/or second order sequences (see Fig. 6). Therefore in the studied boreholes two sequences are present: a Late Miocene, which, with the duration of its formation seems to exceed the third-order scale, and a Pliocene one. The unconformity at the Miocene/Pliocene boundary (SB2) seems to correlate well with the Messinian salinity crisis recognized by Hsti (1978).
Sedimentary features The succession of the changing sedimentary environments records the relative water level changes in the Pannonian lake. Between SB3 and SB2 an apparently continuous sequence developed, including four units with distinct stacking pattern: a basal transgressive, a lower aggradational, a middle progradational, and an upper aggradational units. The last mentioned three units may be recognized also on seismic profiles. An idealised profile of the Pannonian Basin fill (Fig. 14) shows the main sedimentological features, the facies distribution, the stacking pattern with the observed thickness
conditions, the time control, derived from the magnetostratigraphic measurements, the bounding major unconformities, and the relative lake-level changes based on the genetic stratigraphic analysis. Above SB3 a relatively rapid transgression is expressed by the landward shift of facies. Subsequently, deep-water environments were established and a distinctive aggradational unit accumulated. The striking basinward shift of the facies, i.e. a drop in relative water level, is a characteristic part of each lake level curve. The second aggradational unit, up to the Miocene-Pliocene boundary (SB2), records an equilibrium between sedimentation rate and basin subsidence (i.e. there were no major changes in the accommodation potential). In the studied boreholes we have not found any sign of deeper water facies capping shoreface or delta front facies, except the thin basal conglomerates and sandstones. The Pliocene sequence is situated between the SB2 and SB3 unconformities. The predominantly fluvial character is strikingly different to the upper part of the Late Miocene sequence. In the studied boreholes, in terms of sequence stratigraphy, there are no complete sequences in either order. The lowstand deposits are lacking or not recognizable by sedimentological methods in deepwater facies. Only thin transgressive and thick highstand deposits are evident. But, as all the studied boreholes are located in the shallow, marginal part of the Pannonian Basin, we can not exclude the occurrence of complete sequences in the deep sub-basins.
Correlation Six boreholes are presented (Fig. 6) for the intrabasinal correlation. They were dated by magnetostratigraphy, using biostratigraphical and radiometric tie-points, enabling to make a good estimation of the age of the sedimentary cycles (Fig. 5). In the case of the Pliocene and Quaternary deposits, the time span of deposition was estimated mostly by biostratigraphical correlation. The recognized facies were assembled into four main facies groups: terrestrial, fluvial, shoreface and offshore sediments. The temporal change of environment at the studied sites was plotted against time in the diagrams (Fig. 6). The diagram illustrates also the global (eustatic) sea level changes as proposed by Haq et al. (1987). These diagrams may be regarded as a rough estimation of relative lake level variations in the former environs of the studied boreholes. The stacking pattern of each borehole shows a
LATE NEOGENE SEDIMENTS, PANNONIAN BASIN
Lithologic and sedimentary features
Age
Ma
Depositional environment
RADAT,ONA ]A
3.2-1.8 -" ~ S B 1
6.2-3.9 --
Shoreline shift and stacking pattern
sea I
mud. ] sd. [gr.
E
land
~
~
2
351
SBI
ALLUVIAL PLAIN
UNIT 40-140m
SB2
~
SB2
AGGRADATIONAL UNIT 90-1300m
DELTA PLAIN
~z BEACH ! !
DELTA FRONT
PROGRADATIONAL UNIT 70-200m
i BASIN & SLOPE
11.2-9.8~
3
AGGRADATIONAL UNIT 100-400m
SB3 TRANSGRESSIVE UNIT 2-10m
Fig. 14. Complex presentation of the stratigraphic, lithological, sedimentological and facies characteristics for the eastern part of the Pannonian basin fill. The cycle analysis was carried out on the Upper Miocene sequence (except the basin and slope facies sediments). The upper part of the Upper Miocene sequence is cyclic; however the cycle thickness and cycle stratigraphy can be variable. The stacking pattern of the genetic units and the shoreline shift indicate two dramatic changes in the life of the Pannonian lake: right above SB3 a rapid and intense transgression, which gave rise to the lake, and the progradational unit with a distinct shoreline shift towards the sea. The ages of the sequence boundaries were determined by magnetostratigraphic measurements. The lithological and sedimentologicalfeatures were identified on the continuous cores of the studied boreholes. For legend see Figs 8 and 9.
general regressive, upward-shoaling trend. The shift of the site of deposition from terrestrial to offshore, then from offshore to shoreline, or terrestrial facies groups, is time-transgressive in the basin. Thus, the shift from offshore to shoreface (or even to fluvial and to terrestrial) facies
groups reflects the prograding delta- and i n t e r d e l t a infilling process. Considering the magnetostratigraphic data, it is evident, that the p r o g r a d a t i o n a l phases in each well do not coincide in time within the basin. The dominant controls of the p r o g r a d a t i o n a l process,
352
E. JUH/i, SZ E T A L .
probably, were the different subsidence in the individual subbasins and the very high sedimentation rate, which overprinted the influence of the water level changes in this scale in the studied area. Most probably the signals of low frequency lake level changes are overprinted by the high sediments influx. Very faint records of low frequency lake level changes were detected only in few cases (see Cycle analysis and Relative water level sections). Relative water level
Numerical cycle analysis revealed similarities and differences in the thickness change of the lithological components. The same thickness trend of the silt + clay and sand + coal, and the shift between the two pairs (see Fig. l l a and b) leads to the question of interpretation of low and high water level positions. Concerning lake-level variations, most literature supposes climatic changes as one of the most important factors. They determine low lake level as a response of aridification which causes coarse bedload at nearshore areas and chemical precipitation in the basin; while in the case of high lake level due to humid climate it is not expected to load coarse and abundant sediment, because of the reduction of the stream gradient and the stabilisation of the bank by vegetation (Swann 1964; Picard & High 1972, 1981; McGowen et al. 1979; Galloway & Hobday 1983; Allen & Collinson 1986; Talbot & Kelts 1989; Dam & Surlyk 1992, 1993; Surlyk et al. 1993 and others). Opposite interpretations of coarse grained sediments to be products of high water level and the result of increased runoff was suggested by Brough (1928) and Olsen (1984), who also proposed mainly chemical precipitation during low lake level. Data from the Pannonian lake sediments support the latter interpretation. The same thickness trend of the sand and coal layers suggests that they both were formed during rising lake-level conditions (see Fig. l l a , b). A number of papers suggested that coal is produced in great thickness in transgressive environments (Galloway & Hobday 1983; Einsele 1992; Riegel 1991). In the case of the Pannonian lake, however, more sand was transported into and deposited in the lake during transgression and high lake-level. The periods of the detected higher order cycles in the Pannonian sediments are very close to, or are equivalent with the climatic cycles discovered by Milankovitch (1940) and Bacsfik
(1944). The values of the Milankovitch frequency band are: 14-18 ka for precession (Berger 1980, 1988), 41 ka for obliquity (Berger et al. 1989), and 100 and 400 ka for eccentricity (Imbrie & Imbrie 1979 in Fischer & Bottjer 1991). In the early Pliocene (between 5 Ma and 2.8 Ma), which is very close in time to the studied Pannonian sequence, Tiedeman et al. (1994) found the obliquity period (41 ka) as the dominant record, while later (3-1.5 Ma) the precessional periodicities (19-23 ka) were dominant. The average period values of the cycles identified in the Pannonian sediments are: c. 19 ka caused by precession; c. 71 ka caused by obliquity and c. 370 ka is the longer period of eccentricity (Fig. 12c). The >1.0 Ma (see Fig. 12c) cycle, which is very faintly developed in the Pannonian sediments, can be an analogue to the third-order global climatic cycle, which causes eustasy. This is expressed by the lower order curve (see Fig. 12c). The rhythm of the oscillation of the lake is very similar to that of the seas, but no evidence of direct or indirect correlation has been found. For a partly similar situation, an inverse correlation was suggested by Semenenko (1987). Studying the records of the past 300 ka of the Caspian sea, he found, that in those periods when the water-level of the world oceans was high, the water-level of the Caspian sea was low and vice versa (Semenenko 1987). The relative amplitudes of the oscillations, i.e. the impact of the cyclic climatic event on the layer thickness suggests that the major part of the regular changes in the layer-thickness was controlled by the precession and obliquity driven climate changes (fifth and sixth order cycles). In other words, among the climatic factors the higher order cycles had the greatest impact on the sedimentation pattern of the Pannonian lake. The fourth (eccentricity) and third order climatic cycles (eustatic in the world oceans) had only a very faint influence on the sedimentation of the Pannonian lake.
Conclusions (1) Basin subsidence curves indicate the different nature of Pannonian Basin dynamics and the corresponding creation or loss of stratigraphic accommodation space. Early subsidence during the Pannonian (s.l.), between about 5.3 and 4.0 Ma, was interrupted in the southern part of the Great Hungarian Plain, while during the same time interval, basement subsidence and high sedimentation rates continued in the northern sub-basins. After 4 Ma, thermo-isostatic subsidence seems to have dominated in all four successions.
LATE NEOGENE SEDIMENTS, PANNONIAN BASIN The 6-4 Ma interval corresponds to the SB2 (subaerial) unconformity, whereas the H a q et al. (1987) curve in the same period shows transgression and sea-level highstand. Clearly, the dominant factors that determined the style and geometry of Pannonian Basin fill during the Late Neogene were tectonics (including thermoisostatic subsidence) and sediment influx. Although the subsidence analysis alone cannot discount possible eustatic effects, it does illustrate that eustasy played no significant role in the sequence-stratigraphic architecture of the Pannonian Basin. (2) On the basis of the detailed sedimentological studies three regional unconformities were detected in the Late Neogene Pannonian Basin fill, which can correspond to third and second order sequence boundaries. Consequently two sequences build up the Late Neogene Pannonian Basin fill: a Late Miocene, which in terms of Vail et al. (1991) can be interpreted as a second order, and a Pliocene, which is a third order sequence. Based on palaeomagnetic data, it is obvious that neither the transgression nor the progradation, recorded in the Late Miocene sequence, coincide in time within the basin. The dominant controls of the latter process probably were the rapidly changing basin subsidence and the very high s e d i m e n t a t i o n rate. The general slow increase in frequency and thickness of silt and sand beds along the sections would fit the idea that the Pannonian lake was infilled by the basinward migration of the marginal facies. (3) There are no similarities between the sealevel curve (Haq et al. 1987) and that of the Pannonian lake level in second or third order scale, suggesting that the sea level fluctuations had no direct influence on the lake level. However, the high frequency (>1 Ma magnitude) water-level oscillations of the Pannonian lake, which were faintly observable in the sedimentation, could have followed the r h y t h m of the eustatic changes of the world oceans. This points to the careful interpretation of the sedimentary cycles of an endorheic lake in terms of sequence stratigraphy, and also to the problems of direct correlation b e t w e e n lacustrine and marine sequences. The Late Miocene fill of the Pannonian Basin shows a distinct, high-frequency cyclicity caused by climatically driven relative water-level changes. In terms of sequence stratigraphy, the series sedimented during these time intervals can correspond to fourth, fifth and sixth order sequences. The water-level fluctuations are evidently caused by cyclic climatic changes of the Milankovitch frequency band, mainly due to
353
variations in precession and obliquity. Eccentricity caused cycles (fourth order) are poorly detectable in the sediments. The longer term climatic cycles (third order) had a very slight influence only on the sedimentation pattern of the Pannonian lake. (4) The SB2 boundary, at the top of the Late Miocene sequence, seems to reflect a major global or Mediterranean event. It may be correlated in time, and probably, causally to the Messinian salinity crisis and Lago Mare event, as well. A similar conclusion was drawn by Csat6 (1992; 1995). The authors express their deep gratitude to the Hungarian Science Foundation (OTKA T 7372 and T 019679), the US-Hungarian Joint Fund J.F.No.329. and the Integrated Basin Studies EC project which supported this research. The study was completed in the framework of a co-operation between the Geological Institute of Hungary and the Geological Survey of Canada. We are grateful to S. Cloetingh and R. Stephenson (Free University, Amsterdam) for permission to use the backstripping software.
References ALLEN, P. A. & COLLINSON, J. D. 1986. Lakes. In: READING H. G. (ed.) Sedimentary environments and.facies. Blackwell Science, Oxford 63-94. BACSAK, G. V. 1944. Die Erdgeschichte der Letztverftossenen 600 000 Jahre. Commentarii de Studiis
Disputationum Instituti Geologici Regii Hungarici, 6 (5). BALOGH, K., ARVA-SOS,E. & PI~CSKAY,Z. 1986. K/Ar dating of post-Sarmatian alkali basaltic rocks in Hungary. Acta Mineralogica-Petrographica, 28, 75-93, Szeged. BARTHA, E 1971. A magyarorsz~igi pannon biosztratigrgfiai vizsg~ilata. [Biostratigraphic investigation of the Pannonian layers, Hungary]. In: GOCZANE & BENKO J. (eds) A magyarorszdgi pannonkori k@z6dm~nyek kutat6sai. Akad6miai Kiad6, Budapest 11-172 [In Hungarian]. BERCZI, I. & PHILHeS,R. L. 1985. Processes and depositional environments within Neogene deltaiclacustrine sediments, Pannonian Basin, Southeast Hungary. In: KIL~NYI t~. • TELEKI P. G. (eds) Geophysical Transactions, Special Edition. EOtv6s Lor~ind Geophysical Institute of Hungary, 31, 55-74 (also US Geological Survey Open-File Report, 85-360). BERGER, A. 1980. Milankovitch astronomical theory of paleoclimates: a modern review. Vistas in Astronomy, 24, 103-122. - - 1988. Milankovitch theory and climate. Reviews of Geophysics, 26, 624-657. , LOUTRE, M. E & DEHAN2; V. 1989. Influence of
the changing lunar orbit on the astronomical fiequences of pre-Quaternary insolation patterns. Paleoceanography, 4, 555-564. BERGGREN, W. A., KENT, D. V., FLYNN, J. J. & VAN
354
E. JUH/~SZ E T A L
COUVERING, J. A. 1985. Cenozoic geochronology. Geological Society of America Bulletin, 96, 1407-1418. - - , SWISHER, C. C. & AUBRY, M.-R 1995. A revised Cenozoic geochronology and chronostratigraphy. - In: BERGGREN,W. A., KENT, D. V., AUBRY, M.-R & HARDENBOL, J. (eds) Geochronology time scales and global stratigraphic correlation. SEPM Special Publications, 54, 129-212. BOND, G. C. & KOMINZ, M. A. 1984. Construction of tectonic subsidence curves of the early Paleozoic miogeocline, southern Canadian Rocky Mountains: implications for subsidence mechanisms, age of breakup, and crustal thinning. Geological Society of America Bulletin, 95, 155-173. BROUGH, E. 1928. On rhythmic deposition in the Yordale Series. Proceedings of the University Durham Philosophical Society, 8, 116-126. COOKE, H. B. S., HALL, J. M. & RONAI, A. 1979. Paleomagnetic, sedimentary and climatic records from boreholes D6vav~inya and Vdszt6, Hungary. Acta Geologica Hungarica, 22, 86-109. CSAT0 ,I. 1992. The Messinian water-level change and the connected stratigraphic traps in the Pannonian basin, Hungary. In: Abstracts, EAPG 4th Conference, Paris 1-5 June 1992, 110. -1995. Sequence stratigraphic interpretations and modelling in lacustrine rift basins; southern Dead Sea basin, Israel and Pannonian Basin, Hungary. PhD thesis, University of South Carolina, Columbia. DAM, G. & SURLYK, E 1992. Forced regression in a large wave - and storm-dominated anoxic lake, Rheatian-Sinemurian Kap Stewart Formation, East Greenland. Geology, 20, 749-752. & -1993. Cyclic sedimentation in a large wave- and storm-dominated anoxic lake; Kap Stewart Formation (Rhaetian-Sinemurian), Jameson Land, East Greenland. Special Publications of the International Association of Sedimentology, 18, 419--448. EINSELE, G. 1992. Sedimentary Basins. SpringerVerlag. ELSTON, D. R, LANTOS, M. & H/~MOR, T. 1990. Az AIf61d Pann6niai (s.I.) k6pz6dm6nyeinek magnetosztratigr~ififija. Annual Report of Hungarian Geological Institute 1988, 109-134. & -1994. High resolution polarity records and the stratigraphic and magnetostratigraphic correlation of late Miocene and Pliocene (Pannonian, s.l.) deposits of Hungary. In: TELEKI, R G., MATTICK, R. E. & KOKAI, J. (eds) Basin Analysis in Petroleum E~ploration. Kluwer Academic Publishers, Netherlands, 111-142. FISCHER, A. G. & BOTTJER, D. J. 1991. Orbital forcing and sedimentary sequences. Journal of Sedimentary Petrology, 61, 1063-1069. GALLOWAY,W. E. & HOBDAY,D. K. 1983. Terrigenous Clastic Depositional Systems. Springer-Verlag. HAMOR, G. (ed.) 1988. Neogene paleogeographic atlas of Central and Eastern Europe. Hungarian Geological Institute, Budapest. HAQ, B. U., HARDENBOL, J. &; VAIL, P. R. 1987. Chronology of fluctuating sea levels since the -
-
Triassic (250 Myr ago to present). Science, 235, 1156-1167. HOMEWOOD, P., GUILLOCHEAU, E, ESCHARD, R. & CROSS,T. A. 1992. Corrdlations haute r6solution et stratigraphie g6ndtique: Une ddmarche int6gr6e. (High resolution correlations and genetic stratigraphy: An integrated approach.) Bulletin du
Centres de Recherche Exploration-Production Elf-Aquitaine, 16, 357-381. HORVATH, E 1984. Neotectonics of the Pannonian basin and the surrounding mountain belts: Alps, Carpathians and Dinarides. Annales Geophysical, 2, 147-154. Hs0, K. J. 1978. When the Black Sea was drained. Scientific American, 238, 5, 53-63. JA.MBOR A. 1980. Pannonian in the Transdanubian
Central Mountains. A Dundntfli - kOz@hegysdg pann6niai kdpz6dmOnyei. A Magyar/i, llami F01dtani Int6zet EvkOnyve, 62. (ed.) 1987. A Magyarorszdgi kunsdgi emeletbeli
--
k~pz6dm~nyek fOldtani jellernz&'e [Geologische Characterisierung der Ablagerung~en der KunsdgStufe in Ungarn]. A Magyar Allami F61dtani ---,
Intdzet I~vk6nyve 69. BAL,a,ZS, E., BALOGH, K. & 24 OTHERS 1987. General characteristics of Pannonian s.1. deposits in Hungary. h~: Proceedings of the 8th Congress,
Regional Committee on Mediterranean Neogene Stratigraphy, Budapest, 15-22 September, 1985. Annals of the Hungarian Geological Institute, 70, 1 5 5 - 1 6 7 .
JUH,a,SZ, E., MtJLLER, P., R1CKETI'S,B., TOTH-MAKK,/~., HAMOR, T., FARKAS-BULLA,J. & SOT0-SZENTAI, M. 1996. High resolution sedimentological and subsidence analysis of the Late Neogene in the Pannonian Basin, Hungary. Acta Geologica Hungarica, 39, 129-152. --, KOVACS, L. O, MI~LLER, R, TOTH-MAKK, /~., PHn.LWS, L. & LANTOS, M. 1997. Climatically driven sedimentary cycles in the late Miocene sediments of the Pannonian Basin, Hungary. Tectonophysics, 282, 257-276. JUHASZ, GY. 1991. Sedimentoiogical and lithostratigraphical framework of the Pannonian (s.1.) sequence in the Hungarian Plain, Eastern Hungary. Acta Geologica Hungarica, 34, 53-72. 1993. Relativ vizszintingadozfisok rdtegtani-szedimentol6giai bizonyftdkai az AlfOid pann6niai s. 1. tileddk6sszlet6ben [Sedimentological and stratigraphical evidences of water-level fluctuations in the Pannonian Lake]. FOldtani KOzlOny, 123, 379-398. K~,ZMER, M. 1990. Birth, life and death of the Pannonian Lake. Paleogeography, Palaeoclimatology, Palaeoecology, 79, 171-188. KOKAY J., HAMOR T., LANTOSM. & MOLLER P. 1991. A Berhida 3. sz. ftirfis paleomfigneses 6s f61dtani vizsgfilata (The paleomagnetic and geological study of borehole section Berhida-3). A Magyar -
-
,4llami FOldtani Int~zet Evi Jelent&e az 1989. Ovr6l. 45-63 [in Hungarian with English summary]. KORPAS-HODI M. 1983. A Dundnt~li-kOzOphegysdg
dszaki el&ere pann6niai mollusca faundjdnak
LATE N E O G E N E SEDIMENTS, PANNONIAN BASIN paleoOkolGgiai ds biosztratigrGfiai vizsgGlata. [Palaeoecology and Biostratigraphy of the Pannonian mollusca fauna in the Northern foreland of the Transdanubian Central range.] A Magyar Allami F/Jldtani Intdzet I~vkGnyve, 66. 1992. A Szombathely II. sz. ftir~is pannGniai (s.1.) molluscai. The Pannonian (s.1.) molluscs of borehole section Szombathely II. Annual Report of Hungarian Geological Institute 1990, 505-525. & POGACSAS,GY. 1992. Paleogeographic outline of the Pannonian s.1. of the southern DanubeTisza interfluve. Acta Geologica Hungarica, 35, 145-163. LANTOS,M. & ELSTON,D. P. 1995. Low- to high-amplitude oscillations and secular variation in a 1.2 km late Miocene inclination record. Physics" of the Earth and Planetary Interiors, 90, 37-53. - - , HAMOR,T. & POGACSAS,GY. 1992. Magneto- and seismostratigraphic correlations of Pannonian s.l. (Late Miocene and Pliocene) deposits in Hungary. Paleontologia i EvoluciG, 24--25, 35-46. MArrvAs J., BURNS,S. J., MULLER R & MAGYARI. 1996. What can stable isotopes say about salinity? An example from the late Miocene Pannonian Lake. Palaios, 96, 31-39. McGOWEN, J. H., GRANATA, G. E. d~ SENI, S. J. 1979. Depositional framework of the lower Dockum Group (Triassic), Texas Panhandle. Bureau of Economic Geology, University of Texas, Austin, Report of Investigation No. 97. MILANKOVITCH,M. 1940. Kanon der Erdbestrahlung und seine Anwendung auf das Eiszeit Problem. Royal Serbian Academy, Beograd, 133. MITCHUM, R. M. & VAN WAGONER, J. C. 1991. Highfrequency sequences and their stacking patterns: sequence-stratigraphic evidence of highfrequency eustatic cycles. Sedimentary Geology, 70, 131-160. MI3LLER, P. d~;MAGYAR,I. 1992a. Continuous record of the evolution of lacustrine cardiid bivalves in the Late Miocene Pannonian Lake. Acta Palaeontologica Polonica, 36, 353-372. & -1992b. A Prosodacnomydk rGtegtani jelentGsGge a K/3tcse kGrnyGki pannGniai s.l. tiledGkekben. Stratigraphic significance of the Upper Miocene lacustrine cardiid Prosodacnomya (KGtcse section, Pannonian Basin, Hungary). FOldtani KOzlOny, 122, 1-38. NAGYMAROSY,A. & MI~LLER, P. 1988. Some aspects of Neogene biostratigraphy in the Pannonian basin. In: ROYDEN,L.H. & HORVATH,E (eds) The Pannonian basin - a study in basin evolution. AAPG Memoirs, 45, 69-77. OLSEN,E E. 1984. Periodicity of lake-level cycles in the Late Triassic Lockatong Formation of the Newark basin (Newark Supergroup, New Jersey and Pennsylvania). In: BERGER, A., IMBRIE, J., HAYS, J., KUKLA, G. & SALTZMAN, B. (eds) Milankovitch and Climate. Understanding the response to Astronomical forcing. D. Reidel Publ. Co., 1, 129-146. PHILLIPS,R. L., REVESZ,I. & BERCZI,I. 1985. Processes and depositional environments' of Neogene deltaiclacustrine sediments, Pannonian basin, Southeast -
-
-
-
355
Hungary: core investigation summary. US Geological Survey Open-file Report, 85-360. & 1994. Lower Pannonian deltaic-lacustrine processes and sedimentation, BGkGs basin. In: TELEKI, E G., MATrICK, R. E. & KOKAI, J. (eds) Basin Analysis' in Petroleum Exploration. Kluwer Academic Publishers, Netherlands, 67-82. PICARD, M. D. &HIGH, L. R., JR. 1972. Criteria for recognising lacustrine rocks. In: RIGBu J. K. & HAMBLIN,W. K. (eds) Recognition of ancient sedimentary environments. SEPM Special Publications, 16, 108-145. & -1981. Physical stratigraphy of ancient lacustrine deposits. In: ETHRIDGE,E G. & ROMEO, I. M. (eds) Recent and ancient nonmarine depositional environments: models for exploration. SEPM Special Publications, 31,233-259. POGACSAS, GY., LAKATOS, L., UJsz/tszI, K., VAKARCS, G., VARKONY1,L., VARNA1,E & REVESZ, I. 1988. Seismic facies, electro facies and Neogene sequence chronology of the Pannonian Basin. Acta Geologica Hungarica, 31, 175-207. --, MA'rrlCK, R. E., ELSTON, D. E, HAMOR, T., JAMBOR, A., LAKATOS,L., LANTOS,M., SIMON,E., VAKARCS, G., VARKONYI, L. & VARNAI, E 1994. Correlation of seismo- and magnetostratigraphy in southeastern Hungary. In: TELEKI, E G., MATTICK R. E. & KOKAI J. (eds) Basin Analysis" in Petroleum Exploration. Kluwer Academic Publishers, Netherlands, 143-160. --, MIJLLER, P. & MAGYAR, I. 1993. The role of seismic stratigraphy in understanding biological evolution in the Pannonian Lake (SE Europe, Late Miocene). Geologica Croatica, 46, 63-69. --, SZAB0, A. & SZALAY, J. 1992. Chronostratigraphic relations of the progradational delta sequence of the Great Hungarian Plain. Acta Geologica Hungarica, 35, 311-327. RIEGEL, W. 1991. Coal cyclothemes and some models for their origin. In: EINSELE, G., RICKEN, W. & SEILACHER,T. (eds) Cycles and events in stratigraphy. Springer, 733-750. ROYDEN, L. H. 1988. Late Cenozoic Tectonics of the Pannonian Basin System. In: ROYDEN, L. H. & HORVA'rH,E (eds) The Pannonian Basin - a study in basin evolution. A A P G Memoirs, 45, 27-48. ROGL, E & STEININGER, E 1983. Vom Zerfall der Paratethys zu Mediterran und Paratethys. Annales der naturhistorisches Museum Wien, 85/A, 135-163. SCLATER, J. G. • CHRISTIE, P. A. E 1980. Continental stretching: an explanation of the post MidCretaceous subsidence of the central North Sea basin. Journal of Geophysical Research, 85, 3711-3739. SEMENENKO,V. N. 1987. Stratigraphycal correlation of the Upper Miocene and Pliocene of the Eastern Parethetys. Ukrainian Academy of Science, Kiev, 'Naukova Dumka' publisher. [in Russian]. STECKLER,M. S. &WATTS,A. B. 1978. Subsidence of the Atlantic-type continental margin of New York. Earth and Planetary Science Letters, 41, 1-13. STEININGER, E E, BERNOR, R. L. d~ FAHLBUSCH, V. -
-
356
E. J U H A S Z ETAL.
1990. European Neogene marine/continental chronologic correlations. In: LINDSAY, E. H., FAHLBUSCH, V. & MEIN, P. (eds) European Neogene Mammal Chronology. Plenum, New York, 15-46. SURLYK, E, NOE-NYGAARD N. & DAM, G. 1993. High and low resolution sequence stratigraphy in lithological prediction - examples from the Mesozoic around the northern North Atlantic. In: PARKER, J. R. (ed.) Petroleum Geology of Northwest Europe: Proceedings of the 4th Conference. Geological Society, London, 199-214. SWANN, D. H. 1964. Late Mississippian rhythmic sediments of Mississippi Valley. AAPG Bulletin, 48, 637-658. TALBOJ, M. R. & KELTS, K. (eds) 1989. The Phanerozoic record of lacustrine basins and their environmental signals. Palaeogeography Palaeoclimatology Palaeoecology, 701-304. TARI, G.,VARKONu L. & VARNA~,E 1992. Third-order Miocene-Pliocene depositional sequences in eastern Hungary, Pannonian Basin. Stratigraphic framework of the post rift sediments in the
Pannonian Basin based on seismic-reflexion, welllog and detailed paleontologic data. In: Confer-
ence of Sequence Stratigraphy of European Basins, Dijon, France. Abstracts, 260-261. TIEDEMAN, R., SARNTHE[N, M. & SHACKLETON,N. J. 1994. Astronomic timescale for the Pliocene Atlantic ~ 8 0 and dust flux records of Ocean Drilling Program site 659. Paleoceanography, 9, 619-638. VAIL, P. R., AUDEMARD,E, BOWEMAN,S. A., EISNER, P. N. & PEREZ-CRIZ, C. 1991. The Stratigraphic Signitures of Tectonics, Eustacy and Sedimentology - an Overview. In: EINSELE, G., RICKEN, W. & SEILACnER, T. (eds) Cycles and Events in Stratigraphy. Springer-Verlag, Berlin, 617-659. VAKAaCS, G. 1997. Sequence stratigraphy of the Cenozoic Pannonian Basin, Hungary. PhD Thesis, Rice University, Houston, Texas. - - , VALE,R R, TAm, G., PO6ACSAS,GY., MATT1CK,R. E. & SZAB0, A. 1995. Third-order Middle Miocene-Pliocene depositional sequences in the prograding delta complex of the Pannonian Basin. Tectonophysics, 240, 81-106.
Role of unconformity-bounded units in the stratigraphy of the continental record: a case study from the Late Miocene of the western Pannonian Basin, Hungary MARCO
S A C C H I 1, F R A N K
HORVATH
2 & ORSOLYA
MAGYARI
3
1Research Institute G E O M A R E SUD, CNR, Napoli, Italy 2Department o f Geophysics, EOtvOs University, Budapest, Hungary Abstract: We present an up-to-date stratigraphic framework for the Late Miocene (postrift) non-marine strata of the western Pannonian Basin, based on unconformity-bounded units as they are derived from seismic interpretation. The data set used for this study consisted of some 1700 km of conventional, multi-channel reflection seismic profiles across western Hungary integrated by 190 km of high-resolution, single-channel seismic profiles acquired on Lake Balaton in June of 1993. Seismic stratigraphic analysis has been constrained by selected geological mapping, well-logs and borehole data. A magnetostratigraphic record was also available from a corehole in the study area, together with recent K/Ar dating of basaltic rocks from the Balaton highland. Five third-order (with 106 year periodicities) stratigraphic sequences have been recognized at regional scale in the Late Miocene succession of the western Pannonian Basin. We have designated these sequences, from bottom to top, as Sarmatian-1 (SAR-1) and Pannonian-1 (PAN-I) to Pannonian-4 (PAN-4). Reliable time constraints were only available for the two maximum flooding surfaces of sequences PAN-2 and PAN-3, namely mfs-2 (9.0 Ma) and mrs-3 (7.4 Ma), and the boundary of sequence PAN-2 (PAN-2 SB) which is approximately dated at 8.7 Ma. PAN-2 sequence boundary is associated with evidence of relative water-level drop in the Pannonian Lake and significant exposure of lake margins that is widely recorded in the so-called 'marginal facies' of western Hungary. The higher rank unit bounded by PAN-1 SB and PAN-4 SB includes most of the Pannonian s.l. succession of the central Paratethys and approximately correlates with the Tortonian-Messinian of the standard chronostratigraphy. Seemingly, no major palaeoenvironmental impact was perceptible in the western Pannonian Basin during the Messinian salinity crisis of the Mediterranean. However a significant change in the regional stratigraphic patterns may be observed since earliest Pliocene (after PAN-4 SB), possibly associated with the very beginning of a large-scale tectonic inversion within the intraCarpathian area. The case of Late Miocene non-marine strata of Pannonian Basin is a textbook example of how single categories of stratigraphic units do not fit (sometimes do not even approximate) chronostratigraphic correlation. The use of unconformity-bounded units offers new insights into the complex and long debated problem of stratigraphic correlation between Late Neogene deposits of the Pannonian Basin and 'similar' non-marine strata of the Central Paratethys realm. Our study shows that the so-called 'Pontian facies' of western Hungary correspond to an unconformity-bounded unit which is older than the Pontian s. s. facies of the stratotype area (Black Sea basin). Accordingly, we suggest that different stages may be used to discriminate between such similar-in-facies but different-in-age strata. We hence recommend the introduction of a new chronostratigraphic unit ('Danubian' or 'Transdanubian') in the Late Miocene series of Central Paratethys and a three-fold subdivision of the Pannonian (s.1.) strata into Early Pannonian (Pannonian s.s.), 'Middle Pannonian' ('Danubian' or 'Transdanubian') and Late Pannonian (Pontian s.s.) stages.
T h e P a n n o n i a n Basin was part of the Paratethys, a separate b r a n c h of the Tethys o c e a n which d e v e l o p e d d u r i n g t h e O l i g o c e n e to P l i o c e n e f r o m t h e w e s t e r n A l p i n e M o l a s s e B a s i n to central Asia ( L a s k a r e v 1924). T h e gradual separation of this vast epicontinental basin f r o m the o p e n o c e a n was m i r r o r e d by severe provincialism of the aquatic faunas, which has long been a m a j o r source of u n c e r t a i n t y for biostratigraphic
c o r r e l a t i o n b e t w e e n P a r a t e t h y a n and Mediterr a n e a n events. Difficulties resulted mostly f r o m the facies d e p e n d e n c e of faunas (Korp~is-H6di 1983; N a g y m a r o s y & Mt~ller 1988) and facies d i a c h r o n i s m within the various P a r a t e t h y s subbasins (Mtiller & M a g y a r 1992). Biofacies provincialism r e n d e r e d necessary the elaboration of a regional stage system for the P a r a t e t h y s p r o v i n c e ( P a p p et al., 1968, 1985;
SACCHI,M., HORVATH,E & MAGYARI,O. 1999. Role of unconformity-bounded units in the stratigraphy of the continental record: a case study from the Late Miocene of the western Pannonian Basin, Hungary. In:
DURAND,B., JOLIVET,L., HORVATH,E & SI~RANNE,M. (eds) The Mediterranean Basins: Tertiary Extension
within the Alpine Orogen. Geological Society, London, Special Publications, 156, 357-390.
358
M. SACCHI E T A L .
Cicha &Senes 1968; RCMNS 1971; Bfildi 1980; R6gl & Steininger 1983; Nagymarosi & Mt~ller 1988; Steininger et al. 1988,1990; Stevanovid et al. 1990; Miiller & Magyar 1992; R6gl 1996). Paratethys stages were based mainly on mollusc assemblages. Correlation among marine episodes in the Paratethys was established by planktonic microfossils; on the contrary, correlation of nonmarine stages of the Paratethys with standard marine stages still remains substantially obscure (Magyar & Hably 1994; Sacchi et al. 1997). Radiometric age determination, magnetostratigraphy and, more recently, sequence stratigraphy have been used to improve stratigraphic correlation within the Pannonian Basin fill (Tari et al. 1992; Csat6 1993; Ujsz~iszi & Vakarcs 1993; Vakarcs et al. 1994). However, absolute age constraints for the Pannonian strata are still poor and a direct correlation of relative water level changes in the Pannonian Lake with the cycle chart of Haq et al. (1987) has revealed to be driven mostly by conceptual presumption rather than adequate documentation. In this paper we present an up-to-date stratigraphic framework for the Late Miocene (postrift) non-marine strata of western Pannonian Basin (Fig. 1) which relies on unconformitybounded units derived from sequence (or genetic) stratigraphic procedure (Vail et al. 1977, 1990; Galloway 1989; Salvador 1994). Our study was based on the interpretation of about 1700 km of conventional, multi-channel reflection seismic profiles across western Hungary (Fig. 2) complemented by 190 km of high-resolution, single-channel seismic profiles acquired during our expedition on Lake Balaton in June of 1993 (Sacchi et al. 1995, 1998). Seismic stratigraphic analysis has been constrained by geological mapping of selected areas, well-logs and borehole data. The magnetostratigraphic log of Iharosberdny-I well (SW Hungary) was also available, together with recent K/Ar dating of basaltic rocks from the Balaton highland (Lantos et al. 1992; Balogh 1995). We also document that the reconstruction of a reliable stratigraphic framework for the Late Miocene non-marine succession of the Pannonian Basin can be only achieved through an integrated stratigraphic approach. This means full combination of biostratigraphy, magnetostratigraphy, radiometric age determination, sequence (or genetic) stratigraphy and classic field study (Vail 1987; Galloway 1989; Weimer & Posamentier 1993; Shanley & McCabe 1994; Miall 1997). A novelty of our work is the use of highresolution seismic data acquired on Lake Balaton in addition to the existing standard exploration seismics in the western Pannonian
Basin. The acquisition of high-resolution seismics on continental areas is possible when seismic profiling is performed on large rivers or lakes. It is because broad band signals can be generated in the water and the fairly low acoustic impedance contrast between the lacustrine mud and water-saturated deeper layers facilitates the propagation of acoustic waves at depth. Lake Balaton in Hungary is the largest natural water surface in central Europe and it offers ideal conditions to apply high-resolution seismics for the detailed study of the underlying Pannonian strata (Sacchi et al. 1995, 1998).
Geological setting The Pannonian Basin is part of a large back-arc depression superimposed on the Alpine megasuture (Bally & Snelson 1980; Horvfith et al. 1981; Tari 1994) (Fig. 1). Recent studies on the geodynamic evolution of the Pannonian Basin have shown fairly complex mechanisms of basin formation and evolution, which include subduction-related extensional collapse of an overthickened nappe pile and collision-related escape of orogenic terranes (Royden & Horv~th 1988; Tari et al. 1992, 1993; Horvfith 1993, 1995; Tari 1994). Early to Mid-Miocene extensional/strike-slip tectonics caused relatively fast synrift subsidence of narrow basinal areas while a significant part of the Pannonian region still remained intact and elevated. Late Mid-Miocene marked the onset of the post-rift phase when thermal cooling resulted in a generalized subsidence and broadening of the whole Pannonian area. Seismic data coupled with well-log interpretation and core-sample analysis, have shown that the Pannonian depression has been filled up by a fluvial-dominated delta system which prograded into a large lacustrine basin (Pogficsfis 1984; Berczi & Phillips 1985; Mattick et al. 1988; Pogficsfis et al. 1988; Horvfith & Pog~csfis 1988; Juhfisz 1994; Vakarcs et al. 1994). Based on the interpretation of seismic profiles and subsidence analysis of boreholes Horvfith (1993), Tari (1994) and Horvfith & Cloetingh (1996) proposed that, during latest NeogeneQuaternary, the Pannonian Basin underwent significant tectonic inversion. Reverse faulting associated with uplift occurred locally and the Hungarian mountains began to rise, thus causing extensive erosion of large areas covered by Late Neogene deposits. A recent study (Horvfith 1995) suggested that tectonic tranquillity of Pannonian Basin was interrupted even earlier in the synrift phase, at the end of Sarmatian, by compressional (locally transpressional) events.
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN
359
Fig. 1. Tectonic map of the intra-Carpathian region and location of the study area.
Neogene geodynamic evolution was accompanied by widespread volcanic activity in the Pannonian Basin. Mid-Miocene to Pliocene calc-alkaline volcanism and Pliocene-Pleistocene alkali-basaltic volcanism resulted in a number of volcanic edifices scattered over the basinal area (Balogh et al. 1986; Szab6 e t al. 1992). During the Late Pliocene-Quaternary, shallow isolated lakes, wetlands, and mostly continental conditions prevailed throughout the basin.
The problem of stratigraphic subdivision of Pannonian s.l. strata. Traditionally, stages and stage systems of the Paratethys have been derived from biostratigraphy complemented with lithologic data. Particularly, the classic subdivisions of Pannonian strata (Roth 1879; L6renthey 1900; Stevanovid 1951) have been mostly based on benthic mollusc assemblages defined within marginal facies of the basin sequence. H6rnes (1851) first described the Pannonian formations of Hungary as Congeria and Paludina beds. The concept of a Pannonian Stage was
introduced by Roth (1879) to designate a relatively monotonous mostly continental sequence, which developed in the Central Paratethys between the Sarmatian (late Mid-Miocene) and the Pleistocene. Since late nineteenth century, the term Pannonian was adopted to include the stages Pontian (Congeria beds), Levantian (Paludina beds) and Thracian (Belvedere beds) when more accurate stratigraphic resolution was impossible. However, both the terms Pannonian and Pontian were used by the majority of stratigraphers as synonyms of Congeria beds. L6renthey (1900) suggested the term Levantian for the upper part of Pannonian s e n s u Roth (1879). This stage was mainly used to designate youngest (late Pliocene to Pleistocene) fluviatile or terrestrial deposits of uncertain stratigraphic position. L6renthey's proposal (1900) included a two-fold subdivision of the Pannonian stage into Lower and Upper Pannonian substages. During the same period, Halavfits (1903) proposed a three-fold subdivision of the Pannonian s e n s u L6renthey (1900) based on Congeria assemblages. However his proposal had little success among Paratethys stratigraphers and was no longer used after that time (Fig. 3).
360
M. SACCHI E T A L .
Fig. 2. Location of seismic profiles used in this study. The detailed seismic grid acquired during our expedition on Lake Balaton in June of 1993 is shown in Fig. 15.
During the first half of the century nomenclature controversies developed among stratigraphers whether the P a n n o n i a n stage or the Pontian stage should be used. SzfideczkyKardoss (1938) suggested a subdivision of Pannonian s e n s u L6renthey (1900) into a lower part (Pannonian s.s.) and an upper part (Pontian and Dacian). Stevanovid (1951) proposed substituting the upper part of the P a n n o n i a n s e n s u L6renthey (1900) with the stage Pontian, based on a presumed coeval appearance of common molluscan species on both sides of the Carpathians. In 1975, a general agreement upon a regional Stage System for the Paratethys was achieved at the VI International Congress on M e d i t e r r a n e a n Neogene Stratigraphy in Bratislava. Accordingly, the Pannonian stage s e n s u Roth (1879) was subdivided into Pannonian s.s., Pontian, Dacian and Rumanian stages and a correlation was proposed with Late Miocene-Pliocene stages of the standard (Mediterranean) chronostratigraphy. Notwithstanding Stevanovid's (1951) redefini-
Fig. 3. Synopsis of Late Neogene chronostratigraphic units for the Central Paratethys according to different authors and correlation with the standard chronostratigraphic scale (after Sacchi et al. 1997). Note the threefold subdivision of Pannonian (sensu L6renthey 1900) strata we adopt in this study (see also Fig. 11 and text for discussion).
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN tion of the Pannonian Stage has been officially accepted since 1971 by the International Union of Geological Sciences, practically all the three 'versions' of the term Pannonian (sensu Roth 1879; s e n s u L6renthey 1900 and sensu Stevanovi61951) are still in use in Hungary (Magyar & Hably 1994). In the everyday practice, for instance, many geologists, including those in the oil industry, still use the 'long' Pannonian stage s e n s u L6renthey (1900), and refer to Lower Pannonian instead of Pannonian s.s., and to Upper Pannonian instead of Pontian. The use of the term Pannonian in the 'traditional' Hungarian sense, (Pannonian s e n s u L6renthey, 1900), although in disagreement with the official Paratethys stage system, was somehow dictated by the extreme difficulty of correlating the Pontian facies of the stratotype area (Black Sea Basin) to the timeequivalent Pannonian strata in Hungary. But it also led to the disagreeable practice of assuming an implicit chronostratigraphic correlation between the local Upper Pannonian stage and the Pontian stage of the official stage system without adequate documentation. The Late Miocene chronostratigraphic framework proposed in this study is reported in the last column to the right of the comparative correlation chart of Fig. 3 and illustrated in detail in Figs 11 and 12. Our proposal was based on integrated stratigraphic approach with specific contribution of unconformity-bounded units (Salvador 1994). According to our interpretation none of the stage names in current use in Hungary adequately represents the middle part of Pannonian s.l. stage (c. 9.0-7.4 Ma in the chronology adopted in this study). Based on recent stratigraphic research (Mt~ller & Magyar 1992, 1995; Sacchi et al. 1997) and the results of our work, we adopt and recommend a three-fold subdivision of the Pannonian s e n s u L6renthey (1900) into Lower Pannonian (Pannonian s.s.), Middle Pannonian ('Danubian' or 'Transdanubian' in Sacchi et al. 1997, 1998) and Upper Pannonian (Pontian s.s.). Our 'Middle Pannonian' partly corresponds to C o n g e r i a u n g u l a c a p r a e and C o n g e r i a b a l a t o n i c a beds (Middle Pannonian s e n s u Halavfits 1903).
Lithostratigraphic framework of western Pannonian Basin Pannonian strata unconformably overlie the Sarmatian sequence in the deepest troughs of the basin while they are transgressive on older rocks at the basin margins Szentgy6rgyi & Juhfisz (1988). Sandy turbidite units (Szolnok Sandstone) interbedded with pelagic marl
361
(Endr6d Marl) represent the lowermost part of the Pannonian succession in the deep basins (Danube, Zala and Drava). Towards the top, sandstone (15jfalu Sandstone) and marl (Algy6 formation) follow, indicating delta slope and delta plain settings. The prograding delta complex is overlain by alluvial deposits (Hansfig and Zagyva formations). In the marginal part of the basin, the Pannonian succession commonly overlies a major unconformity. Turbidites and slope deposits are generally missing and basal transgressive sequences consist of nearshore conglomerates. In the Transdanubian Central Range the Pannonian s./. strata have been subdivided into two major groups (Jfimbor 1980, 1987, 1989), the Peremarton and the Dun~nt~l groups. The Peremarton group, represented by the Os Variegated Clay, Kisb6r Gravel, and Cs~kvfir and Sz~k Claymarl formations, has been traditionally considered as corresponding to the Lower Pannonian s e n s u L6renthey (1900). Similarly the Dun~ntt~l group, made up of the Kfilla Gravel, Soml6, Tihany, Torony, Hansfig and Tapolca Basalt formations, has been considered as corresponding to the Upper Pannonian s e n s u L6renthey (1900). The Tapolca Basalt formation includes Neogene basalt and tuff (Balogh et al. 1986; Balogh 1995) from tens of eruptive centres (J~mbor et al. 1981) located in the Bakony mountains and the Balaton highland. Volcanic rocks of Tapolca Basalt formation display pronounced alkaline character (Balogh et al. 1986; Szab6 et al. 1992; Harangi & Harangi 1995). Late Neogene strata of Pannonian Basin are overlain by Quaternary fluvial deposits whose thickness ranges from a few tens to several hundred metres.
Seismic and sequence stratigraphy of western Pannonian Basin Three regional profiles across western Pannonian Basin are presented, which run in a NNW-SSE direction from south of Lake Balaton down to the Drava Basin toward the borders of former Yugoslavia (Figs 2 and 4-6). Based on methods and procedures of sequence (and genetic) stratigraphy (Vail 1987; Galloway 1989), five third-order (with ]06 year periodicities) sequences have been recognized at regional scale in the post-rift succession. We have designated these sequences, from bottom to top, as Sarmatian-1 (SAR-I)and Pannonian-1 (PAN-l) to Pannonian-4 (PAN-4). Maximum thickness of the sequence stack is on the order of 3 kin.
362
M. SACCHI E T A L .
Fig. 4. Patterns of third-order stratigraphic sequences across western Hungary (section A). Note the occurrence of the Lower Pannonian-Upper Pannonian boundary (as reported by previous authors) along maximum flooding surfaces (see also Fig. 7 and text for discussion). The major unconformity at the base of the post-rift sequence is often associated with a significant stratigraphic gap whose amplitude increases from WSW (Drava Basin) to E N E (Somogy). This may cause amalgamation of more sequence boundaries (Figs 4-6). A cartoon section, based on the interpreted seismic profiles is illustrated in Fig 7. Reliable time constraints were only available for the stratigraphic interval between 9.0 Ma and 7.4 Ma (Figs 4 and 8) which has been calibrated by the magnetostratigraphic record of Iharosber6ny-I well (Lantos et al. 1992; Ujsz4szi & Vakarcs 1993), revised after Cande & Kent (1992, 1995).
Maximum flooding surface mfs-2 marks the peak of a major flooding event, which occurred in the Pannonian Lake at c. 9.0 Ma. This event was manifested by C o n g e r i a c z j z e k i open lacustrine beds (Sz4k fm) which flooded the basin margins. Mfs-2 represents a quasi-isochronous surface at basin scale that can be strikingly correlated with the top of Pannonian s.s. stage (Lower Pannonian sensu L6renthey 1900). Sequence boundary PAN-2 is associated with a significant drop of base level of erosion within the Pannonian Basin at c. 8.7 Ma, which was accompanied by extensive subaerial exposure of the lake margins. This is documented in the
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN
363
9
9 ~,.-4
,-r
O
364
M. SACCHI E T A L .
Fig. 6. Patterns of third-order stratigraphic sequences across western Hungary (section C). Note the amalgamation of SAR-I and PAN-1 sequence boundaries along a major unconformity at the base of Pannonian strata (see also Fig. 7 and text for discussion).
so-called marginal facies of Transdanubia and evidenced by erosional scours due to a several tens of metres lake level fall (Figs 5 & 9). Maximum flooding sequence mfs-3 (c. 7.3 Ma) represents a second important flooding event within the basin which is again characterized by the occurrence of open lake strata, this time associated with a younger Congeria assemblage (Congeria rhomboidea). Mrs-3 represents another useful quasi time line at basin scale and may be considered a good proxy in western Hungary for the base of Pontian as it is defined in the stratotype area (Black Sea Basin). The stratigraphic (genetic) unit b o u n d e d by maximum flooding surfaces mfs-2 and mrs-3 may be correlated with the lower part of the Pontian sensu Stevanovi6 (1951), and regarded as a sort of anticipation of the Pontian s.s. facies in Hungary. As a consequence, it is likely the case that Pontian s. s. (younger than 7 Ma) strata are
practically missing in outcrop in central western Hungary. Sequence boundary PAN-4 (c. 5.0 Ma) is likely to be associated with stratigraphic gap (see also Juh~sz et al. this volume) and significant tectonic overprint as it suggested by the general tectono-stratigraphic patterns within the Neogene Basin fill (Figs 5 & 7). Sequence boundaries SAR-1 SB to PAN-4 SB of this study basically correlate with sequence boundaries III to VII of Ujszfiszi & Vakarcs (1993). The only exception is PAN-2 SB which we find definitely at a higher stratigraphic position with respect to correspondent sequence boundary SB V of Ujsz~szi & Vakarcs (1993). However the absolute ages reported here have been revised in the light of the updated magnetic polarity scale (Cande & Kent 1995; Berggren et al. 1995) and they significantly differ from those reported by the above authors. This would
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN
365
Fig. 7. Tectonic and sequence stratigraphic framework of Pannonian strata based on the interpretation of 1700 km of seismic sections across western Pannonian basin (see Figs 2 and 4-6). Pannonian units Q and @ of this figure respectively correspond to the 'Late Miocene sequence' and the 'Pliocene sequence' of Juhfisz et al. (this volume).
suggests that the age dating proposed by gjsz~szi & Vakarcs (1993) on the basis of a direct correlation of P a n n o n i a n third-order sequences with the eustatic curve of Haq et al. (1987) is not applicable, at least for the stratigraphic interval between 9.0 Ma and 7.4 Ma. The higher rank unit bounded by PAN-1 SB and PAN-4 SB approximately correlates with the Tortonian-Messinian stages of the standard time scale. Seemingly, no major palaeoenvironmental impact was perceptible in the western Pannonian Basin during the Messinian salinity crisis of the Mediterranean. However a significant change in the regional stratigraphic patterns is observed since the earliest Pliocene (after PAN-4 SB), which was possibly associated with the very beginning of a large-scale tectonic inversion within the intra-Carpathian area (Fig. 7). The major unconformity at the base of the Pannonian s.l. strata and PAN-4 SB subdivide the Neogene strata of western Pannonian Basin into two main tectono-stratigraphic units (Fig. 7) which namely correspond to the 'Late Miocene sequence' and the 'Pliocene sequence' of Juhfisz et al. (this volume).
The concept of Lower Pannonian-Upper Pannonian boundary from the perspective of unconformity-bounded units The observation that the Lower P a n n o n i a n U p p e r P a n n o n i a n ( s e n s u L 6 r e n t h e y 1900) boundary is basically a time-transgressive facies b o u n d a r y is not new (Pogficsfis et al. 1988; Jfimbor 1989). In particular, Pogficsfis et al. (1988) had pointed out that the Lower PannonJan-Upper Pannonian (LP-UP) boundary gradually rejuvenates from about 9 Ma to about 6 Ma, as depositional sequences of the Pannonian Basin prograde from NNW to SSE at a regional scale. Nevertheless, our study (Figs 4 ~ & 10) shows that a 'virtual' L P - U P lithostratigraphic boundary is likely to occur at each maximum flooding surface, i.e. it is cyclically repeated at the top of each transgressive system tract of consecutive third-order sequences). The patterns of stacking sequences evidence, in other words, that the L P - U P lithostratigraphic boundary does not move in time gradually, but rather rejuvenates with discrete 'jumps' by stepping one maximum flooding surface after another in the direction of prograding sequences. As a
366
M. SACCHI E T A L .
Fig. 8. Magnetostratigraphic calibration of Pannonian strata at Iharosberdny-I well site (after Lantos et al. 1992; Cande & Kent 1995). The adopted three-fold subdivision of Pannonian s e n s u L6renthey (1900) is based on regional maximum flooding surfaces mfs-2 (9.0 Ma) and mrs-3 (7.4 Ma) (see also Figs 11 and 12 and text for discussion).
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN
367
Fig. 10. Lithostratigraphic versus sequence stratigraphic patterns in the western Pannonian Basin. Note the cyclic 'repetition' of the boundary between 'Lower Pannonian type' and 'Upper Pannonian type' lithofacies along maximum flooding surfaces. Mfs-2 (9.0 Ma) corresponds to the top of Pannonian s.s. (Pannonian sensu L6renthey 1900). Mfs-3 (7.4 Ma) is as a good proxy for the base of Pontian s.s. in western Hungary. The unconformity bounded unit between mfs-2 and mfs-3 represents our 'Mid-Pannonian' stage (after Sacchi et al. 1997). consequence, it is likely the apparent paradox that one could detect a number of 'distinct' L P - U P facies boundaries, which are clearly repeated in time and space, and rejuvenate from NNW to SSE across the same regional stratigraphic section (Fig. 10). The reason that a specific boundary is interpreted as 'the Lower P a n n o n i a n - U p p e r Pannonian boundary' (and then correlated with the Pannonian s.s.-Pontian boundary) seems to lie on the local occurrence of 'proper' conditions in the sedimentary environment, e.g. sharp lithological contrast, associated with transition from 'Pannonian' versus 'Pontian' faunal assemblages. These observations suggest the concept that a boundary between 'facies of Pannonian affinity' and 'facies of Pontian affinity' cannot be correlated unambiguously to a single, distinct lithostratigraphic boundary throughout the Pannonian Basin. In order to avoid confusion and misunderstanding, we recommend that the use of Lower P a n n o n i a n - U p p e r Pannonian boundary in a formal lithostratigraphic sense should be abandoned, or significantly revised.
Difficulties in the application of classic lithostratigraphic and biostratigraphic units for stratigraphic correlation through the Pannonian Basin are mirrored by substantial uncertainty concerning chronostratigraphic position of the boundary between the Pannonian s.s. and the Pontian s.s. stages of the Central Paratethys. Chronostratigraphic miscorrelation due to diachronism of biofacies has already been detected in the Pannonian Basin by Mtiller & Magyar (1992, 1995) who have shown that the 'biostratigraphic Pontian of the P a n n o n i a n Basin' (Pontian sensu Stevanovid 1951), developed nearly 2 million years earlier than the 'biostratigraphic Pontian' of the Black Sea Basin (Pontian s.s. ).
A proposal for a three-fold subdivision of Pannonian
s.1. s t r a t a
Following the results of MUller & Magyar (1992, 1995) we propose that the Pontian of western Hungary (Miiller & Sz6noky 1990), although
Fig. 9. Palaeogeographic sketch-section across western Pannonian Basin at Pan-2 SB (c. 8.7 Ma) showing subaerial exposure at the basin margin due to relative drop of Pannonian lake level. Note the erosional slope at the 'shelf-break' and associated depositional, epigenetic and/or diagenetic features in response to the lowering of base level (after Magyar 1988).
368
M. SACCHI E T A L .
similar to the Pontian facies of the Black Sea, is better understood as a stratigraphic unit sandwiched between the Pannonian s.s. and the Pontian s.s. strata (Sacchi 1997). Such a rock unit is bounded by maximum flooding surfaces mfs-2 (9.0 Ma) and mrs-3 (7.4 Ma) and might be expressed, in terms of chronostratigraphy, by a new stage inbetween the Pannonian s.s. and the Pontian s.s. stages (Fig. 11). Accordingly we recommend a three-fold subdivision of the Pannonian s e n s u L6renthey (1900) into Early Pannonian (Pannonian s.s.), Mid-Pannonian ('Danubian' or 'Transdanubian' in Sacchi et al. 1997, 1998) and Late Pannonian (Pontian s.s.). The introduction of such a 'Mid-Pannonian' may be helpful in enhancing stratigraphic resolution and avoiding prolonged confusion and misunderstanding concerning the apparent diacrhonous nature of the boundary between the Pannonian s.s. and the Pontian s.s. facies of Central Paratethys. It would also fit, as a matter of fact, the present-day 'chronostratigraphic gap' existing between these stages (Figs 11 and 12). A stratotype for our Mid-Pannonian stage might be adequately represented by the Tihany-Fehdrpart section (MOiler & Szdnoky
1990) which is presently used as Pontian faciostratotype in Hungary (Figs 12 & 24). The concept of an intermediate stage between 'Lower Pannonian' and 'Upper Pannonian' is not new. Since early times of stratigraphic study on Pannonian strata, Halavfits (1903) had already defined a 'Middle Pannonian', corresponding to C o n g e r i a u n g u l a c a p r a e and C o n g e r i a balatonica beds, which partly correlates with our Danubian (or Transdanubian) stage.
Acquisition and processing of highresolution seismic data on Lake Balaton During June of 1993, a geophysical survey (cruise GMS93-02) was carried out on Lake Balaton. The aim of the cruise was to acquire high-resolution single-channel seismic reflection data on the Pannonian s.l. succession beneath the Balaton Quaternary deposits. Significant advantage towards a meaningful application of marine-type high-resolution seismics as a tool for the study of Pannonian strata has been offered by the geographic position of
Fig. 11. Late Miocene chronostratrigraphy of Central Paratethys (Pannonian basin). We propose the introduction of a new stage ('Danubian' or 'Transdanubian' in Sacchi et al., 1997, 1998) and a three-fold subdivision of the Pannonian strata sensu L6renthey (1900) into Early Pannonian (Pannonian s.s.), 'MidPannonian' ('Danubian' or 'Transdanubian') and Late Pannonian (Pontian s.s.) stages. Our Mid-Pannonian stage partly corresponds to Congeria ungulacaprae and Congeria balatonica beds (Middle Pannonian sensu Halav~its 1903).
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN
369
Fig. 12. Sequence stratigraphic concept of the 'Mid-Pannonian' ('Danubian' or 'Transdanubian') stage proposed in this study, and the framework of Late Miocene stratotypes of the Paratethys. The Tihany-Feh6rpart section, which is currently used as Pontian faciostratotype in Hungary (Mfiller and Sz6noky 1990), represents a possible stratotype for our 'Mid-Pannonian' stage (see also Fig. 24 and text for discussion).
the lake itself. The Balaton is located in fact at the southern foot of the Bakony Mountains, where the Palaeozoic-Mesozoic basement crops out and the Pannonian sequence pinches out towards NNW. Consequently, the late Miocene stratigraphic record, which in deepest parts of the Pannonian Basin may reach several thousand metres, is here represented (even though incomplete), within a few hundred metres (Fig. 13).
The survey area Lake Balaton covers an area of c. 600 km 2. It has a S W - N E elongated shape (about 78 km in length and up to 14 km in width) and its surface is at 105 m above sea level. It is the largest lake in central Europe, although very shallow. Water depth of the lake is 3-4 m along its entire axis, with the only exception of the tip of the Tihany peninsula where a local 'hole' of 12 m depth is present. The Balaton is part of a relatively young hydrographic system that has formed in the latest Pleistocene-Holocene during the postWtirm deglaciation of Central Europe. The oldest lacustrine deposits have been proven to be as old as approximately 15 000 years Be. by palynological analysis (Nagy-Bodor 1988) as
well as 14C-dating (Cserny et al. 1995). The lake presumably formed after the joining of individual ponds as climatic conditions changed from cold-arid to warm-humid during the Quercus-Fagus vegetation phase (Cserny & Nagy-Bodor 1996). The present-day lake is quite an artificial feature in the sense that its coastline is mostly a constructed e m b a n k m e n t and the water table is kept at a constant level by regulating the water discharge. The origin of the lake has been long debated among geomorphologists (L6czy 1913; Cholnoky 1918; Erddlyi 1961, 1962; Wein 1967; Mike 1980; Marosi & Szil~rd 1981). Several explanations have been proposed, most of which include various erosional processes superimposed on some tectonic control in the broad shape and location of the lake itself (Fig. 14).
Acquisition and processing The 1993 seismic survey was carried out on board of the RV VizvOdelem.The use of a differential GPS with slave stations located at the lakeshore ensured ship positioning with accuracy on the order of 2-3 m. Ship position was recorded each 30 s; fixes were taken each 10 minutes. The total length of acquired profiles is
370
M. S A C C H I E T A L .
Fig. 13. Sketch-section across western Pannonian Basin based on regional seismic profiles (see Figs 2 and 4-6) and compiled after Tari (1994). The unconformity bounded unit between mfs-2 and mfs-3 represents the 'MidPannonian' stage we introduced in this study. Note the erosional truncation of Pannonian strata towards N W due to the Plio-Quaternary uplift of the Bakony mountains (see also Fig. 7 and text for discussion).
Fig. 14. Cholnoky's (1918) model on the origin of Lake Balaton.
around 190 km. The seismic grid is shown in Fig. 15; details of navigation courses are summarized in Table 2. Seismic acquisition was obtained with a U n i b o o m type source (300 J) operating at a shot rate of I s. The streamer was equipped with a group of eight equally spaced hydrophones within an active length of 4.6 m. Boomer and streamer were towed with a lateral offset of 3 m. The average towing speed was 3.5 nml/h (c. 6.5 kin/h). Seismic signal has been acquired on a two-channel VHS hi-fi analogue tape recorder, channel 1 recording the response of the hydrophone group, channel 2 the trigger signal. A bandpass filter of 100 Hz-5 kHz and Time Varying Gain have been applied before printout on an E P C graphic recorder. The recording window was set on 250 ms. Resolution was on the order of 0.5-1.0 m. A scheme of the acquisition system is shown in Fig. 16; acquisition parameters are listed in Table 1. The shallow-water conditions within Lake Balaton imposed severe operating problems as the length of the hydrophone group was in the same the order of the water depth. Thus, water b o t t o m reflections were not summed in phase in the case of high frequencies; in addition direct arrivals interfered with water bottom reflections. In a second stage, the original analogue data set was converted to digital format. The length of digitized record is 300 ms with a sampling rate of 32.052 kHz.
LATE M I O C E N E S T R A T I G R A P H Y , P A N N O N I A N BASIN
371
Tnble 1. Lake Balaton 1993 seismic survey: parameters o f the acquisition system Boomer Hydrophone
Hydrophone group
Recording system
Power Shot rate Sensitivity Pre-amplification Output resistance Input voltage Sensitivity Bandwidth Plotter Analogue recorder
300 J 1s -103 dB V 1 ~bar -1 40 dB 2 kW 9V -63 dB V -1 p b a r 1 100 Hz-10 kHz EPC-3200 PAL VHS hi-fi
Fig. 15. Sketch map of Lake Balaton area showing the location of high-resolution seismic profiles acquired in June of 1993 and main localities and boreholes cited in the text.
Raw data have been converted to SEG-Y format and processed on P r o M A X . The standard processing sequence started with Trace D C Removal and Bandpass Filtering. Minimum phase Butterwoth filter was chosen with 100-1000 Hz corner frequencies and 12 dB/oct slopes. The effect of spherical divergence was corrected by True Amplitude Recovery, using velocities 1480 m/s, 1600 m/s and 2000 m/s in the time ranges of 0-20 ms, 20-50 ms and 50-300 ms respectively (Fig. 17a, b). Remarkable improvement has been achieved by FK Filtering. Fan filters were used with a spatial extent of 50 traces. Fan filter velocities were + 500 m/s and _+ 5000 m/s, corner frequencies were 50 Hz and 1500 Hz. Primary reflections revealed to be often masked by other types of reflections. The most ubiquitous nonprimary event has a 2-3 ms delay and exhibits reverse
polarity. This was due to boomer plate reverberation. Weaker non-primary events with 8-12 ms delay correspond to base-of-mud multiples. Phase correction was applied first with 2 ms operator length, then predictive deconvolution with the same operator length; 2.5 ms prediction distance suppressed the b o o m e r plate reverberation. Lake Balaton mud deposits show an average thickness of the order of 6-8 m. Base-of-mud multiples were also removed by predictive deconvolution (Fig. 17c, d). Time gates were determined to exclude direct arrivals from both predictions while base-of-mud reflections were excluded from the longer prediction. Bandpass filtering was applied after the deconvolution. Coherency filtering improved the signal/noise ratio but e n h a n c e d the multiples, for this reason c o h e r e n c y filtering was applied before the phase
M. SACCHI E T A L .
372
Table 2. Summary o f navigation courses and seismic profiles acquired.on Lake Balaton in June o f 1993 Date of acquisition 17.06.93 17.06.93 17.06.93 17.06.93 17.06.93 18.06.93 18.06.93 18.06.93 19.06.93 19.06.93 19.06.93 19.06.93 19.06.93 20.06.93 20.06.93 20.06.93 20.06.93 20.06.93 21.06.93 21.06.93
Seismic profile
Heading
Length (km)
L-1 (test) L-2 (test) L-11 (test) L - 12 L - 13 L -4 L-9 L-6 L - 10 L - 10/11 L - 11 L-11/12 L - 3/4 LW - 1 LW - 2 LW - 3 LW - 4 LW - 5 LW - 6 L -7
W-E E-W N-S S-N N-S E-W N-S W-E N-S S-N N-S S-N E-W NE-SW S-N N-S S-N W-E W-E W-E
2.0 2.5 11.5 12.0 9.0 16.0 8.0 19.5 8.0 9.5 11.5 9.0 7.5 11.0 6.5 7.5 5.5 15.0 16.0 11.0
Lake Balaton 1993 Survey Acquisition System
"fzzq
~
vLr~ Channel 1 Channel 2
~.._,V(~H~ ;L Re(~
Fig. 16. Scheme of the acquisition system during the 1993 high-resolution survey on Lake Balaton.
correction. The final step of the processed sections includes Trace Mixing (Fig. 17d).
Seismic and sequence stratigraphy of the Pannonian strata beneath Lake Balaton T h e 1993 survey i n v e s t i g a t e d L a k e B a l a t o n subsurface to a m a x i m u m t i m e - d e p t h of c. 180 ms d o w n to t h e acoustic b a s e m e n t , w h i c h is r e p r e sented by Sarmatian marl and marly limestone
(Sacchi et al. 1995, 1988). C o n s t r a i n t for seismic stratigraphic i n t e r p r e t a t i o n was given by boreh o l e s T i h a n y - 6 2 , Si6fok-3, Balatonf61dv~irMHSz, and fieldwork on selected areas (B~intapuszta, Papv~is~irhegy, Tihany, K6v~ig66rs) in the s u r r o u n d i n g s o f the lake (Fig. 15). Calibration of seismic profiles was o b t a i n e d b y the i n t e r s e c t i o n of seismic profile LW-1 with interpolated stratigraphy between boreholes Tihany-62 and BalatonfOldv~ir-MHSz. S e q u e n c e
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN stratigraphic interpretation of the Balaton seismic profiles (Figs 18-22) was aided by the study of the regional seismic sections across western Hungary and Iharosber6ny-I well-log and magnetostratigraphy (Figs 4-6 & 13). Interpreters should keep in mind that vertical exaggeration for the seismic data set is about ten-fold. Two major unconformities, both associated with significant stratigraphic gap have been
(a)
(b)
373
detected in the subsurface of Lake Balaton: (a) an upper unconformity at the base of the Balaton Quaternary deposits; (b) a lower unconformity at the top of the pre-Pannonian (Middle Miocene) sequence. These two unconformities separate, from bottom to top, three major seismic-stratigraphic units (Fig. 23): (1) PrePannonian strata; (2) Pannonian s.l. (Late Miocene) sequence; (3) Late PleistoceneHolocene deposits of Lake Balaton.
374
M. SACCHI ETAL.
(c)
(d) Fig. 17. Sample of high-resolution seismic profile across Lake Balaton (section L-6, detail): (a) analogue (EPC printout) record; (b) digitized section after trace DC removal, bandpass filtering and TAR; (c) processed section without coherency and F-K filtering; (d) final processed section, including trace mixing.
Pre-Pannonian strata Pre-Pannonian (Sarmatian) strata are dramatically truncated at the top by a mature (polycyclic) erosional surface. Stratigraphy at Tihany-62 site (Fig. 24) indicates that this
surface marks a significant stratigraphic gap (about 12 to 9 Ma) spanning from the Upper Sarmatian to Congeria czjzeki beds (upper part of Pannonian s.s.). The gap includes amalgamation of two third-order sequence boundaries, namely SAR-1 SB (top of Sarmatian sequence)
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN
375
9
~E ~13 r
o~9- ~
~
9
~1~+ ...~
r,~
z.-~
"go~ 9
O,.~
9
.o .~ a,
376
M. SACCHI E T A L .
9,,.-i
r~
9
~.~
, ,...~
~~ z..~
::
>~0
,
,x::l " ~~ 9,0~ ~.~ 9 m 0 "
~
~9
~D
~
-~,~.~ r~
~
"~ ~o ,
,.~
E.S : ~
>
"7.
LATE MIOCENE STRATIGRAPHY, P A N N O N I A N BASIN
377
O +,,a
9
~Z ..=
e~t)
9
o
Z
u?, 9 ~,,,,,~ t~
o
,,,,~
O ~ O
eq
378
M. SACCHI E T A L .
9
i
r~
2;
O +,,a
O
O
9
+6 Z ,,...z,
.~t-q
E=
, ,,,,~
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN
379
-6 O
. ,...,,~ 9
6 9
. ,,,,~
"5 . ,,,,,.
,a=
9
Z~
~Z
. ,,.,~
e4"~ e'.l
380
M. SACCHI ETAL.
Fig. 23. Stratigraphic framework of Neogene-Quaternary strata beneath Lake Balaton and correlation with Neogene sequence stratigraphic units of the western Pannonian Basin (see also Fig. 7 and text for discussion).
and PAN-1 SB (top of first Pannonian thirdorder sequence) (Fig. 23). The duration of this hiatus increases from SW to NE across Lake Balaton area, to include the whole Sarmatian. Good exposures at this stratigraphic level are found at Bfintapuszta, where Pannonian s.s. lacustrine strata directly overlie Badenian marine deposits. Incised valleys and associated fluvial terraces and channel-fills are also recognized from the seismic record (Fig. 25). Channel-fills within valleys are bounded at the top by a transgressive surface and may be interpreted as lowstand systems tract (LST) deposits (Shanley & McCabe 1994).
Pannonian s.l. (Late Miocene) succession beneath Lake Balaton Maximum thickness of this unit is in the order of 200 m. The lowermost part of the Pannonian sequence is represented by the transgressive systems tract (TST) deposits of the Szfik formation (Congeria czjzeki beds). These deposits consist of open lacustrine grey clay-marl and siltstone which directly onlap the pre-Pannonian
basement. The top of Szfik formation is bounded by mrs-2 (c. 9.0 Ma) that correlates with the top of Pannonian s.s. stage. The Soml6 formation and the lower part of Tihany formation follow, which are interpreted, in turn, as early progradational and late progradational-aggradational highstand systems tract (HST) deposits. The Tihany formation developed partly in nearshore areas, mainly lagoons, at shallower depth than the underlying Soml6 formation. Stratigraphic architecture of these deposits is characterized by forestepping strata which downlap above the underlying TST deposits and are accompanied by local development of small coarse-grained prograding deltas (Figs 18, 20, 24). Good exposures of this sequence are found at Tihany-Feh6rpart. Notwithstanding these strata are reported as Pontian faciostratotype in Hungary (Mialler & Szdnoky 1990), it is to be noted they are Danubian (or Transdanubian) in age according to the chronostratigraphy adopted in this study (Fig. 24). A subtle third-order sequence boundary (PAN-2 SB) has been also detected within the Tihany formation, the upper part of which corresponds to the LST of the sequence above.
LATE MIOCENE STRATIGRAPHY, P A N N O N I A N BASIN
381
o,..~ ~ ,...--.,
"~cq
r.g/ ,,--,-x
9 ..~
r.~ O
m,..~
9
9
*..o 9
9
O
382
M. SACCHI E T A L .
Fig. 25. Enlarged detail of seismic profile L-11/12 showing evidence of v-shaped valleys incised within the prePannonian (Sarmatian) strata and filled up by lowstand systems tract (LST) deposits of sequence PAN-2 (see Fig. 22). The unconformity at the base of Pannonian strata marks a major stratigraphic gap spanning from c. 12 Ma to c. 9 Ma) and includes amalgamation of SAR-I+ PAN-1 sequence boundaries. Note small mound-shaped features along PAN-2 SB (see also Fig. 27 and text for discussion).
Sequence boundary PAN-2 may be followed as a regional unconformity on seismic profiles all through western Hungary. It is related with a significant lowering of base level in the Pannonian Lake and is possibly associated with tectonic and volcanic activity (Tapolca basalt and Kabhegy formations). PAN-2 SB crops out at the top of Tihany Peninsula. Its occurrence (c. 8.7 Ma) predates the onset of basaltic eruption of the Tihany Volcano (c. 7.8 Ma, Balogh 1995). Stratal patterns towards igneous bodies inferred from seismic profiles suggest that most near-surface magmatic intrusions were coeval and/or slightly postdated the deposition of the Sz~k formation (c. 9.0 Ma). This is also suggested by 'anomalous' truncation of strata at the top of the TST, possibly due to local uplift related to magmatic intrusion (Figs 18, 24 & 26). Scattered over the top of Tihany Peninsula, a number of silicified carbonate mounds are found that are currently believed to be P l i o c e n e Pleistocene in age. These deposits (geyserite of the Hungarian literature) have been so far interpreted as purely chemical deposits related to post-volcanic activity (L6czy 1913). According
to our interpretation they are better interpreted as silicified travertine mounds developed at warm/hot springs. Furthermore, a number of mound shaped features have been detected from the seismic record which can be traced down to the Balaton subsurface along PAN-2 SB (Figs 19, 22, 23, 24, 27). We propose these features may be correlated with the Tihany travertine mounds which, accordingly, would be Late Miocene in age (Sacchi et al. 1995, 1998). According to our sequence stratigraphic framework, we also suggest that the K~lla formation, a coarse-grained foreshore deposit that crops out in the K~I Basin, north of Lake Balaton, is a facies heteropy of the lower part of Tihany fm., being itself part of a HST progradational-aggradational unit that formed along the Pannonian lakeshore between 9.0 and 8.7 Ma. We speculate that significant part of K~lla deposits in this area were possibly removed by erosion during the lake level fall which caused PAN-2 SB to occur. One puzzling feature of the K~lla formation is certainly the k6tenger (sea of stones), a 2 m thick silicified sandstone and conglomerate bed. We
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN
383
Fig. 26. Enlarged detail of seismic profile L-4 showing evidence of near-surface magmatic intrusion beneath Lake Balaton. Stratal patterns within the Pannonian succession, suggest the intrusions are coeval and/or slightly postdate mrs-2 (9.0 Ma).
suggest the k6tenger may be interpreted as a groundwater silica-cemented regolith (groundwater silcrete), (A. Mindszenty pers. comm.) and regarded as an epigenetic feature that developed at the phreatic-vadose interface in response to the base-level drop within the Pannonian Lake at Pan 2 SB (8.7 Ma). In other words the k6tenger may be seen as a 'sequence stratigraphic equivalent' of the silicified travertine mounds of Tihany peninsula (Fig. 9). The occurrence of silcrete would also suggest relatively warm (dry/humid seasonal) climatic conditions during that time.
Fig. 27. Enlarged detail of seismic profile L-6 showing mound-shaped features along PAN-2 SB (c. 8.7 Ma). We interpreted these features as travertine build-up, similar to those ~ropping out at the top of Tihany Peninsula (Geyserite of the Hungarian literature).
Late Pleistocene-Holocene deposits of Lake Balaton This unit is mostly represented by the deposits of Lake Balaton; it also includes thin patches of older (Late Quaternary) fluvial clastics at its base that have been documented by drillings
384
M. SACCHI E T A L .
(Cserny & Corrada 1989), but cannot be solved seismically. Lake Balaton deposits display an average thickness of c. 5 m and consist of silt and subordinately clay and fine sand with a carbonate content of 50-70% (Cserny & Corrada 1989). Seismic response of these deposits is characterized by parallel-continuous reflectors. The unconformity at the base of this unit corresponds to a subaerial erosional surface which dramatically truncates the underlying Pannonian strata (Figs 18-22, 24). It marks a major stratigraphic hiatus that practically encompasses the whole Pliocene and most of the Pleistocene (Fig. 23).
Tectonic interpretation Seismic profiles across western Hungary indicate that the Late Miocene strata of western Pannonian Basin underwent significant post-rift tectonic deformation (Figs 4-6, 18-22, 24). Regional scale tectonic deformation is evidenced by severe tilting of the whole Neogene sequence of western Pannonian Basin towards ESE. This is clearly seen on seismic profiles that consistently show E S E dipping strata which are erosionally truncated at the top. In order to explain this
tectono-stratigraphic setting Tari (1994), Horvfith (1995) have suggested that the Pannonian Basin underwent significant late-stage tectonic inversion and uplift with consequent subaerial erosion of Late Neogene strata in Transdanubia and Northern Hungary (Tari 1994; Horvfith 1995). Based on our interpretation we propose that the onset of this tectonic inversion could be placed at the very beginning of the Pliocene (Fig. 7). An older compressive/transpressive tectonic phase can be detected from regional seismic profiles at about the end of Sarmatian (Horvfith 1995). This may account for the significant and almost ubiquitous stratigraphic gap that we have documented at the base of Pannonian strata. Small scale tectonic deformation in the area of Lake Balaton is expressed by S W - N E strike-slip faults and associated folds which postdate the 'Middle Pannonian' succession of Transdanubia (Figs 18-22, 24, 28). Evidence of minor tectonic activity also exists between mrs-2 and PAN-2 SB (9.0-8.7 Ma). This may suggest causal relations between tectonics and coeval volcanic activity in the area of Balaton highland. SW-NE-trending faults had possibly an important role in controlling the recent hydrographic pattern in the Balaton area (even the location and shape of the
Fig. 28. Main faults detected from the interpretation of Lake Balaton high-resolution seismic profiles. Tectonic deformation mostly postdates 'Middle Pannonian' and consists of WSW-ENE strike-slip faults and associated folds. Fault patterns suggest a possible tectonic control on shape and location of the lake itself.
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN lake itself) and the distribution of igneous rocks in the outcrop.
Summary and conclusions Recent stratigraphic research has focused on the sequence architecture in shallow marine deposits of continental margins. Although a consensus seems to be emerging, concerning the application of sequence stratigraphic concepts to the continental record (Shanley & McCabe 1994; Miall 1997), very few well-documented examples of non-marine sequences have yet been described. The case study of the Late Miocene continental record of western Pannonian Basin represents a contribute towards sequence stratigraphy applied to non-marine strata. It also offers new insights into the complex and debated problem of correlation between regional unconformity-bounded units (Salvador 1994) and the more consolidated lithostratigraphic and biostratigraphic units which derive from local or regional use in a given geological province. Furthermore our results suggest that absolute age constraints for the Pannonian strata are still poor and the correlation of stratigraphic sequences of western Pannonian Basin with the cycle chart of Haq et al. (1987) is not documented. The 1993 high-resolution seismic survey on Lake Balaton offered an outstanding complementary data set for our purpose as it served as outcrop-scale link between the regional sequence stratigraphic procedure and the fieldwork. According to the data presented and discussed in this paper, a number of conclusions can be summarized as follows.
Sequence stratigraphic f r a m e w o r k o f western Pannonian Basin Five third-order (with 106 year periodicities) sequences have been recognized at regional scale in the post-rift succession of western Pannonian Basin (Figs 4-8). We have designated these sequences, from bottom to top, as Sarmatian-1 (SAR-1) and Pannonian-1 (PAN-l) through Pannonian-4 (PAN-4). Reliable time constraint was only available for the stratigraphic interval between 9.0 Ma and 7.4 Ma which has been calibrated by the Iharosber6ny-I magnetostratigraphic record, revised after Cande & Kent (1995). Maximum flooding surface mrs-2 marks the peak of a major flooding event which occurred in
385
the Pannonian Lake at c. 9.0 Ma. This event was manifested by Congeria czjzeki open lacustrine beds (Szfik fm.) which flooded the basin margins. Mrs-2 represents a quasi time line at basin scale that can be strikingly correlated with the top of Pannonian s.s. stage (Lower Pannonian sensu L6renthey 1900). Sequence boundary PAN-2 is the result of a major water level drop (several tens of metres) in the Pannonian Lake at c. 8.7 Ma. Evidence for subaerial exposure of the lake margins documented by significant erosion along PAN-2 SB, suggests a scenario where a number of depositional, epigenetic and/or diagenetic features developed in response to the lowering of base level. These features are widely represented in the 'marginal facies' of Transdanubia in the area of Lake Balaton. Among them are: continental travertine (geyserite of the Hungarian literature), evaporitic dolomite, swamp deposit, calcrete and palaeosols of Tihany formation, residual rocks associated with epigenesis in the vadose zone, such as silica-cemented regolith (k6tenger silcrete) of the Kfilla formation (Fig. 9). Maximum flooding surface mfs-3 (c. 7.4 Ma) represents a second important flooding event within the Pannonian Basin during Late Miocene. It is again characterized by occurrence of open lake strata, this time associated with a younger Congeria assemblage (Congeria rhomboidea). Mfs-3 represents another useful chronostratigraphic surface at basin scale which can be considered a good proxy in western Hungary for the base of Pontian as it is defined in the stratotype area (Black Sea Basin) (Fig. 12). The higher rank unit bounded by PAN-1 SB and PAN-4 SB approximately correlates with the Tortonian-Messinian of the standard time scale and correspond to the 'Late Miocene sequence' of Juhfisz et al. (this volume).
Chronostratigraphic implication Maximum flooding surfaces mfs-2 (9.0 Ma) and mfs-3 (7.4 Ma) represent two quasi-isochronous surfaces at basin scale. Our study showed (Figs 8, 11, 12 & 13) they individuate a package of strata in between the Pannonian s.s. and the Pontian s.s. stages of the Paratethys (lower part of Pontian sensu Stevanovid 1951). This correlation suggest the opportunity to introduce a three-fold subdivision of the Pannonian stage sensu L6renthey (1900) into Early Pannonian (Pannonian s.s.), Mid-Pannonian ('Danubian' or Transdanubian' in Sacchi et al. 1997, 1998) and Late Pannonian (Pontian s.s.).
386
M. SACCHI ETAL.
The concept of an intermediate stage between Early Pannonian and Late Pannonian is not new. Since an early stratigraphic study on Pannonian strata, Halav~its (1903) had already defined a 'Middle Pannonian' succession, corresponding to Congeria ungulacaprae and Congeria balatonica beds, which partly correlates with our Danubian (or Transdanubian) stage (Fig.
11). High-resolution seismic survey on Lake Balaton and the outcrop-scale stratigraphic f r a m e w o r k o f Pannonian s.1. strata Two major unconformities, both associated with significant stratigraphic gap have been detected in the subsurface of Lake Balaton: (a) an upper unconformity at the base of the Balaton Quaternary deposits; (b) a lower unconformity at the top of the pre-Pannonian (Middle Miocene) strata. These two unconformities separate, from bottom to top, three major seismostratigraphic units (Figs 18-24): (1) pre-Pannonian strata; (2) Pannonian s.l. (Late Miocene) succession; (3) Late Pleistocene-Holocene deposits of Lake Balaton. (1) Pre-Pannonian (Sarmatian) strata are dramatically truncated at the top by a mature (polycyclic) erosional surface that marks a significant stratigraphic gap (about 12-9 Ma) spanning from the Upper Sarmatian to the upper part of Pannonian s.s. The gap includes amalgamation of SAR-1 SB (top of Sarmatian sequence) and PAN-1 SB (top of first Pannonian third-order sequence).V-shaped valleys with associated fluvial terraces and channel-fills are recognized from the seismic record. (2) The Pannonian s.l. strata are essentially represented by part of sequence PAN-2 and include from bottom to top the Sz~ik, Soml6 and Tihany formations. The Sz~ik formation is interpreted as open lacustrine TST deposits which onlap the underlying Sarmatian basement. The top of this formation is represented by mfs-2 (9.0 Ma) and correlates with the top of Pannonian s.s. stage. The Soml6 formation and the lower part of Tihany formation correspond to early progradational and late progradational-aggradational highstand systems tract (HST) deposits. These strata have been designated as Pontian faciostratotype in Hungary. However they are Danubian (or Transdanubian) in age according to the chronostratigraphy adopted in this study. The upper part of Tihany formation is separated from the underlying strata by PAN-2 SB (8.7 Ma) and corresponds to the LST of the
above sequence PAN-3. Unless associated with travertine mounds PAN-2 SB is hardly detectable on seismic profiles. Similarly, this subtle sequence boundary is easily detected in the outcrop only where associated with distinctive depositional/diagenetic features (i.e. paleosols, silcrete). The development of PAN-2 SB slightly predates the onset of basaltic eruption of the Tihany Volcano (c. 7.8 Ma, Balogh 1995). Stratal patterns towards igneous bodies inferred from seismic profiles suggest that most nearsurface magmatic intrusions were coeval and/or slightly postdated the deposition of the Szfik fm (c. 9.0). (3) Lake Balaton is part of a relatively young hydrographic system which evolved in the latest Quaternary, during the post-Wtirm deglaciation of Central Europe. The unconformity at the base of the Pleistocene-Holocene deposits of Lake Balaton, corresponds to a subaerial erosional surface which dramatically truncates the underlying Pannonian strata (Figs 18-22, 24). This unconformity marks a major stratigraphic hiatus that practically encompasses the whole Pliocene and most of the Pleistocene.
Tectonic interpretation The Neogene succession of the western Pannonian Basin is affected by significant post-rift tectonic deformation. A regional tilting of the western Pannonian Basin fill towards ESE is evidenced by seismic profiles, consistently showing ESE-dipping strata which are erosionally truncated at the top. Previous studies (Tari 1994; Horv~ith 1995) have proposed this regional tectono-stratigraphic setting was induced by a late-stage tectonic inversion and uplift with consequent subaerial erosion of Late Neogene strata in Transdanubia and Northern Hungary. Based on our interpretation we suggest this tectonic inversion may have started at the very beginning of Pliocene (after PAN-4 SB). An older compressive/transpressive tectonic phase can be also detected from regional seismic profiles at about the end of Sarmatian (Horv~ith 1995). This may account for the significant and almost ubiquitous stratigraphic gap that we have documented at the base of the post-rift strata of western Pannonian Basin. The tectonic pattern in the area of Lake Balaton is expressed by SW-NE strike-slip faults and associated folds which postdate the 'Middle Pannonian' succession of Transdanubia (Figs 18-22, 24, 28). We kindly acknowledge the officers and the crew of the RV VizvOdelem (Central Transdanubia Water
LATE MIOCENE STRATIGRAPHY, PANNONIAN BASIN Management Office) for their skilled help and assistance during the seismic survey on Lake Balaton, L. Mirabile and his staff (Oceanography Institute, Istituto Universitario Navale, Naples) for their expertise in the data acquisition, T. McGee (Thalassic Data Limited, Vancouver) for his help during the acquisition and t h e preliminary data processing and T. Cserny and A. Jfimbor (Geological Institute of Hungary, Budapest) for allowing us access to existent seismic and borehole data in the study area. We sincerely thank I. Magyar, A. Mindszenty, P. Mtiller and O. Sztan6 for their help during the field work, together with M. B. Cita, B. D'Argenio, P. D6v6nyi, A. Gal~cz, M. K~izm6r, M. Lantos, F. Molisso, A. Nagymarosy, P. Sclafani and G. Tari for their support at various stages of this work. Thanks are also due to A. Bally for his precious suggestions on the interpretation of Balaton seismic profiles and T. Jacquin who revised the manuscript. The research work has been developed as part of a PhD project that the first author is carrying on at the Department of Geophysics of EOtv6s Lorfind University, Budapest. The digital conversion and processing of the original analogue seismics acquired on Lake Balaton were edited by O. Magyari. Financial support was provided by the IBS Project (Contract JOU2-CT92-0110), the Italian-Hungarian Cooperation Agreement (CNR-MTA) for the period 1995-1997, the Geomare Sud Institute, CNR, Naples, Italy, and the National Science Foundation of Hungary (OTKA 4181).
References BALDI, T. 1980. The early history of the Paratethys. FOldtani K6zIOny, 110, 456472. BALLY, A. W. & SNELSON, S. 1980. Realms of subsidence. In: MIALL, A. D. (ed.) Facts and principles of world petroleoum occurrence. Memoirs of the Canadian Society of Petroleum Geologists, 6, 9-94. BALOGH, K. 1995. K/Ar study of the Tihany .Volcano, Balaton Highland, Hungary. Report on the work supported by the European Community in the flame of program 'IBS'. Institute of Nuclear Research. Hungarian Academy of Sciences. Debrecen. , fikRVA-SOs,E., PI~CSKAY,Z. & RAVASZ-BARANYAI, L. 1986. K/Ar dating of Post-Sarmatian alkali basaltic rocks in Hungary. Acta Mineralogica Petrographica, 28, 75-93. BgRCZI, I. & PHILLIPS,R. 1985. Processes and depositional environments within Neogene deltaiclacustrine sediments, Pannonian Basin, Southeast Hungary. Geophysical Transactions Special Edition, 31, 55-75. BERGGREN, W. A., HILGEN, E J., LANGEREIS, C. G., KENT, D. V., OBRADOVICH,J. D., RAFFI, I., RAYMO, M. E. & SHACKLETON,N. J. 1995. Late Neogene chronology: New perspectives in high-resolution stratigraphy. Geological Society of America Bulletin, 107, 1272-1287. CANDE, S. C. & KENT,D. V. 1992. A new Geomagnetic Polarity Time Scale for the Late Cretaceous and
387
Cenozoic. Journal of Geophysical Research, 97, 13917-13951. & -1995. Revised calibration of the geomagnetic polarity time scale for the Late Cretaceous and Cenozoic. Journal of Geophysical Research, 100, 6093-6095. CHOLNOKY, J. 1918. A Balaton hidrogr6fi6ja. A Balaton tudom~inyos tanulm~inyoz~is~inak eredm6nyei I/2. CICHA, I. & SENES,J. 1968. Sur la position du Mioc6ne de la Parat6thys Centrale dans le cadre du Tertiaire de l'Europe. Geologica Carpathica, 19, 95-116, Bratislava. COMMITTEE OF HUNGARIAN STRATIGRAPHY 1983. Lithostratigraphic Formations of Hungary. In: FOLOP, J. (ed.) Hungarian Geological Survey Formation Chart. CSAT0, I. 1993. Neogene sequences in the Pannonian Basin, Hungary. Tectonophysics, 226, 377-400. CSERNY,Y. & CORRADA,R. 1989. Complex geological investigation of Lake Balaton (Hungary) and its results. Acta Geologica Hungarica, 32, 117-130. -& NAGY-BODOR, E. 1996. The Pre-Quaternary Morphology and Quaternary Geohistory of Lake Balaton. In: JUHASZ, A. & PECSI, M. (eds) lAG European Regional Geomorphological Conference, Hungary, April 9-12, 1996. Geomorphological studies on the Transdanubian Mountains, 70-73. - - , HERTELENDI,E. & TARJAN, S. 1995. Results of isotope-geochemical studies in the sedimentological and environmental geologic investigations of Lake Balaton. Acta Geologica Hungarica, 38, 355-376. ERDELYI, M. 1961. Ktils6-Somogy vizf61dtana. Hidrol6giai k6zl6ny, 41, 445-528. 1962. Ktils6-Somogy vizf61dtana. Hidrol6giai kOzlOny, 42, 56-65. FEJFAR, 0. 1988. Tableau stratigraphique de l'Eocene superieure au Pleistocene. In: Symposium de Gtinzburg (unpublished). GALLOWAY, W. E. 1989. Genetic stratigraphic sequences in basin analysis I: architecture and genesis of flooding surface-bounded depositional units. AAPG Bulletin, 73, 125-142. GYALOG, L. (ed.) 1996. A fOldtani tdrk~pek jelkulcsa Os a rOtegtani egys~gek rOvid leirdsa [Key of geological maps and short description of the lithostratigraphical units] MAFI alk. kiadv. [Hungarian Geological Survey], Budapest. HALAVATS, J. 1903, Die Fauna der pontischen Schichten der Umgebung des Balatonsees. In: Resultate der wissenschaftlichen Erforschung des Balatonsees 1/1 Palaontologischen Anhag 4/2, Holzel, Wien, 1-80. HAQ, B. U., HARDENBOL,J. & VAIL P. R. 1987. Chronology of fluctuating sea level since the Triassic (250 million years ago to present). Science, 235, 1156-1167. HARANGI, Sz. & HARANGI, R. 1995. Late Neogene magmatism in the NW Pannonian Basin. In: HORVATH, E, TARI, G. & BOKOR, CS. (eds) Extensional collapse of the Alpine Orogene and Hydrocarbon Prospects in the Basement and Basin Fill of
388
M. SACCHI E T A L .
the Western Pannonian Basin. A A P G International Conference and Exhibition, Nice, France, Guidebook to field trip No. 6, Hungary, 61-74. HORNES, M. 1851. Die fossilen Mollusken des Terti~irBeckens von Wien. Jahrbuch Geol. Reichsanst., 2. HORVATH, E 1993. Towards a mechanical model for the formation of Pannonian Basin. Tectonophysics, 226, 333-357. 1995. Phases of compression during the evolution of the Pannonian Basin and its bearing on hydrocarbon exploration. Marine and Petroleum Geology, 12, 837-844. -& CLOETINGH, S. 1996. Stress-induced late-stage subsidence anomalies in the Pannonian Basin. Tectonophysics, 266, 287-300. & POGACSAS,Gv. 1988. Contribution and Seismic Reflection Data to Chronostratigrphy of the Pannonian Basin. In: ROYDEN, L. H. & HORVATH, F. (eds) The Pannonian Basin. A Study in Basin Evolution. AAPG. Memoirs, 45, 97-105. , BERCKHEMER,H. & STEGENA,L. 1981. Models of Mediterranean back-arc basin formation. Phil. Trans. R. Soc. London, s. A, 300, pp. 383M02. JAMBOR, A. 1980. A Dun~inttili-k/3z6phegysdg pann6niai k6pz6dm6nyei. Magyar dllami FOldtani IntOzet EvkOnyve, 62, 1-259. -(ed.) 1987. Kuns6gi Emelet - A Magyarorsz6gi Kunsdgi emeletbeli k4pz6dmOnyek fOldtani jellemzOse Annals of the Hungarian Geological Institute, 69. 1989. Review of the geology of the s. 1. Pannonian formations of Hungary. Acta Geologica Hungarica, 32, 269-324. , PARTI2NYI,Z. & SOLTI, G. 1981. Geological features of basalt volcanics in Hungary. Geological Institute of Hungary, 1979, 225-239. JtJHAsz, G. 1994. Comparison of the sedimentary sequences in Late Neogene subbasins in the Pannonian Basin, Hungary / Magyarorsz~igi neogdn medencerdszek pannGniai s. 1. tiledGksorfinak 6sszehasonlitGelemzdse. FOldtani KOzlOny, 124, 341-365. , PHILLIPS, L., MOLLER, P., RICKETTS, B., TOTHMAKK, A., LANTOS,M. & KovAcs, L.O. 1999. Late Neogene sedimentary facies and sequences of the Pannonian Basin, Hungary. This volume. KORPAS-HODI, M. (ed.) 1983 DunGntfdi KOz~phegys~g -Paleoecology and biostratigraphy of the pannonian mollusca fauna in the northern foreland of the Transdanubian Central Range. Annals of the Institute of Geology, Hungary. 66. KRIJGSMAN,W., GARCI2S,M., LANGEREIS,C. G., DAAMS, R., VANDER MEULEN,A. J., AGUSTi, J. & CABRERA, L. 1996. A new chronology for the middle to late Miocene continental record in Spain. Earth and Planetary Science Letters. 142, 367-380. LANTOS, M., HAMOR, Y. & POGACSAS, G. 1992. Magneto - and seismostratigraphic correlations of Pannonian s. 1. (late Miocene and Pliocene) deposits in Hungary. Paleontologia i EvoluciG, 24-25, 35-46. LASKAREV,V. 1924. Sur les equivalents du Sarmatien superieur en Serbie. Recueil de travaux offert a M. Jovan Cvijic, Beograd. -
-
-
-
-
-
LoczY, L. 1913. A Balaton tudomdnyos tanulmfnyozdsdnak eredm~nyei. I. A Balaton kOrnyGkGnek geolGgiai kGpzGdmGnyei 6s ezeknek vidGkek szerinti telepedGse. LGREYrHEY, I. 1900. Foraminifera der pannonischen Stufe Ungarns. Neues Jarbuch fiir Mineralogie, Geologie und Paliiontologie, 2, 99-107. MAGYAR, I. 1988. Mollusc fauna and flora of the Pannonian quartz sandstone at Mindszentk~illa, Hungary. Annales Universitatis Scientiarun Budapestiensis de R. EOtvOs Nominatae, Sectio Geologia - 28, 209-222. 1995. Late Miocene mollusc biostratigraphy in the eastern part of the Pannonian Basin (Tisz~intul, Hungary). Geologica Carpathica, 46, 29-36. & HABLY, L. 1994. Stratigraphic position of late Neogene paleobotanical sites in Hungary: Miocene or Pliocene? Acta Paleobotantica Polonica, 34, 195-203. MAGYAR,I., GEARY,D., SI)T(3-SZENTAI,M., LANTOS,M. & MOLLER, P. in press. Integrated biG-, magnetoand chronostratigraphic correlations of the Late Miocene Pannonian Lake Deposits. Acta Geological Hungarica. MAROSI, S. & SZIL,/~RD,J. 1981. A Balaton kialakulfisa. FOldrajzi KOzlem~nyek, 1-30. MATrlCK, R., PHILLIPS,R. & RUMPLER,J. 1988. Seismic stratigraphy and depositional framework of sedimentary rocks in the Pannonian Basin in southeastern Hungary. In: ROYDEN, L.H. & HORVATH, E (eds), The Pannonian Basin: a study in basin evolution. A A P G Memoirs, 45, 117-145. MIALL, A. D. 1997. The Geology of Stratigraphic Sequences. Springer. MIKE, K. 1980. Osmedernyomok a Balaton k6rnydk6n. FOldrajzi Ortesit6, 29, 313-334. MULLER, P. & MAGYAR, I. 1992. Stratigraphic significance of the Upper Miocene lacustrine cardiid Prosodacnomya (K6tcse section, Pannonian Basin, Hungary) / A Prosodacnomy~ik r6tegtani jelent6sdge a K6tcse k6rny6ki panndniai s.1. tileddkekben. FOldtani KOzlOny, 112, 1-38, Budapest. & -1995. Designation of the base of Pontian stage in the Pannonian Basin. Romanian Journal of Stratigraphy, 76. Abstracts of the 10th Congress of RCMNS, 1, 83-84. --& SzONOKu M. 1990. Faciostratotype TihanyFehdrpart (Hungary), ("Balatonica Beds" by L6renthey, 1905). In: STEVANOVId, P., NEVESSKAYA, L., MARINESCU, E, SOKAd, A. & JAMBOR, fi~. (eds) Chronostratigrapie und Neostratotypen, Neogen der Westliche ("Zentrale") Paratethys, Pontien, Jazu and Sanu, Zagreb-Beograd, 427-436. NAGY-BODOR, E. 1988. A Balaton pannGniai 6s holocGn kGpzGdmGnyeinek palynolGgiai vizsgfilata. FOldt. int. Evi Jel. 1986-r61., 568-580. NAGYMAROSY,A. & MOLLER, P. 1988. Some aspects of Neogene biostratigraphy in the Pannonian Basin. In: ROYDEN, L. H. & HORVATH,F. (eds) The Pannonian Basin. A Study in Basin Evolution. AAPG. Memoirs, 45, 69-78. PAPP, A. , GRILL, R., JANOSCHEK, R., KAPOUNEK, J., -
-
-
-
LATE MIOCENE STRATIGRAPHY, P A N N O N I A N BASIN KOLLMAN, K. & TURNOVSKY,K. 1968. Nomenclature of the Neogene of Austria. Verh. Geol. Bundesanst., Jg. 1968, 1%27. --, JAMBOR, A. & STEININGER, E E (eds) 1985. Chronostratigraphie und Neostratotypen, Mioziin der Zentralen Paratethys 7, Pannoniaen. AkadGmiai KiadG, Budapest. POGACSAS,GY. 1984. Seismic stratigraphic features of Neogene sediments in the Pannonian Basin. Geophysical Transactions, 30, 373-410. , LAKATOS, L., UJSZASZI, K., VAKARCS, G., VARKONYI, L., VARNAI, P. & REVESZ, I. 1988. Seismic facies, electro facies and Neogene sequence chronology of the Pannonian Basin. Acta Geologica Hungarica, 31, 175-207. REGIONAL COMMITTEE ON MEDITERRANEAN NEOGENE STRATIGRAPHY 1971. Stratotypes of Mediterranean Neogene Stages. Giornale di Geologia, 37, 1-266. ROTH, L. 1879. Geologische Skizze des KroisbachRuster Bergzuges und des sudlichen Teiles des Leita-Gerbiges. FOldtani KOzlOny, 9, 139-150, Budapest. ROYDEN, L. H. & HORVATH, E (eds) 1988. The Pannonian basin. A study in basin evolution. A A P G Memoirs, 45. ROGL, E 1996. Stratigraphic correlation of the Paratethys Oligocene and Miocene. Mitt. Ges. Geol. Bergbaustud. Osterr. 41, 65-73. & STEININGER,E E 1983. Vom Zerfall de Tethys zu Mediterran und Paratethys. Die neogene Pal~iogeographie und palinspastik des zirkummediterranen Raum. Ann. Naturhis. Museum Wien, 85/A, 135-163. , BERNOR, R. L., DERMITZAKIS,M. D., MOLLER, C. & STANCHEVA,M. 1991. On the Pontian correlation in the Aegean (Aegina Island). Newsletters in Stratigraphy, 24, 137-158. SACCHI, M. 1997. Two-fold versus three-fold subdivision of Pannonian s.l. strata: a proposal for a new stage in the Upper Miocene Series of the Central Paratethys. In: Abstracts of the Interim Colloquium RCMNS, Catania (Italy), 4-9 Nov. 1997, 87-88. , CSERNY,T., DOVI~NYI,P., HORVATH,E, MAGYARI, O., MCGEE, T. M., MIRABILE, L. & TONIELLI, R. 1998. Seismic stratigraphy of the Late Miocene sequence beneath Lake Balaton, Pannonian Basin, Hungary. Acta Geologica Hungarica, 41, 63-88. , HORVATH, E, MAGYAR, I. & MOLLER, P. 1997. Problems and progress in establishing a Late Neogene Chronostratigraphy for the Central Paratethys. Neogene Newsletter, 4, 37-46. , --& MAGYARI, 0. 1995. High resolution seismics in the Lake Balaton: insights into the evolution of western Pannonian Basin. /n: HORVATH,E,TARI, G. & BOKOR, CS. (eds), Extensional collapse of the Alpine Orogene and Hydrocarbon Prospects in the Basement and Basin Fill of the Western Pannonian Basin. A A P G International Conference and Exhibition, Nice, France, Guidebook to field trip No. 6, Hungary, 171-185.
389
SALVADOR,A. (ed.) 1994. International stratigraphic guide. International Subcommission on Stratigraphic classification of International Commission on Stratigraphy. The Geological Society of America. SHANLEY, K. W. & MCCABE, P. J. 1994. Perspectives on the sequence stratigraphy of continental strata. A A P G Bulletin. 78, 544-568. STEININGER, E E, BERNOR, R. L., FAHLBUSCH,g. ~; MEIN, E 1990. European Neogene marine/continental chronologic relations. In: L1NDSAY,E. H., FAHLBUSCH, V. & MEIN, P. (eds) European Neogene Mammal Chronology. Plenum Press, New York, 15-46. , Mt3LLER, C. & ROGL, E 1988. Correlation of Central Paratethys, Eastern Paratethys and Mediterranean stages. In: ROYDEN, L. H. & HORVATH, E (eds) The Pannonian Basin: a study in basin evolution. A A P G Memoirs, 45, 79-87. Sa'EVANOV~C, R M. 1951. Pontische Stufe im engeren s i n n e - Obere Congerienschichten Serbiens und der angrenzenden Gebiete. Serbische Akademie der Wissenschaften, Sonderausgabe 187, Mathematisch-Naturwissenschaftliche Klasse no 2, Beograd. , NEVESSKAYA,L. A., MARINESCU,E, SOKAC, A. & JAMBOR, A. (eds) 1990. Chronostratigraphie und Neostratotypen, Neogen der Westlichen ("Zentrale") Paratethys 8, Pontien. Jazu and Sanu, Zagreb-Beograd. SZAB0, Cs., HARANGI, Sz. & CSONTOS,L. 1992. Review of Neogene and Quaternary volcanism of the Carpathian-Pannonian region. Tectonophysics, 208, 243-256. SZADECZKY-KARDOSS,E. 1938. Geologie der rumpfungarlandischen kleinen Tiefeben. A soproni Bfinya 6s ErdGmGrnOki FGiskola KGzl. 10(2). SZENTGYORGYI, K. & JuIqASz, GY. K. 1988. Sedimentological charasteristics of the Neogene sequence in SW Transdanubia, Hungary. Acta Geologica Hungarica. 31, 209-225. TARI, G. 1994. Alpine tectonics of Pannonian Basin. PhD thesis, Rice University, Houston, Texas. , BALDI, T. & BALDI-BEKE, M. 1993. Paleogene retroarc flexural basin beneath the Neogene Pannonian Basin: a geodynamic model. Tectonophysics, 226, 433-455. - - , HORVATH,E & RUMPLER,J. 1992. Styles of extension in the Pannonian Basin. Tectonophysics, 208, 203-219. UJszAszI, K. & VAKARCS G. 1993. Sequence stratigraphic analysis in the South Transdanubian region, Hungary. Geophysical Transactions, 38, 69-87. VAIL, E R. 1987. Seismic stratigraphy interpretation procedure. A A P G Studies in Geology 27. - - , AUDEMARD,E, BOWMAN,S., EISNER, P. & PEREZCRUZ, C. 1990. The stratigraphic signatures of tectonics, eustasy and sedimentology. An overview. In: EINSELE,E., RICKEN,A. & SEILACHER,A. (eds) Cycles and events in stratigraphy, Springer-Verlag, New York, 617-659. --, MITCHUM, R. JR & THOMPSON, S. 1977. Seismic stratigraphy and global changes of the sea level,
390
M. SACCHI E T A L .
part 3: Relative changes of the sea level from the coastal onlap. In: PAYTON, C. E. (ed.) Seismic Stratigraphy - A p p l i c a t i o n s to Hydrocarbon exploration, AAPG Memoirs, 26, 63-82. VAKARCS, G., VAIL, P. R., TARI, G., POGACSAS, GY., MATrICK, R. E. & SZAB0, A. 1994. Third-order Miocene-Pliocene depositional sequences in the prograding delta complex of the Pannonian Basin. Tectonophysics, 240, 81-106. VASS, D., KOVAC,,M. & LEXA, J. 1988. Molasse Basins and volcanic activity in west Carpathian Neogene.
Its evolution and geodynamic character. Geologica Carpathica, 39, 539-561. --, REPCOK, I., BALOGH, K & HALMAI, J. 1987. Revised radiometric time-scale for central Paratethyan Neogene. Annales of the Hungarian Geological Institute 70, 423434. WEIMER, P. & POSAMENTIER,H. W. (eds) 1993. Siliciclastic sequence stratigraphy - Recent developments and applications. AAPG Memoirs, 58. WHN, GY. 1967. D61kelet-Dungmttil hegysdgszerkezete. FOldtani K6zlOny, 97, 371-395.
Two-dimensional modelling of stratigraphy and compaction-driven fluid flow in the Pannonian Basin R. T. V A N B A L E N 1, L. L E N K E Y 2, F. H O R V / k T H 2 & S. A. P. L. C L O E T I N G H
1
1Tectonics~Structural Geology Group, Department o f Sedimentarv Geology, Inst. o f Earth Sciences, Vrije Universiteit, De Boelelaan 1081, 1081 H V Amsterdam, The Netherlands 2Geophysical Department, EOtvOs L o r a n d University, L u d o v i k a tOr 2, Budapest, Hungary Abstract: During the Pliocene-Quaternary time interval, the peripheral parts of the Pan-
nonian Basin system have been uplifted and subsidence in the basin centre accelerated, causing a distinctive truncation pattern in the basin stratigraphy. Stress analyses indicate that the Pannonian Basin system, originally formed in an extensional regime, is subjected to a compressive stress since the early Pliocene. Results of forward modelling of basin subsidence and sedimentary filling along a cross-section through the southern part of the Pannonian Basin demonstrate that a change of the basin shape due to the compressive stress can successfully explain the observed pattern of differential uplift and subsidence occurring since the early Pliocene. In addition, the forward modelling of subsidence and fill provides constraints for the depth of lithospheric necking during extension, the palaeowater-depth history and lake-level changes in the southern part of the Pannonian Basin. Compaction-driven fluid flow modelling shows that the first significant overpressures in the southern part of the Pannonian Basin developed during progradation of a large deltaic system, at a time when sedimentation rates increased rapidly. Due to the stress-induced acceleration of subsidence during Pliocene to Quaternary times, sedimentation rates increased again, causing a further increase of overpressure. The Pliocene stress induced uplift of the basin flanks combined with a preceding lake-level fall created a larger gravity potential of the groundwater table, enhancing the influx of meteoric water into the basin. This can explain observed diagenetic patterns in the southern part of the Pannonian Basin. The P a n n o n i a n Basin is a N e o g e n e intram o n t a n e basin system bounded in the west by the Alps, in the north, east and southeast by the Carpathians and in the southwest by the Dinarides. The intra-Carpathian Basin system comprises a central part, the G r e a t H u n g a r i a n Plain, and several peripheral basins: the Styrian Basin, the Vienna Basin, the Little H u n g a r i a n Plain (the D a n u b i a n Basin), the Transcarpathian Basin and the Transsylvanian Basin (Fig. 1). The P a n n o n i a n Basin system originates from contemporaneous tectonic escape in the Alps (Ratschbacher et al. 1991), asthenospheric upwelling (Stegena et al. 1975; Becker 1993) and s u b d u c t i o n roll-back along the Carpathian front in a back-arc setting, comparable to the present-day A e g e a n (Horv~ith & B e r c k h e m e r 1982). A detailed discussion on the origin of the P a n n o n i a n Basin can be found in Horv~ith (1993). A n overview of basement structures resulting from the basin-forming mechanisms is given in Fig. 2. The basement of the Pannonian Basin consists of Palaeogene deposits and Palaeozoic to Cretaceous rocks, stacked on top of each other as imbricate nappes during the Cretaceous Alpine collision (Csontos et al. 1992; Tari et al. 1992; Grow et al. 1994). The Neogene extension began
in the early Mid-Miocene with the opening of rift- and pull-apart basins and the formation of metamorphic core complexes along reactivated Alpine overthrusts (Rumpler & Horvfith 1988; Tari et al. 1992). The style of extension and the direction of normal faulting varies in the basin system. The strike-slip faults, depicted in Fig. 2, act as transfer zones for the different types and amounts of extensional strain. The total amount of extension in an E - W direction exceeds 100 km and is approximately equal to the amount of c o n t e m p o r a n e o u s shortening in the O u t e r Carpathian flysch belt (Csontos et al. 1992; Tari et aL 1992). Due to changes in the motions of lithospheric plates, the stress field in the Pannonian Basin changed to compressive in Pliocene. The maximum horizontal stress in the southern part of the basin has a S W - N E orientation (Csontos et aL 1991; Gregersen 1992; MUller et al. 1992). Changing intraplate stresses can cause substantial differential subsidence in s e d i m e n t a r y basins, which is recorded as a relative sea-level change in the stratigraphy (Cloetingh et al. 1985). We investigate the effect of the compressive stress on the stratigraphy and compaction driven pore fluid overpressures in the sub-basins of the southern part of the Pannonian Basin
VANBALEN,R. T., LENICEY,L., HORV~,T.,E & CLOETINGH,S. A. P. L. 1999. Two-dimensional modelling of stratigraphy and compaction-driven fluid flow in the Pannonian Basin. In: DURAND,B., JOLIVET,L., HORVATH, E & Ss M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 391-414.
392
R . T . VAN B A L E N E T A L .
Fig. 1. (a) The Pannonian Basin and its surroundings. The names of various sub-basins and mountain ranges have been indicated. (b) Isopach map of the Pannonian Basin. A - A ' denotes the modelled cross-section. The contour lines depict the depth to the pre-Neogene basement (modified after Horv~ith 1993).
MODELLING IN THE PANNONIAN BASIN
393
Fig. 2. Fault pattern resulting from the tectonic escape from the Eastern Alps, subduction roll-back along the Carpathian front and mantle upwelling (modified after Horvfith 1993). Dextral and sinistral strike-slip faults accommodate the general eastward movement of the basin from the eastern Alps towards the Carpathian front. The amount, the style and the direction of extension in between the strike-slip faults vary considerably. A-A' denotes the modelled cross-section.
(Drava, Sava and Mako troughs and the B6k6s Basin) using forward modelling techniques. The location of the modelled cross-section is shown in Fig. 1. Below, we first give an introduction to the Neogene sedimentary fill, palaeo-water depths and lake-level variations, as these are important parameters for the tectono-stratigraphic forward modelling and the compaction-driven fluid-flow modelling. Subsequently, the latestage anomalous subsidence and the lithospheric stress field are discussed. Next, we present and discuss results from the tectono-stratigraphic forward models. Finally, we provide an introduction to overpressure mechanisms in the southern part of the Pannonian Basin and show results of forward modelling of compactiondriven overpressure. The results of forward modelling are compared to predictions obtained by a neural network analysis.
Neogene sedimentary fill of the southern part of the Pannonian Basin The Neogene Pannonian Basin was filled by a large deltaic system originating from the rising Carpathians and Alps. The reconstructed positions of the depositional shelf break of the Neogene deltaic system (Ujszaszi & Vakarcs, 1993; Vakarcs et al. 1994) demonstrate the gradual fill of the basin system from the rims towards the central and southern parts. The Neogene stratigraphy in the southern part of the basin can be subdivided into seven depositional units. From bottom to top these are: a basal unit (B), a deep basin unit (DB), a prodelta unit (PD; Szolnok formation), a delta front-delta slope unit (DS; Algy6 formation), a delta plain-delta front unit (DP; T6rtel formation), an alluvial plain unit (AP; Zagyva and Nagyalf61d formations) and Quaternary unit (Q) (B6rczi 1988; Horv~ith et al. 1988; Mattick et al. 1988; K~izmer 1990; Juh~isz 1991; see
394
R.T. VAN BALEN E T A L .
Table 1. Depositional units in the southern part of the Pannonian Basin Palaeo-water depth (m) Unit name Q AP DP DS PD DB B
Age (Ma)
Sedimentary environment
shallow
deep
Effective sand content (vol%)
0.0-2.4 2.4-3.9 3.9-5.5 5.5-6.3 6.3-8.2 8.2-10.5 10.5-17.5
alluvial plain alluvial plain delta plain delta slope pro-delta deep basin basal
0 0 0 200 n.a. n.a. 0
0 0 0 400 800 >1000 0-2000
0.2 0.2 0.8 0.8 0.2 0.0 1.0
Table 1). These depositional units have diachronous ages, as they are related to a deltaic system. The basal unit was deposited during the initial part of the syn-rift stage. The deposits in this unit in the southeastern part of the Pannonian Basin system consist of redeposited conglomerates alternating with marl beds in the deep subbasins and abrasive conglomerates around the basement highs (B6rczi & Phillips 1985; B6rczi 1988; K~zm6r 1990; Juh~sz 1991). In the southeastern part of the Pannonian Basin, the next depositional unit, the DB unit, consists of argillaceous and calcarous marls (B6rczi & Phillips 1985; B6rczi et al. 1988; K~izm6r 1990; Juh~isz 1991). The first deltaic sediments in the deep sub-basins consist of turbidites (the PD unit). This facies lacks in the shallower parts of the basin; it was, for example, not deposited on the highs in between and flanking the sub-basins. The PD unit consists of grey argillaceous marls, siltstones and light-grey sandstone beds. In general, the marl intercalations decrease upwards (K~zm6r 1990). The next depositional unit consists of delta slope deposits (DS unit), mainly containing mudstone (siltstone and argillaceous marl), with a few interbedded sandstone bodies (Juh~sz 1991). These sediments are widespread over the basin (K~izm6r 1990; Juh~sz 1991), indicating that the prograding delta covered both the deep sub-basins and the basin highs, filling the whole basin almost completely by early Pliocene (K~zm6r 1990). The next unit is deposited in a delta plain environment (DP unit). This is a sand rich unit. The sandstone bodies occur as distributary mouth bars and channel fill rhythms of about 20-50 m thickness. The extension and continuity of these bodies are rather restricted, but they can merge laterally with one another (Juh~isz 1991). The last depositional units consist of alluvial plain sediments (AP and Q units). It contains thin bedded siltstone, claystone and sandstone with the dominance of the fine-grained fraction (Juh~isz 1991).
The modelled profile has a SW-NE orientation and transects the deep sub-basins in the southern part of the Pannonian Basin (the Sava, Drava and Mako troughs and the B6k6s Basin), see Fig. 1. Because the modelling transect is parallel to the trend of the depositional shelf break at 5.5 Ma (Vakarcs et al. 1994), we assume that along the profile line the depositional units are also chronostratigraphic units. The ages of the different units are constrained by the sequence stratigraphic analyses of the B6k6s Basin by Vakarcs et al. (1994; Table 1). Palaeo-water depths
The tectono-stratigraphic forward modelling requires an estimation of palaeo-water depths for the depositional units. Due to endemic fauna, it is hard to assess palaeo-water depths using biostratigraphy (Nagymarosy & MUller 1988). Furthermore, for sediments in the deepest parts of the sub-basins knowledge about palaeo-water depths is even more problematic, as these rocks were almost not drilled. Therefore, indirect indications have to be used in order to estimate palaeo-water depths of sediments in the deep sub-basins and in shallower parts of the basin system. During the mid- and late Miocene the B6k6s Basin was starved, as other sub-basins closer to the sediment source area captured the sediment derived from the uplifted and eroding Carpathians and Alps (K~zm6r 1990; Grow et al. 1994). The combination of low sedimentation rate and high subsidence rate must have produced great water depths (1000-1500 m) in the central part of the B6k6s Basin. The difference between present-day burial depths of abrasive conglomerates around basement highs and turbiditic conglomerates in the Mako trough and B6k6s Basin is more then 4 km (Juh~sz 1991); both are syn-riff deposits. After correction for differential subsidence due to thermal subsidence and isostasy, this indicates that the deepest part of
MODELLING IN THE PANNONIAN BASIN the Mako and B6k6s sub-basins may have obtained palaeo-water depths of about 2 km during and immediately after the rifting stage. The calcarous marls of the DB unit have a pale yellow colour on the basement highs and become darker (and eventually black) as the basin gets deeper (B6rczi et al. 1988; Juhfisz 1991). This, combined with the abundant occurrence of pyrite (K~izm6r 1990; Juhfisz 1991), the decreasing content of carbonate towards the deep sub-basins (Csat6 1993) and the lack of bioturbation, indicates a deep, oxygen-deficient, reducing environment with a stratified water column in the deep sub-basins (Kfizm6r 1990). Therefore, also sedimentological evidence supports very deep palaeo-water depths (>1000 m) during the deposition of the marls of the DB unit. Assuming that the alluvial plain sediments were deposited at zero palaeo-water depth, the thickness of the delta slope deposits gives an estimate for the palaeo-water depths of the prodelta turbidites. The difference in burial depth between the base of the alluvial plain deposits and the pro-delta turbidites (after corrections for compaction, lake-level variation and isostasy) gives an estimate of 700 m for the palaeo-water depths during deposition of the pro delta (PD) unit. Likewise, our estimate for the palaeo-water depth of the DS unit is 400 m, for the delta plain sediments and for the A P unit 0 m (see Table 1). Using similar reasoning, the palaeo-water depths for the depositional units at the flanks of the sub-basins are also estimated. The results are depicted in Table 1. Lake-level
variations
During the extension stage, the Pannonian Basin is part of the Paratethys and connected to the world seas. Eustatic sealevel changes (e.g. Haq et al. 1987) must have influenced the stratigraphy during this time interval. Elevation of the Dinarides and the eastern and southern parts of the Alps at the Sarmatian-Pannonian stage boundary causes the isolation of the Pannonian Basin from the world seas, it becomes a lake (Royden etal. 1983; Csontos etal. 1992). The isolation is associated with a drop in salinity (Jfimbor 1989; Kfizm6r 1990; Tari et al. 1992; Vakarcs et al. 1994) and the creation of an endemic fauna (Nagymarosy & Mt~ller 1988; Kgzm6r 1990). According to seismic sequence stratigraphic analyses combined with palaeomagnetic dating, a correlation exists between the eustatic sealevel curve of Haq et al. (1987) and the relative lake-level variations in the Pannonian Basin (Tari et al. 1992; Csat6 1993; Ujszaszi & Vakarcs 1993; Vakarcs et al. 1994). This can
395
be explained by changes in climatic conditions (drainage in the uplifting Carpathians) and by upstream effects of changes in the ultimate base level of the proto-Danube (Tari et al. 1992). The most dramatic lake-level change occurs at 6.3 Ma (Messinian): the lake-level drops and subsequently rises 100 to 200 m (Tari et al. 1992; Vakarcs et al. 1994). However, also based on seismic correlations, Mattick et al. (1994) have argued that most of the identified sequence boundaries are caused by delta-lobe switching. In this study we follow the inferences of Tari et al. (1992) and Vakarcs et al. (1994).
The Pliocene to Recent stress field and anomalous subsidence Due to lithospheric plate motions, the intraplate stress field has changed to a compressive regime since early Pliocene (Csontos et al. 1991; Gregersen 1992; Maller et al. 1992). Focal mechanism solutions show maximum horizontal stress orientations perpendicular to the Carpathian arc, except for the southeastern part of Hungary where the orientation is roughly N E - S W (Gregersen 1992). In addition, along the Mid-Hungarian-Balaton strike-slip zone positive flower structures are evidence for Pliocene-Quaternary wrench tectonics related to a N-S-oriented compressional stress field (L6rincz & Szab6 1993). These inferences are in agreement with bore-hole breakout and in-situ stress measurements (Mt~ller et al. 1992; Grt~nthal & Stromeyer 1992; Becker 1993; Fig. 3). In the major part of the Pannonian Basin the break-out data are scattered, possibly reflecting detachment of the sedimentary cover from the lithospheric basement. However, in the southeastern part of the Pannonian Basin system a general trend of the maximum horizontal stress orientation of SW-NE can be identified, parallel to our modelling profile line. Finite element modelling of the stress field in and around the Pannonian Basin indicates that a combination of North Atlantic ridge push, forces resulting from the northward movement of Africa and Arabia, subduction forces at the eastern border of the Pannonian Basin and a decreased rigidity of basement in the centre of the basin system can explain the observed stress field (Grt~nthal & Stromeyer 1992; Bada et al. 1995). The decreased rigidity is related to thermal weakening due to asthenospheric upwelling. Anomalous
late stage subsidence
During Pliocene to Quaternary times, subsidence decreases or uplift takes place in the
396
R.T. VAN BALEN E T A L .
Fig. 3. The stress map of the Pannonian Basin (after Mtiller et al. 1992; Becker 1993) showing the consistently SW-NE directions of maximum horizontal stress in the southern part of the basin system. A-A' denotes the modelled cross-section.
external parts of the Pannonian Basin system and subsidence in the basin centre accelerates (e.g. Royden et aL 1983; L~z~rescu et al. 1983; Polonic 1985; D e m e t r e s c u & Polonic 1989; Vakarcs et al. 1994; Horvfith & Cloetingh 1996). Studies of the Danube river terraces and travertine horizons show clear evidence for a 200-300 m uplift of the Transdanubian Central Range (Ronai 1974; Gfibris 1994). Furthermore, Quaternary erosion has removed large parts of Late Pliocene sediments in this area (Ronai 1985) with exceptions of parts which were covered by contemporaneous basaltic lava flows (G~bris 1994; Horv~ith & Cloetingh 1996). Geodetic measurements show that the area of the B6k6s Basin is presently subsiding at a rate of 4 mm a -a whereas Transdanubia rises up to 1.3 mm a -I (Joo et al. 1990). A map of compiled anomalous subsidence data by Horvfith & Cloetingh (1996) is shown in Fig. 4. Further insights into the subsidence history of the Pannonian Basin can be obtained by constructing basement subsidence curves. These curves are obtained by subtracting the deposited sediment thickness from the present-day thickness, in order to obtain the depth of the basement
in the past. Corrections are made for compaction of sediment and changes in palaeo-water depth and eustasy. No isostatic correction is made, as isostasy is one the components contributing to basement subsidence. Basement subsidence curves for several wells in the neighbourhood of the modelling profile line clearly show three different subsidence phases: a syn-rift phase, subsidence due to sediment loading during the deposition of delta-slope deposits and an acceleration or deceleration of subsidence since the early Pliocene (Fig. 5). Backstripping data were obtained from various sources: data for the centre of the Drava Trough (Molve field) are after Bari6 et al. (1991), artificial well data for the flank of the Drava trough are after seismic data in Ujszaszi & Vakarcs (1993), the MI-1 and Bes-1 well stratigraphies are after B6rczi (1988), Usz-1, Doboz-1, Sarkad-1 and Tot-1 are after Horvfith et al. (1988) and Hod-1 and Mako-1 are after Mattick et al. (1988). All stratigraphic data are approximate. Ages for the depositional units are after Vakarcs et al. (1994). The positions of the wells in our modelling transect are indicated in Fig. 6. The wells located in and at the flank of the Drava Trough show uplift since early
MODELLING IN THE PANNONIAN BASIN
397
Fig. 4. The pattern of Pliocene-Quaternary anomalous subsidence and uplift in the Pannonian Basin (modified after Horvfith & Cloetingh 1996). In general, the central part of the basin system shows continuous or enhanced subsidence, whereas the flanks of the system are uplifted. Pliocene (Fig. 5). The wells located in and at the flanks of the Mako trough and B6k6s Basin have experienced an acceleration of subsidence starting at early Pliocene.
Tectono-stratigraphic forward modelling The adopted numerical model for basin evolution is based on the pure shear stretching principle, originally proposed by McKenzie (1978). In this model, thinning of the lithosphere results in post-rift subsidence due to thermal contraction. The pure-shear model has been extended in order to account for two-dimensional heat flow, flexural isostasy (including the effects of changing in-plane stresses), and necking of the lithosphere during extension (Kooi & Cloetingh 1992; Van Balen & Cloetingh 1993). The latter process can be described as thinning of the lithosphere around its strongest part(s), commonly termed the level of necking. The main parameters for the tectono-stratigraphic forward modelling are: pre-rift crustal and subcrustal thicknesses, thinning factors, the depth to the level of necking, sea-(lake-)level history, palaeowater depths and intraplate stresses. The forward modelling predits a stratigraphic
cross-section which is compared to data. Before the modelling, the parameters have to be constrained as much as possible. By fitting the predicted profile to observations, an iterative procedure can provide further constraints, i.e. the modelling gives further limits to parameter values. The basement of the Pannonian Basin consists of Cretaceous Alpine nappes (Tari et al. 1992) and, therefore, the crust was overthickened before the Neogene extension took place (Horv~th & Berckhemer 1982; Csontos et al. 1992; Tari et al. 1992). The pre-rift thickness of the crust and subcrustal parts of the lithosphere are assumed to be equal to the current thicknesses in the eastern Alps, giving a crustal thickness of about 42.5 km and a subcrustal thickness of about 82.5 km before extension (Table 2). The current crustal thickness below the Great Hungarian Plain, without the sedimentary cover, is on average 20 km, whereas the subcrustal thickness is 40 km and less (Horvfith 1993). The present-day crustal strength of the Pannonian Basin, as demonstrated by rheological modelling of earthquake depths by Horv~th & Cloetingh (1996), is concentrated around a depth range of 6-10 km (Fig. 7). Assuming that the
398
R.T. VAN BALEN E T A L .
Fig. 5. Basement subsidence curves for several wells along or in the vicinity of our modelling profile line. Three phases of subsidence can be recognized in all wells: a syn-rift phase, subsidence due to increase of sediment loading during the arrival of the Pannonian delta and the stress induced accelaration of subsidence. The wells located in or at the flank of the Drava trough (depicted in upper panel) show Late Pliocene uplift or cessation of subsidence. Basement subsidence curves for wells located in or at the flanks of the Mako trough and B6k6s basin (depicted in the lower panel) show Late Pliocene acceleration of subsidence. See text for further discussion and data sources.
upper crust has strengthened due to cooling during the post-rift phase, the strongest part of the lithosphere is 1-2 km shallower immediately after rifting. Subtracting an additional 2 km for present-day average sedimentary cover thickness, results in a depth of the strongest part of the crust of 3 km at the beginning of the post-rift
phase. Taking into account a crustal extension factor of 2 to 3 (Royden et al. 1983; Horv~ith et al. 1988; Royden & D6v6nyi 1988) and assuming that the strongest part of the crust coincides approximately with the level of necking during extension (Van Balen & Cloetingh 1994), it follows that the pre-rift depth of necking is probably located at a depth between 5 and 20 km (in the absence of isostasy, the level of necking itself is, by definition, not displaced during extension). We have, therefore, modelled the subsidence along the profile using two end-member values of 7.5 and 15 km for the depth of lithospheric necking during extension. Constraints for the stratigraphic modelling are given by well data (B6rczi 1988; B6rczi & Phillips 1985; Mattick et al. 1988; Rumpler & Horv~ith 1988), a Neogene basement depth map (Horv~ith 1985), palaeogeographic maps for the distribution and thickness of the main sedimentary sequences in the southeastern part of the Pannonian Basin (Juh~isz 1991) and stratigraphic cross-sections through the Drava Trough (Bari~ et al. 1991). In addition, a seismic sequence stratigraphy interpretation of the B6k6s Basin (Vakarcs et al. 1994), palaeomagnetic ages of the depositional units (determined by mollusc faunas) in the Danube-Tisza interfluve area (Korp~s-H6di et al. 1992) and a sequence stratigraphic interpretation of the south Transdanubian region (Ujszaszi & Vakarcs 1993) provide further constraints for the thicknesses and ages of the sedimentary sequences, see Table 1. We have converted the two-way travel time scale of the seismic section of Vakarcs et al. (1994) to a depth scale using seismic velocity-depth logs measured in wells in the B6k6s Basin (Doboz-I, B6k6s-2) and the Mako trough (Hod-I) (e.g. Posgay et al. 1996). As our modelling transect is parallel to the trend of the depositional shoreline break at 5.5 Ma in the southeastern part of the Pannonian Basin (Korp~is-H6di et al. 1992; Ujszaszi & Vakarcs 1993; Vakarcs et al. 1994), boundaries between the lithostratigraphic units are assumed to be proxys for chronostratigraphic boundaries. Because the transition from deposition of basal sandstone to deep basin marls reflects the deepening and drowning of the sub-basins during rifting, this stratigraphic boundary is also almost synchronous. The stratigraphic cross-section based on the data sources mentioned above is shown in Fig. 6. Clearly recognizable are the truncations at the basin margins in Pliocene sequences (the DP and A P units). An important objective for the stratigraphic forward modelling is to reproduce the same stratigraphic cross-section, with special
MODELLING IN THE PANNONIAN BASIN
399
Fig. 6. Constructed stratigraphic cross-section along our model profile line. The position of hydrocarbon pools (Dank 1988) and backstripped wells (see text) are also indicated. V indicates the position of the seismic crosssection in Vakarcs et al. (1994) which is almost perpendicular to our section; other symbols are abbreviated well names. See text for further discussion and data sources.
emphasis to the truncations. Knowledge about rheology and amounts of crustal and lithospheric thinning are used to constrain the modelling. As a result of the forward modelling, predictions for lake-level variations, palaeowater depths during deposition, intra-plate stress history and the level of lithospheric necking depth during extension are obtained by matching the predicted with the known stratigraphic cross-section.
Modelling results
Fig. 7. Depths of Recent earthquakes in and around the Pannonian basin (modified after Horv~th & Cloetingh 1996). Earthquakes occurring within and outside the Pannonian Basin system are indicated. The shape of this frequency curve resembles the shape of lithospheric strength profiles.
The results of the stratigraphic forward modelling are depicted in Fig. 8. Also shown in the same figures are the estimates obtained by the modelling for the lake-level variations and palaeo-water depths for the deepest part of the Mako trough (the position of the Hod-I well). For the lake-level variations we have taken into account that they varied in concert with the eustatic sea-level changes, although not necessarily with the same amplitude (see the previous section on lake-level variations). The most dramatic lake level change occurs a r o u n d the
400
R.T. VAN BALEN E T A L .
Messinian (6.3 Ma): lake level drops 100-200 m and subsequently rises by about the same amount (Tari et al. 1992; Vakarcs et al. 1994). Along our modelling transect, the lake level fall occurs during deposition of the PD unit and lake
(a)
(b)
level rises when the delta slope deposits of the DS unit are deposited. At early Pliocene, after deposition of the DS unit, the Pannonian Basin becomes completely filled and the Pannonian lake no longer exists. As discussed before, the
MODELLING IN THE PANNONIAN BASIN
401
Fig. 8. The stratigraphy predicted by our forward modelling. The insets show the adopted lake-level curve, the palaeo-water depth curve for the deepest part of the Mako trough (the position of the Hod-I well) and the applied intraplate stress history. The lower panel depicts the adopted crustal and subcrustal extension factors. (a) Result of the applying only lake level variations, this model can not explain the thickness distribution of the AP unit. (b) Improved model applying an increase in the level of compressive intraplate stress starting at Late Pliocene; this model result better fits the observed stratigraphy. (c) The result of a model applying a deeper level of necking (17.5 km). This model requires higher palaeo-water depths than the previous models.
Table 2. Adopted numerical values for tectono-stratigraphicforward modelling
Initial crustal thickness Initial subcrustal thickness Density crust, density mantle Thermal diffusivity Thermal expansion coefficient Young's modulus lithosphere Poisson ratio lithosphere Isotherm defining effective elastic thickness
palaeo-water depths are less well constrained. Especially the basal sandstone and the d e e p basin marls of the DB unit were deposited at unconstrained water depths (2000-1000 m). The a d o p t e d numerical forward m o d e l l i n g tool applies differential lithospheric extension. Crustal extension factors, although constrained by crustal thickness maps given by Horvfith (1993), cannot be determined on the scale of our modelling. Therefore, they are determined by a modelling iteration on the basement morphology. Subcrustal extension factors are even less
42.5 km 82.5 km 2.9 g cm-3, 3.3 g cm-3 7.5 • 10-7m2 s 1 3.2 • 10-5 ~ 7 • 101~Pa 0.25 350~
constrained, as the distribution of subcrustal thicknesses is uncertain and thermal relaxation causes the subcrust to thicken during the postrift phase. As can be inferred from the subcrustal thickness maps in Horvfith (1993), the distribution of subcrustal extension factors should be smooth. Subcrustal extension is the prime cause for post-rift (thermal) subsidence (McKenzie 1978; R o y d e n & K e e n 1980). Therefore, subcrustal extension factors can be assessed by fitting the thickness of post-rift sediments. Empirically we found that a subcrustal extension
402
R.T. VAN BALEN E T A L .
factor of 5.0 below the Great Hungarian Plain results in the observed amount of post-rift subsidence. This value is in agreement with a coherency analyses for the amount of heat flow, syn- and postrift subsidence (assuming local isostasy) by Royden & DOv6nyi (1988), which resulted in crustal extension factors ranging from 1.6 to 2.9 and subcrustal extension factors ranging from 6.7 to 20 for wells in the Great Hungarian Plain. The difference in the total amount of thermal subsidence and the thermal subsidence rate for subcrustal extension factors in the range of 5 to 20 is minor (McKenzie 1978). We apply, arbitrarily, subcrustal extension factors which are constant below the sub-basins and linearly decay towards the rims of the basin system. As the amount of post-rift subsidence is not a model-fitting parameter, we apply the same amount and distribution of subcrustal extension in all presented models. Adopted values for other modelling parameters are given in Table 2. In the first model a 7.5 km deep level of lithospheric necking during extension is applied. Crustal extension factors and palaeo-water depths are obtained by a fitting iteration. In this first simulation, we determine whether a lakelevel variation can explain the truncations of the DP and A P units. Therefore, lake-level varies in accordance with the eustatic sea-level curve. The result is depicted in Fig. 8a. The modelling requires a palaeo-water depth maximum of 2300 m during the extension stage in order to yield a good fit to the observed stratigraphy. The required crustal and subcrustal extension factors are displayed in the lower panel of the figure. The result shows that the assumed lake-level variations can indeed explain the truncation patterns at the flanks of the Mako and B6k6s subbasins and the thickness of the DP unit. However, the thickness of the AP unit in the deepest parts of these sub-basins does not fit the observed thickness (Fig. 6). The adopted lakelevel fall causes a predicted thickness of about 500 m for this unit which is about half of the observed thickness. This cannot be corrected by adopting a different magnitude of palaeo-water depth, as it is minor during this time interval. Therefore, we conclude that a lake-level fall can not explain the truncation and thickness distribution of the AP unit. The implications of the adopted palaeo-water depths will be discussed below. In the next model we determine whether an increase in the level of compressive intraplate stress can explain the truncation of the AP unit and its thickness distribution. We adopted an increase in the level of compressive intraplate
stress starting at 3.9 Ma, reaching a value of 3 kbar at 2.4 Ma and subsequently decreasing to a present-day value of 1.5 kbar. As shown by an inspection of Fig. 8b the overall stratigraphy and basin shape predicted by the model compares well with the observed large scale pattern in the basin geometry and basin fill. The Pliocene increase in the level of compressive stress can explain the truncation of the A P unit. The predicted thickness of this unit in the deepest parts of the sub-basin better fits the observed thickness than the previous model which applied a lake-level change. The depth of lithospheric necking controls the shape and depth of the syn-rift basin. Shallow levels of necking cause undeep, wide syn-rift basins. Deep levels of necking induce deep synrift basins with a pronounced topography (Kooi & Cloetingh 1992; Van Balen & Cloetingh 1993). Due to the higher amount of syn-rift subsidence, models invoking deeper levels of necking require lower amounts of extension to generate the same total subsidence (syn- and post-rift) and, therefore, require larger palaeo-water depths to fit the observed thicknesses of syn- and first post-rift deposits. The result of a model adopting a 15 km deep level of necking during extension is depicted in Fig. 8c. This model invokes the same stress and lake-level history as the previous model. The adopted crustal extension factors and palaeo-water depths are displayed in the same figure. This model requires indeed a smaller amount of crustal extension and larger palaeo-water depths. Discussion
The modelling results shows that an increase in the level of compressive stress starting at early Pliocene can explain the observed truncation of the AP unit and its thickness in the deep parts of the Mako and B6k6s sub-basins. The compressive intraplate stress causes an uplift of the basement high located between the Drava and Mako sub-basins and the eastern flank of the B6k6s sub-basin (Fig. 9). This uplift and subsidence pattern is in accordance with the general uplift and subsidence occurring in the Pannonian Basin system (Fig. 4) (Horv~th & Cloetingh 1996). The tectono-stratigraphic forward modelling applying a 7.5 km deep level of lithospheric necking requires very deep palaeo-water depths (2.3 km) during the syn-rift and initial part of the post-rift phase for the deepest parts of the Mako and B6k6s sub-basins. This is in accordance with the syn-rift basin morphology, but is higher than previous estimates based on sedimentological evidence (e.g. Kfizm6r 1990; Pog~.cs~,s et al.
MODELLING IN THE PANNONIAN BASIN
403
Fig. 9. The compressive stress-induced subsidence and uplift pattern at the stress maximum (2.4 Ma). The maximum stress-induced subsidence occurs in the Mako trough. The eastern flank of the B6k6s basin and the high in between the Drava and Mako troughs are uplifted, causing a relative lake-level fall and stratigraphic offlap of the AP unit.
1994). This problem cannot further be resolved. Because the palaeo-water depths are already very deep for a lithospheric necking depth of 7.5 kin, a deeper level of necking is unlikely. The presented model applying a 15 km deep level of necking requires a palaeo-water depth of 2.5 km. We, therefore, conclude that the Pannonian Basin extended with a shallow level of necking (5-10 km) during rifting. This implies that the strength of the subcrustal lithosphere was negligible during the rifting process (Van Balen & Cloetingh 1993). This is in accordance with the results of Gaudemer et al. (1988) and Cloetingh et aL (1995), showing that the bulk lithospheric strength of a wide orogenic belt is almost completely determined by the strength of the upper crust, due to the restricted cooling possibilities and thickness of the crust.
Forward modelling of compaction-driven fluid flow and overpressure Using the results of tectono-stratigraphic forward modelling presented in the first part of this chapter, we have analysed the history of compaction-driven fluid flow and overpressures in the southern part of the Pannonian Basin system. Special emphasis is given to the Pliocene-Quaternary evolution. During this time interval the
subsidence in the central part of the basin system accelerates, causing sedimentation rates to increase. Sedimentation rates exert a first-order control on compaction driven overpressures (Bredehoeft & Hanshaw 1968; Bethke & Corbet 1988). Therefore, an increase in the amount of overpressure is expected to have occurred during the late stage evolution of the Pannonian Basin (Van Balen & Cloetingh 1994). Fluid overpressures
High overpressures are known to occur in the deep buried marly sequences in the Mako Trough and B6k6s Basin (Szalay 1982, 1988; Spencer et al. 1994). Generally, hydrostatic fluid pressures are present down to depths of about 1800 m in the Mako and B6k6s sub-basins (Szalay 1982, 1988; Clayton et al. 1990; Spencer et al. 1994). Hydrostatic pressures also occur in sediments overlying the basement highs flanking the sub-basins. Only few pressure measurements were made in the B6k6s Basin (Spencer et al. 1994). The amount of overpressuring measured at the northeastern flank of the B6k6s Basin is up to 15 MPa. At the western flank of the Mako Trough, in the Mako-I well, the maximum measured overpressure is around 45 MPa (Szalay 1982; Spencer et al. 1994). Overpressured oil flowed into this well during drilling
404
R.T. VAN BALEN E T A L .
(Horv~th et al. 1994). Therefore, part of this overpressure might be due to buoyancy forces. The overpressures measured by drill-stem tests in the H6d-I well in the centre of the Mako Trough are 18 MPa at 3.8 km, 25 MPa between 4.7 and 4.8 km and 40 MPa at 5.0 km (Szalay 1982). The inferred overpressures from the depth-porosity trend of pelitic sediments in the same well also shows a steadily increasing overpressure starting at a depth of about 2.5 km to a value of over 30 MPa at 4.8 km (Szalay 1988). Several overpressure mechanisms for the Mako Trough and B6k6s Basin have been proposed: undercompaction, hydrocarbon generation, aquathermal expansion due to the high heatflow and CO2 generation caused by thermal decomposition of carbonatic rocks in the basement (Spencer et al. 1994). Undercompaction is probably the prime mechanism for shallow overpressures. As overpressure increases dramatically at depths with temperatures of about 125~ hydrocarbon generation has been inferred to be a major contributor to overpressure at larger depths (Spencer et al. 1994). However, as shown by Szalay (1982, 1988), an almost perfect equilibrium relationship exists between porosity and effective pressure, implying that overpressure increase was contemporaneous with porosity decrease and denoting mechanical compaction as the prime overpressure causing mechanism. The overpressures have probably contributed considerably to the migration of hydrocarbons. The major Neogene source rock, the Miocene marl of the DB unit, has a very low permeability and, probably, high capillary pressure, which makes buoyancy an unlikely hydrocarbon migration driving force. Instead, the high overpressure occurring in the same beds has probably promoted migration in lateral and vertical directions, including downward (Spencer et al. 1994). The maximum overpressures, occurring at depths in excess of 3 km, measured in the northwestern part of the Drava Trough range from about 15 MPa in Molve field to about 10 MPa in the Stari Gradac field, promoting a secondary migration of hydrocarbons from WNW to ESE direction. Hydrostatic pressures occur down to a depth of 2 km (Barid et al. 1991).
depth versus porosity E
ECU v
"(30O
+ = sandstone z~ = marl J
(a)
.
,
J
L
]
.
,
L
,
,
,
,
,
L
,_J_•
J
9
i
10 20 porosity (%)
permeability versus porosity
-"o
7-
/
Q'u~
o Modelling parameters
For modelling purposes the lithology in the cross-section has been simplified to a mixture of idealized 'marl' and 'sandstone' types of sediment, based on the sedimentological description of the Hod-I well by Mattick et aL (1988; see Table 1). The porosity versus depth trends for
'i
(b)
~,-t- - sandstone - marl 0 2 3 log hor. permeability x 10^-16 m^2
Fig. 10. (a) Porosity versus depth trends for marls and sandstones in the southern part of the Pannonian Basin (after Szalay 1982; Spencer et aL 1994). (b) The relationship between porosity and horizontal permeability (after Szalay 1982).
MODELLING IN THE PANNONIAN BASIN
405
Table 3. Porosity and permeability constants' First estimate
qb0 bp ak bk kx/k z
Calibrated
Marl
Sandstone
Marl
Sandstone
0.65 8.82 • 10-8 16 10.5 11.7
0.475 4.5 • 10~8 16 18 1.62
0.65 8.82 • 10-8 20 8 11.7
0.475 4.5 • 10-8 18 8
these two lithologies (Szalay 1982; Spencer et al. 1994) are shown in Fig. 10a. The porosity-depth data are translated to porosity-effective pressure data and fitted to an exponential function required for the forward modelling (Van Balen & Cloetingh 1993, 1994): qb = qb0 exp-bp Pe with (b = porosity, 60 = surface porosity, Pe = effective pressure, bp = compaction coefficient. The deep porosities for marls (>3 kin) are not used, because these porosities are abnormally high due to fluid overpressures and, possibly, secondary porosity (A. Szalay pers. comm. 1993). The relationship between porosity and permeability depicted in Fig. 10b is reconstructed from permeability-depth data and the porosity-depth data for the two lithologies given by Szalay (1982) and Spencer et aL (1994). These data are fitted to functions given by: logk = -ak + bkqb with ak, bk = empirical constant, + = porosity, k = permeability. The permeability anisotropy is accounted for by adopting a vertical to horizontal permeability ratio. Porosity and permeability vary strongly in the subsurface. Thin interlayered shale beds control to a large extent the vertical permeability in a rock sequence (BjOrlykke 1993). Such beds may not have been sampled for permeability measurements. Furthermore, fractures were probably not taken into account in the laboratory measurements and presumably porosity and permeability were assessed at atmospheric pressure and surface temperature conditions. Therefore, we consider the derived porosity and permeability functions provided by the data as a first estimate. During the modelling process, the functions are further calibrated and the sand to marl ratios are adjusted until the predicted overpressures match the observations. The first estimate and finally obtained values are shown in Table 3. Essentially, the permeability values
1.62
determined by calibration show much lower values than those derived from the data. The minimum permeability for marl, for example, differs by four orders of magnitude (10 -16 v. 10- 2o m2). However, the minimum permeability value for marl and the values determined for sandstone are in keeping with values for shales given by Freeze & Cherry (1979), Bethke (1985) and Harrison & Summa (1991). The porosity-effective pressure functions did not require calibration. Adopted values for viscosity, compressibility and density of pore fluid and densities of sediment grains are given in Table 4. M o d e l l i n g results
The results of modelling of compaction-driven fluid flow and overpressures are shown for five different time slices (Fig. 11). These time slices are related to important changes in the basin evolution in the southern part of the Pannonian Basin system. The overpressures and flow directions at 6.3 Ma are depicted in Fig. l l a . Up to this stage, sedimentation rates in the sub-basins have been relatively low. Therefore, the overpressures, mainly occurring in the marls of the DB units, are subdued. The maximum overpressure is about 9.5 MPa (B6k6s Basin). The predicted compaction-driven fluid velocity field, shown in the same figure, demonstrates that the pore fluids are mainly expelled in lateral directions. The upward flow directions occur mainly in the upper 0.5 km. The maximum flow velocity is about 5.1 mm a -1. In order to show the large lateral component of fluid fluxes, the velocity fields depicted in Fig. 11 are not scaled. Therefore, true velocity directions are shown in these Table 4. Additional fluid modelling parameters" Pore fluid visocisity Pore fluid compressibility Fluid density Grain density
5 • 10-4 Pa s 4.3 • 10-1~Pa -1 1.024 g cm -3 2.7 g cm ~
406
R . T . VAN B A L E N E T A L .
MODELLING IN THE PANNONIAN BASIN
407
Fig. 11. The history of compaction driven overpressures and fluid flows predicted by our forward modelling. Vertical exaggeration of the basin cross-section is about 25. The velocity fields have not been scaled. (a) Overpressures and flows just before the arrival of the Pannonian delta (6.3 Ma). (b) The increase in sedimentation rates during the delta progradation causes overpressure to increase considerably (5.5 Ma). (c) The compressive stress induced acceleration of subsidence in the basin center causes sedimentation rates to increase again, inducing even higher overpressures (3.9 Ma). (d) Overpressure and flows predicted for the present day situation. The predicted overpressures in the centre of the Mako trough match with the inferred overpressures (Fig. 12). See text for a discussion of the results.
figures. The basin cross-section is, however, scaled with a factor of about 25. If the velocity field would also be scaled, it would show only vertical flow directions. During deposition of the delta slope deposits (DS unit), sedimentation rates increase considerably. As a consequence, overpressures increase to a maximum value of 17.8 MPa in the DB unit (Fig. 11b) at the end of deposition of this unit. Compared to the previous time frame, fluid-flow velocities have decreased slightly. After deposition of the A P unit, at the peak of compressional intraplate stress, overpressures have reached a maximum value of 35.7 MPa, due to the stress-induced increase of sedimentation rates (Fig. 11c). Maximum compaction-driven flow velocities are 2.3 mm a -1. Our prediction for the present day compaction driven overpressures and flows is shown in Fig. 11d. Maximum overpressure, occurring in the deep basin marls of the Mako Trough, is around 40.9 MPa. Flow velocities have decreased slightly to a maximum value of 2.2 mm a -1. The predicted overpressures at the centre of the Mako Trough are in reasonable agreement with measured and inferred overpressures in the Hod-I well (Fig. 12). In general, the overpressures are maximal in the DB unit, which is caused by the combination of low permeability and high porosity at deposition. The overpressures
decrease upwards and downwards from the maximum overpressure zone. The overpressure decrease with depth is high enough to promote downward migration of hydrocarbons into the permeable basal unit and basement (Mako Trough: 11 MPa; B6k6s Basin, 8.5 MPa). Additionally, our modelling predicts slightly less present-day overpressuring in the B6k6s Basin than in the Mako Trough.
Discussion on forward overpressure modelling results During the increase of the level of compressive intraplate stress the southwestern margin of the M a k o Trough was uplifted. This may have caused degassing of pore fluids due to pressure release. Furthermore, the uplift increased the gravity potential of meteoric waters which could, therefore, p e n e t r a t e deeper into the basin. Isotopical, geochemical and diagenetic data show that meteoric water has penetrated into the PD unit around the basement highs of the Derecske Basin (M~ity~s & Matter 1998), which was also uplifted during the P l i o c e n e - Q u a t e r n a r y (Vakarcs et al. 1994; Horv~th & Cloetingh 1996). The tectonic uplift of these highs combined with a preceding lakelevel fall caused an increase of the gravity potential enhancing the influx of these waters.
408
R.T. VAN BALEN E T A L . overpressure ( MPa ) 0
10
20
30
40
0
mud weight data
Trough and B6k6s Basin (PusztafOldvfir field; Dank 1988; Fig. 6). Therefore, the pore fluids flow towards the hydrocarbon pools. Thus, the compaction-driven fluid flow may have contributed to the secondary migration of hydrocarbons from the deeper parts of the sub-basins towards the highs. Our modelling results suggest therefore that, in addition to known hydrocarbon occurrences, more fields can be found on the eastern flank of the B6k6s Basin.
t-,-
Artificial neural n e t w o r k analyses o f overpressure data porosity
estimated X ~ ~ , , ~ , ~ . ~
Fig. 12. Overpressure profile in the centre of the Mako trough (Hod-I well) inferred from porosity trends and drill-stem tests (crosses) (after Szalay 1982) and the predicted present-day overpressure at the same position by our forward modelling. The predicted overpressure is in reasonable agreement with the inferred and measured overpressure.
The hydrocarbon generation started between 9 and 6 Ma (Horv~ith et al. 1987; Szalay 1988), implying that during the stress-induced change of the fluid regime the hydrocarbons are mature and actively expelled in the southern part of the Pannonian Basin. As the marl sequences are the major source rock for hydrocarbons in the Neogene sediments and the overpressure maximum occurs exactly in this unit, the increase of overpressure has probably enhanced the primary migration considerably. Our modelling also shows that the overpressures decrease with depth in the deepest part of the basin, promoting a downward migration of hydrocarbons, enhancing the prospectivity of the deepest unit. The predicted compaction-driven fluid velocity field shows that the pore fluids are expelled from the deep sub-basins mainly in lateral directions towards the basement highs. The largest velocities occur in the delta plain unit at the southwestern margin of the Mako Trough and the northeastern margin of the Bdkds Basin. The major hydrocarbon occurrences in the southeastern part of the Pannonian Basin system are located at the southwestern margin of the Mako Trough (AlgyO field) and the high inbetween the Mako
In order to compare the predicted present day overpressure to known overpressures in the area of interest, we have performed an analyses of overpressure data, based on neural network technology (Van Balen & Cloetingh 1995). Artificial neural networks are software tools inspired by brain models, which are used for pattern recognition purposes (Sejnowski et al. 1988). They are capable of learning, i.e. they can find relationships in data. Artificial neural networks learn by 'investigating' a set of training data. Once the relationship in these data has been found they can interpolate and extrapolate to new values. For data analyses purposes, artificial neural networks provide an alternative to statistical regression methods because they can find functional relationships in data sets without any a priori knowledge about their form (linear, exponential, etc.). We applied a neural network to mapping, interpolation and extrapolation of fluid overpressures in the southeastern part of the Pannonian Basin. The trained networks are used to assess overpressures along the crosssection depicted in Fig. 13a. The locations of the wells from which the overpressure data are used are shown in Fig. 13b. The number of input variables for the analysis is determined by parameters influencing the fluid overpressure in sedimentary basins in general and the Pannonian Basin in particular. We have defined five such parameters: burial depth, composition, thickness of overlying Quaternary deposits and geographical distances (2 parameters, plane coordinates). In a separate test, all five parameters were found to be significantly influencing the results. The data set for the network analysis is compiled from different sources. Fluid overpressure data for several wells are reported in Szalay (1982, 1988) and Clayton et al. (1990). The wells used in this study are indicated in Fig. 13b. In this part of the Pannonian Basin there is only one well which reaches a depth of 5.4 km: Hod-I
MODELLING IN THE PANNONIAN BASIN
409
Fig. 13. (a) The Pannonian basin and its surroundings (after Horvfith 1993). The inset shows the position of (b). A-A' denotes the modelled cross-section. (b) The locations of the wells used in this analyses plotted on the isopach map. X and Y are arbitrary geographical coordinate axes. For scale (b) located in the centre of the Mako trough. All the other wells are much shallower, typical depths are 2-3 km. As can be seen in Fig. 13b, the deep part of the B6k6s Basin is not penetrated by a well. For this sub-basin overpressure information is only available for the outer, shallower part. As will be shown below, neural network
analysis enables extrapolation down to a depth of 6 km in this sub-basin. Sediment compositions (the percentage of sandstone in the deposits) for the lithostratigraphic units in the southeastern part of the Pannonian Basin are based on Mattick et al. (1988). The distribution, thickness and depth of these
410
R.T. VAN BALEN ETAL.
lithostratigraphic units are given by Juhfisz (1991). The thickness and distribution of the Quaternary deposits are presented by R6nai (1974). Finally, the geographical coordinate axes we have defined are indicated in Fig. 13b. Using these sources, we have assembled a database with 193 records. During training of the network 150 patterns (data records) are used, while the remaining 43 patterns are used to compare the predicted and known overpressures during the testing phase. A neural network with a topology consisting of one hidden layer with three neurons is found to be the optimal configuration (Van Balen & Cloetingh 1995). A comparison between overpressure profile predicted by the neural network analyses and the forward modelling is shown in Fig. 14. The results are in fairly good agreement. We, therefore, conclude that results from both methods confirm each other. Discussion
As shown in both numerical forward modelling and neural network analyses, the overpressure decreases at the deepest parts of the sub-basins, in the basal sandstone unit. The amount of overpressure is, however, not equal in the sub-basins. Both methods predict an overpressure in the B6k6s Basin which is slightly less than the overpressure in the Mako Trough. This can be explained by spatial variations in the thickness of the Quaternary deposits. As the Quaternary is thicker in the Mako Trough, sedimentation rates during Quaternary above this sub-basin were higher than in the B6k6s Basin, inducing faster compaction. The anomalous thickness of the Quaternary deposits can be explained by compression induced lithospheric deflection since early Pliocene. Due to the distribution of vertical lithospheric loads, the maximum of stress induced subsidence is located exactly above the Mako Trough.
Conclusions On basis of this modelling study several important conclusions can be made regarding palaeowater depths, lake-level fluctuation, lithospheric necking depth during extension, intraplate stress history, extension factors, timing of overpressures and the directions of fluid flow. These are presented below. The tectono-stratigraphic forward model invoking a lithospheric necking depth of 7.5 km is best in accordance with earthquake, gravity and lithospheric data, and requires reasonable
lake-level variations, palaeo-water depths and stretching values. The relatively shallow necking depth in the Pannonian Basin probably reflects the pre-rift weak bulk rheology of the lithosphere due to Cretaceous and Palaeogene crustal thickening. The stratigraphic forward modelling requires a palaeo-water depth of 2300 m during deposition of the deep basin marls in the centre of the Mako Trough. Although this value is large, it is in keeping with observations. A lake-level fall can explain the truncation and thickness distribution of the delta plain (DP) depositional unit. The truncation of Pliocene alluvial plain sediments (AP unit) can not be explained in terms of lake-level changes This agrees with the fact that during this time interval the Pannonian lake became extinct. Therefore, lake-level changes could not affect the stratigraphy anymore. Palaeo-stress analyses, borehole break-out data, focal mechanism solutions and in-situ stress measurements indicate that the Pannonian Basin is in a compressive state of stress since early Pliocene. Our modelling results show that an increase in the level of compressive stress during Pliocene followed by a slight decrease during Quaternary can successfully explain the observed truncation patterns and thickness distribution of the alluvial plain (AP) and Quaternary (Q) depositional units. The compressive stress induces differential movements across the basin system, consisting of uplift at the basin system flanks and subsidence in the sub-basins. Forward modelling of compaction-driven fluid flow and overpressures in the Pannonian Basin shows that overpressures have increased twice due to accelerations of sedimentation rates. The first overpressure increase event occurs during deposition of the delta slope deposits (DS unit) between 6.3 and 5.5 Ma. Starting in the early Pliocene (3.9 Ma), an increase in the level of compressive stress causes subsidence and sedimentation rates to increase again, which enhances overpressuring substantially. This has several implications for the diagenesis of sediments and migration of hydrocarbons. G. Spadini and P. Szafian are thanked for stimulating discussions. G. Vakarcs and an anonymous reviewer provided useful reviews of the manuscript. This research was funded by the Hungarian Research Fund (OTKA) and IBS (Integrated Basin Studies) project, part of the JOULE II research programme funded by the Commission of European Communities (contract no. JOU2-CT 92-0110). Publication No. 970110 of the Netherlands Research School of Sedimentary Geology.
M O D E L L I N G IN T H E P A N N O N I A N BASIN
411
Fig. 14. (a) Overpressure through the Mako Trough and B6kds Basin produced by the 5 • 3 • 1 neural network. Thick lines are contour lines for the overpressures (MPa), thin lines represent the stratigraphy. A n overpressure reversal occurs in the deepest part of the sub-basins. Generally, the overpressures in the Mako Trough are less than in the B6k6s Basin. (b) Overpressure profile predicted by numerical forward modelling (close up of Fig. 1 ld), showing similarities to the overpressure profiles obtained by the neural network analyses. Thick lines are contourlines for the overpressure (MPa), thin lines represent the modelled stratigraphy. The modelled overpressure profile also shows the overpressure reversal and a slightly less overpressure in the B6k6s Basin inferred from the neural network analyses.
References BADA, G., GERNER, P., SZAFIAN, P., CLOETINGH, S. & HORVATH, E 1995. Sources of recent tectonic stress in the Pannonian region derived from finite element modelling. Terra Abstracts (EUG 8), supplement to Terra Nova, 7, 173. BARIC, G., MESIC, M., JUNGWIRTH, I., & SPANIC, D. 1991. Gas and condensate-fields in the north-west of the Drava depression, Yugoslavia. In: SPENCER, A. M. (ed.) Generation, Accumulation and Production of Europe's Hydrocarbons I. European Association of Petroleum Geologists, Special Publications, 1. Oxford University Press, Oxford, 323-340. BECKER, A. 1993. Contemporary state of stress and neotectonic deformation in the Carpathian-Pannonian region. Terra Nova, 5, 375-388.
BI~RCZI, I. 1988. Preliminary sedimentological investigation of a N e o g e n e depression in the Great Hungarian Plain. In: ROYDEN, L. H. & HORVATH, E (eds) The Pannonian Basin. American Association of Petroleum Geologists Memoir, 45, 107-116. -& PrtILLIPS, R. L. 1985. Processes and depositional environments within Neogene deltaiclacustrine sediments, Pannonian Basin, southeast Hungary. Geophysical Transactions, 31, 55-74. - - - , HAMOR, H., JAMBOR, A. & SZENTGYORGYI, K. 1988. Neogene sedimentation in Hungary. In: ROYDEN, L. H. & HORVATH,E (eds) The Pannian Basin. American Association of Petroleum Geologists Memoir, 45, 27-48. BETHKE, C. M. 1985. A numerical model of compaction-driven groundwater flow and heat transfer and its application to the paleohydrology of
R.T. VAN BALEN ET AL.
412
intracratonic sedimentary basins. Journal of Geo90, 6817-6828. - & SORBET, T. E 1988. Linear and nonlinear solutions for one-dimensional compaction flow in sedimentary basins. Water Resources Research, 24, 461-467. BJORLYKKE, K. 1993. Fluid flow in sedimentary basins. Sedimentary Geology, 26, 137-158. BREDEHOEFr, J. D. & HANSHAW, B. B. 1968. On the maintenance of anomalous fluid pressures I, Thick sedimentary sequences. Geological Society of America Bulletin, 79, 1097-1106. CLAYTON,J. L., SPENCER, C. W., KONCS, I. & SZALAY,A. 1990. Origin migration of hydrocarbon gases and carbon dioxide, B6k6s Basin, southeastern Hungary. Organic Geochemistry, 15, 233-247. CLOETINGH, S., MCQUEEN, H. & LAMBECK, K. 1985. On a tectonic mechanism for regional sea level variations. Earth and Planetary Science Letters, 75, 157-166. , VAN WEES, J. D.,VAN DER BEEK, R A. & SPADINI, G. 1995. Role of pre-rift Theology in kinematics of extensional basin formation: constraints from thermomechanical models of Mediterranean and intracratonic basins. Marine and Petroleum Geology, 14, 793-807. CSAT0, I. 1993. Neogene sequences in the Pannonian Basin, Hungary. Tectonophysics, 226, 377-400. Csoyros, L., NAGYMAROSY,A., HORVATH,E & KOVAC, M. 1992. Tertiary evolution of the IntraCarpathian area: a model. Tectonophysics, 208, 221-241. , TARI, G., BERGERAT,E & FODOR, L. 1991. Evolution of the stress fields in the Carpatho-Pannonian area during the Neogene. Tectonophysics, 199, 73-91. DANK, V. 1988. Petroleum geology of the Pannonian Basin, Hungary: an overview. In: ROYDEN,L. H. & HORVATH, E (eds) The Pannian Basin. American Association of Petroleum Geologists Memoir, 45, 319-331. FREEZE, J. A. & CHERRY, J. A. 1979. Groundwater. Prentice Hall, Englewood Cliff, New Jersey. GABRIS, G. 1994. Pleistocene evolution of the Danube in the Carpathian Basin. Terra Nova, 6, 495-501. GAUDEMER, Y., JAUPART, C. 8,: TAPPONNIER, P. 1988. Thermal control on post-orogenic extension in collision belts. Earth and Planetary Science Letters, 89, 48-62, GREGERSON. S. 1992. Crustal stress regime in Fennoscandia from focal mechanism. Journal of Geophysical Research, 97, 11 821-11 827. GROW, J. A., MATTICK,R. E., BERCZI-MAKK, A., PERO, C., HAJD(J, D., POGASCAS, G., V,~,RNAI, P. & VARGA, g. 1994. Structure of the B6k6s Basin inferred from seismic reflection, well and gravity data. In: TELEKI, P. G., MATTICK, R. E. & KOKAI, J. (eds) Basin analysis is in petroleum explor-
physical Research,
ation, a case study from the Bikes Basin, Hungary. Kluwer Academic Publishers, Dordrecht, 1-38. GR13NTHAL, G. & STROMEYER, D. 1982. The recent crustal stress field in Central Europe: trajectories
and finite element modeling. Journal of Geophysical Research, 97, 11 805-11 820. HARRISON,W. J. & SUMMA,L. L. 1991. Paleohydrology of the Gulf of Mexico Basin. American Journal of Science, 291, 109-176. HAQ, B. U., HARDENBOL, J. & VAIL, P. 1987. Chronology of fluctuating sea levels since the Triassic. Science, 235, 1156-1167. HORVATn, E 1985. Thickness of Neogene-Quaternary basin fill (map). Hungarian Geological Institute, Budapest, Hungary. - 1993. Towards a kinematic model for the Pannonian Basin. Tectonophysics, 226, 333-357. & BERCKHEMER, H. 1982. Mediterranean backarc basins. In: BERCKnEMER, H. & HSO, K. (eds) Alpine-Mediterranean geodynamics. Geodynamic Series 7. American Geophysical Union and Geological Society of America, 141-173. -& CLOETINGH,S. 1996. Stress-induced late-stage subsidence anomalies in the PannonianBasin. Tectonophysics, 266, 287-300. - - , DOVENYI, P., PAPP,S., SZALAY,A. & HARGITAY,M. 1987. Tectonic control on migration and trapping of hydrocarbons in the Pannonian Basin. In: DOLIOEZ, B. (ed.) Migration of hydrocarbons in sedimentary basins. Editions Technip, Paris, 667-681. --, SZALAY, A. & ROYDEN, L. 1988. Subsidence, thermal, and maturation history of the Great Hungarian Plain. In: ROYDEN, L. H. & HORVATH, E (eds) The Pannian Basin. American Association of Petroleum Geologists Memoirs, 45, 355-372. JAMBOR,/~. 1989. Review of the geology of the s.1. Pannonian formations of Hungary. Acta Geologica Hungarica, 32, 269-324. Joo, I., CZOBOR,A., GAZSO, M. & NEMETH,Z. 1990. On recent crustal movements in the Pannonian Basin.
Acta Geodynamica, Geophysical et Montanista Hungarica, 25, 231-242. JUnASZ, G. 1991. Lithostratigraphical and sedimentological framework of the Pannonian (s.l.) sedimentary sequence in the Hungarian Plain (Alf61d), Eastern Hungary. Acta Geologica Hungarica, 34, 53-72. KAZMER, M. 199(I. Birth, life and death of the Pannonian Lake. Palaeogeography, Palaeoclimatology, Palaeoecology, 79, 171-188. KOOl, H. & CLOETINGH,S. 1992. Lithospheric necking and regional isostasy at extensional basins 2. Stress induced vertical motions and relative sea level changes. Journal of Geophysical Research, 97, 17 573-17 591. KORPAs-HODI, M., POGA('SAS, G. & SIMON, E. 1992. Paleogeographic outlines of the Pannonian s.1. of the southern Danube-Tisza interfluve. Acta Geological Hungarica, 35, 145-163. LAZARESCU, V., CORNEA, I., RAOULESCU, E & POPESCU, M. 1983. Moho surface and recent crustal movements in Romania: geodynamic connections. Proceedings of the Institute of Geology and Geophysics in Bucarest, 63, 83-91. LORINCZ, K. & SZAB0, P. 1993. Seismic analysis of
MODELLING IN THE PANNONIAN BASIN multi-phase tectonism in the central part of the Pannonian Basin in Hungary. In: SPENCER,A. M. (ed.) Generation, Accumulation and Production of Europe's Hydrocarbons IlL European Association of Petroleum Geologists, Special Publications, 1. Oxford University Press, Oxford, 311-323. MATYAS, J. & MATTER, A. 1998. Diagenetic indicators of meteoric flow in the Pannonian Basin, SE Hungary. Marine and Petroleum Geology, in press. MATTICK, R. E., PHILIPS, R. L. & RUMPLER, J. 1988. Seismic stratigraphy and depositional framework of sedimentary rocks in the Pannonian Basin, southeastern Hungary. In: ROYDEN, L. H. & HORVATH, E (eds) The Pannian Basin. American Association of Petroleum Geologists Memoirs, 45, 117-146. - - , RUMPLER,J., UJFALUSY,a . , SZANYI,B. & NAGY,I. 1994. Sequence stratigraphy of the B6k6s Basion. In: TELEKI,P. G., MATTICK,R. E. & KOKAt, J. (eds) Basin analysis in petroleum exploration, a case study from the B~k~s Basin, Hungary. Kluwer Academic Publishers, Dordrecht, 39~55. MCKENZIE, D. R 1978. Some remarks on the development of sedimentary basins. Earth and Planetary Science Letters, 40, 25-31. MOLLER, B., ZOBACK, M. L., FUCHS, K., MASTIN, L., GREGERSEN, S., PAVONI, N., STEPHANSON, 0. & LJUNGGREN, C. 1992. Regional patterns of tectonic stress in Europe. Journal of Geophysical Research, 97, 11 783-11 803. NAGYMAROSY,A. & Mf)LLER, P. 1988. Some aspects of Neogene biostratigraphy in the Pannonian Basin. In: ROYDEN, L. H. & HORVATH, E (eds) The Pannian Basin. American Association of Petroleum Geologists, Memoirs, 45, 58-68. POGACSAS, G., MATTICK,R. E., ELSTON,D. P., HAMOR, T., JAMBOR, A., LAKATOS,L., LANTOS,M., SIMON, E.,VAKARCS,G.,VARKONYI,L. & VARNAI,R 1994. Correlation of seismo- and magnetostratigraphy in southeastern Hungary. In: TELEKI, P. G., MATTICK,R. E. & KOKAI,J. (eds) Basin analysis" in petroleum exploration, a case study from the BOkOs Basin, Hungary. Kluwer Academic Publishers, Dordrecht, 143-160. POLOMC, G. 1985. Neotectonic activity at the eastern border of the Pannonian depression and its seismic implications. Tectonophysics, 117, 109-115. POSGAY,K. T., BODOKY,T., HEGEDOS, E., KOVACSVOLGYI, S., LENKEY,L., SZAFIAN,P., TAKACS,E., IMAR, Z. & VARGA, G. 1996. Asthenospheric structure beneath a Neogene basin in SE Hungary. Tectonophysics, in press. RATSCHBACHER,L., FRISCH,W., LINZER, H. & MERLE, O. 1991. Lateral extrusion in the Eastern Alps, part 2: structural analyses. Tectonics, 10, 257-271. RONAI, A. 1974. Size of Quaternary movements in Hungary's area. Acta Geologica Academiae Scientiarum Hungaricae, 18, 39-44. 1985. Limnic and terrestrial sedimentation and the N/Q boundary in the Pannonian Basin. In:
-
-
413
KRETZOI, M. & PECSJ, M. (eds) Problems of the Neogene and Quaternary. Akad6miai Kiad6, Budapest, 21-49. ROYDEN, L. H. & DOVI~NYI, P. 1988. Variations in extensional styles at depth across the Pannonian Basin. In: ROYDEN,L. H. & HORVATH,E (eds) The Pannian Basin. American Association of Petroleum Geologists Memoirs, 45, 235-255. & KEEN, C. E. 1980. Rifting processes and thermal evolution of the continental margin of eastern Canada determined from subsidence curves. Earth and Planetary Science Letters, 51, 343-361. --, HORVATH, E • NAGYMAROSY,A. 1983. Evolution of the Pannonian Basin system 2: subsidence and thermal history. Tectonics, 2, 91-137. RUMPLER,J. & HORVATH,E 1988. Some representative seismic reflection lines from the Pannonian Basin and their structural interpretation. In: ROYDEN,L. H. & HORVATH, E (eds) The Pannian Basin. American Association of Petroleum Geologists Memoirs, 45, 153-171. SEJNOWSKI,T. J., KOCH, C. & CHURCHLAND,P. S. 1988. Computational neuroscience. Science, 241, 1299-1306. SPENCER, C. W., SZALAY,a . & TATAR, E. 1994. Abnormal pressure and hydrocarbon migration in the Bdk6s Basin. In: TELEKI, P. G., MATTICK, R. E. & K(SKAI,J. (eds) Basin analysis in petroleum exploration, a case study from the BdkOs Basin, Hungary. Kluwer Academic Publishers, Dordrecht, 201-219. STEGENA, L., GECZu B. & HORVATH, E 1975. Late Cenozoic evolution of the Pannonian Basin. Tectonophysics, 26, 71-90. SZALAY,A. 1982. A rekonstrukci6s szemldletii fOldtani kutatds lehetOsOgei a szOnhidrog~n-perspektiv6k elOrejelzds~ben [Possibilities of the reconstruction of basin evolution in the prediction of hydrocarbon prospects]. PhD thesis, Hungarian Academic of Sciences, Budapest. 1988. Maturation and migration of hydrocarbons, SE PannonianBasin. In: ROYDEN, L. H. & HORVATH, E (eds) The Pannian Basin. American Association of Petroleum Geologists Memoirs, 45, 347-354. TARI, G., HORVATH, E & RUMPLER, J. 1992. Styles of extension in the Pannonian Basin. Tectonophysics, 208, 203-219. UJSZASZI, K. & VAKARCS, G. 1993. Sequence stratigraphic analysis in the south Transdanubian region, Hungary. Geophysical Transactions, 38, 69-87. VAKARCS, G., VAIL, P. R., TARI, G., POGACSAS, GY., MATTICK, R. E., SZABO, A. & BOWMAN, S. 1994. Third-order sequence Miocene-Pliocene sequences in the prograding delta complex of the Pannonian Basin. Tectonophysics, 240, 81-106. VAN BALEN,R. & CLOETINGH,S. 1993. Stress-induced fluid flow in rifted basins. In: HORBURY,A. D. & ROBINSON, A. G. (eds) Diagenesis and Basin Development. A A P G Studies in Geology series, 36, 87-98. -
-
414 -
-
R.T. VAN B A L E N ETAL. & 1994. Tectonic control of the sedimentary record and stress induced fluid flow: constraints from basin modelling. In: PARNELL, J. (ed.)
Geofluids: origin, migration and evolution of fluids in sedimentary basins. Geological Society, London, Special Publications, 78, 9-26.
-
-
& -1995. Neural network analyses of stressinduced overpressures in the Pannonian Basin. Geophysical Journal International, 121, 532-544.
The present-day tectonics of the Aegean as deduced from seismicity DENIS HATZFELD L a b o r a t o i r e de G ~ o p h y s i q u e Interne et Tectonophysique, I R I G M , U J F - C N R S , B P 53, 38041 G r e n o b l e C e d e x 9, France (e-mail: D e n i s . H a t z f e l d @ o b s . ujf-grenoble.fr)
Abstract: We present a review of the seismicity and focal mechanisms for the Aegean (historical, teleseismically and microseismically computed), which we compare with the recent active faults. A very consistent trend in the T-axes within the Aegean across blocks of different size and orientation suggests that the surface tectonics of the Aegean is controlled by the deformation of the lithosphere as a whole. Some recently active faults (mostly the NW-SE-striking faults that are located west of the Aegean sea) are not associated with seismicity, some seismicity is associated with new E-W-striking faults of modest displacement that are mostly seen in continental Greece. We suggest that, due to the important amount of internal deformation of the Aegean lithosphere and the inferred rotations, the localization of the deformation changes with time and a precise timing of the recent faults activity is necessary to correlate with present-day deformation. We also suggest that, for continental crust, the work to create new faults that strike perpendicular to the main strain direction (as deduced from focal mechanisms and GPS measurements) is less than the work to reactivate faults that strike obliquely to the strain direction. We propose that a diffuse system of normal faults or grabens starts to develop, parallel to the Gulf of Corinth, and links the western termination of the North Aegean Trough to the northern end of the Kefallinia fault. In continental regions, deformation is distributed over wide areas and, if the amount of deformation is significant, the location of the active surface features changes with time. We observe block rotations and creation or death of active faults while there is no change in the boundary conditions (McKenzie & Jackson 1983; Roberts & Jackson 1991). The picture of the active faults is, therefore, strongly dependent upon the considered age of the structure (Mercier et al. 1987). Some faults seem to be 'locked', others seem to slip aseismically, whereas earthquakes occur in places that do not seem related to active faults. Because tectonics, palaeomagnetism and seismology may be related to different periods of time, contradictions occur that may not be due to the different techniques considered but to the dating of the event. For these rapidly deforming regions, it is necessary to ensure that the presentday tectonics is described by observations related to present-day active structures. The A e g e a n is a region of fast and intense deformation whose dimensions are only several h u n d r e d kilometres wide (Jackson 1994). Apparent contradictions are observed between active faults (Jackson 1994; Jolivet et al. 1994; Armijo et al. 1996) and seismicity (Jackson & McKenzie 1988). A n d different geodynamic models have been suggested (e.g. McKenzie 1978; Taymaz et al. 1991; Le Pichon et al. 1995; Armijo et al. 1996) that do not agree with each other. In this paper, we summarize some of the seismological observations that have been
already published elsewhere (e.g. Hatzfeld et al. 1997) and assume they constrain both the present active tectonics and the actual strain pattern. We compare the seismicity with recent faults inferred from surface observations and propose a rough sketch of the principal presentday active faults.
Geodynamic background T h e A e g e a n region is located b e t w e e n the African and European lithospheric plates that converge at a rate of about 1.5 cm a -1 (Fig. 1). T h e motion across the Hellenic trench is, however, of about 5 cm a -1, probably due to the westward motion of Turkey and the southwestward motion of the Aegean (McKenzie 1978; Jackson 1994; Le Pichon et al. 1995). This motion is related to the active lithospheric subduction beneath the Hellenic Trench (Papazachos 1973), but it does not describe completely the tectonics of the Aegean, which experiences also internal deformation as evidenced by geological observations (Angelier 1979), geodetic displacements (Veis et al. 1992), focal mechanisms (e.g. Hatzfeld et al. t993) and paleomagnetic rotations (Kissel & Laj 1988). Shortening is mostly observed along the external part of the Hellenic arc from Epirus to Rhodos, generally oriented perpendicular to the active boundary. Extension, trending N-S, affects the internal part of the Aegean, but trends differently along the Hellenic arc. This
HATZFELD,D. 1999. The present-day tectonics of the Aegean as deduced from seismicity. In: DURAND,B., JOLIVET,L., HORVATH,E & St~RANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 415-426.
416
D. HATZFELD
Fig. 1. Sketch of the main active boundaries and NEIC-seismicity (mb > 4.5) until 1996. Principal places described in the text: A, Amvrahikos; C, Corinth; E, Evia; Ka, Kavala; Kr, Kremasta; M, Mygdonian; S, Saros; Te, Thessaly; Tr, Trikhonis; V, Volos; The seismicity defines four branches for the North Aegean Trough that continue toward the Kefallinia zone. The Saros, Thessaly and Evia basins have low seismicity.
extension affecting the Aegean probably started 15 Ma ago (Mercier 1977; Jolivet et al. 1994) or even 22 Ma ago (Gautier & Brun 1994), but the relation with the westward motion of Turkey is unclear because the motion along the western termination of the North Anatolian fault started probably only 5 Ma ago (Jolivet et al. 1994; Le Pichon et al. 1995; Armijo et al. 1996). The most prominent faults, more recent than Miocene time, have been summarized by Mercier et al. (1976), Angelier (1979), Lyberis (1984), Pavlides & Mountrakis (1987), Mascle & Martin (1990), Roberts & Jackson (1991), Jolivet et al. (1994), Armijo et al. (1996). Faults are mostly normal within Central Greece, Pelo-
ponnese and on islands of the Aegean sea, and dextral strike slip in the North Aegean sea. But the details in the sketch of the active faults differ from one author to another, especially for Central Greece and the North Aegean sea. In this paper we try to identify active faults by comparing tectonic information with seismicity and infer a present-day tectonic map.
Data Seismicity
Shallow earthquakes are related to fractures that occur within the brittle part of the crust. In
TECTONICS & SEISMICITY IN THE AEGEAN order to bring relevant information to geodynamical problems, however, the seismicity maps should follow some basic rules. The location of the earthquakes should be mapped with an accuracy that is comparable to the scale of the considered tectonic problem and seismicity should be representative of long term deformation at a scale that is comparable to dimensions of large tectonic faults. Earthquakes of magnitude greater than 5.5 are usually recorded at teleseismic distances (greater than 30~ in a spherical Earth with an accuracy strongly dependent on the velocity structure at the focus of the earthquake and on the number and the azimuthal coverage of the recording stations. We used the NEIC catalogue that is probably complete for magnitude greater than 3.5 since 1963. Earthquakes of moderate magnitude (above 3.0) are recorded in regional seismological networks. The epicentral distances are usually smaller than 30 ~ which is the range where refracted waves are observed with weak onsets. They can provide good locations but generally a poor control in the depth of the earthquakes. Temporary experiments conducted with dense networks usually provide reliable information but for small magnitude earthquakes that occur during a short period of time and which might be not representative of the long term deformation. During the years 1984-1993, we installed several times, in different regions, temporary networks of portable seismological stations to locate the seismicity more precisely (especially the depth of the earthquakes) and compute more focal mechanisms. The duration of the experiments was generally 6-8 weeks and we installed 30-100 stations over regions that were approximately 150 • 150 km 2 in area. Because the instrumental seismicity covers a period of time that is short compared to loading processes of faults, it could be not representative of tectonic activity. We therefore also examined the historical record of seismicity of the Aegean (Papazachos & Papazachou 1997), which is incomplete, but provides complementary information to the recent measurements. Focal mechanisms
Fault plane solutions are also computed with different methods and different datasets. Mechanisms computed at teleseismic distances are generally good for events of magnitude larger than 5.5, when the focal sphere is well sampled. Mechanisms computed with first motion polarities at regional distances are questionable because, due to refracted waves,
417
they sample only a limited part of the focal sphere. Mechanisms computed for microearthquakes are well controlled when enough polarities are available (more than 8-12 readings sampling a minimum of three quadrants). If the earthquake is well recorded in several long period stations, it is possible to compare the observed seismograms with synthetics that are computed for body waves at teleseismic distances. This method, which gives a good control on the parameters of the source properties (depth, focal mechanism, source function), is possible only for magnitudes greater than 5.5 and therefore does not provide many solutions in regions of moderate seismicity. Centroid moment tensor solutions are routinely performed automatically by Harvard (Dziewonski et al. 1988) for earthquakes of magnitude above 5.0. As in any automatic procedure, some unreliable solutions can be computed accidentally.
Shallow seismicity In this section we will examine three different data sets that are complementary. (1) The teleseismically located seismicity by NEIC shows a complex pattern (Fig. 1). Most of the seismicity is observed along the active boundary of western Greece and the Hellenic Trench but does not extend further south. Important seismicity is also observed in the North Aegean sea and is clearly associated with the different branches of the North Aegean trough. The sea of Crete is almost aseismic. In continental Greece, beside the active region located around the Gulf of Corinth, seismicity is spread over wide areas and not clearly limited to obvious surface faults. Most of the seismicity is located south of a line which joins Volos to Kefallinia. (2) The seismicity that we recorded during our temporary experiments (Fig. 2) is not a homogeneous picture of the seismic activity of the Aegean, but it is the juxtaposition of several detailed studies. However, it confirms some of the aspects of the teleseismically located seismicity. We confirm the concentration associated with the Hellenic Trench and the continental collision. The sea of Crete is almost free of earthquakes and we located only a few events around the Cyclades islands. In the Peloponnese (Hatzfeld et al. 1990) and in western Greece (Hatzfeld et al. 1995), we cannot easily associate earthquakes with individual surface faults. The seismicity is diffuse with depth ranging between 5 and 20 km. In a few places in Central Greece, however, as around the Gulf of Corinth
418
D. HATZFELD
Fig. 2. Seismicity map recorded during the microearthquake experiments that we conducted from 1984 to 1993. This is not an homogeneous picture of seismicity but complements Fig. 1.
(Hatzfeld et al. 1990, Rigo et al. 1996), around the Gulf of Volos and in Thessaly (Hatzfeld et al. 1997), or around the Mygdonian graben (Hatzfeld et al. 1987) seismicity seems to underline recent, sometimes not very well developed, normal fault systems that are striking E-W. The very i m p o r t a n t N W - S E normal faults that bound the Gulf of Saros, the Gulf of Evia, the basin of Thessaly, are not related to microseismic activity. (3) These main features (low seismicity in the Sea of Crete, around the Cyclades, the Gulf of Soras, the Gulf of Evia and Thessaly) are confirmed over a longer period of time by the historical seismicity (Papazachos & Papazachou 1997). Particularly, we do not observe earth-
quakes related to the Saros basin and Evia basin, whereas we observe earthquakes related to Volos or to the Kavalla fault (Fig. 3). It seems to us, therefore, that there is not a unique relationship between the faults that are generally mapped as recently active in Greece and seismicity. Some faults are seismically active and some are not. Some seismicity is located in regions where no recent important surface deformation is mapped. Focal mechanisms
The first integrated study of focal mechanisms in the Mediterranean was the pioneering work of McKenzie (1972, 1978), which established the
TECTONICS & SEISMICITY IN THE AEGEAN
419
Fig. 3. Historical seismicity from 550 Bc until AD 1900 (Papazachos & Papazachou 1997). Geographical names are as Fig. 1. Little seismicity is related to Evia, Thessaly and the Saros basin. Kavalla experienced strong events.
global kinematic framework of the Mediterranean. At that time, most of the fault plane solutions were computed using only first motion polarities read on long period records. In this paper, we will use three different sets of data that will help to indicate the strain pattern of the area. (1) We present 61 solutions computed using the body waves modelling technique from three papers (Taymaz et al. 1990, 1991; Baker et al. 1997) which cover the Aegean. They represent the most reliable mechanisms in this area. (2) We compare these solutions with 183 CMT solutions automatically computed by Harvard (Dziewonsky et al. 1988) since 1977.
(3) We use the 746 microearthquake mechanisms that we computed during our different experiments from 1984 to 1993 (Fig. 5). Again, comparing the source properties of magnitude 6 and magnitude 2 e a r t h q u a k e s needs some caution, and we will never consider a single microearthquake mechanism as representative of the strain pattern if it is not consistent with a nearby large event or with other mechanisms that belong to the same area. In order to visualize better the strain pattern as defined by the mechanisms, we display also the T-axes that represent the direction of lengthening and the P-axes that represent the direction of shortening.
420
D. HATZFELD
Fig. 4. Map of the 61 reliable mechanisms computed by body waves modelling (Baker et al. 1997; Taymaz et al. 1990, 1991) in black, and 183 CMT solutions published by Harvard in grey. Reverse faulting is restricted to a very narrow zone along the active boundary. Strike slip is associated with all the different branches of the North Aegean Trough and the Kefallinia fault. Normal faulting is present in most of the Aegean.
A first examination of the mechanisms, both for strong (Fig. 4) or small (Fig. 5) earthquakes clearly show that reverse faulting is observed along the Hellenic Trench, normal faulting within the A e g e a n , and dextral strike slip mechanisms are associated with the different branches of North Aegean Trough and with the Kefallinia Fault. Within the internal Aegean, the T-axes (Fig. 6) trend consistently N-S from Macedonia to the Gulf of Corinth, and from western Greece to Turkey. But, conversely, toward the active
boundary, from Albania to Rhodes, there is a progressive change in the direction of the Taxes. Actually, the T-axes trend parallel to the Hellenic arc, and are therefore different in orientation from those within the interior Aegean. This change in trend suggests that internal deformation affects the A e g e a n (Hatzfeld et al. 1993, 1997). The P-axes (Fig. 7) that are associated with reverse faulting along the Hellenic Trench trend consistently from Albania to South Peloponnese. There is no noticeable rotation of the
TECTONICS & SEISMICITY IN THE AEGEAN
421
Fig. 5. Map of the 746 mechanisms computed during the different microearthquake experiments between 1984 and 1993.
P-axes (or of the slip vectors) across the Kefallinia fault which represents the transition between continental collision and oceanic subduction (Kahle et al. 1993; Hatzfeld et al. 1995). It is therefore unlikely that western Greece has rotated for a long time around a pole that is located near the active boundary (Baker et al. 1997). Further east, between the Kythira strait and eastern Crete, there is a significant change in the trend of the P-axes to a more N-S direction. If we assume a rigid African plate, this change in the orientation of the P-axes for earthquakes located along the boundary implies either a pole of rotation located close to the active boundary (Le Pichon & Angelier 1979) or more likely, as
inferred from the T-axes trend, that the Aegean region is deforming. Discussion T h e strain p a t t e r n
The strain pattern deduced from focal mechanism resembles the strain pattern deduced from geodetic measurements (Veis et al. 1992) or deduced from microtectonic observations for the present time (Mercier etal. 1987). It is homogeneous over all the Aegean, across blocks of various orientation and dimensions and supports the idea that the forces that move the
422
D. HATZFELD
Fig. 6. Map of the T-axes that dip shallower than 45~. Thick arrows are for large earthquakes, thin arrow are for microearthquakes. The direction of lengthening is homogeneous in the North Aegean and rotates slightly clockwise from west to east within the Aegean. The T-axes trend along strike for the Hellenic arc.
crustal blocks are applied to their base by a lithosphere that deforms as a continuum ( Le Pichon & Angelier 1979; England et al. 1985; Jackson et al. 1992; Hatzfeld et al. 1997). This strain pattern is slightly different to the strain p a t t e r n deduced from microtectonic observations for the Pliocene time (Mercier et al. 1987), which implies either a rotation of the Aegean within a constant strain field or a rotation of the strain field. The tectonics
It seems that there is not a unique relationship between the seismicity and the most important
faults displayed by Jackson (1994), Jolivet et al. (1994) or Armijo etal. (1996). Some of the faults are indeed seismically active as the North A e g e a n Trough, the Gulf of Corinth, the Kefallinia fault, the southern Peloponnese and western Crete. These faults are either NE-SWstriking strike-faults, or normal faults of various direction. The Kavalla fault, which is also a E N E - W S W striking strike-slip fault (Lyberis 1984), experienced several very strong earthquakes during historical time, while instrumental seismicity is rather low. On the other hand, some of the faults recently active are not seismic, such as those which bound the western Saros basin, the North Evia basin,
TECTONICS & SEISMICITY IN THE AEGEAN
423
Fig, 7, Map of P-axes that dip shallower than 45~ Same symbols as Fig. 6. We note a constant trend along western Greece across the Kefallinia fault, and a progressive anticlockwise rotation toward Rhodos.
the Argos basin, or Thessaly. This is true for the historical seismicity (Fig. 3), for the NEIC-bulletins seismicity (Fig. 1) and for the microseismicity (Fig. 2). Finally, we note seismic activity in regions where no important active faults are mapped. This is true for western Greece, from Albania to Peloponnese, where little is known about active faults. This is also true for young faults of modest displacement in the Mygdonian graben (Mercier et al. 1983), in the Gulf of Volos (Caputo & Pavlides 1993), in Central Attiki (Armijo et al. 1996) or in Trikhonis lake and in the Amvrahikos Gulf (Brooks et al. 1988; Doutsos et al. 1987). These young faults are all pure normal faults striking roughly E - W and therefore
perpendicular to the strain pattern and they experienced strong e a r t h q u a k e s during the present century or in historical time.
Strength o f the continental crust and o f the
faults We therefore propose that the present-day tectonics is probably different than the Pliocene tectonics. Most of the existing very large N W - S E - s t r i k i n g normal faults which were active during the Pliocene are not seismically active at the present time (Fig. 8), and large earthquakes occur only exceptionally (as the 1894 Atalanti Martinon earthquake). On the other hand, new E-W-striking faults
424
D. H A T Z F E L D 42 ~
22 ~
20 ~
24 ~
26 ~
o
28 ~
42 ~
EUROPE
Albania ~
o
%~ <M
Ka %
.
9 //
40 ~
~
~,~
40 ~
SVRKEV-.
A
9
~
.dz
38 ~
38 ~
:e,oo.nes
....
r;6,"
~t
"~hodos
/
36 ~ Crete
36 ~
/ 34 ~
34 ~ Trertctt
20 ~
22 ~
24 ~
26 ~
28 ~
Fig. 8. Rough interpretative tectonic sketch deduced from tectonic faults (after Caputo & Pavlides 1993; Mercier et al. 1987; Jackson, 1994; Jolivet et al. 1994; Armijo et al, 1996) that experience present-day seismicity. We plot, as presently active, only the faults that are associated with historical or instrumental seismicity. Thin lines are aseismic faults, thick lines are seismic faults. Most of the NW-SE-trending surface faults seem to be inactive at the present time. We note that the western termination of the North Anatolian Fault is a complex system that is connected to the Northern termination of the Kefallinia fault similarly as originally proposed by McKenzie (1972).
of small d i m e n s i o n s start to d e v e l o p in the Mygd o n i a n g r a b e n , in t h e V o l o s basin, in A t t i k i , a r o u n d t h e T r i k h o n i s lake a n d t h e A m v r a h i k o s Gulf. In s o m e places t h e y c r o s s - c u t t h e existing N W - S E - s t r i k i n g faults as in T h e s s a l y ( C a p u t o & P a v l i d e s 1993) or t h e M y g d o n i a n g r a b e n ( M e r c i e r et al. 1983). T h e s e faults strike p e r p e n d i c u l a r to t h e m a i n N - S - t r e n d i n g e x t e n s i o n t h a t affects t h e A e g e a n a n d are l o c a t e d o n l y in c o n t i n e n t a l G r e e c e . M o r e o v e r , m o s t of t h e s e r e c e n t E - W - s t r i k i n g faults are l o c a t e d at t h e
t e r m i n a t i o n of dextral strike-slip faults such as t h e d i f f e r e n t b r a n c h e s of the N o r t h A e g e a n T r o u g h a n d t h e Kavalla fault.
Conclusion A careful e x a m i n a t i o n of reliable seismicity a n d focal m e c h a n i s m s clearly shows t h a t t h e r e is n o t a p e r f e c t fit w i t h t h e active faults as t h e y are g e n e r a l l y d i s p l a y e d . M o s t of t h e N W - S E striking faults are n o t seismically active a n d n e w
TECTONICS & SEISMICITY IN THE A E G E A N E - W - s t r i k i n g faults start to d e v e l o p in continental G r e e c e . We have g o o d e v i d e n c e that important changes in the tectonics o c c u r r e d since late Pliocene ( M e r c i e r et al. 1987; A n g e l i e r et al. 1982; A r m i j o et al. 1996) w h i c h could be d u e either to a c h a n g e in the b o u n d a r y conditions, or m o r e likely to the i m p o r t a n t a m o u n t of internal d e f o r m a t i o n (and r e l a t e d rotation) within the Aegean. T h e r e is, t h e r e f o r e , a possibility that s o m e of the late Pliocene or e a r l y P l e i s t o c e n e faults, that are oblique to the strain field, are no l o n g e r active. N e w faults of m o d e s t dimensions, that m o r e easily a c c o m m o d a t e t h e d e f o r m a t i o n , d e v e l o p at the p r e s e n t time. T h e global picture of p r e s e n t active faults is t h e r e f o r e different to that for the late Pliocene. T h e s e n e w E - W - s t r i k i n g faults are m o s t l y l o c a t e d in c o n t i n e n t a l G r e e c e a n d T u r k e y , w h o s e crust is certainly c o n t i n e n t a l and probably w e a k , r a t h e r t h a n in the n o r t h e r n A e g e a n sea w h i c h has b e e n s t r e t c h e d and t h i n n e d by a factor of 2, and p r o b a b l y o c e a n i z e d (Le P i c h o n & A n g e l i e r 1979), and w h i c h m i g h t be stronger. As a c o n s e q u e n c e , the w e s t e r n t e r m i n a t i o n of the N o r t h A e g e a n t r o u g h is c o n n e c t e d to the K e f a l l i n i a fault by a n e w diffuse s y s t e m of n o r m a l faults or grabens that start to d e c o u p l e P e l o p o n n e s e f r o m central G r e e c e , as p r o p o s e d by M c K e n z i e (1972). The microearthquake seismicity is the result of numerous seismological experiments that were conducted in collaboration with the Universities of Athens (K. Makropoulos) and Thessaloniki (P. Hatzidimitriou, D. Panagiotopoulos, V. Karakostas) and supported by the EEC, contracts Simulation-121 and -353, contract EPOCH CT-91-0043. We benefitted from interesting discussions with J. Jackson, H. Lyon-Caen, J. Martinod, P. Molnar and B. Parsons.
References
ANGELIER,J. 1979. Recent Quaternary tectonics in the Hellenic arc: examples of geological observations on land. Tectonophysics, 52, 267-275. --, LYBERIS, N., LE PICHON, X., BARRIER, E. & HUCHON, P. 1982. The neotectonic development of the Hellenic Arc and the Sea of Crete: a synthesis. Tectonophysics, 86, 159-196. ARMIJO, R., MEYER, B., KING, G. C. R, RIGO, a . & PAPANASTASSIOU,D. 1996. Quaternary evolution of the Corinth Rift and its implications for the late Cenozoic evolution of the Aegean. Geophysical Journal International, 126, 11-53. BAKER, C., HATZFELD, D., LYON-CAEN,H., PAPADIMITRIOU,E. & RXGO,A. 1997. Earthquake mechanisms of the Adriatic sea and western Greece. Geophysical Journal International, 131, 559-594.
425
BROOKS, M., CLEWS,J. E., MELIS,N. S. & UNDERHILL,J. R. 1988. Structural development of Neogene basins in western Greece. Basin Research, l, 129-138. CAPUTO, R. & PAVLIDES,S. 1993. Late Cainozoic geodynamic evolution of Thessaly and surroundings (central-northern Greece). Tectonophysics, 223, 339-362. DOUTSOS, T., KONTOPOULOS, N. & FRYDAS, D. 1987. Neotectonic evolution of northwestern Greece, Geologisches Rundschau, 76, 433-450. DZIEWONSKI,A., FRIEDMAN,A. & WOODHOUSE, J. H. 1983. Centroid-moment tensor solutions for January-March, 1983. Physics of the Earth and Planetary Interiors, 33, 71-75. ENGLAND, P., HOUSEMAN, G. & SONDER, L. 1985. Length scales for continental deformation in convergent, divergent, and strike slip environments: analytical and approximate solutions for a thin viscous sheet model. Journal of Geophysical Research, 90, 3551-3557. GAUTIER, P. & BRUN,J.-P. 1994. Ductile crust exhumation and extensional detachments in the central Aegean (Cyclades and Evvia island). Geodinamica Acta, 7, 57-85. HATZFELD, D., BESNARD, M. MAKROPOULOS, K. & HATZIDIMITRIOU,P. 1993. Microearthquake seismicity and fault plane solutions in the southern Aegean and its tectonic implications. Geophysical Journal International, 115, 799-818. - - , CHRISTODOULOUA. A., SCORDILISE. M., PANAGIOTOPOULOS D. G. c~; HATZIDIMITRIOU, P. M. 1987. A microearthquake study of the Mygdonian graben (Northern Greece). Earth and Planetary Science Letters, 81, 379-396. - - , KASSARAS,I., PANAGIOTOPOULOS,D., AMORESE, D., MAKROPOULOS, K., KARAKAISIS, a. & COUTANT, O. 1995. Microseismicity and Strain Pattern in Northwestern Greece. Tectonics, 14, 773-785. - - , MARTINOD, J., BASTET, G. & GAUTIER, P. 1997. An analog experiment for the Aegean to describe the contribution of gravitational potential energy. Journal of Geophysical Research, 102, 649-659. , PEDOTI'I, G., HATZID1MITR1OU,P. & MAKROPOULOS, K. 1990. The strain pattern in the western Hellenic arc deduced from a microearthqauke survey, Geophysical Journal International, 101, 181-202. JACKSON,J. A. 1994. The Aegean deformation. Annual Review of Geophysics, 22, 239-272. & MCKENZm, D. 1988. The relationship between plate motions and seismic moment tensors, and the rates of active deformation in the Mediterranean and the Middle East. Geophysical Journal International, 93, 45-73. --, HAINES, E & HOLT, W. 1992. The horizontal velocity field in the deforming Aegean Sea region determined from the moment tensor of earthquakes. Journal of Geophysical Research, 97, 17 657-17 684 JOLIVET, L., BRUN, J-P., GAUTIER,P., LALLEMAND,S. & PATRIAT,M. 1994.3-D Kinematics of extension in
426
D. HATZFELD
the Aegean from early Miocene to the present: insight from ductile crust. Bulletin de la Soci~t~ GOologique de France, 165, 195-209. KAHLE, H-G, MOLLER, M. V., MUELLER, S. & VEIS, G. 1993. The Kephalonia transform fault and the rotation of the Apulian plateform: evidence from satellite geodesy. Geophysics Research Letters, 20, 651-654. KISSEL, C. & LAJ, C. 1988. The tertiary geodynamical evolution of the Aegean arc; a paleomagnetic reconstruction. Tectonophysics, 146, 183-201. LE PICHON, X. & ANGELIER, J. 1979. The Hellenic arc and trench system: a key to the neotectonic evolution of the Eastern Mediterranean region. Tectonophysics, 60, 1-42. , CHAMOT-ROOKE,N., LALLEMANT,S., NOOMEN,R. & VEIS, G. 1995. Geodetic determination of the kinematics of Central Greece with respect to Europe: Implications for Eastern Mediterranean Tectonics. Journal of Geophysical Research, 100, 12675-12690. LYBERIS, N. 1984. Tectonic evolution of the North Aegean Trough. In: ROBERTSON, A. H. E & DIXON, J. E. (eds) The Geological Evolution of the Eastern Mediterranean. Geological Society, London, Special Publications, 17, 708-725. MASCLE, J. & MARTIN, L. 1990. Shallow structure and recent evolution of the Aegean Sea: A synthesis based on continuous reflection profiles. Marine Geology, 94, 271-299. MCKENZIE, D. P. 1972. Active Tectonics of the Mediterranean Region. Geophysical Journal of the Royal Astronomical Society, 30, 109-185. - 1978. Active tectonics of the Alpine-Himalayan belt: the Aegean Sea and surrounding regions.
Geophysical Journal of the Royal Astronomical Society, 55, 217-254. JACKSON, J. 1983. The relationship between strain rates, crustal thickening, palaeomagnetism, finite strain and fault movements within a deforming zone. Earth and Planetary Science Letters, 65, 182-202. MERCIER, J-L. 1977. L'arc 6gden, une bordure ddform6e de la plaque eurasiatique; rdflexions sur un exemple d'dtude n6otectonique. Bldletin de la SociOt~ G~ologique de France, 19, 663-672. , CAREY E., PmL1e H. & SOREL D. 1976. La ndotectonique plio-quaternaire de l'arc 6gden externe et de l a m e r Egde et ses relations avec la sismicitd, Bulletin de la SociOtO G~ologique de France, 7, 355-372. --, CAREY-GAILHARDIS, E., MOtIYARIS, N., -
-
&
SIMEAKIS, C., ROUDOYANNI,T. & ANGELIDHIS, C. 1983. Structural analysis of recent and active faults and regional state in the epicentral area of the 1978 Thessaloniki earthquake (Northern Greece). Tectonics, 2, 577-600. --, DELIBASIS, N., GAUTIER, A., JARRIGE, J. J., LEMEILLE,E, PHILIP,H., SI~BRIER,M. & SOREL,D. 1979. La n6otectonique de l'Arc Eg6en. Revue de
G~ographie Physique et de G~ologie Dynamique, 21, 67-92. - - , SOREL, D. • SIMEAKIS,K. 1987. Changes in the state of stress in the overrinding plate of a subduction zone: the Aegean arc from the Pliocene to the Present. Annales Tectonicae, 1, 20-39. PAPAZACHOS,B. C. 1973. Distribution of seismic foci in the Mediterranean and surrounding area and its tectonic implication, Geophysical Journal of the Royal Astronomical Society, 33, 419-428. PAPAZACHOS B. & PAPAZACHOUK. 1997. The Earthquakes of Greece. Ekdoseis Ziti, Thessaloniki. PAVLIDES, S. B. & MOUNTRAKIS, D. M. 1987. Extensional tectonics of northwestern Macedonia, Greece, since the late Miocene. Journal of Structural Geology, 9, 385-392. RIGO, A., LYON-CAEN,H., ARMIJO, R., DESCHAMPS,A., HATZFELD, D., MAKROPOULOS, K., PAPADIMITRIOU, P. & KASSARAS I. 1996. A microseismic study in the western part of the Gulf of Corinth (Greece): implications for large-scale normal faulting mechanisms. Geophysical Journal International, 126, 6634588. ROBERTS, S. & JACKSON,J. A. 1991. Active normal faulting in Central Greece: an overview. In: ROBERTS, A. M., YIELDING, G. & FREEMAN, B. (eds) The Geometry of Normal Faults. Geological Society, London, Special Publications, 56, 125-142. SPAKMAN,W., WORTEL, M. J. R. & VLAAR, N. S. 1988. The Hellenic subduction zone: a tomographic image and its geodynamical implications. Geophysical Research Letters, 15, 60453. TAYMAZ, T., JACKSON, J. A. & WESTAWAY, R. 1990, Earthquake Mechanisms in the Hellenic Trench near Crete, Geophysical Journal International, 102, 695-732. - - , JACKSON,J. A. & MCKENZlE, D. 1991. Active Tectonics of the North and Central Aegean Sea, Geophysical Journal International, 106, 433-490. VEIS, G., BILLIRIS, H., NAKOS,B. & PARADISSIS,D. 1992, Tectonic strain in Greece from geodetic measurements, Academy of Sciences of Athens, 67, 129-166.
Ductile extension and the formation of the Aegean Sea L. J O L I V E T
t & M. PATRIAT 2
1DOpartement des Sciences de la Terre, Universit~ de Cergy-Pontoise, 8 le Campus, A v e n u e du Parc, 95011 Cergy-Pontoise cedex, France, U R A C N R S 1759 eLaboratoire de GOologie, Ecole Normale Sup&ieure, 24 rue L h o m o n d , 75231 Paris cedex 05, France, U R A C N R S 1316 Abstract: The Aegean Sea and Tyrrhenian sea areas offer almost unique examples of
actively collapsing orogens where the internal velocity field, kinematic boundary conditions and seismic activity can be compared and where older structures formed during the early stages of the same process are exhumed. We study these processes along two major transects. The first runs from Mt Olympos to the centre of the Cyclades, it is perpendicular to the direction of extension and offers the opportunity to study the brittle-ductile transition by direct observation of rocks and deformation. A gradient of finite exhumation along the transect has brought various crustal levels to the surface, and the localization of deformation during exhumation along major detachments has preserved older penetrative structures formed below. As these outcrops are distributed along the strike of a single crustal-scale tilted block the observed structures probably corresponds to a single tectonic event, and we can sample the associated deformation at increasing depth along the transect. We propose a stratification of deformation regimes associated to extension: localized normal faults in the brittle crust, distributed extensional shear bands with a partition between domains of noncoaxial and coaxial flow below 8-10 km, and distributed coaxial flow below 20 km. Flat-lying extensional shear bands extend brittle normal faults within the brittle-ductile transition. The second transect runs from Crete to Naxos and Mykonos and is parallel to the main direction of extension. It shows the transition from tectonic accretion and syn-orogenic extension near the subduction front to post-orogenic extension in the backarc region. The exhumation of metamorphic rocks proceeds in two stages: synorogenic 'cold' exhumation along detachments in the upper part of the accretionary complex, with a good preservation of HP-LT parageneses, and post-orogenic exhumation in a warmer environment in the backarc region with a complete retrogression of HP-LT parageneses. The 3D finite strain field is then compared to recent space geodesy data and the mechanism of crustal collapse discussed. E x t e n s i o n has b e e n active in the A e g e a n r e g i o n for a long e n o u g h time (25 Ma) for ductile structures f o r m e d u n d e r the same extensional r e g i m e to crop out at the surface. T h e y are thus accessible to direct o b s e r v a t i o n in various m e t a m o r phic core c o m p l e x e s such as Naxos in the c e n t r e of the Cyclades (Lister et al. 1984) (Figs 1, 2 & 3). T h e y offer t h e o p p o r t u n i t y to study a c o m p l e t e section of an e x t e n d i n g crust f r o m t h e d e e p ductile to the u p p e r brittle crust. In this p a p e r we d e m o n s t r a t e the structural relations b e t w e e n v a r i o u s levels of t h e crust focussing on the b r i t t l e - d u c t i l e transition and discuss the kinematic b o u n d a r y conditions and processes which allowed the f o r m a t i o n of those structures. E x t e n s i o n in t h e b a c k - a r c r e g i o n (Cyclades) o c c u r r e d c o n c u r r e n t l y with the f o r m a t i o n of an a c c r e t i o n a r y c o m p l e x n e a r the s u b d u c t i o n front (Le P i c h o n et al. 1994; T a y m a z et al. 1991) (Fig. 4). D u r i n g t h e e x t e n s i o n a l process frontal accretion and e x t e n s i o n both m i g r a t e d s o u t h w a r d as the s u b d u c t i o n slab r e t r e a t e d (Le P i c h o n & A n g e l i e r 1981). This implies that the m a t e r i a l
first a c c r e t e d in the w e d g e was t h e n t r a n s f e r r e d to the b a c k - a r c region. D u r i n g this m i g r a t i o n d e e p l y b u r i e d m a t e r i a l was e x h u m e d in two d i f f e r e n t settings, t h e a c c r e t i o n a r y c o m p l e x itself and the w a r m b a c k - a r c d o m a i n (Jolivet et al. 1994b). T h e m e t a m o r p h i c history of rocks is different in b o t h cases a n d we describe it along a N - S t r a n s e c t f r o m C r e t e to t h e c e n t r a l Cyclades. K i n e m a t i c b o u n d a r y conditions are well constrained in the e a s t e r n M e d i t e r r a n e a n and the active velocity field inside the A e g e a n d o m a i n precisely k n o w n f r o m space g e o d e s y (Fig. 5). It is thus possible to c o m p a r e the p r e s e n t velocity field to the finite strain field o b t a i n e d after 25 M a of c o n t i n u o u s d e f o r m a t i o n (Jolivet et al. 1994a). We show that b o t h fields are c o m p a t i b l e and that the s a m e k i n e m a t i c s has p r e s i d e d to the deform a t i o n during the last 25 Ma. D i r e c t o b s e r v a t i o n of t h e in situ b r i t t l e - d u c t i l e transition d u r i n g extension in the c o n t i n e n t a l crust is impossible with available t e c h n i q u e s and no consensus exists as to h o w the n o r m a l faults
JOLIVET,L. & PATRIAT,M. 1999. Ductile extension and the formation of the Aegean Sea. In: DURAND,B., JOLlVET, L., HORVATH,F. & SERANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, 156, 427--456.
428
L. JOLIVET & M. PATRIAT
Fig. 1. Major active structures in the Aegean domain and P - T paths in the main metamorphic core complexes (Avigad et al. 1992; Jolivet et al. 1996). Insert: present-day kinematics of extrusion of the Anatolian block along the North Anatolian Fault (thick dotted lines for small circles and large black dot for rotation pole) (Le Pichon et al. 1994). Thin lines represent the direction of finite stretching (Jolivet et al. 1994a).
DUCTILE EXTENSION, AEGEAN SEA
429
Fig. 2. Bathymetric and topographic map of the Aegean region (source etopo 5) and sense of shear in exhumed ductile metamorphic core complexes and sense of shear along the brittle-ductile transition deduced from the attitude of normal faults.
seen in the upper crust root in the middle or lower crust. The succession of deformational events shown by exhumed metamorphic rocks gives a rather precise idea of the vertical superposition of deformation regimes (Sibson 1983) but does not constrain the nature of the brittle-ductile transition at a larger scale. One of the pending problems is how large normal faults and tilted blocks are relayed at depth by distributed ductile flow, and whether or not the simple shear model of extension is applicable to the whole crust (Wernicke 1981) or limited to its middle and upper part (Andersen & Jamtveit 1990; Jackson 1987; Lister & Davis 1989). Large-scale detachments seen above extensional metamorphic core complexes separate domains with respectively brittle and ductile deformation but they do not represent the brittle-ductile transition (Lister & Davis 1989). They are the ultimate evolution of a more diffuse shear zone which encompasses a large
crustal thickness. A large part of the finite deformation seen along those detachments is the result of exhumation of deep crustal rocks and thus it cannot give a precise idea of the in situ brittle-ductile transition. Seismological studies in regions of active extension show that seismogenic normal faults are planar down to the base of the brittle crust (Jackson & White 1989; Braunmiller & Nabelek 1996). Decoupling from the lower ductile crust is probably achieved through aseismic slip (Braunmiller & Nabelek 1996). Other more detailed studies in areas of active extension such as the Gulf of Corinth suggest that steep normal faults in the upper crust root in shallow dipping (e.g. 10 ~ extensional shear zones at depth around 10 km (Eyidogan & Jackson 1985; King et al. 1985; Rigo et al. 1996). Those d6collements are seismically active and the sense of motion along them is synthetic to that along major normal faults (Rigo et al. 1996). The active seismic zone
430
L. JOLIVET & M. PATRIAT
Fig. 3. Bathymetry and topography along the Olympos-Naxos transect, and isocontours of crustal thickness (Le Pichon & Angelier 1981). Fault pattern is also shown (Gautier & Brun 1994a; Masclc & Martin 1990; Rigo et al. 1996). Arrows represent the sense of shear observed in exhumed metamorphic rocks (Godfriaux 1965; Schermer 1990, 1993; Schermer et al. 1990; Urai et al. 1990; Buick 1991; Faure, et al. 1991; Godfriaux and Ricou 1991 ; Lee & Lister 1992; Gautier, et al. 1993; Gautier 1994; Gautier & Brun 1994a; Jolivet et al. 1996 and this work).
marks the lower limit of crustal seismicity and likely represents the transition from brittle to ductile behaviour. The depth of the transition between the seismic and aseismic crust varies between 10 and 20 km (Molnar & Chen 1983) and is most likely controlled by the rheology of
quartz and feldspar which have their brittle-plastic transition at respectively 300-350~ and 450-500~ (Voll 1976; Kerrich et al. 1977; Sibson 1983). Independantly of seismological studies, field observations of Mediterranean core complexes
DUCTILE EXTENSION~ A E G E A N SEA
Fig. 4. Synthetic cross sections through the Aegean domain and the two types of retrograde P - T paths, (Jolivet et al. 1994b).
Fig. 5. Displacement field recorded from space geodesy, (Le Pichon et al. 1994). Dotted lines represent the small circles about the rotation pole Anatolia/Eurasia.
431
432
L. JOLIVET & M. PATRIAT
(Alpine Corsica and the Aegean Sea) shows the existence of similar flat-lying ductile shear zones active within the pressure and temperature domain of the greenschist facies which is in the range of the above temperatures (Daniel et al. 1996; Gautier & Brun 1994a,b; Jolivet et al. 1991, 1994a,b). The geological history of these shear zones makes them good candidates for comparison with the seismically active detachments seen in regions of active extension (Jolivet et al. 1994a). The question is however the following: can we go back to the initial geometry while they were active at depth if their present attitude is the result of a succession of deformations during exhumation? or, is it possible to find regions were deep deformations have been preserved (frozen) without much reworking during the process of extension? Generally core complexes show an internal part where such deep deformations were preserved because deformation has a tendancy to localize along the detachment, and the degree of localization increases during exhumation, while rocks pass through the brittle-ductile transition. Observing various core complexes with various finite rates of extension (and exhumation) might allow reconstruction of the vertical stratification of structures in an extending crust. The Aegean domain offers this possibility along the Olympos-Naxos transect (Fig. 3), where a succession of metamorphic core complexes exhumed during the same progressive process of extension crop out, and where, next to the transect, the same phenomenon is active at depth below the Gulf of Corinth.
Regional geology Aegean
Sea
Crustal thinning has been active in the Aegean Sea in the back-arc region of the Hellenic trench since the Early Miocene (McKenzie 1972, 1978; Dewey & Seng6r 1979; Le Pichon & Angelier 1981; Taymaz et al. 1991). Present-day kinematics is governed by rigid extrusion of the Anatolian block along the North Anatolian Fault toward the subduction zone free edge (Fig. 5), and internal deformation of the same block leading to extension in the backarc domain (McKenzie 1972, 1978; Le Pichon 1981; Le Pichon et al. 1994; Armijo et al. 1996). Incipient collision of the African plate and the Anatolian block south of Crete is responsible for arc-parallel extension close to the trench while N-S extension prevails in the back-arc domain (Angelier et al. 1982; Armijo et al. 1992). Before 3 Ma N-S extension was prevailing in the entire
Aegean Sea (Lyberis et al. 1982; Mercier et al. 1979, 1987). Metamorphic rocks were exhumed during the process of extension and the geometry of ductile extension has been studied by several authors (Faure & Bonneau 1988; Buick 1991; Faure et al. 1991; Lee & Lister 1992; Gautier et al. 1993; Sokoutis et al. 1993; Gautier & Brun 1994a; Jolivet et al. 1994a, 1994b, 1996). From Crete to the Cyclades islands and to the northern Aegean Sea the direction and sense of motion along major detachments were elucidated. The results are shown on Fig. 2. Extension was superimposed on an earlier episode of crustal thickening which led to the formation of the Hellenides culminating 45 Ma ago (Avigad & Garfunkel 1991; Avigad et al. 1992; Bonneau & Kienast 1982; Jacobshagen et al. 1978; Wijbrans & McDougall 1988; Wijbrans et al. 1993). The products of this early episode are a stack of nappes of alpine type and highpressure and low-temperature metamorphic rocks known as the Cycladic Blueschists (Blake et al. 1981; Bonneau & Kienast 1982; Okrusch & Br6cker 1990; Avigad & Garfunkel 1991). Crustal collapse started in the Late Oligocene-Early Miocene probably as a consequence of the initiation of the Hellenic trench 40 Ma ago (Berckhemer 1977; Gautier & Brun 1994a; Spakman et al. 1993). Syn- versus post-orogenic
extension
Post-orogenic extension affected most of the Aegean domain while crustal thickening was still active in Crete and Peloponnese (Bassias & Triboulet 1994; Bonneau 1984; Fassoulas et al. 1994a; Jolivet et al. 1994b, 1996; Seidel et al. 1982; Theye & Seidel 1991). Extension is recorded in those regions as soon as the Early and MidMiocene and was responsible for the exhumation of younger high pressure and low temperature metamorphic rocks in the upper part of a convergent accretionnary wedge. It is thus important to distinguish between syn - and post-orogenic extension. Along the Olympos-Naxos transect a similar process of syn-orogenic extension has probably been active and responsible for the exhumation of HP-LT metamorphics during the Eocene crustal stacking event (Avigad & Garfunkel 1991; Wijbrans et al. 1993) as will be discussed later. Crustal thickness
Crustal thickness in the Aegean domain has been estimated from gravity data (Makris 1978; Le Pichon & Angelier 1981) and published as
DUCTILE EXTENSION, AEGEAN SEA maps of Moho depth and crustal thickness which show that the North Aegean Trough and Sea of Crete are the thinnest regions. Crustal thickness decreases gradually from the Hellenic arc to the central Aegean Sea along strike (Fig. 3). Along the Olympos-Naxos transect it varies from 38 km below Mt Olympos to 30 km below the island of Naxos. North and south of Naxos the crust is considerably thinner especially in the Sea of Crete where it can be as thin as 20 km or less. 45-50 km is a fair minimum estimation of the pre-extension thickness by comparison to the Hellenic Chain further west. Metamorphic
core complexes
The first description (Lister et al. 1984) of a cordilleran-type metamorphic core complex in the Aegean Sea was made in Naxos (Fig. 2). Later the islands of Naxos, Paros, Evia, Andros, Tinos, Mykonos, Ikaria were studied by various teams (Urai et al. 1990; Buick 1991; Faure et al. 1991; Lee & Lister 1992; Gautier et al. 1993; Gautier & Brun 1994b; Vandenberg & Lister 1996) and a complete description of the finite strain field during the Late Oligocene Early Miocene high-temperature overprint is available in the form of stretching lineation maps in all these islands (Fig. 2). The island of Thassos and the southern Rhodopian massif were also studied and metamorphic core complexes revealed there also (Dinter & Royden 1993; Sokoutis et al. 1993). Except for the two latter regions and the island of Ios, kinematic indicators related to extension are top-to-the-north. A recent study of palaeomagnetic directions in the Miocene intrusions of Mykonos and Naxos (Morris & Anderson 1996), suggests postductile deformation rotations about vertical axes that have to be considered when discussing the Miocene evolution. Ductile deformation was also studied in the region of Mt. Olympos (Schermer 1990, 1993; Schermer et al. 1990; Godfriaux & Ricou 1991). The most obvious deformation in this region is contemporaneous with the Eocene HP-LT event and dated around 50-60 Ma. Most kinematic indicators are top-to-the-west consistent with the geometry of the crustal stacking episode.
The Olympos-Naxos transect and the brittle-ductile transition Geological
context
The studied transect runs N W - S E parallel to the western branch of the Hellenic arc. Major normal faults, active or recent, separates the arc
433
from the deep basins of the Aegean Sea to the northeast (Fig. 3). The whole region is tectonically divided by a series of normal faults of similar trend (Taymaz et al. 1991; Jackson 1994). Several crustal scale tilted blocks are readily recognized, separated by sedimentary basins (the Gulf of Evvia or the Gulf of Corinth) (Jolivet et al. 1994a; Papanikolaou et al. 1988) (Fig. 4). Studies of the subsurface structure between the blocks (Papanikolaou et al. 1988), as well as studies of the active deformation of the whole area show that major normal faults dip toward the northeast and that blocks are tilted toward the southwest (Taymaz et al. 1991; Jackson 1994). The case is particularly clear for the Gulf of Corinth which is clearly asymmetric with north-dipping normal faults which root in the seismically active decollement (King et al. 1985; Rigo et al. 1996). The metamorphic core complexes belonging to the transect studied here all belong to the same tilted block starting in the Mt Olympos region, running through Evia, Andros, Tinos and Mykonos. At the latitude of Mykonos the morphological expression of the block is less clear and the islands of Mykonos, Naxos and Paros are located on the same wide topographic high (Figs i and 2). P-T-t evolution The Late Oligocene-Early Miocene deformation reworks the Eocene high-pressure-lowtemperature event (Gautier & Brun 1994a) (Figs 1 & 6). A gradient of retrogression toward high temperature parageneses is observed from NW to SE. Little retrogression in the greenschist facies is observed in the Mt Olympos region (Schermer 1990; Schermer et al. 1990), while pervasive retrogression in the greenschist facies characterizes southern Evvia, Andros and part of Tinos, and amphibolite facies overprint is observed from southern Tinos to Mykonos and Naxos where partial melting and formation of migmatites is observed (Alther et al. 1982; Buick & Holland 1989; Jansen 1977; Wijbrans & McDougall 1988). Retrogression P - T - t paths show these different thermal histories (Figs 1 & 6). For the three examples of Mt Olympos, Tinos and Naxos, published P - T - t paths are available (Buick & Holland 1989; Okrusch & Br6cker 1990; Schermet 1990; Schermer et al. 1990). In Mt Olympos the P - T conditions of peak metamorphism are around 8 kbar-300~ and radiometric dates for this event cluster between 53 and 61 Ma (Schermer 1990, 1993; Schermer et al. 1990). The last event recorded in the radiometric data is a fast cooling around 100-150~ at
434
L. JOLIVET & M. PATRIAT
Fig. 6. Retrograde P-T-t paths from Mt Olympos, Tinos and Naxos (Buick & Holland 1989; Okrusch & Br6cker 1990; Schermer 1990; Schermer et al. 1990) and in Crete (Jolivet et ah 1996).
16-23 Ma associated with normal faulting. This shows undoubtedly that the HP-LT metamorphic rocks had already been exhumed to shallow levels when the Aegean extension started in the early Miocene. Peak P - T conditions in Tinos fall in the eclogite facies (15-18 kbar, 450-500~ and ages around 45 Ma were obtained (Okrusch & Br6cker 1990; Br/3cker et al. 1993). A subsequent greenschist-amphibolite overprint is recorded around 21-23 Ma. As will be discussed later this late overprint is associated to ductile reworking of the earlier HP foliation, motion along the detachement and intrusion of a granite at 19 Ma. The deformation contemporaneous with the greenschist reworking has been interpreted to be (Gautier & Brun 1994a; Jolivet et al. 1994a) the result of crustal scale extension. The Aegean extension is thus recorded here as ductile to brittle structures, the deepest structures being well within the greenschist facies. The shape of the P - T - t path shows that a fast exhumation occurred before 21-23 Ma; whether
this is due to extension or to some other process removing overburden is not clear. Peak P - T conditions are less well constrained in Naxos because of a pervasive overprint in the amphibolite facies and partial melting in the core of the dome. P - T conditions around 10 kbar and at least 550-600~ were attained 45 Ma ago (Jansen 1977; Buick & Holland 1989; Buick 1991). The retrogression path involves a peak toward high temperatures (700~ 5 kbar) and is also associated with extensional deformation and formation of the metamorphic core complex. In a first approach these data show that when the Aegean extension started the metamorphic rocks of Naxos had already been equilibrated in high-temperature-low-pressure conditions while those of Tinos or Mt Olympos had escaped from the deep and hot environment. In the Early Miocene the blueschist of Mt Olympos were already in the surficial parts of the crust, while those in Tinos were at intermediate depth (greenschist facies) and those of Naxos deeper still and affected by amphibolite facies and anatexis. The observed pattern is consistent with thinning of the crust from NW to SE. The conclusion of this preliminary analysis is that from Mt Olympos to Naxos one can observe rocks exhumed by extension from a rather superficial setting in the NW to a deep one in the SE. In the following a short summary of the deformation observed in each region is presented. For more detailed studies the reader is referred to Shermer (Schermer 1990, 1993; Schermer et al. 1990) and Godfriaux and Ricou (Godfriaux 1965; Godfriaux & Ricou 1991) for Mt Olympos, to Melidonis (1980), Bavay & Romain-Bavay (1980), Gautier and Brun (Gautier 1994; Gautier & Brun 1994a) Patriat (1996) for the islands of Evvia, Andros and Tinos, to Faure et al. (1991) and Lee & Lister (1992) for Mykonos, and to Gautier et al. (1993) and Buick (1991) or Urai, et al. (1990) for the islands of Naxos and Paros. Mt Olyrnpos
The region is characterized by the Mt Olympos tectonic window which shows the contact between the Pelagonian units (sensu lato) above the Cycladic blueschist (Ambelakia unit) and the parautochtonous Olympos-Ossa unit (Figs 3 & 7). HP-LT parageneses are observed both above and below the tectonic contact which is described as a syn-HP thrust. Ductile deformation localized in a narrow zone along the contact and most syn-HP kinematic indicators indicate top-to-the-southwest motion (Fig. 3).
DUCTILE EXTENSION, AEGEAN SEA
Fig. 7. Cross-sections through Mt Olympos, Tinos and Paros-Naxos (Schermer 1990; Gautier et Gautier 1994; Gautier and Brun 1994a, b). Rare kinematic indicators corresponding mostly to semi-brittle structures on the eastern side of Mt Olympos indicate top-to-the-northeast motion but it cannot be excluded that they represent simply layer-parallel extension due to a late bending of the foliation during normal faulting. The early nappe stack is reworked by extensional tectonics best expressed by the large recent northeast-dipping normal fault which cuts the topography to the northeast of Mt
435
al.
1993;
Olympos and Mt Ossa (Fig. 8a). An older extensional event is recorded along a flat-lying detachment (Schermer 1993) (Fig. 7). No major ductile deformation is associated to this detachment. It can be concluded that the Aegean extension in Mt Olympos is expressed mostly by brittle structures, high-level detachments and normal faults indicating a top-to-the-northeast motion of the hanging wall.
436
L. JOLIVET & M. PATRIAT
9
,....
E.-.
5~ ,,~
~.,,,,.
~,,..~
r.~
. dans l'Ossa et le Bas-Olympe (Thessalie-Grece). Bulletin de la SociOtO GOologique de France, 2, 252. - & -1978. Ddcouverte de microfaunes pal6ogenes dans le flysch m6tamorphique de Spilia (Ossa, Grece). Comptes Rendus de l'Academie des Sciences, Paris, 286, 555-558. DINTER, D. A. & ROYDEN, L. 1993. Late Cenozoic extension in northeastern Greece. Geology, 21, 45-48. --, MACFARLANE, A., HAMES, W., ISACHSEN, C., BOWRING, S. & ROYDEN, L. 1995. U-Pb and 4~ geochronology of the Symvolon granodiorite: Implications for the thermal and structural evolution of the Rhodope metamorphic core complex, northeastern Greece. Tectonics, 14, 886-908. EVANS, B. W. 1990. Phase relations of epidote blueschists. Lithos, 25, 3-23. FASSOULAS, C., KILIAS, A. & MOUNTRAKIS, D. 1994. Postnappe stacking extension and exhumation of high-pressure/low-temperature rocks in the island of Crete, Greece. Tectonics, 13, 127-138. FERRIERE, J. 1976. Etude preliminaire d'un secteur des massifs cristallins internes (): la region de Volos, Gr6ce continentaleorientale. Bulletin de la SociOtO GOologique de France, 2, 265-272. , GODFRIAUX, I. & SCHMITF,A. 1988. Les series metamorphiques infrapieriennes et pagasitiques; bordure orientale de la zone (isopique) Pelagonienne (Thessalie, Gr6ce)? Deltio tes Ellenikes Geologikes Etaireias, 20, 193-206. FRANZ, L. & OKRUSCH, M. 1992. Aragonite-bearing blueschists on Arki island, Dodecanese, Greece. European Journal of Mineralogy, 4, 527-537. GAUTIER, P & BRUN, J.-P. 1994. Ductile crust exhumation and extensional detachments in the central Aegean (Cyclades and Evvia Islands). Geodinamica Acta, 7, 57-85. GODFRIAUX, I. 1970. Etude geologique de la region de l'Olympe (Grbce). Annales of Geologie du Pays Hellenique, 19, 1-271. --, FERRIERE, J. & SCHM1TT,A. 1988. Le development en context continental d'un metamorphisme HP/BT; les "schistes bleus" tertiaires Thessaliens. Deltio tes Ellenikes Geologikes Etaireias, 20, 175-192. HAMES, W. E. & HODGES, K. V. 1993. Laser 4~
evaluation of slow cooling and episodic loss of 40Ar from a sample of polymetamorphic muscovite. Science, 261, 1721-1724. HANES, J. A. (1991). K - A r and 4~ geochronology: Methods and Applications. In: HEAMAN,L. & NUDDEN, J. N. (eds) Short course handbook on applications of radiogenic isotope systems to problems in geology. Mineralogical Association of Canada, 27-57. HARRISON, T. M. 1981. Diffusion of 40Ar in Hornblende. Contributions to Mineralogy and Petrology, 78, 324-331. -& McDOUGALL, I. 1982. The thermal significance of potassium feldspar K-At ages inferred from 40Ar/39Ar age spectrum results. Geochimica et Cosmochimica Acta, 46, 1811-1820. HETZEL, R., RING, U., AKAL, C. & TROESCH, M. 1995. Miocene NNE-directed extensional unroofing in the Menderes Massif, southwestern Turkey. Journal of the Geological Society, London, 152, 639-654. HODGES, K. W., HAMES, W. E. & BOWRING, S. A. 1994. 40Ar/39Ar age gradients in micas from a hightemperature-low-pressure metamorphic terrain: Evidence for very slow cooling and implications for the interpretation of age spectra. Geology, 22, 55-58. IGME 1983. Geological map of Greece, scale 1:500,000. Second edition. Institute of Geology and Mineral Exploration, Athens. JACOBSHAGEN, V. & WALLBRECHER, E. 1984. PreNeogene nappe structure and metamorphism of the North Sporades and the southern Pelion peninsula. In: DIXON, J. E. & ROBERTSON,A. H. E (eds) Geological Evolution of the Eastern Mediterranean. Geological Society, London, Special Publications, 17, 591-602. JOHN, B. E. & HOWARD, K. A. 1995. Rapid extension recorded by cooling-age patterns and brittle deformation, Naxos, Greece. Journal of Geophysical Research, 100 (B7), 9969-9979. JOLIVET, L., BRUN, J.-P, GAUTIER, P., LALLEMANT,S. & PATRIAT, M. 1994. 3D-kinematics of extension in the Aegean region from the early miocene to the Present, insights from the ductile crust. Bulletin de la SociOtO GOologique de France, 165, 195-209. --, GOFFf2, B., MONII~, P., TRUFFERT-LUXEY, C., PATRIAT, M. & BONNEAU, M. 1996. Miocene detachment in Crete and exhumation P - T - t paths of high-pressure metamorphic rocks. Tectonics, 15, 1129-1153. LIATI,A. & SEIDEL,E. 1994. Sapphirine and hOgbomite in overprinted kyanite-eclogites of central Rhodope, N. Greece: first evidence of granulitefacies metamorphism. European Journal of Mineralogy, 6, 733-738. - & -1996. Metamorphic evolution and geochemistry of kyanite eclogites in central Rhodope, northern Greece. Contributions to Mineralogy and Petrology, 123, 293-307. LIPS, A. L. W., WHITE, S. H. & WIJBRANS, J. R. 1998. 4~ laserprobe direct dating of discrete
TIMING OF METAMORPHISM & TECTONISM, A E G E A N deformational events: a continuous record of Early Alpine tectonics in the Pelagonian Zone, NW Aegean area, Greece. Tectonophysics, 298, 133-153. LISTER, G. S. & RAOUZAIOS,A. 1996. The tectonic significance of a porphyroblastic blueschist facies overprint during Alpine orogenesis: Sifnos, Agean Sea, Greece. Journal of Strucural Geology, 18, 1417-1435. MALUSKI, H., BONNEAU, M. & KIENAST, J. R. 1987. Dating the metamorphic events in the Cycladic area: 39Ar/4~ data from metamorphic rocks of the island of Syros (Greece). Bulletin de la Soci~t~ GOologique de France, (8), III, 5, 833-842. - - , VERGELY,P,. BAVAY,O., BAVAY,P. & KATSIKATSOS, G. 1981. 39Ar/40Ar dating of glaucophanes and phengites in Southern Euboa (Greece) geodynamic implications. Bulletin de la Soci(t~ G~ologique de France, (7), XXIII 5, 469-476. McDOUGALL, I. &HARRISON,T. M. (1988). Geochronology and Thermochronolgy by the 4~ Method. Oxford University Press, New York. OKRUSCH, M. & BROCKER,M. 1990. Eclogites associated with high-grade blueschists in the Cycladic archipelago, Greece: a review. European Journal of Mineralogy, 2, 451-478. - - , RICHTER,P. & KATSIKATSOS,G. 1984. High-pressure rocks of Samos, Greece. In: DIXON, J. E. & ROBERTSON, A. H. E (eds) The geological evolution of the Eastern Mediterrenean, 529-536. RAOUZAIOS, A., LISTER, G. S. & FOSTER, D. 1996. Oligocene exhumation and metamorphism of eclogite-blueschists from the island of Sifnos, Cyclades, Greece. Geol. Soc. Australia, Abstracts of 13th AGC, 141, 358. REDDY, S. M., KELLEY, S. P. & WHEELER, J. 1996. A 4~ laser probe study of micas from the Sesia Zone, Italian Alps: implications for metamorphic and deformation histories. Journal of Metamorphic Geology, 14, 493-508. RIDLEY,J. 1984. Evidence of a temperature-dependent 'blueschist' to 'eclogite' transformation in highpressure metamorphism of metabasic rocks. Journal of Petrology, 25, 852-870. SCAILLET,S., FERAUD, G., LAGABRIELLE,Y.,BALLEVRE, M. & RUFFET, G. 1990. 4~ laser-probe dating by step heating and spot fusion of phengites from the Dora MalTa nappe of the western Alps, Italy. Geology, 18, 741-744. SCHERMER, E. R. 1990. Mechanisms of blueschist creation and preservation in an A-type subduction zone, Mount Olympos region, Greece. Geology, 18, 1130-1133. 1993. Geometry and kinematics of continental basement deformation during the Alpine orogeny, Mt. Olympos region, Greece. Journal of Structural Geology, 15, 571-591. , Lux, D. R. & BURCHFIELD,B. C. 1990. Temperature-time history of subducted continental crust, Mount Olympos region, Greece. Tectonics, 9, -
-
1 1 6 5 - 1 1 9 5 .
SCHLIESTEDT, M. &. MATTHEWS,A. 1987. Transformation of blueschists to greenschist facies rocks as
473
a consequence of fluid infiltration, Sifnos (Cyclades), Greece. Contributions to Mineralogy and Petrology, 97, 237-250. SCHMIqq', A. 1983. Nouvelles contributions a l'~tude g~ologique des Pieria, de l'Olympe, et de l'Ossa (Greece du Nord). Thesis at Facult6 Polytechnique de Mons, Belgium. SEIDEL, E., KREUZER, H. & HARRE, W. 1982. A Late Oligocene/Early Miocene high pressure belt in the External Hellenides. Geologisches Jahrbuch, E23, 165-206. SISSON,V. B., & ONSTOTT,T. C. 1986. Dating blueschist metamorphism: A combined 40Ar/39Ar and elctron microprobe approach. Geochimica et Cosmochimica Acta, 50, 2111-2117. SKARPELIS,N. 1989. Metamorphosed ultramafic rocks in the blueschist (arna unit) of the external metamorphic belt of the Hellenides (Peloponnesos and Kythira Island, Greece). Bulletin of the Geological Society of Greece, 23, 345-358. THEYE, T., SEmEL, E. & VIDAL, O. 1992. Carpholite, sudoite and chloritoid in low-grade high-pressure metapelites from Crete and the Peloponnese, Greece. European Journal of Mineralogy, 4, 487-507. TRIBOULET, C. & BASSIAS, Y. 1988. Geothermobarometry on zoned amphiboles and progressive metamorphism in glaucophane-bearing metabasites from the Parnon massif (external Hellenides, Peloponnesus, Greece). Comptes Rendus de l'Academie des Sciences, Paris, 306, Serie II, 991-996. VAN DER MAAR, E A. & JANSEN, J. H. B. 1983. The geology of the polymetamorphic complex of Ios, Cyclades, Greece, and its significance in the Cycladic Massif. Geologisches Rundschau, 72, 283-299. VANDENBERG,L. C. & LISTER, G. S. 1996. Structural analysis of basement tectonites from the Aegean metamorphic core complex of Ios, Cyclades, Greece. Journal of Structural Geology, 18, 1437-1454. WALCOTT, C. R. & WHITE, S. H. 1998. Constraints on the kinematics of post-orogenic extension imposed by stretching lineations in the Aegean region. Tectonophysics, 298, 155-175. WAWRZENITZ, N. ~; MPOSKOS,E. 1997. First evidence for Lower Cretaceous HP/HT-metamorphism in the Eastern Rhodope, North Aegean region, North-East Greece. European Journal of Mineralogy, 9, 659-664. WHITNEY, D. L. ~; DILEK, Y. 1997. Core complex development in central Anatolia, Turkey. Geology, 25, 1023-1026. WIJBRANS, J. R. & McDOUGALL, I. 1986. 4~ dating of white micas from an Alpine highpressure metamorphic belt on Naxos (Greece); the resetting of the argon isotopic system. Contributions to Mineralogy and Petrology, 93, 187-194. & - - 1988. Metamorphic evolution of the Attic Cycladic metamorphic belt on Naxos (Cyclades, Greece) utilizing 4~ age spectrum -
-
A . L . W . LIPS ETAL.
474
measurements. Journal of Metamorphic Geology, 6, 571-594. & YORK, D. 1990. Single grain argon laser probe dating of phengites from the blueschist to greenschist transition on Sifnos (Cyclades, Greece). Contributions to Mineralogy and Petrology, 104, 582-593. ,
-
-
, VAN WEES, J. D., STEPHENSON, R. A. &
CLOETINGH, S. A. R L. 1993. Pressure-temperatute-time evolution of the high-pressure metamorphic complex of Sifnos, Greece. Geology, 21, 443-446.
Tethyan sutures of northern Turkey A R A L I. O K A Y & O K A N TI21YSUZ
I T U Eurasian Institute of Earth Sciences and Maden Faki~ltesi, Jeolofi BOli~mii, Ayaza~,a, 80626, istanbul, Turkey (e-maik okay@itu.edu.tr) Abstract: The two main Tethyan sutures of Turkey, the Izmir-Ankara-Erzincan and the
Intra-Pontide sutures, are reviewed through several well-studied transects crossing the suture regions. Both sutures have formed during the Early Tertiary continental collisions following northward subduction of Tethyan oceanic lithosphere. The Izmir-AnkaraErzincan suture is represented along most of its c. 2000 km length by Paleocene and younger thrust, which emplace the upper crustal rocks of the northern continent over that of the southern continent with an intervening tectonic layer of Cretaceous subductionaccretion complexes. These thrusts constitutes a profound stratigraphic, structural, magmatic and metamorphic break, of at least Carboniferous to Palaeocene age and form the main boundary between Laurasia and Gondwana in the Turkish transect. Voluminous subduction-accretion complexes of Triassic and Cretaceous ages occur respectively to the north and south of the suture giving the antithetic subduction polarities during these two periods. This, and evidence for a major accretionary orogeny of Late Triassic age north of the Izmir-Ankara-Erzincan suture suggest that two separate oceanic lithospheres, of Carboniferous to Triassic (Palaeo-Tethys) and of Triassic to Cretaceous ages (Neo-Tethys) respectively have been consumed along the suture. The final continental collision along the Izmir-Ankara-Erzincan suture was slightly diachronous and occurred in the earliest Palaeocene to the west and in the Late Palaeocene to the east. The c. 800 km long IntraPontide suture is younger in age and have formed during the Early Eocene and younger continental collisions linked to the opening of the Western Black Sea Basin as an oceanic back-arc basin. At present the North Anatolian Fault, which came into existence in the Late Miocene, follows the course of the older Intra-Pontide suture.
Sutures r e p r e s e n t the b o u n d a r i e s of f o r m e r lithospheric plates. In an idealized case of collision of two continental plates, the suture will be manifested as a major fault or fault zone in the brittle seismogenic part of the crust and by a shear zone in the ductile lower crust and lithospheric mantle. The sutures will form profound stratigraphic, palaeogeographic, structural, magmatic and metamorphic breaks and as such will be easy to recognize and distinguish from o t h e r major intraplate faults or fault zones. Here, the term 'suture' is used if the former plate boundary can be defined as a single fault or fault zone less than one kilometre in width, suture zone refers to a belt more than one kilometre in thickness, occupied by former oceanic crustal rocks. In the case of the suture zone, the former boundary between two plates cannot be shown as a single fault line on a map of 1:500 000 scale or smaller. The structures that represent sutures frequently show major along-strike variation in type and age, and most are probably post-collisional. For example, the suture in the Western Alps is represented from west to east, by the Late Cretaceous thrusts between the Sesia Z o n e and the Penninic units (e.g., Compagnoni 1977), by the segment of the post-collisional dextral strike-slip fault (the Insubric line) b e t w e e n the
Sesia Z o n e and the Eastern Alps (e.g., Schmid et al. 1989), and by the Late Cretaceous thrusts b e t w e e n the A u s t r a - A l p i n e and P e n n i n i c nappes along the Arosa Schuppen Z o n e (e.g., Ring et al. 1988). The only c o m m o n link b e t w e e n these three different structures is that they are believed to separate rocks deposited on different margins of an ocean. The delineation of a suture in the field is, hence, partly subjective, as the decision which stratigraphic units belong to which former plate can be controversial especially in metamorphic areas (e.g., see Michard et al. 1996, for a recent discussion on the problems associated with the delineation of the Alpine suture). H e r e , we describe s o m e of the T e t h y a n sutures and neighbouring former continental margins in Turkey. The description is aimed in answering some of the p r o b l e m s associated with sutures, such as can the boundaries of f o r m e r plates be d e s c r i b e d by single m a j o r faults, or by areally extensive suture zones. To what extent do the former continental margins fit into a simple W i l s o n i a n cycle of rifting, passive margin sedimentation, subduction and collision? H o w can variation of deformation, m e t a m o r p h i s m and m a g m a t i s m along the sutures be explained?
OKAY,A. I. & T(;rYSOZ,O. 1999. Tethyan sutures of northern Turkey. In: DURAND.B., JOLIVET,L., HORVATH,E & SERANNE,M. (eds) The Mediterranean Basins: Tertiary Extension within the Alpine Orogen. Geological Society, London, Special Publications, J56o 475-515
476
A.I. OKAY & O. TISYS(JZ
Major continental blocks of Turkey The Tethyan Ocean, which existed between Laurasia and Gondwana as a westward narrowing embayment since the Carboniferous, was not a single continuous oceanic plate during its long evolution (S,eng6r 1987; Ricou 1994; Stampfli 1996). Throughout its history, small continental fragments were rifted off from either side of the Tethyan Ocean and moved towards the other side, creating new oceanic lithosphere in their wakes, and eventually colliding with the opposite continental margin. Therefore in most regions, the former plate boundary between the two megacontinents, Laurasia and Gondwana, cannot be represented by a single suture. This is most obvious in Turkey, where several sutures isolate drifting continental fragments between Laurasia and Gondwana (S,eng6r & Yllmaz 1981). The final amalgamation of these fragments into a single continental mass occurred in Turkey as late as Late Tertiary, when the Arabian plate collided with the Anatolian plate. Figure 1 shows the sutures and major continental fragments in Turkey and the surrounding regions. There are six major lithospheric fragments in Turkey: the Strandja, the istanbul and the Sakarya Zones, the Anatolide-Tauride Block, the Karsehir Massif and the Arabian Platform (Fig. 1, ~engOr & Yllmaz 1981; ~engOr et al. 1982; Okay 1989a; Okay et al. 1994). The first three zones, which show Laurasian affinities, are classically referred to as the Pontides. They are separated by the |zmir-Ankara-Erzincan suture from the Klr~ehir Massif and the Anatolide-Tauride Block, the latter is in contact with the Arabian Platform along the Assyrian-Zagros suture (Fig. 1). Although separated by the Assyrian suture, the Anatolide-Tauride Block shows a similar Palaeozoic stratigraphy to that of the Arabian Platform, and hence to the northern margin of Gondwana. The Klrsehir Massif, which consists mainly of metamorphic and granitic rocks with Cretaceous isotopic ages, is in contact along the controversial Inner Tauride suture with the Anatolide-Tauride Block, while the Intra-Pontide suture constitutes the former plate boundary between the Sakarya and istanbul zones. Here, we describe two of the major sutures of northern Turkey, large sections
of which we have mapped or seen in the field. These are the |zmir-Ankara-Erzincan suture, which extends for c. 2000 km from the Aegean Sea to Georgia, and the c. 800 km long IntraPontide suture. First we briefly give the relevant geological features of the continental blocks, which are separated by these sutures. The Istanbul Zone
The Istanbul Zone is a small continental fragment, about 400 km long and 70 km wide, located in the southwestern margin of the Black Sea (Fig. 1). It is made up of a Precambrian crystalline basement overlain by a continuous, welldeveloped transgressive sedimentary sequence extending from Ordovician to Carboniferous, which was deformed during the Carboniferous Hercynian Orogeny (Aydln et al. 1986; Dean et al. 1997; G6r0r et al. 1997). The stratigraphy of the |stanbul Zone is shown in Fig. 2 on two sections, one from the western part near istanbul, and the other from the eastern part around Zonguldak (Fig. 1). These sections illustrate the notable facies differences along the length of the istanbul Zone, especially marked during the Carboniferous. In the Istanbul region the Carboniferous deposits consists of Vis6an pelagic limestones and radiolarian cherts, which pass up into thick turbidites probable of Namurian age, while in the east, around Zonguldak, Visean is represented by shallow marine carbonates overlain by Namurian to Westphalian paralic coal series (Dill 1976; Kerey et al. 1986). Westphalian megafloras show close affinities to those of western Europe and Donetz basins (Charles 1933). The Upper Carboniferous siliciclastic rocks in the Zonguldak area were derived from the north (Kerey et al. 1986) and were deposited in a large deltaic basin located on the southern margin of the Laurasian plate. These features, as well as palaeomagnetic results from the Palaeozoic rocks of the Istanbul Zone (Evans et al. 1991) underscore its Laurasian affinity. In the |stanbul Zone Hercynian deformation started earlier and is stronger to the west (MidCarboniferous) than to the east (latest Carboniferous). In the |stanbul region it is characterized by minor folds, relatively rare and discontinuous
Fig. 1. Tectonic map of northeastern Mediterranean region showing the major sutures and continental blocks. Sutures are shown by heavy lines with the polarity of former subduction zones indicated by filled triangles. Heavy lines with open triangles represent active subduction zones. The Late Cretaceous oceanic crust in the Black Sea is shown by grey tones. Small open triangles indicate the vergence of the major fold and thrust belts. BFZ denotes the Bornova Flysch Zone. Modified after SengOr (1984), Okay (1989a), and Okay et al. (1994, 1996).
T E T H Y A N S U T U R E S OF N O R T H E R N T U R K E Y
477
478
A.I. OKAY & O. TOYSOZ ISTANBUL istanbul
ZONE Zonguldak .-.-.-.-.-.-.-.-.-.-.-
-.-.-.-.-.-
i~_--~_--~_--~i~-~i~:~i ~i-:~6
Eocene Paleocene
a,.ede,o.ma,,on
!1--1-1--1--1"-1--'1--1--14 I-I-I-I-I-I-I-II'l--l--l--l--l--l---
-/ --I-I
Cretaceous
Jurassic
. -. .-.-.-. -_-.-. -. . . . . . _ . - - s _ _ _ _ _ _ _
9
Triassic
--,--,--,--,-5-5--F-F 51 Cimmeride deformation _;_;_;_; ..... .;_
~_;_%.%o_;822m_';~
.~.~.~.~o.o.o.o.o
Permian -o-o-o-o-o-o-o-o-.
Nm1r o-o-o-c ~,,,~,~, > 1000 m~o ;%?;?$.%.~ ;?$?$?;?;?;%?;?,
Carboniferous :::::::::::::::>3000
m: I
~ : l l t : [ : [ i l : : ] : : :
I
I Devonian
t
7-F-f~prTv7=7=:=7 -
Silurian
~
Ill
I [ I i
I t
I I
I
I
t
t
I 350 m t
I
I
I
I
~
-r op ra
============================ :iiiiiiiiiiiiiii250m[;!il; --.'O'.'O'_'O_O'_'d,'O.'O.d_',.
.-o-o~o-o-o-o~o-o-~ o-o-o-o-o-o-o-o-o.
Ordovician
Hercynian deformation
i:::::I
Flysch
[:.iiii;iiiii] Sandstone
Ophiolite
~
Marl Shale
Metamorphic units
-o-o-o-o-o-o-o-o-
". . . .
Marmara
~
~'x, G G ? 5 ~ ~
[ - - 1 T e r t i a r y deposits ~ - ~ T e r t i a r y STRANDJA MASSIF ~]U. Cretaceous andesite ~
granitoid ~ - x ] U .
~
~
t
t
J
b
~.-"
,
J -- i
Cretaceous granitoid ~
Triassic sedimentary rock s
.~; ; ; ,...~ . ~
~
i
ISTANBUL ZONE ~ U. Cretaceous limestone, marl v- volcm~ic facies
Triassic_Jurassic metasediments ~
~-~-]Permian metagranites
S
~
~ - -
~-- i ~,o),nuK,.-7 t - i
I 7- I
B asalt,radiolarite,serpentinite Cret.-Palaeocene accretionary complex SAKARYA ZONE ~ Jurassic-Eocene sequence ~
~
Ordovician-Carboniferous sedimentary rocks
~
Metamorphic rocks and metaophiolite of unknown age ano ongln
Pre-Jurassic basement (mainly Triassic subductionaccretion units) 0 I
20 t
40 ~
60 I
80 kin I
Fig. 21. Geological map of the Marmara Sea region.
followed by the flip of the subduction zone from a south- to north-dipping subduction, which is required to explain the start of the subductionrelated volcanism in the Outer Eastern Pontides during the Turonian. The collision between the Eastern Pontide magmatic arc and the Anatolide-Tauride Block resulted in thrust imbrication of the active margin. The collision is well dated as Late Palaeocene from the age of the foreland basin sequence in front of the north-vergent thrust sheets in the Inner Eastern Pontides. This is corroborated by the end of the arc magmatism in the outer Eastern Pontides during the Late Paleocene (Elmas 1995; Okay & Sahinttirk 1998). Data from the Tauride margin, where the strongest deformation occurred during the Early Maastrichtian, are more ambiguous but does not exclude a Late Palaeocene collision.
The Intra-Pontide suture The Intra-Pontide suture forms the c. 400 km long boundary between the istanbul Zone and the Sakarya Zone. It also extends for approximately another 400 km farther west through the
Sea of Marmara defining the contact between the Rhodope-Strandja Massif and the Sakarya Zone, and bending south may join the izmir-Ankara-Erzincan suture in the central Aegean Sea (Fig. 1). The Intra-Pontide suture can, thus, be divided into two segments, the eastern segment between the istanbul and Sakarya zones and the western segment between R h o d o p e - S t r a n d j a Massif and the Sakarya Zone.
Intra-Pontide suture between the Istanbul and Sakarya zones The Intra-Pontide suture in this segment can be subdivided into a 400 km long east-west-trending collisional suture, and the two limiting, north-trending transform faults, which during the Cretaceous also formed a plate boundary (Figs 13 and 21). Although the suture constitutes a profound stratigraphic, metamorphic, magmatic and structural boundary, the lithological units and structures along the suture are poorly known. Along most of its length the suture is defined by the North Anatolian Fault, a major post-Miocene strike-slip fault with a cumulate
504
A.I. OKAY & O. TOYSLIZ
r-xJ
r-.-/ /--,-/
r-~
/
/'xj
'
rx../
1
,
r,~
,'-'x-/ r-x../
r".../ f'x..r..._/r~. ~ X X
x
--xJx x
X
X
X
X
X
X
X
~-'-zs
x
38x
x
k
X
x
x
%/
•
x
•
X
A-
X
•
X X /~
X
•
X
>32x
X
x
X
•
--
•
--
30
2
x
o Mudurnu .
.
.
.
....)J_
-0 [
2
4
6
I
I
I
8 km'-----
I
--
--
--
,
--
SAKARYAZONE Paleocene ~ Paleocene [~'.~-:~ U. Cret. ~_----~ L.Cret. ~_ ~ U.Jurassicl i--r ~ m
M. JurassicI x
m
I ] Lacustrinelimestone Sandstone,shale, marl ~ U . Shale,sandstone,tuff, marl Pelagic limestone F-~ Limestone,tuff
~--j
xj Basic tuft',basalt
Md~K~ )cene and younger ~
Metabasite,marble
Cretaceousturbidites ~ Basalt, radiolarite accretionarycomplex ~ Slightlycrystallized clastic rocks with Dewmianfossils
Gabbro,amphibolite Ultramaficrock Marble, phyllite
Fig. 22. Geological map of the region around the Intra-Pontide suture south of Almaclk mountains (modified from Abdiisselamo~lu 1959; Yflmaz et al. 1982).
dextral offset of 25-40 km ($eng6r 1979; Barka 1992). The North Anatolian Fault disguises the pre-Miocene relations between the ]stanbul and Sakarya zones. Furthermore, there are large metamorphic areas along the suture, whose provenance and age are unknown. A good example of the problems associated with the Intra-Pontide suture is provided by the geology of the Almaclk mountains east of Adapazarl studied by Abdtisselamo~lu (1959), GOztibol (1980), Yllmaz et al. (1982) and Greber (1996). A geological map based on these works is given in Fig. 22. In the Almaclk region the North Anatolian
Fault defines the suture and divides the region into two parts (Fig. 22). To the south of the North Anatolian Fault there is a well-developed Jurassic to Eocene sequence of the Sakarya Zone (Fig. 12, Saner 1980; G6ziibol 1980; Altlner et al. 1991), which indicates continuous marine deposition from Mid-Jurassic to Early Eocene. To the north the Intra-Pontide suture is defined by a few kilometre wide North Anatolian Fault zone, where elongate tectonic slivers of serpentinite, pre-Jurassic basement of the Sakarya Zone and terrigeneous Neogene deposits outcrop (Fig. 22). In the Almaclk mountains north of the North Anatolian Fault there is a thrust stack separated
TETHYAN SUTURES OF NORTHERN TURKEY by north-south-trending thrust traces (Fig. 22). Sanbudak et al. (1990) have explained the discordance between the general east-west trend of the regional structures and the north-south thrust traces by the Miocene flake rotation of the Almaclk thrust stack in the North Anatolian Fault Zone. At the base of the Almaclk thrust stack there is a metamorphic sequence of quartzite, phyllite, micaschist and marble, over 1000 m in thickness. The metasedimentary sequence is overthrust by a dismembered and metamorphosed ophiolite of peridotite, pyroxenite, gabbro and basalt (Fig. 22, Abdtisselamo~lu 1959; Ydmaz et al. 1982; GOziibol 1980). The common greenschist to amphibolite facies metamorphic grade shown by the dismembered ophiolite and metasedimentary sequence indicates that the ophiolite was tectonically emplaced over the sedimentary sequence prior to the regional metamorphism. A slightly metamorphic clastic-limestone sequence with scarce Devonian fossils (Abdtisselamo~lu 1959) lies with a problematic contact on the metamorphosed ophiolite (Fig. 2). Abdtisselamo~lu (1959) and GOztibol (1980) regard the metasedimentary sequence and the meta-ophiolite as of pre-Devonian age. On the other hand, Ydmaz et al. (1982) consider the meta-ophiolite of
ISTANBUL ZONE Ulus Basin Cretaceousturbidites Carboniferousand Olistostromes Triassicolistoliths
NNW A
lokmt
. . . . . ~ _ ~ ~ 9 ..,_,_,_ N
~
~
Cretaceous age and regard the contact with the overlying Devonian rocks as tectonic. The only consensus on the age of the metamorphic rocks is the poorly preserved unconformable cover of Maastrichtian to Palaeocene neritic limestones (Greber 1996). A similar problem exists in the Armutlu peninsula farther west along the suture, where ages ranging from Precambrian to Devonian have been suggested for a similar pre-Upper Cretaceous metamorphic sequence (Akartuna 1968; GOnctio~lu & Erendil 1990; Ydmaz et al. 1995). As there are no data on the depositional or isotopic ages of these metamorphic complexes, the tectonics and geological evolution of the Intra-Pontide suture remain highly uncertain. O n s h o r e c o n t i n u a t i o n s o f the West B l a c k Sea a n d West C r i m e a n faults
These fossil transform faults form the north-south-trending boundaries of the istanbul Zone with the Strandja Massif to the west and the Sakarya Zone to the east (Fig. 1). Although they do not represent collision-related sutures, nevertheless they are former plate boundaries and form distinct stratigraphic, metamorphic, magmatic and structural discontinuities.
SAKARYA ZONE Triassic-Liassicshale ESE andsiltstone WestCrimean Mid-Jurassic granite ~ c o n t . Qfaultl !
, " -_"~~
