1992 - McClay - Thrust Tectonics
1992 - McClay - Thrust Tectonics
1992 - McClay - Thrust Tectonics
K.R. McClay is Professor of Structural Geology at Royal Holloway and Bedford New College, University
of London.
The cover photograph is of the McConnell Thrust in Mississippian Limestones at Mt Head, Highwood
Range, Southern Canadian Rocky Mountains, Alberta, Canada.
Also available
Edited by KKMcClay hK
K.R. McClay
Department of Geology
Royal Holloway and Bedford New College
University of London
Si C H A P M A N & HALL
London • New York • Tokyo • Melbourne • Madras
Published by Chapman & Hall, 2-6 Boundary Row, London SEl 8HN
Chapman & Hall, 29 West 35th Street, New York, NYlOOOl, USA
Chapman & Hall Japan, Thomson Publishing Japan, Hirakawacho Nemoto Building, 7F,
1-7-11 Hirakawa-cho, Chiyoda-ku, Tokyo 102, Japan
Chapman & Hall Australia, Thomas Nelson AustraUa, 102 Dodds Street, South Melbourne, Victoria 3205,
Australia
Chapman & Hall India, R. Seshadri, 32 Second Main Road, CIT East, Madras 600 035, India
Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted
under the UK Copyright Designs and Patents Act, 1988, this pubhcation may not be reproduced, stored, or
transmitted, in any form or by any means, without the prior permission in writing of the publishers, or in the
case of reprographic reproduction only in accordance with the terms of the licences issued by the Copyright
Licensing Agency in the UK, or in accordance with the terms of licences issued by the appropriate
Reproduction Rights Organization outside the UK. Enquiries concerning reproduction outside the terms
stated here should be sent to the publishers at the London address printed on this page.
The publisher makes no representation, express or implied, with regard to the accuracy of the information
contained in this book and cannot accept any legal responsibiHty or Uability for any errors or omissions that
may be made.
A catalogue record for this book is available from the British Library
Errata
Contributors viii
Acknowledgements x
Introduction xi
Part One
Theoretical aspects and thrust mechanics
Erosional control of active compressional orogens
C. Beaumont, P. Fullsack and J. Hamilton 1
Dynamic and kinematic growth and change of a Coulomb wedge
S.D. Willett 19
A developmental stage of a foreland belt
E.G. Bombolakis 33
One-dimensional models for plane and non-plane power-law flow in shortening and elongating
thrust zones
5. Wojtal 41
Part Two
Physical modelling
Centrifuge modelling of the propagation of thrust faults
/. M. Dixon and Shumin Liu 53
Physical models of thrust wedges
Huiqi Lui, K.R. McClay and D. Powell 71
Generation of curved fold-thrust belts: Insight from simple physical and analytical models
5. Marshak, M.S. Wilkerson and A.T. Hsui 83
Thrust faults in inverted extensional basins
K.R. McClay and P.G. Buchanan 93
Part Three
Thrust geometries and thrust systems
Rates of folding and faulting determined from growth strata
7. Suppe, G. T. Chou and S. C. Hook 105
Role of shear in fault-propagation folding
/. Mosar and J. Suppe 123
2-D reconstruction of thrust evolution using the fault-bend fold method
R. Zoetemeijer and W. Sassi 133
Kinematic models of deformation at an oblique ramp
T.G. Apotria, W.T. Snedden, J.H. Spang and D.V. Wiltschko 141
Stress controls on fold thrust style
W.R. Jamison 155
Kinematics of large-scale asymmetric buckle folds in overthrust shear: an example from the
Helvetic nappes
M.G. Rowan and R. Kligfield 165
Forelimb deformation in some natural examples of fault-propagation folds
J.L. Alonso and A. Teixell 175
The geometric evolution of foreland thrust systems
M.P. Fischer and N.B. Woodward 181
Some geometric problems of ramp-flat thrust models
J.G. Ramsay 191
The duplex model: Implications from a study of flexural-slip duplexes
P.W.G. Tanner 201
Palaeomagnetic techniques applied to thrust belts
A.M. McCaig and E. McClelland 209
Evolution of crystalline thrust sheets in the internal parts of mountain chains
R.D. Hatcher, Jr. and R.J. Hooper 217
Part Four
Pyrenees
Evolution of a continental collision belt: ECORS-Pyrenees crustal balanced cross-section
J.A. Munoz 235
Thrusting and foreland basin evolution in the Southern Pyrenees
C. Puigdefabregas, J. A. Munoz and J. Verges 247
South Pyrenean fold-and-thrust belt: Role of foreland evaporitic levels in thrust geometry
/ . Verges, J.A. Munoz and A. Martinez 255
Contrasting rotations within thrust sheets and kinematics of thrust tectonics as derived from
palaeomagnetic data: an example from the Southern Pyrenees
J. Dinares, E. McClelland and P. Santanach 265
Part Five
Alps
The Alps - a transpressive pile of peels
H. Laubscher 277
Structural evolution of the western Chartreuse fold and thrust system, NW French Subalpine Chains
R.W.H. Butler 287
Kinematics of a transverse zone in the Southern Alps, Italy
G. Schonborn 299
Hangingwall geometry of overthrusts emanating from ductile decollements
P. Jordan and T. Noack 311
The Venetian Alps thrust belt
C. Doglioni 319
Part Six
Himalayas
Thrust geometries, interferences and rotations in the Northwest Himalaya
P.J. Treloar, M.P. Coward, A.F. Chambers, C.N. Izatt and K.C. Jackson 325
Balanced and retrodeformed geological cross-section from the frontal Sulaiman Lobe, Pakistan:
Duplex development in thick strata along the western margin of the Indian Plate
I.A.K. Jadoon, R.D. Lawrence and R.J. Lillie 343
Part Seven
NW American Cordillera
The Monashee decollement of the southern Canadian Cordillera: a crustal-scale shear zone linking the
Rocky Mountain Foreland belt to lower crust beneath accreted terranes
R.L. Brown, S.D. Carr, B.J. Johnson, V.J. Coleman, F.A. Cook and J.L. Varsek 357
The Skeena fold belt: a Hnk between the Coast Plutonic Complex, the Omineca belt and the Rocky
Mountain fold and thrust belt
C.A. Evenchick 365
Geometric evidence for synchronous thrusting in the southern Alberta and northwest Montana thrust
belts
S.E. Boyer 377
The analysis of fracture systems in subsurface thrust structures from the Foothills of the Canadian
Rockies
M. Cooper 391
Thrust tectonics and Cretaceous intracontinental shortening in southeast Alaska
CM. Rubin andJ.B. Saleeby 407
Glossary
K.R. McClay 419
Index 435
CONTRIBUTORS
The papers presented in this volume were part of the Thrust Tectonics Conference held at Royal Holloway and
Bedford New College, University of London from 4-7 April 1990. The Conference and production of this
volume were greatly assisted by the generous sponsorship of the Geology Department, Royal Holloway and
Bedford New College, Brasoil UK Ltd, BP Petroleum Development PLC, Chevron Petroleum UK Ltd, Clyde
Petroleum, Enterprise Oil, Esso Exploration and Production Ltd and Shell Petroleum UK.
The staff and students of the Geology Department, Royal Holloway and Bedford New College, in particular
Tim Liverton, Luis D'el Rey Silva, Martin Insley, James Dobson, Kevin D'Souza and Sandra Muir, are
thanked for their assistance and support in running the conference.
The authors who submitted manuscrips for consideration for publication in this volume are thanked for their
efforts and particularly for following the editor's guidelines and submitting in Macintosh format. Minor
grammatical changes in the texts remain the responsibility of the editor.
The manuscrips were ably and promptly reviewed by the persons listed below, many of whom kindly
reviewed two or more papers:
Finally a special vote of heartfelt thanks goes to Andrew D. Scott and Ruth Durrell for their invaluable and
cheerful assistance in organizing the conference and in the production of this volume. This book was compiled
and typeset by R. Durrell, A. Scott and K. McClay, Geology Department, Royal Holloway and Bedford New
College, University of London using Microsoft Word 4.0® and Pagemaker 4.0®. Kevin D'Souza kindly
produced the PMTs. Phototypesetting was done at the Typesetting Bureau of the University of London
Computer Centre.
Ken McClay
Geology Department
Royal Holloway and Bedford New College
University of London
Thrust Tectonics: An introduction
K.R. McClay
Since the first Thrust and Nappe Tectonics Conference in London in 1979 (McClay & Price 1981), and the Toulouse Meeting
on Thrusting and Deformation in 1984 (Piatt et al. 1986) there have been considerable advances in the study of thrust systems
incorporating new field observations, conceptual models, mechanical models, analogue and numerical simulations, together
with geophysical studies of thrust belts. Thrust Tectonics 1990 was an International Conference convened by the editor and
held at Royal Holloway and Bedford New College, University of London, Egham Surrey, from April 4th until April 7th 1990.
There were one hundred and seventy participants from all continents except South America. The conference was generously
sponsored by Brasoil U.K. Limited, BP Exploration, Chevron U.K. Limited, Clyde Petroleum, Enterprise Oil, Esso
Exploration and Production UK Limited, and Shell U.K. Exploration and Production.
One hundred and five contributions were presented at the meeting, - seventy six oral presentations (together with poster
displays) and an additional twenty nine posters without oral presentation (McClay 1990, conference abstract volume).
The thirty six papers presented in this volume represent a distillation of the material presented at the conference and are
grouped into thematic sections that follow the main topics of discussion at the meeting. To an extent the grouping of papers
into specific thematic sets is somewhat arbitrary as there are many common linkages between the various sections. In
particular many papers focus upon both the geometries of thrust systems and their kinematic evolution - a recurring theme
throughout the volume. Space limitations prevented publication of all of the papers presented at the meeting but it is hoped
that this volume will provide the reader with an appreciation of the current research themes and concepts of thrust systems.
The contents of the volume are summarized below.
The first section of this volume i s headed' Theoretical aspects and thrust mechanics'. Beaumont et al. describe an elegant
finite element model of the evolution of the Southern Alps of New Zealand. The model incorporates tectonic data, erosional
and orthographic data and demonstrates surprisingly that the climate, erosion and fluvial systems are coupled to the orogenic
evolution such that the structural and metamorphic style of this thrust belt may be controlled by the fluvial erosion and climate
on the western flank of the Southern Alps. Willett presents a finite-element model of Coulomb wedges in order to simulate
the growth of accretionary prisms and foreland fold and thrust belts. The model demonstrates self-similar growth of the
Coulomb wedge and investigates the kinematics and particle velocities of critical wedges growing by accretion. Changes in
the boundary conditions of the wedge (e.g. a change in the basal friction) changes the wedge geometry and thus may permit
out of sequence thrusting (increase in basal friction) or extensional collapse of the wedge whilst in compression (decrease in
basal friction). Bombolakis examines the partitioning of elastic strains and strain-rate dependent ductile deformation in the
emplacement of the frontal parts of thrust systems. He demonstrates that significant thrust movement in these environments
can occur by aseismic processes. Wojtal presents a one-dimensional model for plane and non-plane power-law flow in
shortening and elongating thrust zones. He concludes that viscous creep is significant along external thrust faults and that
fluid assisted diffusive mass transfer processes (pressure solution) are important deformation mechanisms in the fine-grained
banded cataclasites of such thrust systems.
The second section of this volume titled 'Physical modelling' contains four papers on analogue modelling of thrust
structures. Dixon & Liu investigate the formation of foreland fold and thrust belts using scaled centrifuge models. They
demonstrate that folding precedes thrusting, and that the thrust periodicity is inherited from the buckling wavelength of the
antecedent fold systems. The models also demonstrate the 3D coalescence of small thrusts to form larger thrust systems. Liu
et al. present a detailed study of sandbox models of thrust systems found in accretionary wedges and foreland fold and thrust
belts. The boundary conditions of the models were systematically varied and the results are in broad agreement with Coulomb
wedge theories (cf. Davis et al. 1983; Dahlen 1990). Low basal friction models are characterised by both foreland-vergent
and hinterland-vergent thrust faults (i.e. backthrusts) as well as low critical tapers of the resultant Coulomb wedge.
Intermediate basal friction models show a dominance of foreland-vergent thrust systems whereas models in which the
coefficient of friction at the base is equal to that within the wedge, depart from ideal Coulomb wedges and show greater tapers
than predicted by theory. Marshak et al. investigate the generation of curved thrust belts using viscous analogue models.
K.R.MCCLAY
finite difference models and sandbox models. They conclude that curved fold and thrust belts may be generated by differential
displacement such as that resulting from the interaction with obstacles or along strike variations in sediment thicknesses.
McCIay & Buchanan investigate the geometries of thrust systems formed by inversion of extensional basins. Sandbox
models demonstrate the complexities of thrusts produced by inversion - steep angle thrusts, footwall shortcut thrusts,
backthrusts and both 'forward-breaking' and 'break-back' thrust sequences.
The third section of this volume is titled - 'Thrust geometries and thrust systems' and focuses upon geometric and kinematic
problems of thrust systems. Suppe et al. set out the theory for the geometry of growth folding associated with kink band
folding, fault-bend folding and fault-propagation folding. They demonstrate that the stratal geometries of growth sequences
may be used to determine fault slip rates and illustrate their results with both seismic and map examples of growth fold systems.
Mosar & Suppe introduce the theory for fault-propagation folding with variable shear. They show how the geometries of
fault-propagation or tip-line folds may be significantly altered by layer parallel shear. Zoetemeijer & Sassi describe a
computer method for 2D reconstruction of thrust systems using the fault-bend fold model. The method can incorporate
syntectonic sedimentation and erosion. The problems of three dimensional thrust geometries developed over oblique ramps
are tackled by Apotria etal. They describe a detailed kinematic analysis of fault-bend fold interactions at frontal and oblique
ramps and illustrate their results with both cross-sections and map views. The stress controls on thrust related folding are
addressed by Jamison. He proposes that the style of folding in a thrust belt is a result of the competition between buckling
and faulting and as such is dependent upon the depth of burial and tectonic regime. The hypothesis is illustrated with examples
from the Rocky Mountain thrust belt. Rowan & Kligfleld present a kinematic analysis of asymmetric detachment folds from
the Wildhom nappe, Helvetic Alps. They propose that overthrust shear between the nappe boundaries generated buckle folds
and rotational strains within the nappe system. Alonso & Teixell describe an analysis of fault-propagation folds in which
the forelimbs are thinned by heterogeneous simple shear localized at ramps and not penetrative through the whole of the thrust
sheet. The progressive geometric evolution of thrust faults and related folds is examined by Fisclier & Woodward. They
propose, using examples from both Wyoming and the Appalachians, that fold-thrust systems do not evolve in a self-similar
fashion but argue that variations of fold-thrust structures along strike in a thrust belt do not represent temporal variations in
evolution but rather reflect intrinsic variations in the mechanical response of the stratigraphic package to contraction.
Ramsay argues against unquestioning acceptance and wholesale application of preconceived geometric models for thrust-
fold systems - particularly those where the deformation is concentrated in the hangingwalls of thrust faults. He describes
examples where footwall deflection is important and advocates many more detailed studies of natural thrust systems. The
geometries and kinematics of duplex structures are addressed by Tanner who proposes that the Boyer & Elliott (1982) model
does not universally apply and that many natural duplex systems show indications of synchronous fault movement. The use
of palaeomagnetic studies to determine rotations about vertical axes in thrust belts is reviewed by McCaig & McClelland
whereas Hatcher & Hooper review the main characteristics of thrust sheets that involve crystalline basement in the interior
parts of mountain belts.
The fourth section on the 'Pyrenees' concerns recent research which integrates structural, sedimentological and geophysical
results. An overview and balanced crustal cross-section of the Pyrenean continental collision belt is presented by Munoz.
He points out the discrepancy between the cross-sectional length of the upper crust and that of the lower crust and argues for
subduction of Iberian lower crust northwards under the European crust. The structural, stratigraphic and sedimentological
evolution of the southern Pyrenean foreland basins is synthesized by Puigdefabregas et al. The evolution of the foreland
basin sequences is described with reference to the patterns of thrusting. Verges et al. investigate the role of the foreland
evaporitic sequences in controlling the evolution and emplacement of the frontal thrust systems and demonstrate their
influence on the strike of the thrust system and on the location of oblique ramps. The palaeomagnetic study of Dinares et
al. demonstrates that up to 45° rotation about vertical axes occurs in thrust sheets associated with oblique ramps in the southern
Pyrenean thrust belt.
The fifth section of this volume contains five papers on the 'Alps'. Laubscher addresses the problems of linked com-
pression and strike-slip tectonics in the central Alps and quantifies the E-W strike-slip components. Butler describes the
western Chartreuse fold and thrust system and emphasizes the role of folding and preexisting extensional structures in the
evolution of the belt. The detailed geometries and kinematics of a transverse zone in the southern Alps are analysed by
Schonborn. Jordan & Noak develop models for the evolution of the hangingwall geometries of thrust systems developed
above ductile decollements and apply these to examples in the Jura. The final paper in this section by Doglioni describes the
structure of the Venetian Alps.
The sixth section of this book contains two papers on the 'Himalayas'. Treloar etal. summarize the kinematic evolution
of the northwest Himalayas linking displacement patterns, thrust sequences with the analysis of rotations of the thrust sheets
using palaeomagnetic data. Jadoon et al. describe the geometry and kinematics of the Sulaiman fold belt, Pakistan.
INTRODUCTION
The final section contains five papers on the 'North American Cordillera'. Brown etal. describe the Monashee decollement
in the southern Canadian Cordillera. Using new LITHOPROBE data they are able to demonstrate the existence of a linked
detachment from the Rocky Mountain foreland fold and thrust belt through to the accreted terranes on the western margin of
the Canadian Cordillera. Evenchick describes the newly recognized Skeena fold belt and demonstrates a kinematic linkage
between the Coast Plutonic complex and the Omineca and Rocky Mountain fold and thrust belts. The geometric and kinematic
evolution of duplex systems is addressed by Boyer with reference to duplexes in the southern Canadian Rocky Mountains
and in the Rocky Mountains of Montana. He argues that these duplexes evolved through synchronous thrust movement - not
by a sequential, forward-breaking 'piggy-back' sequence. Cooper describes a detailed subsurface fracture analysis of thrust-
related folds in the Foothills of the Canadian Rocky Mountains. The last paper in this section by Rubin & Saleeby describes
the thrust belt features in the interior of the orogen between the Intermontane and Insular terranes of the North American
Cordillera.
It will be evident to a reader of this volume that the theoretical and conceptual models for the geometries and kinematic
evolution of high-level foreland fold and thrust belt systems are well developed and highly sophisticated (Suppe etal. 1991,
this volume; Mosar & Suppe 1991, this volume). However, detailed field observations are still needed to test their applicability
particularly with respect to the question of whether thrust structures evolve in a self-similar fashion (Fischer & Woodward
1991, this volume) (cf.fault related folds) or whether they evolve through different structural forms (e.g. from detachment
folds - to fault propagation folds - to fault bend folds). It is also apparent that whilst the external zones of thrust belts are
relatively well studied, the interior parts of orogens are less well understood and require much more detailed research on the
more complex thrust systems found therein. Field work will need to be integrated with geophysical data and sections of the
whole crust and lithosphere will need to be constructed. Petrofabric, strain and P-T-t data are needed in order to interpret the
processes operating during thrust emplacement. The mechanics of thrust belt evolution are now being addressed using
Coulomb wedge models (cf. Willett 1991, this volume), variationsof which may provide answers for the problems of fault
sequencing in thrust belts. It is apparent that, from both analogue modelling (Dixon & Liu 1991, this volume; Liu et al. 1991,
this volume) and from field studies (Boyer 1991, this volume), synchronous thrust movement is a significant element in the
kinematic evolution of foreland fold and thrust belts. This therefore will have important implications for thrust sequences
and the balancing and restoration of cross-sections in thrust terranes.
Field based research, however, is the ultimate test of the theoretical, conceptual, numerical and analogue models. It is clear
that the better understanding of thrust belts - for example that of the Canadian Rocky Mountains (e.g. Price 1981; Evenchick
1991, this volume; Brown er al. 1991, this volume) and of the Pyrenees (e.g. Munoz 1991, this volume; Puigdefabregas et
al. 1991, this volume), stems from detailed fieldwork that integrates structural geology, stratigraphy, sedimentology and
geophysical data. Future research on thrust tectonics will still need to be firmly based upon multidisciplinary field studies
if progress is to be made in the understanding of the geometries, kinematics and mechanics of thrust systems.
References
Boyer, S.E. 1991. Geometric evidence for synchronous thrusting in the McClay.K.R. 1990. {ed.)Thnisttectonics 1990. Programme with abstracts
southern Alberta and northwest Montana thrust belts, this volume. volume, Royal Holloway and Bedford New College, University of
Boyer, S.E. & Elliott, D. 1982. Thrust systems. American Associcition of London, I06p.
Petroleum Geologists Bulletin, (,(>, 1196-1230. McClay, K.R. & Price, N.J. 1981. (eds). Thrust and Nappe Tectonics.
Brown, R, L., Carr, S. D., Johnson, B, J., Coleman, V. J.. Cook, F. A. & Geological Society of London Special Publication, 9, 544p.
Varsek.J.L. 1991. The Monashee decollement of the southern Canadian Mufioz, J. A. 1991. Evolution of a continental collision belt: ECORS-
Cordillera: a crustal-scale shear zone linking the Rocky Mountain Pyrenees crustal balanced cross-section, this volume.
Foreland Belt to lower crust beneath accreted terranes, this volume. Mosar, J. & Suppe, J. 1991. Role of shear in fault-propagation folding, this
Dahlen, F.A. 1990. Critical taper model of fold-and-thrust belts and accre- volume.
tionary wedges. Annual Reviews of Earth and Planetary Sciences. 18,5.S- Piatt, J.P., Coward, M.P., Deramond, J. & Hossack, J. 1986. (eds) Thrusting
99. and Deformation, Journal of Structural Geology, 8, 215 - 483.
Davis, D., Suppe, J. & Dahlen, F.A. 1983. Mechanics of fold-and-thrust belts Price, R.A. 1981. The Cordilleran foreland thrust and fold belt in the southern
and accretionary wedges. Journal of Geophysical Research, 94,10,347- Canadian Rocky Mountains. In: McClay, K.R. & Price, N.J. (eds). Thrust
54. and Nappe Tectonics, Geological Society of London Special Publica-
Dixon, J. M. & Shumin Liu. 1991. Centrifuge modelling of the propagation tion, 9, 427-448.
of thrust faults, this volume. Puigdefabregas, C , Mufioz. J. A. & Verges, J. 1991. Thrusting and foreland
Evenchick, C. A. 1991. The Skeena fold belt: a link between the Coast basin evolution in the Southern Pyrenees, this volume.
Plutonic Complex, the Omineca Belt and the Rocky Mountain fold and Suppe, J., Chou, G. T. & Hook, S. C. 1991. Rates of folding and faulting
thrust belt, this volume. determined from growth strata, this volume.
Fischer, M. P. & Woodward, N. B. 1991. The geometric evolution of Willett, S. D. 1991. Dynamic and kinematic growth and change of a
foreland thrust systems, this volume. Coulomb wedge, this volume.
Liu Huiqi, McClay, K. R. & Powell, D. 1991. Physical models of thrust
wedges, this volume.
PART ONE
Theoretical aspects and thrust mechanics
Erosional control of active compressional orogens
Christopher Beaumont, Philippe Fullsack & Juliet Hamilton
Abstract: Denudation has long been acknowledged as a process that contributes to the unroofing
of compressional orogens. It has, however, mainly been considered as a passive process, not one
that can dictate or control the tectonic evolution. This view prevails despite the knowledge that
the style of deformation is controlled by the interplay of gravitational and tectonic stresses: an
interplay that is sensitive to the mass removed by denudation.
This paper presents a model designed to investigate the syn-tectonic style of a compressional orogen
that is subject to large-scale denudation. The model couples tectonics and erosion in the following
manner. The rate of lithospheric deformation, including surface uplift, is calculated by a plane-
strain rigid plastic and/or viscous model of the lithosphere's response to tectonic compression. A
digital erosion model acts on the current planform surface of the orogen to redistribute mass by a
combination of short-range (hillslope) diffusion and long-range (fluvial) transport.
The fluvial network is recharged by a model of orographic rainfall in which the distribution is
related to the current model topography. Tectonic uplift is added to the planform surface at the
beginning of each erosional timestep. The current planform topography after erosion is averaged
along the direction normal to compression and is then used to define the surface for the next step
of the tectonic model.
Preliminary results compare favourably with the orogen-scale precipitation, morphology and
tectonics of the Southern Alps of South Island, New Zealand. The potential importance of at-
mospheric coupling to orogenic evolution is illustrated by results from contrasting models of the
Southern Alps in which: (1) there is no erosion; (2) rainfall is derived from a westerly source, and;
(3) rainfall is derived from an easterly source. If the model is correct, the structural and metamor-
phic style of this orogen is controlled by fluvial denudation of its western flank charged by the
orographic precipitation.
Can erosion control the evolution of compressional orogens the component of the stress tensor within the wedge that is
at the scale of the orogen? This paper investigates the way in related to the weight of the overburden. Additional deformation
which orographic rainfall and consequent erosion of a follows because the yield strength of the Coulomb (frictional)
mountain belt may be linked to the tectonics of the underlying material decreases as overburden is removed.
deforming orogen. If this analysis is correct, palaeoclimatic It may, alternatively, be argued that a viscous, not Coulomb
information may be required before the evolution of ancient orplastic, theology is more appropriate for orogenic mechanics
orogens can be fully understood. at the largest scale. If this is true, the viscous sheet model
The basis of the coupling between erosion and deforma- (England & McKenzie 1982; Vilotte et al. 1986, among
tion was discussed by Jamieson & Beaumont (1988, 1989). others) demonstrates the relative importance of gravitational
They linked the concept of critical topography (Chappie and tectonic forces on the velocity and strain-rate distribu-
1978; Stockmal 1983; Davis etal. 1983; Dahlen 1984, among tions within the sheet through their ratio, the Argand number.
others) applied at the orogen scale, which they termed critical For finite Argand numbers gravitational forces limit orogenic
shape, to syn-tectonic deformation within the orogen when thickening, a plateau develops and high strain rate regions are
the critical shape is destroyed by erosion. The purpose of this limited to the periphery of the plateau. When erosion removes
paper is to quantify these concepts. A critical orogen will mass from the plateau or its flanks, gravitational stabilization
respond to erosion by a perturbation to the internal velocity is reduced and strain rates correspondingly increase where
distribution that will act to restore the critical shape. Persist- erosion has occurred. The vertical strain, coupled with
ent erosion leads to exhumation of material from within the isostatic adjustment, induces enhanced uplift as a response to
orogen in proportion to the mass removed by erosion. The the erosion. If erosion persists, the coupled tectonic response
process may control how rocks from deep within an orogen induces exhumation of the crust in the viscous sheet model
are exposed at the surface. that is equivalent to that of the eroding Coulomb wedge.
By considering erosion of a critical Coulomb wedge in The strain within a deforming compressional orogen is
steady state, Dahlen & Barr (1989) and Barr & Dahlen (1989) related to the internal state of stress, irrespective of the
illustrated the consequences of the dynamic steady state for rheology. At crustal and lithospheric scales gravitationally
the velocity distribution, and pressure-temperature-time (PTt) induced stress is as important as tectonic stress. It follows that
paths of rocks as they pass through the wedge. Erosion the syn-tectonic deformation of an orogen will be modified
destroys the critical state of the deforming wedge by reducing by erosion.
1
C. B E A U M O N T £7/iL.
Coupled Oroaen and Erosion Model Here the assumption is made that when the deformation is
0R06RAPI
large, elastic strains are negligible, and a rigid-plastic rheology
-~~-/\ TECTONIC is used to model brittle behaviour of an orogen. The theory
/ / COMPRESSION
PLANFORM DIGITAL
is briefly described (see also Willett 1991, this volume) and
LANDFORM MOI L/ /I it is shown to be formally analogous to a viscous rheology.
On the basis of this analogy the Stokes equation for a viscous
PLANE STRAIN fluid is used to solve incrementally for the velocity and strain
DYNAMIC rate for each step of the deformation.
OROGEN
Rigid-plastic rheology
• Orogen Evolution
• Surface Processes and Morphology A rigid-plastic material does not deform until stresses reach
• Topogrophy-Climate Coupling a yield level, at which point it deforms according to a plastic
• Consequences for Tectonics and
P-T-t Paths
flow rule, the stresses never exceeding the yield level. In the
Levy-Mises theory (Malvern 1969), the rigid and plastically
deforming regions are governed by:
governing equations are therefore: of the erodable surface boundary layer and subscript s de-
notes short range slope processes, u^ and h^ may be combined
rifV2v - V/7+pg = 0 (equilibrium) (4) as a single transport coefficient K = u^ h , but it is useful to
distinguish between them, partly because they represent
V- v = 0 (incompressibility) (5) different processes, u^ the ease of material transport once it
has become fragmented and, h^ the steady-state thickness of
the boundary layer in which cohesion has been destroyed by
subject to the boundary conditions. Equations (4) and (5) are weathering. As Carson & Kirkby (1972) explain, h^ is gov-
solved for u in the current configuration which is then updated emed by the rock type, the surface processes at work and the
before the next step of the solution. relative timescales of downslope transport versus weather-
The equilibrium distribution of the pseudo viscosity, ri^, ing. In our model the poorly understood problem of weath-
is found by iteration. A high level of stress, generated by ering is avoided and the thickness of the weathered layer is
assuming ri,, = r\„ everywhere and applying the boundary represented by parameters h for crustal rocks that have not
conditions, is relaxed progressively by 'softening' locations been previously transported and h^ for alluvium created by
where this level of stress exceeds the yield. Both plastic and the model. An improved model will include the processes
viscous regions of the lithosphere can be included in the which determine the spatial dependence of h^.
solution by assigning a fixed 'true' viscosity where appropri- If it is also assumed that there is no tectonic transport of the
ate. This technique is used to examine the effect of a basal surface material, that volume is conserved and that the effects
viscous boundary layer. of solution are negligible, the transport equation can be
Computations were made using a modified form of the combined with the continuity equation.
finite element code ADINA (ADINA 1987).
Typical solutions on a CRAY-XMP require 40s to deform
a crustal scale 30 x 60 element grid of 4-noded elements by dt ^ (7)
400 m.
to give the linear diffusion equation for the rate of change of
local height, the denudation, in response to erosion by the short
Surface transport model range processes.
vary during Af. If neighbour/ is higher than^, the volume of This explicit method of calculation is valid if the change
material transported between the two cells {Qs)ij in time Ar is in height is small by comparison with the overall topography,
the space-time integrated form of equation 6 in which all 2
quantities in capital letters are assumed to be averaged at the a constraint that places an upper bound on Vt«T=c^ lus^s •
cell scale. The effects of 'dry' and 'wet' (that is, water enhanced)
transport can be considered separately in the short-range
transport model. The overall model remains diffusive with
equation 8 modified to
(Qs)ij •"^'^T[JPI,^^' (9)
\>R
dt'' \^ + —)h•'s V2/! (11)
(AH /AL)ij is the average slope between the cell centres. The
factor r;^/2 results from integration ofthe sediment flux along
a representative length ofthe cell boundary. The total length,
where Tj^ is the average rainfall accumulation rate and D^ is
4ci, is shared equally among the eight possible transport
the rainfall rate required to keep the diffusing layer saturated.
directions (Fig. 2). There is no transport upslope to higher cells.
The\)/j / v'ff enhancement term can be interpreted as contributing
The equivalent space-time integration of the continuity
to increase the downslope transport velocity and/or to in-
equation gives the change in height of celly, (D^)y , (Fig. 2)
crease the steady-state value of h .
once all of the (Qs)ij components have been calculated,
The relative rates of 'wet' and 'dry' transport are not well
8 known, therefore this paper does not attempt to distinguish
(10) them but instead uses only one average transport coefficient
(Ds)^ =
for each of crustal rocks and alluvium (Table 1). It should,
however, be remembered that the transport velocity or the
EROSIONAL CONTROL OF ACTIVE COMPRESSIONAL OROGENS
=1
iSw)i= SwiO)-- (16)
ciAt 1 = 1
is the average water flux in the form of water vapour over cell
Figure 3. Illustration of the orographic rainfall model in which the clouds (. The flux Sw (0) can also be considered as the effect of a
diagrammatically represent a long-term integrated precipitation source in the water column, height h^ (0), that is advected over the
form of water vapour that is progressively depleted as it passes over one row
of the cellular topography. The corresponding precipitation distribution is
shown in the diagrammatic graph.
C. BEAUMONT ET AL.
dq dq
v-?. (21)
dt dt I
AH
i^C/V =?;4e,'f 2J
•fjjk (23)
L o n g R a n g e (Fluvial') T r a n s p o r t M o d e l
Long R a n g e (FluviaH Transport M o d e l
Sediment Transport Capacity
Erosion
Deposition
Deposition
-'•^ - {D^l Erosion
Figure 5. Illustration of the difference between the integrated fluvial Figure 6. Illustration of the consequence for erosion or sedimentation of
transport capacity onto and away from a topographic cell. disequilibrium in sediment transport onto and away from a topographic cell.
evaluating (Q,.)y^ in the order of cells ranked by descending of the tectonics and erosional setting of the Southern Alps.
elevation and simultaneously calculating (Qf )jk , (Q/ y^ , While their models differ from our quantitative coupled
and (AQ/ )ji( avoids the need to calculate and store any model, many of the underlying concepts, which were in part
information about the overall pattern ofthe drainage network. developed during discussions with Peter Koons, have the
Calculation of fluvial erosion on cell / (Fig. 5) requires same basis.
summation of the sediment transported onto the cell by the n The plane-strain finite element model (Fig. 8) represents
rivers draining higher cells in the network, a 700 km wide region that is normal to and spans the Pacific-
iQf)lN = ^iQf)nl
(25)
The maximum transport capacity of the river draining/ to k
eqh
(Q )„ Figure 5 illustrates the example where the
7
C. BEAUMONT ET AL.
Alpine
Pacific Plate
Fault
(Deforming)
V E
Shear Zone
Indian P l a t e
25 k m (Rigid)
V = 0
? Pacific P l a t e
Figure 8. Plane-strain finite element model used to represent the tectonics of compressional deformation of central South Island, New Zealand. The grid
has 59 X 31, 4-noded quadrilateral elements.
Indian plate boundary through central South Island. The net oligoclase gneiss, which on the basis of the phase as-
width of the model was chosen to be greater than the plate semblage kyanite-quartz-talc indicates an equilibration depth
boundary zone (Walcott 1984), therefore, the relative plate ~ 25km (Cooper 1980). This depth was therefore taken as the
velocity component normal to the Alpine fault (Fig. 7), depth of the model shear zone. Although this depth remains
presently ~ 2 cm/yr, is absorbed by crustal deformation poorly determined, the choice is consistent with interpreta-
within the modelled region. The corresponding ~ 4cm/yr tions that typical continental lithosphere tends to delaminate
relative plate velocity parallel to the boundary is not included at mid-crustal levels. The choice of model depth can also be
in the plane-strain model. This paper therefore considers only tested a posteriori for consistency with observations because
the effect of compressional deformation on the growth of the the predicted width of the zone of deformation and uplift is
Southern Alps in this example of the coupling of tectonics and related to the assumed geometry and can be compared with
erosion. the width of the Southern Alps.
It is assumed that to the right of the ramp of the Alpine The 2km thick region of the model above the rigid Indian
fault, shown dipping at 45° at the centre of the model (ED, plate (i.e. above region FE) represents crust and cover rocks
Fig. 8), crust of the Pacific plate is progressively decoupled that may thicken and detach to form a thin-skinned fold and
from the underlying lithosphere along a sub-horizontal shear thrust belt when subject to stress transmitted across the Al-
zone, located above the base of the model CD. Crust of the pine fault. This region is included in order to investigate how
Indian plate, in the footwall of the Alpine fault below ED, is deformation might propagate westward with the growth of
not included in the model because it remains rigid. Instead, the Southern Alps were there no erosion. That this region has
the model follows the conceptual interpretation, first pro- not deformed requires explanation.
posed by Wellman (1979) and subsequently by Adams (1980, A plastic-viscous rheology is used to model a frictional
1985), that detached mid and upper Pacific plate crust is crustal rheology with a basal shear zone. When the internal
uptumed and rapidly upthrust along the ramp of the Alpine angle of friction of the Drucker-Prager plastic material is
fault. taken to be 20°, the yield strength in compression, in crust
density p = 2800 kg/m', increases linearly with depth to ap-
proximately 300 MPa at 20km. This yield strength with
Crustal properties
depth relationship approximates a uniform gradient form of
The finite element model includes the Pacific plate crust in Byerlee's Law (Brace & Kohlstedt 1980) in a thrust envi-
and above the basal shear zone and east of the Alpine fault. ronment for a pore pressure ratio. A, = 0.36.
The depth of the model (BC, Fig. 8) was chosen on the The transition from brittle-frictional behaviour to the
assumption that continued deformation and erosion has ex- ductile regime, represented by a linear viscous (Newtonian)
posed rocks from as deep as the basal shear zone at the surface rheology, is assumed to occur in a thin, basal shear zone,
in the hanging wall adjacent to the Alpine fault; above E. The thickness d^ = 250m along DC, of effective viscosity, 7]^ =
highest grade metamorphic rocks from this location are gar- 3x lO'^Pa.s. It could be argued that the model should include
8
EROSIONAL CONTROL OF ACTIVE COMPRESSIONAL OROGENS
an ~ 5 km thick transition zone of decreasing effective This tectonic interpretation implies that the sense of shear
viscosity below the brittle-ductile transition, which may be as will be dextral along FED but sinistral along DC. In the
deep as 20 km in a quartzo-feldspathic crust with a 20°C/km absence of any additional net plate tectonic forces transmit-
geothermal gradient (Ord & Hobbs 1989). This transition ted through the crust as a stress guide, i.e. along BC and AF,
would require more, poorly determined, model parameters it is the opposing basal shear stresses that give an overall
and is itself only inferred on the basis of laboratory data. initial force balance to the model. An alternative kinematic
Instead, the crustal rheology is purposely simplified. The interpretation (Allis 1980,1986) holds that the underthrusting
present model focuses on the interplay between integrated is opposite in sense; that the Indian plate underthrusts the
frictional strength of the crust, specified by (p, and the op- Pacific plate as it does south of New Zealand beneath the
posing strength of the basal shear zone, characterized by the Puyseger trench and probably beneath Fiordland. This choice
shear stress, Xj < T\eV Ids, that it transfers from the boundary of polarity reverses the sense of shear along DC and will,
to the base of the frictionally dominated crust; where everything else being equal, tend to transfer deformation
•u < -u^ = 2cm/yr is the relative velocity across the shear zone eastward away from the Alpine fault. It also requires addi-
and is the specified boundary velocity. The model results are tional forces, probably transmitted as shear from out of the
therefore sensitive to the ratio rjg / ds = 3 x 10'^/250 which plane of the model, to maintain an overall force balance.
determines TJ and not to the independent values of rjg and ds, Although Allis's interpretation cannot be precluded and may
to which no particular importance should be assigned. be cortect for the southern part of South Island we choose the
The shear zone thickness decreases linearly with decreas- simpler one to illustrate our coupled tectonic-erosion model.
ing depth along the ramp DE and is 40 m thick beneath the This model may therefore apply only to the central Southern
region of the compressed finite elements along FE. This re- Alps.
duction in ds could be compensated by a corresponding re-
duction in ri(, but this was not done in order to increase the
Surface processes model and coupling with the tectonic
shear coupling in the colder, near surface, part of the crust. An
model
improved model will consider a frictional base in the shallow
parts of the model. It is assumed for the purposes of this illustrative model that
the current morphology and drainage pattern of the Southern
Alps is not significantly influenced by the ancestral drainage
Boundary conditions
pattern that predated the Late Cenozoic uplift as transpression
The behaviour of the lower crust and mantle below the developed. Model experiments show that this assumption is
modelled region dictates the basal boundary conditions. valid for random height topography of ~ 20 m but the
Simple buoyancy calculations suggest that typical continen- maximum ancestral topography that is consistent with this
tal lithosphere below ~ 25 km is negatively buoyant and will assumption has not yet been determined. The initial surface
subduct if detached from the mid and upper crust. In this model was therefore assumed to be flat and at sea level. Small
it is assumed that the lower part of the Pacific plate has detached random height perturbations were then added to remove the
and, following Adams (1985), that it has passively underthrust degeneracy in the cellular grid heights and to create a random
the Indian lithosphere to the west in a similar way to the drainage pattern.
oceanic lithosphere further north except that the process is The model is progressively stepped in a coupled way,
aseismic. This interpretation requires that the basal horizon- altemately calculating the tectonic velocity and displacements,
tal velocity along DC be equal to the Pacific plate velocity and followed by the effect of the two components of erosion
equal to the uniform horizontal velocity along the boundary acting on the curtent topography. Specifically, the uplift
BC. These velocities are both taken to be Vo = -2cm/yr and the distribution normal to strike, M„ (JC), predicted by the tectonic
boundary on the Indian plate along APED is fixed with model for timestep n, length \T, is divided into m equal
-u = 0. The model therefore considers compression of the substeps, u'^ {x), length A?, for the surface processes calcu-
Pacific plate crust as it is 'pushed' by the net normal force lation. The surface transport was then calculated over the m
along BC and is 'dragged' by the shear, T, , that develops along substeps, each step involving uplift of the topography by an
DC. This interpretation is entirely equivalent to that which increment «„ (x) /m, such that uplift varies in the cross-strike
takes \) along BC to be zero and considers the indentation of direction but is uniform along strike. The substeps are
the Pacific plate by the rigid Indian plate, Vo = 2 cm/yr, which required to ensure that fluxes for each step are small by
acts as a tectonic wedge in the manner described by Price comparison with the total topography. The precipitation
(1986). The former boundary conditions are preferted only distribution was calculated separately for each substep because
because the location of the Alpine fault remains fixed in the the topography changes with substep.
model. It may actually be rotating in an absolute reference Feedback coupling between the surface processes model
frame (Allis 1986). and the tectonics is calculated prior to the next step,«-i-1, of
Points along boundaries AF and BC are free to move the tectonic model. The current average along-strike topog-
vertically but this is not important because they are far enough raphy from the surface model defines the new surface for the
removed from the plate boundary that deformation does not tectonic model. The tectonic model, therefore, incorporates
propagate to them. The upper boundary of the model is a free the changing average mass of the growing mountains. Their
surface on which the shear stress is zero. weight influences the gravitational component of the subsur-
C. BEAUMONT ETAL.
Model 1 N o Erosion
NW
d=0 SE
\
t =0
_L J I I L
LI H ' :: ^ - -- - -
-^ ^ ^ ; = = I
^= ; = 11 ^ =
V j—j ^
4 km
0.2 Myr.
2 cm/yr -^
J_ J \ I L
" ^ ill ^ ^ X ^ ^ ^ ^ - ^ ~ ~
^^ X ::: ^ ^ ^ - _ __ _<
^l:::: ; :
d — 16 km
t ~ 0.8 Myr.
I J L
5 km - I
20 km
Figure 9. Model 1 results. Evolution of the plane-strain model in the absence of erosion. Velocity vectors are relative to a fixed Indian plate and show
the progressive transformation of horizontal to vertical motion in the deforming plate boundary. Thrusting over the ramp of the Alpine fault, which acts as
a finite height 'backstop', produces a westerly vergent tectonic wedge. Similarly, easterly vergent 'backthrusting' creates a second outward facing tectonic
wedge on the eastern flank of the orogen. Bold horizontal arrows show the total amount of shortening, d.
face stress and ultimately modifies the pattern of deformation no climate coupling or erosion is first examined and then two
and uplift during the n -i-1 'th tectonic step by comparison with cases where denudation is dominated by fluvial transport
the n 'th step. with orographic rainfall derived from either a western or
eastern source are considered.
Model results
Model I: No erosion
The three models discussed here are only preliminary results The geometry and velocity distribution of model 1 are shown
and were chosen to illustrate how climate coupling and (Fig. 9) at the onset of deformation and again after - 0.2 and
denudation may control the evolution of the Southern Alps ~ 0.8 Ma, at which time the overall shortening is 4 and 16 km.
for one particular interpretation of the tectonic style and The results illustrate that the model does respond by upturning
properties of the orogen. The tectonic properties of all three the crust and thrusting it along the ramp of the Alpine fault.
models (Table 1) are identical. The evolution when there is The initial velocity field comprises a region of passive hori-
10
EROSIONAL CONTROL O F ACTIVE COMPRESSIONAL OROGENS
Model 1 N o Erosion
d = 4 km
t ~ 0.2 Myr.
- •=•^s^tlff-f-^^- ^rf~*-
/ /
/ /
/ o^
/
/ / /
/ / /
/ /
/ /
/ /
/ /
•20 k m /
Figure 10. The same model result as shown in Figure 9 (middle panel) but with velocity vectors shown relative to a fixed Pacific plate. The easterly vergent
tectonic wedge caused by backthrusting is seen clearly in this reference frame.
zontal transport of the Pacific plate crust without deformation a 60 km wide zone normal to strike. Although some of this
and a region of uplift and deformation in a triangular wedge excess mass almost certainly resides in thickened crust lo-
or 'plug' bounded by the ramp and its reflection about a cated to the east of the Alpine fault (Woodward 1979; AUis
vertical axis through D (Fig. 8). The initial width of the 1986), a significant proportion has apparently been eroded.
uplifting region is therefore approximately twice its depth Note that the model must be improved to include isostasy
when the dip of the ramp is 45°. before firm conclusions can be made. Further evidence of
The initial style of deformation resembles the first of erosional control of structure is the absence, except for a few
Wellman's (1979, Fig. 4) conceptual models of possible nappes, of a thin-skinned fold and thrust belt west of the Alpine
faulting patterns for the Southern Alps. His concern that this Fault. The model indicates that this westward propagating
style would lead to an unrealistically steep eastern margin for wedge would develop rapidly and achieve a significant size
the Southern Alps was unfounded as is seen from the topog- after only 16 km of shortening. It would be a fully fledged
raphy after finite deformation (Fig. 9, middle panel). fold and thrust belt after 50 km of shortening.
The central region of the orogen continues to uplift in a
relatively uniform manner, while the leading (western) and
Model 2: wind from the west
trailing (eastern) zones take the form of outwardly propagat-
ing tectonic wedges which increase in size with the uplift. The second model (Figs 11 & 12) includes the effects of
The taper of the wedges depends on the internal frictional surface erosion by slope processes and a fluvial system
strength of the model crust, the basal dip, and the shear stress recharged with orographic precipitation from a western source.
applied to the base of the frictional crust by the viscous shear Model parameter values are given in Table 1. The evolution
zone. The results differ from the critical Coulomb wedge of the tectonic model (Fig. 11) shows the strike average re-
models (e.g. Dahlen 1984) because the model considers the sponse as the surface uplifts and the orographic precipitation
development of a two-sided orogen and because the base of develops. The volume of sediment transported by denudation
the model is viscous not frictional. The equilibrium topog- equals the rate of tectonic influx at - 0.5 Ma, d = 10 km of
raphy therefore depends on the convergence velocity, in ad- shortening, after which the orogen remains on average in a
dition to the current geometry. The results have properties in dynamically near steady state as discussed below.
common with Koons' (1990) conceptual models but differ Erosion on the western flank of the mountains dominates
because no significant basal decollement develops for this because the precipitation is greatest on that side and because
amount of shortening. The importance of the eastward the boundary condition on the surface transport model rep-
propagating wedge becomes clear when velocities are plotted resents the Tasman Sea {x - 332km, Fig. 9) as an infinite sink
with respect to the Indian plate (Fig. 10). for eroded sediment. This choice is appropriate because most
It can be seen that only a small amount of shortening, of the sediment that enters the Tasman is transported to the
< 6 km, is required to create a mountain system with the same south. The boundary condition on the east is the same, while
average height and width as the present Southern Alps. the north and south boundaries are reflective.
Shortening by a more realistic amount, 50 km for example, The strike-averaged precipitation distribution has the same
would have produced a much larger orogen by the addition of character as that shown by Griffiths & McSaveney (1983,
approximately 1250 km^ per unit length of crust to the South- their Fig. 2; see also Whitehouse 1988) a maximum to the
em Alps orogen, enough to thicken the crust by 20 km over west of the main divide but some precipitation on the eastern
11
C. BEAUMONT EJ AL.
d = 16 km
t ~ 0.8 Myr. ^-"
i i"~i i i n 1
5 km ''I U ^ - ^ = = =
-« 20 km —•
4:r = - = - =
Figure 11. Model 2 results. Evolution of the plane-strain model when orographic precipitation is derived from a westerly source. Velocity vectors are relative
to afixedIndian plate. The dotted lines show the projected position of the initial surface. The westerly vergent tectonic wedge of Figure 9 has been suppressed
by erosion. The western sink for mass also stunts the growth of the orogen as a whole and a near steady state is reached with only a small easterly vergent
tectonic wedge. It is clear that erosion has modified structural and metamorphic style throughout the model orogen, not just locally. Bold horizontal arrows
show the total amount of shortening, d.
flank. Agreement of the model precipitation with short-term surface projected by its cumulative vertical displacement.
observations is not independent confirmation that the model The distance between the model surface and the dashed line
is correct because the timescales are totally different. The is, therefore, a measure of the total erosion in the spatially
model rainfall must represent processes at > 10^ yrs. fixed model coordinate system. It would equal the exhumation
The dotted lines (Fig. 11) show the position of the original experienced by rocks currently at the surface were the uplift
12
EROSIONAL CONTROL OF ACTIVE COMPRESSIONAL OROGENS
H-llllllllllQ IIIIIIMIItH
Figure 12. Map of Model 2 topography at 0.5 Ma after 10 km of shortening. The contrast in the drainage pattern and morphology east and west of the divide
is also characteristic of the central Southern Alps. The contour interval is 600 m, heights above 3600 m are stippled, and the area is 104 x 72 km. The cell
size for the calculations was 2 x 2 km. Only the major rivers are shown.
vertical. However, the horizontal velocity advects the uplift- tectonics of ancient orogens on the basis of distribution of
ing crust progressively to the west and rocks reach the surface metamorphic grade. The metamorphic grade is, however, a
horizontally offset from their original position. Consequently, measure of the integrated erosion in the tectonic 'upstream'
once steady state has been achieved, the highest grade meta- direction and the horizontal gradient of exhumation is the
morphic rocks, those exhumed from the greatest depth, will quantity most closely related to the spatial distribution of
reach the surface of the model adjacent to the Alpine Fault. surface erosion.
This result agrees with the observed distribution of meta- The contour map of the topography (Fig. 12) for model 2
morphic grade, which is a maximum at the Alpine fault and after 0.5 Ma has a number of features that can be related to
decreases to the east. It also follows that rocks which have the topography of central South Island in the area shown in
undergone the greatest exhumation do not reach the surface Figure 7. It is important to remember that the morphology
where the erosion rate is greatest even in a steady-state orogen. and river system distribution has been self-selected by the
This is an important point to remember when interpreting the model as it evolves from the initial low amplitude random
13
C. B E A U M O N T £7 AL.
d = 4 km \
\
rf=10km \ •••••••••
i ^ ^ X^ ^ - - - = =
^^ V:: N N . it;—3
A^::::: =
\ .'
d = 16 k m \ •' ^.^ , '••-.
t ~ 0.8 Myr. M I I ' I T I " ^ " ^ "N -^-^-^-^-^^--=_ ^
Figure 13. Model 3 results. Evolution of the plane-strain model when orographic precipitation is derived from an easterly source. The source has the same
water vapour flux at its eastern boundary as Model 2 had at its western boundary. Velocity vectors are relative to a fixed Indian plate. The dotted lines show
the projected position of the initial surface. Bold horizontal arrows show the total amount of shortening, d.
topography. The topography therefore contains information created an along-strike ridge-valley system. The ridges have
on the dynamic interaction of uplift with the surface proc- steep noses, lower gradient centres and rise again to the east,
esses which will be analyzed in a future paper. where they climb to the divide. The valleys have strongly
West of the drainage divide, the front of the model orogen concave profiles. These are the same features that Koons
is cut by linear streams, spaced from 6-12 km apart, that have (1989) recognized as characteristic of the western face of the
14
EROSIONAL CONTROL OF ACTIVE COMPRESSIONAL OROGENS
Southern Alps and reproduced with a planform diffusive Model 3: wind from the east
mass transport model. In this model he specified the river
The third model (Figs 13 & 14) is identical to model 2 except
spacing and valley topography. Here the river spacing is self-
that the precipitation is now derived from an eastern source,
selected, and need not be imposed, when fluvial transport is
which is exactly opposite to the prevailing conditions in the
included in the model. The spacing of these small rivers,
Southern Alps. Model 3 is therefore a hypothetical example
which is similar to that of the central Southern Alps, results
designed to indicate how the Southern Alps may have devel-
from the competition between hillslope and fluvial processes
oped with reversed symmetry in the climate.
during uplift. Initially, all of the one hundred rivers equally
As the surface uplifts and orographic rainfall develops, the
drain and erode the western face of the model orogen, but an
eastern flank of the orogen is subject to the greatest precipi-
instability occurs with increasing slope and fluvial discharge
tation and erosion. Erosion on the tectonic divide and on the
which leads to finite incision of rivers with the spacing shown
west flank of the orogen is therefore reduced by comparison
in Figure 12. This instability may be related to slope con-
with model 2. Consequently, the western flank of the orogen
cavity (Smith & Bretherton 1972) but our model is compli-
continues to uplift and a major range of peaks is created which
cated by the additional effects of diffusion and erosion/
sit astride the divide. All of these peaks exceed the height of
deposition kinetics. At least one aspect of the river selection,
Mount Cook within 0.5 Ma of the onset of compression (Fig.
growth and incision requires that when a cell height is per-
14). They continue to grow and begin to assume Himalayan
turbed, the change in AQy from the cell be larger than the
stature after approximately 1 Ma of convergence.
corresponding change in Q^ onto the cell. The same compe-
tition may explain why the main rivers on the eastern slope of River incision of the western flank is less well developed
the model orogen are more widely spaced than their west- (Fig. 14) than in model 2 (Fig. 12) because the fluvial
ward-drainage counterparts. The eastward-draining rivers discharge is smaller. The instability that leads to incision is
have smaller equilibrium sediment carrying capacities for therefore delayed until the slopes are steep enough for the
equivalent sized watersheds. Therefore, only rivers that rivers to achieve sufficient power. In Figure 14, the instabil-
acquire large discharge by developing large watersheds can ity has not fully developed along the entire western flank. The
achieve the same degree of incision as their western equiva- steepness of the western face is, however, preserved by the
lents. flux boundary along the western side which mimics the
effects of longshore transport in the Tasman Sea.
The main peaks that resemble Mount Cook are located on
Creation of Himalayan-size mountains occurs with in-
or close to the drainage divide (Fig. 12). The true counterparts
creasing speed as the extraction efficiency of the precipitation
of the model ridges east of the divide and normal to it may be
is increased. For example, when an is increased by a factor
the southerly trending ridges of the Southern Alps. It is
of three, precipitation is nearly exhausted east of the drainage
suggested that the distributed plate-boundary shear strain has
divide which now develops only 10 km from the model
rotated what were initially ridges trending perpendicular to
coastline (Fig. 15). The divide is crowned by 8000 m peaks
the orogen and propose to test this hypothesis with a complete
within 0.8 Ma of the onset of compression.
planform model of thin-sheet tectonics coupled with erosion
(Ellis et al. in press). Correspondingly, as OR increases in models with an east-
em precipitation source, the implied belt of highest grade
The near steady-state nature of the model orogen (Fig. 11)
metamorphic rocks migrates eastward away from its position
for t > 0.5 Ma was judged qualitatively from the change in
against the Alpine fault and may even cross the drainage
strike-averaged topography by comparison with the same
divide. With declining precipitation on the western flank,
measure of the erosion between t = 0.5 Ma and t = 0.8 Ma
neither erosion nor horizontal advection are sufficient to ex-
(Fig. 11). The results indicate that the volume or mass of the
pose deeply buried rock against the model Alpine fault. The
orogen is approximately in dynamic equilibrium. Individual
influence of erosion patterns on exhumation raises the possi-
topographic features are not, however, in equilibrium. For
bility that the present location of the Alpine fault is partly
example, the westward draining rivers become longer with
controlled by the surface processes.
increasing time.
The present model cannot provide a quantitative assess-
ment of the morphological evolution because the topography
is neither advected nor strained by the tectonic deformation Conclusions
in the horizontal plane. The results are therefore no longer
Both the model and the results as preliminary and only
valid when the effects of horizontal advection become sig-
general conclusions are valid. The results do, however,
nificant. For example, the rate of advection of the topography
indicate that in instances where the surface mass transport
toward the Alpine fault is almost equal to the predicted rate
occurs over distances and at rates comparable to the tectonic
of increase in the length of the westward-draining rivers.
mass flux, orogenic evolution may be significantly modified
Once advection of the topography has been added to the
by climatic conditions at the scale of the orogen. One
model, it will be possible to assess whether a statistically
implication is that because precipitation is often asymmetric
steady-state topography can exist and track the birth, growth
across an orogen, given sufficient time, the orogen may adopt
and death of the Mount Cook-like mountains as they are
the same asymmetry in its tectonic style.
advected across the model.
Proof of climate influence on the tectonics of ancient
15
C . BEAUMONT £T/iL.
Figure 14. Map of Model 3 topography at 0.5Myr after 10 km of shortening. The contour interval is 600 m, heights above 3600 m are stippled, and the
area is 104 x 72 km. The cell size for the calculations was 2 x 2 km. Only the major rivers are shown.
orogens will necessarily be based on circumstantial evidence. of the major characteristics of the southern Alps, then com-
South Island, New Zealand, however, offers a dramatic parison with models 1 and 3 illustrates the significance of the
contemporary example of these process at work. Control of coupling between the atmosphere and the solid earth.
this orogen is shared between plate tectonic and surface Timescales of 0.5 - 5.0 Ma may be sufficient for profound
processes. The dynamic linkage between tectonics and climatic effects in collisional orogens.
erosion is provided by the atmosphere. Prevailing climatic Not all orogens are, or were, as strongly modified by
conditions ensure a sufficient source of orographic rainfall climate coupling as the Southern Alps - an extreme example
which, when coupled with the tectonically induced steep chosen to demonstrate this thesis. It is possible to suggest
western slopes, locks the orogen in a dynamically near that, if the Himalayan orogen was subject to intervals of
steady-state mass balance. steady-state evolution, these may have been responsible for
These models are only first-order approximations of the the exhumation of the belt of highest grade metamorphic
evolution of orogens in general, and the Southern Alps, in rocks.
particular. If it is accepted that model 2 does reproduce some
16
EROSIONAL CONTROL OF ACTIVE COMPRESSIONAL OROGENS
of = 16 km \
t ~ 0.8 Myr.
5 km
•20 km
Figure IS. Model 3 results. Evolution of the plane-strain model when orographic precipitation is derived from an easterly source. The model is identical
with the one shown in Figure 13 (bottom panel) except that the rainfall extraction efficiency has been increased by a factor of 3.
This work benefited greatly from discussions with Sean Willett, Peter Koons of the manuscript. The research has been funded by the Natural Sciences and
and Becky Jamieson. We would also like to thank Clem Chase for providing Engineering Research Council of Canada through Operating and Lithoprobe
us with information on his precipitation landsculpting model prior to Supporting Geoscience grants to C. Beaumont.
publication and for his and Peter Koons' comments on a preliminary version
References
Adams, J. 1985. Largescaletectonicgeomorphology of the Southern Alps, Dahlen, F. A. & Barr, T. D. 1989. Brittle frictional mountain building 1,
New Zealand. In: Morisawa, M. and Hack, J. T. (eds). Tectonic deformation and mechanical energy budget. Journal of Geophysical
Geomorphology, Allen and Unwin, 105-28. Research, 94, 3906-22.
1980. Contemporary uplift and erosion of the Southern Alps, New 1984. Noncohesive critical Coulomb wedges: An exact solution.
Zealand. Geological Society of America Bulletin, Part 11,91, 1-114. Journal ofGeophysical Research, 89, 10125-33.
Adina. 1987. ADINA - a finite element program for automatic dynamic Davis, D., Suppe, J. & Dahlen, F. A. 1983. Mechanics of fold-and-thrust
incremental nonlinear analysis. Report ARD 87-1, ADINA R andD Inc. belts and accretionary wedges. Journal of Geophysical Research, 88,
AUis, R. G. 1986. Mode of crustal shortening adjacent to the Alpine Fault, 1153-72.
New Zealand. Tectonics, 5, 15-32. Ellis, S., Fullsack, P. & Beaumont, C. 1990. Incorporation of erosion into
1981. Continental underthrusting beneath the Southern Alps of New thin sheet numerical models of continental collision, EOS (in press).
Zealand. Geology, 9, 303-7. England, P. C. & McKenzie, D. P. 1982. A thin viscous sheet model for
Armstrong, A. C. 1980. Soils and slopes in a humid environment. Catena, continental deformation. Geophysical Journal of the Royal Astronomi-
1, 327-38. cal Society, 70, 295-321.
Barr, T. D. & Dahlen, F. A. 1989. Brittle frictional mountain building 2, thermal Flemings, P. & Jordan, T. E. 1989. A synthetic stratigraphic model of
structure and heat budget. Journal of Geophysical Research, 94, 3923- foreland basin development. Journal ofGeophysical Research, 94,3851 -
48. 66.
Begin, S. B., Meyer, D. F. & Schumm, S. A. 1981. Development of lon- Griffiths, G. A. & McSaveney, M. J. 1983. Distribution of mean annual
gitudinal profiles of alluvial channels in response to base-level lowering. precipitation across some steepland regions of New Zealand. New
Earth Surface Processes andLandfotms, 6, 49-68. Zealand Journal of Science, 26, 197-209
Brace, W. F. & Kohlstedt, D. L. 1980. Limits on lithospheric stress imposed Jamieson, R. A. & Beaumont, C. 1989. Deformation and metamorphism in
by laboratory experiments. Journal ofGeophysical Research, 85,6248- convergent orogens: a model for uplift and exhumation of metamorphic
52. terrains. In: Daly, J.S., Cliff, R.A. & Yardley, B.W.D. (eds) Evolution
Carson, M. A. & Kirkby, M. J. 1972. Hillslope Form and Processes, of Metamorphic Belts, Geological Society of London Special Publica-
Cambridge University Press, 475p. tion, 43, 117-29.
Chappie, W. A. 1978. Mechanics of thin-skinned fold and thrust beUs. & 1988. Orogeny and metamorphism: a model for deformation
Geological Society of America Bulletin, 84, 1189-98. and pressure-temperature-time paths with application to the central and
Chase, C. G. 1989. Fluvial landsculpting: why topography is like a fractal. southern Appalachians. Tectonics, 7, 417-45.
Geological Society of America, Abstracts with Programs, A 38. Koons, P. O. 1989. The topographic evolution of coUisional mountain belts:
1988. Fluvial landsculpting and the fractal dimension of topography. a numerical look at the Southern Alps, New Zealand. American Journal
EOS, 69, 1207. of Science, 289, 1041-1069.
Coleman, S. M. & Watson, K. 1983. Ages estimated from a diffusive 1990. The two-sided orogen: collision and erosion from the sand box
equation model for scarp degradation. Science, 221, 263-5. to the Southern Alps, New Zealand. Geology, 18, 679-682.
Cooper, A. F. 1980. Retrograde alteration of chromian kyanite in metachert Malvern, L.E. 1969. Introduction to the Mechanics of a Continuous
and amphibolite whiteschist from the Southern Alps, New Zealand, with Medium. Prentice-Hall, New Jersey.
implications for uplift on the Alpine Fault. Contributions to Mineralogy Norris, R. J., Koons, P. O. & Cooper, A. F. 1990. The obliquely-convergent
and Petrology, IS, 153-64. plate boundary in the South Island of New Zealand: implications for
Culling, W.E.H. 1965. Theory of erosion on soil-covered slopes. Journal ancient collision zones. Journal of Structural Geology, 12, 715-725.
of Geology, 73, 230-54. Ord, A. & Hobbs, B. E. 1989. The strength of the continental crust, de-
1960. Analytical theory of erosion. Journal of Geology,6S,336-44. tachment zones and the development of plastic instabilities.
Tectonophysics, 158, 269-89.
17
C. BEAUMONT ET AL.
Smith.T. R.&Bretherton,F. p. 1972. Stability and the conservation of mass Whitehouse, I. E. 1988. Geomorphology of the central Southern Alps, New
in drainage basin evolution. Water Resources Research, i, 1506-1529. Zealand: the interaction of plate collision and atmospheric circulation.
Stockmal.G.S. 1983. Modelingoflarge-scaleaccretionary wedgedeforma- Zeilschrift Geomorphologie Supplement Bd-69. 105-16.
tion. Journal of Geophysical Research, 88, 8271-87. 1987. Geomorphology of a compressional plate boundary. Southern
Vilotte, J. P., Madariaga, R., Daignieres, M. & Zienkiewicz, O. 1986. Nu- Alps, New Zealand./«: Gardiner, V.(ed.)//)/£';7?ono/j(3/Geomo)p/io/og>'
merical study of continental collision: influence of buoyancy forces and 1986, 897-923.
an initial stiff inclusion. GeophysicalJournal of the Royal Astronomical Woodward, D. J. 1979. The crustal structure of the Southern Alps, New
Society, 84, 279-310. Zealand, as determined by gravity. In: Walcott, R. I. and Cresswell, M.
Walcott, R. I. 1984. The kinematics of the plate boundary zone through New M. (eds) The Origin of the Southern Alps. Royal Society of New Zealand
Zealand: acomparisonofshort-andlong-termdeformations. Geophysical Bulletin, 18, 95-8.
Journal of the Royal Astronomical Society, 79, 613-33.
Wellman, H. W. 1979. An uplift map for the South Island of New Zealand,
and a model for uplift of the Southern Alps. In: Walcott, R. I. & Cresswell,
M. M. (eds) The Origin of the Southern Alps, Royal Society of New
Zealand Bulletin, 18, 13-20.
18
Dynamic and kinematic growth and change of a Coulomb wedge
Sean D. Willett
Abstract: Defomiation and structural relationships in accretionary prisms and fold and thrust belts
are the result of dynamic changes in the size, geometry, or strength of the deforming wedge and
its boundary conditions. The concepts of critical slope or taper that have been successful in
explaining the static geometry and state of stress in a Coulomb wedge can be expanded through
the use of finite element models to consider the kinematics and dynamics of a deforming Coulomb
wedge. The numerical technique adopts a Coulomb failure criterion and isotropic plastic flow in
a velocity-based Eulerian formulation. This formulation allows for very large deformation to be
accommodated by a numerical mesh that remains fixed in space, deforming only to follow the
movement of the upper surface.
Critical wedge theory defines deformational domains bounded by the critical wedge solutions.
Imposed changes in boundary conditions or geometry can move the mechanical state of a wedge
off a critical line into either the sub-critical or stable domain, in which a wedge is unstable during
accretion, leading to transient deformation as the wedge adjusts to a new critical geometry.
With steady boundary conditions the accretion process leads to self-similar growth. A zone of high
strain rate representing the frontal step-up thrust and the decollement develops and separates
underthrust sediment from the deforming wedge. An increase in basal strength produces large
internal deformation as the wedge increases its taper. A decrease in basal strength concentrates
deformation at the toe of the wedge. A large decrease in basal strength may lead to extensional
collapse. A complex geological history involving repeated cycles of growth and collapse could
produce tectonic exhumation of the deeply buried interior of the wedge, even in the absence of
erosion.
In compressional tectonic regimes much of the convergence, (1983) argued that a Coulomb yield stress was more appropri-
at least at upper crustal depths, is manifested by shortening ate for thrust belts than the depth-independent von Mises
and thickening of a fold and thrust belt. However, at less than yield criterion used by Chappie (1978) and Stockmal (1983).
crustal scale, styles of deformation and the kinematics of the They presented solutions for the state of stress in both non-
development of a fold and thrust belt can vary greatly. The cohesive (Davis ef a/. 1983)andcohesive(Dahlenffa/. 1984)
classic model of a forward propagating sequence of thrust Coulomb thrust belts based on approximately wedge shaped
faults remains an important component of most or all thrust cross-sectional geometries, modified to include the effects of
belts and plate margin accretionary wedges, but other large pore fluid pressures. Subsequent work addressed changes in
scale deformational processes including out-of-sequence cohesion associated with compaction and lithification (Zhao
faulting (Morley 1988), sediment underplating (Moore et al. et al. 1986). Dahlen (1984) presented an exact analytic
1982; Piatt et al. 1985; Piatt 1986, 1987; Westbrook et al. solution for the state of stress in a non-cohesive Coulomb
1988) and even extension (Piatt 1986) have been recognized wedge. This Coulomb wedge theory has been widely applied
as contributing to the observed structure of thrust belts or as it provides a simple explanation for many features of
compressional orogens as a whole. Along with a recognition accretionary wedges and fold and thrust belts.
of the complexity of the large scale structural and temporal The other class of continuum models was designed to
relationships in thrust belts has been an acceptance of the fact address the kinematics of the deformation internal to a thrust
that thrust belts, and even entire orogens, behave as a me- belt or convergent orogen. A dynamic model of flow of a
chanical entity. This has led to the development of a number linear viscous fluid inside a rigid comer was proposed by
of continuum models for the mechanics of thrust belts. Cowan & Silling (1978) to explain scale models and exhuma-
Continuum models have generally addressed one of two tion in convergent settings. This comer flow model was
aspects of thrust belt mechanics; (1) the geometry and state of developed further by Cloos (1982, 1984) who proposed
stress within the thrust belt; or (2) the kinematics of material viscous flow as a mechanism for emplacing exotic blocks in
transport through the thrust belt. Chappie (1978) presented a melange terranes, and by Shi & Wang (1988) who calculated
model for the state of stress in a thrust belt assuming it acted pressure-temperature paths in the Barbados accretionary
as a plastic material with the stress everywhere at the yield prism. Cloos & Shreve (1988a, 1988b) developed a similar
stress. Stockmal (1983) used slip line theory to expand this model for viscous flow in a confined channel under an
model and to calculate both stress and internal velocities accretionary wedge. These linear viscous models are attrac-
based on the thrust belt geometry and assumptions regarding tive because, in contrast to the plasticity solutions, particle
the slip velocity on the underthrusting plate. Davis et al. velocities and flow paths are easy to calculate. However,
19
S. D . WiLLETT
20
GROWTH AND CHANGE OF A COLOUMB WEDGE
21
S. D . WiLLETT
ing important increases in efficiency. The cost of the numeri- The second invariants of stress, J',, and rate of deforma-
cal efficiency is in the loss of elastic strains and displacements. tion, IJ, defined by equations (8) and (10), respectively, are L,
One such rate formulation is the Levy-Mises theory (Mal- norms. By taking the norm ofboth sides of equation (11) and
vern 1969) that describes a rigid-plastic material. Finite assuming the material is incompressible, we obtain these
element techniques based on this theory have been developed invariants such that stress is independent of the rate of
for engineering applications in metal forming (Zienkiewicz deformation and equation (11) becomes simply the von
& Godbole 1974) and for geodynamic models of lithosphere Mises yield criterion (eqn. 7). The constitutive law defined
deformation (Vilotte etal. 1982, 1984, 1986). In the Levy- by equation (11) relates deviatoric stress to rate of deforma-
Mises theory an increment of plastic strain or, equivalently, tion in the forni of a viscous flow constitutive law.
the rate of deformation, is assumed to be proportional to the
Oli He Oil (12)
deviatoric stress tensor(o'ij).
with an effective, non-linear viscosity, \IQ, defined by.
(5)
V. = ^ (13)
yfh
where D is the rate of deformation tensor, defined in terms
of the velocity components v as.
The constitutive equation must include the isotropic part
of the stress tensor as well as the deviatoric stress which is in
3v. 3v. equation (11). By including the pressure (or mean stress: P =
D . . = 7 9x. (6) Oii/3) it is possible to obtain the complete constitutive equa-
3x.
tion for a viscous fluid.
22
GROWTH AND CHANGE OF A COLOUMB WEDGE
23
S. D . WlLLETT
include the rigid basal boundary which is fixed with no dip (P Constant basal strength model ,' • i', :u n ^
= 0°) and the backstop which is vertical in all cases. The
velocities imposed on the base and backstop are constant in The strength of the basal decollement is the principal param-
space and time. The backstop has no condition imposed on eter in these models. By holding the basal strength constant
the tangential component of stress or velocity (i.e. it is with time, a wedge is produced that is always critical and lies
frictionless), so it can be interpreted as a symmetry condition. somewhere on line 1 in Figure 2. The critical wedge with the
Other parameters that are held constant include the thickness maximum slope for a given basal dip (P) and wedge strength
of the incoming sedimentary section (4 km) and the strength (0) is obtained when the basal strength is equal to the wedge
of the wedge interior (internal angle of friction, 0 = 30°). interior strength (0 b "= 0' Fig- 2). A model of this case is
Although the rale of convergence is specified along the shown in Figure 4. The geometry and instantaneous deforma-
base of the wedge and controls the rate of deformation and tion rates are shown with increasing time, or, equivalently,
growth of the wedge, it is not a dimensional parameter. There increasing convergence. The wedge that develops against the
is no time dependence to plasticity, so even though time backstop shows the self-similar growth characteristic of
appears in the equations, its use is only to calculate the Coulomb wedges. The upper surface develops a characteris-
amount of displacement. Time and the convergence velocity tic slope that is maintained as the wedge grows. In contrast to
scale each other such that only the product, the total conver- the analytic theory (Dahlen 1984) the upper surface is not
gence, is important to the deformation. In other words, a linear, but has some curvature to it. This curvature is most
decrease in convergence velocity is equivalent to a propor- likely a response to the vertical, frictionless backstop, which
tional increase in time. Rate of deformation (strain rate) also is not present in the analytic theory. The shear stresses on a
scales with time and convergence velocity so that only the frictionless surface must be zero which requires the upper
relative deformation rates are important. . .• surface to be approximately normal to the backstop. The
No VE
10 km
Figure 4. Model of growing Coulomb wedge with basal strength equal to wedge interior strength (0^, = 0 = 30°). Nodal velocities relative to the overriding
plate are shown as vectors. Relative strain rate (second invariant of rate of deformation) contoured with larger relative values shaded. Results are shown for
total convergence of: (a) 3 km; (b) 12 km; (c) 24 km.
24
GROWTH AND CHANGE OF A COLOUMB WEDGE
No VE
10 km
R c l a t l v a S t r a i n Rata
Figure 5. Model of growing Coulomb wedge with basal strength less than wedge interior strength (^ - 15°; <p = 30°). Nodal velocities relative to the
overriding plate are shown as vectors. Relative strain rate (second invariant of rate of deformation) contoured with larger relative values shaded. Results are
shown for total convergence of: (a) 3 km; (b) 22 km; (c) 35 km.
upper surface is also somewhat steeper than predicted by the frontal accretion mechanism requires significant pure shear
analytic theory, but this discrepancy is probably also due to deformation of the wedge interior in order to maintain the
the effect of the backstop. Aside from these differences, the critical taper. This internal deformation is not observed in the
geometry and stress field is in good agreement with the model since sediment deformation is localized to the transi-
analytic solution. tion between undeformed sediment and the wedge interior.
The kinematics of deformation corresponding to this stress Growth of the back of the wedge is accomplished by
field are shown as the instantaneous velocity field. The 'subducting' undeformed sediment under the wedge and
velocities show the undeformed sediments moving with accreting it directly to the base, a pattern that is similar to the
constant velocity equal to the convergence velocity until they deformation style described by Brandon (1984) and Silver et
encounter the toe of the wedge (Fig. 4). Sediments near the a/. (1985).
surface are immediately deformed as the material turns up If the base of the wedge is significantly weaker than the
into the wedge. Sediments that are progressively deeper in wedge interior adistinctly different kinematic pattern emerges
the section pass further under the wedge before being pulled (Fig. 5). As predicted by the critical wedge theory a weaker
into the deformed wedge. Once into the wedge, material base implies a lower surface slope. The wedge again grows
moves nearly vertically upward and experiences little addi- in a self-similar fashion, but with a much smaller taper angle.
tional deformation. This velocity field produces a distinctive Changes in the kinematic pattern are more subtle. The
pattern of strain rate with the highest strain rates found in a maximum strain rales again extend from the toe of the wedge
band from the toe of the wedge back to the comer at the base along the lower boundary to the comer at the base of the
of the backstop. These high strain rates correspond to the backstop. However, the zone of high strain rates is much
zone where sediment passes from the underthrust plate into wider, extending far back into the wedge. The velocities do
the wedge. It is difficult to interpret continuum velocity fields not show the sharp transition from horizontal to vertical
in terms of geological structures and the kinematics associ- apparent in Figure 4. A particle path in the model of Figure
ated with discrete faults. However, the kinematics of the 5 changes slowly from horizontal towards vertical, but may
model in Figure 4 do resemble the displacements found in the never reach vertical. There is also no zone of undeformed,
scale models of Cowan & Silling (1978). underthrust sediments. This overall kinematic style is sug-
The kinematic style is also suggestive of a mechanism that gestive of a frontal accretion mechanism with sediment
could be interpreted as sediment underplating. The self- accreted at the toe of the wedge and pure shear shortening
similar growth necessary to maintain the critical taper is maintaining the critical taper.
achieved with very little deformation of the wedge interior. A The models shown in Figures 4 & 5 are examples of
25
S. D . WiLLETT
No VE
10 km
R«Ja(lv«
Figure 6. Model of Coulomb wedge experiencing an increase in basal strength, (a) Iniiial critical wedge develops wiih basal strength (jy - 15°. Total
convergence of 38 km. (b) Basal strength increases from 0^= 15° to 0^,=30° and deformation jumps back to backstop. Total convergence of 38 km. (c) Deformation
propagates forward. Total convergence of 41 km. (d) Total convergence of 44 km. (e) New critical taper is achieved. Total convergence of 50 km.
26
GROWTH AND CHANGE OF A COLOUMB WEDGE
No VE
10 km
Extension
Figure 7. Mode! of Coulomb wedge experiencing a decrease in basal strength, (a) Initial critical wedge develops with basal strength 0,, = 30°. Total
convergenceof 12 km. (b)Basa! strength decreases linearly to 0,, = 20°. Total convergence of 12.5 km. Most of the wedge is in the stable fieldand experiences
no deformation although continued convergence causes deformation of the toe region, (c) Basal strength decreases linearly to 0,, = 10°. Total convergence
of 13 km. (d) With continued weakening of the base ( ^ - 6°), wedge again becomes critical in a tensional mode with horizontal extension. Total convergence
of 13.5km.(e)Thebasalstrengthisheldconstantat0^, = 6° and the wedge again becomes stable. Total convergence of 14.5 km. Continued convergence
will result in the wedge growing from the toe backward until a new critical taper is achieved.
27
S. D . WiLLETT
compressive critical wedges experiencing contractional-styles field and strain rates in the wedge interior go to zero (Fig. 7c).
of deformation. At any time during the wedge development, Deformation continues only at the toe of the wedge where the
stresses within the entire wedge are at the Coulomb yield continued convergence forces the wedge to continue accreting
stress. The wedge maintains the same geometry as it grows new material. This deformational pattern is in contrast to the
self-similarly consistent with the critical wedge theory. As deformational pattern in the sub-critical domain III (Fig. 6b),
such, the state of each model is characterized by a single point where the wedge is also rigid, but deformation continues
on line I in Figure 2b. However, if one of the parameters in adjacent to the backstop. The difference is the large taper
Figure 2 should change, the state of the wedge would depart angle in the wedge of Figure 7, relative to the basal strength,
from line I and critical wedge behaviour. In particular, an which allows the wedge to slide over its base and accrete
increase or decrease in the basal strength i^i, moves the wedge incoming sediments. If the basal strength were to remain
state into the sub-critical domain III or stable domain IV, constant at the value in Figure 7c, a new wedge with a shallow
respectively. These two cases are considered next. surface angle would form against the 'backstop' formed by
the rigid, previously developed wedge. As the new material
piles onto the old wedge, the surface angle would be system-
Increase in basal strength model atically reduced, bringing its state of stress back onto yield.
An increase in the basal strength changes the state of the If the basal strength continues to decrease faster than
wedge such that the surface slope is less than is required to be newly accreted material can reduce the surface slope the
critical. This combination of surface slope and basal strength wedge may again go critical in a state of tension (Fig. 7d). The
place the state of the wedge in domain III of Figure 2b, the combination of surface slope and basal strength of the wedge
sub-critical field, implying stresses in the wedge are less than places it on line II in Figure 2b, the second critical wedge
the Coulomb yield stress. The wedge interior in the sub- solution. The entire wedge fails again, but in a tensile mode
critical domain must be rigid and deformation of the wedge associated with extension and downslope flow. The entire
as a whole must cease. The wedge is not steep enough to wedge is not in tension, because of the continued conver-
propagate over its basal decollement so the decollement also gence and accretion at the toe that keeps the toe in local
'locks' and relative displacement between the wedge and the compression. While the basal strength continues to decrease
lower plate goes to zero, thus forcing the relative convergence the wedge remains in this mode of defonnation. The state of
to be taken up at the backstop. This behaviour is demon- the wedge can not move above line II in Figure 2, so the
strated by the model of Figure 6. The wedge in Figure 6a surface slope decreases as the basal strength decreases to
develops with a basal strength parameter 0 b = 15°, as in the keep the state of the wedge on line II. However, if the basal
model of Figure 5. At this point the basal strength parameter strength remains constant, or increases again, the state of the
<^^ instantaneously increases to 30°, the value used in the wedge immediately moves back into the stable field IV (Fig.
model of Figure 4. As expected, deformation stops in most 7e) and deformation ceases. The wedge continues to deform
of the wedge, as stresses drop below yield and the material only to accrete new material at the toe as in Figures 7b & 7c.
becomes rigid (Fig. 6b). Deformation continues only adja-
cent to the backstop, where the rigid backstop forces the
material to fail and begin building a new wedge with a steeper Discussion
surface. As uplift of the rear of the wedge increases the
surface slope, the deformation propagates forward (Figs 6c, The numerical models illustrate four distinct styles of defor-
6d). When the entire surface is at the new critical slope, mation in growing Coulomb wedges, each associated with a
stresses in the entire wedge are again at yield and the wedge specific geometry and strength state domain in a-p-0-0b
again grows self-similarly (Fig. 6e). space. The models of Figures 4 & 5 have constant boundary
conditions and so remain on the critical solution line I in
Figure 8 at the positions labelled by squares (e) and (a),
Decrease in basal strength model
respectively. Deformation in these models is characterized
The complementary case, a decrease in basal strength, im- by plastic deformation of the entire wedge at all times with the
plies very different deformation behaviour. A decrease in highest strain rate concentrated in a band from the wedge toe
basal strength moves the state of the wedge into the stable along the wedge base to the comer with the backstop. A
domain IV of Figure 2b. In the stable domain, as in the sub- stronger base apparently favours transport of undeformed
critical domain, stresses drop below the Coulomb yield sediment further under the wedge before accretion in an
stress, which stops any widespread plastic deformation of the underplating kinematic style.
wedge interior. However, in this case the surface slope and The evolution of the state of the model with increasing
wedge taper are sufficiently large to permit the wedge to slide basal strength (Fig. 6) is shown schematically by the square
stably over its basal decollement forcing relative conver- points in Figure 8, each of which corresponds to a stage of
gence to be taken up at the toe of the wedge. This behaviour development in Figure 6. The initial wedge develops at (a) on
is demonstrated by the model of Figure 7. The wedge in the critical line I; the increase in basal strength immediately
Figure 7a develops with a basal strength parameter (^ b = 30°. transfers the wedge state along the horizontal path to square
Subsequently, the basal strength systematically decreases (b) in domain III. Subsequent deformation steepens the
(Figs 7b, 7c). Most of the wedge quickly enters the stable surface slope moving the state vertically up to its final state
28
GROWTH AND CHANGE OF A COLOUMB WEDGE
29
S. D . WlLLETT
30
GROWTH AND CHANGE OF A COLOUMB WEDGE
References
Barr. T. D. & Dahlen, F. A. 1989. Steady-State Mountain Building 2. Malvern, L.E. \969. Introduction to the Mechanics ofaContinuous Medium.
Thermal Structure and Heat Budget. Journal of Geophysical Research, Prentice-Hall, New Jersey.
94, 3923-3947. Moore, J. C , Watkins, J. S., Shipley, T. H., McMillen, K. J., Bachman, S. B.,
Beaumont, C , Hamilton, J. & Fullsack, P. 1991. Erosional Control of Active & Lundberg, N. 1982. Geology and tectonic evolution of a juvenile
Compressional Orogens, this volume. accretionary terrane along a truncated convergent margin: Synthesis of
Bird, P. & Piper, K. 1980. Plane Stress Finite Element Models of Tectonic results from Leg 66 of the Deep Sea Drilling Project, southern Mexico.
Flow in Southern California. Physics of the Earth and Planetary Interi- Geological Society of America Bulletin, 93, 847-861.
ors,2\, 158-195. Morley, C. K. 1988. Out-Of-Sequence Thrusts. Tectonics. 7, 539-561.
Borja, R. I. & Dreiss, S. J. 1989. Numerical Modelling of Accretionary Piatt, J. P., Leggett, J. K., Young, J., Raza, H. & Alam, S. 1985. Large-scale
Wedge Mechanics: Application to the Barbados Subduction Problem. sediment underplating in the Makran accretionary prism, southwest
Journal of Geophysical Research, 94, 9323-9339. Pakistan. Geology, 13, 507-511.
Brandon, M. T. 1984. A study of deformational processes affecting unlithified 1986. Dynamics of orogenic wedges and the uplift of high-pressure
sediments at active margins: A field study and a structural model, Ph.D. metamorphic rocks. Geological Society of America Bulletin, 97, 1037-
thesis. University of Washington, Seattle, 160p. 1053.
Chappie, W. M. 1978. Mechanics of thin-skinned fold-and-thrust belts. 1987. The uplift of high-pressure-low-temperature metamorphic rocks.
Geological Society of America Bulletin, 89, 1189-1198. Philosophical Transactions of the Royal Society of London A, 321, 87-
Cloos, M. 1982. Flow melanges: Numerical modelling and geologic con- 103.
straints on their origin in the Franciscan subduction complex, California. Shi, Y. & Wang, C. 1988, Thermal Structure of the Barbados Accretionary
Geological Society of America Bulletin, 93, 330-345. Complex. Pure and Applied Geophysics, 128, 749-766.
1984. Flow melanges and the structural evolution of accretionary Silver, E. A., Ellis, M. J., Breen, N. A. & Shipley, T. H., 1985. Comments
wedges. Geological Society of America Special Paper, 198, 71 -79. on the growth of accretionary wedges. Geology, 13, 6-9.
& Shreve, R. L. 1988a. Subduction-Channel Model of Accretion, Stockmal, G. S. 1983. Modelling of Large Scale Accretionary Wedge
Melange Formation, Sediment Subduction, and Subduction Erosion at Deformation. Journal of Geophysical Research, 88, 8271-8287.
Convergent Plate Margins: 1. Background and Description. Pure and Taylor, C. & Hughes, T. G. 1981. Finite Element Programming oftheNavier-
Applied Geophysics, 128, 455-500. Stokes Equations. Pineridge Press Ltd., Swansea, UK, 244p.
& 1988b. Subduction-Channel Model of Accretion, Melange Vilotte, J. P., Daignieres, M. & Madariaga, R. 1982. Numerical Modelling
Formation, Sediment Subduction, and Subduction Erosion at Conver- of Interplate Deformation: Simple Mechanical Models of Continental
gent Plate Margins: 2. Implications and Discussion. Pure and Applied Collision. Journal of Geophysical Research, 87, 10709-10728.
Geophysics,\2&,50\-5A5. , , & Zienkiewicz, O. 1984. The role of a heterogeneous
Cowan, D. S. & Silling, R. M. 1978. A Dynamic, Scaled Model of Accretion inclusion during continental collision. Physics of the Earth and Plan-
at Trenches and Its Implications for the Tectonic Evolution of Subduction etary Interiors, 36, 236-259.
Complexes. Journal of Geophysical Research, 83, 5389-5396. , , & . 1986. Numerical study of continental collision:
Dahlen, F. A. 1984. Noncohesive Critical Coulomb Wedges: An Exact influence of buoyancy forces and an initial stiff inclusion. Geophysical
Solution. Journal of Geophysical Research, 89, 10125-10133. Journal of the Royal Astronomical Society, 84, 279-310.
, Suppe, J. & Davis, D. 1984. Mechanics of Fold-and Thrust Belts and Westbrook,G.K.,Ladd,J.W.,Buhl,P.,Bangs,N.&Tiley,G.J. 1988.Cross
Accretionary Wedges: Cohesive Coulomb Theory. Journal of Geo- section of an accretionary wedge: Barbados Ridge complex. Geology,
physical Research, 89, 10087-10101. 16,631-635.
& Barr, T. D. 1989. Steady-State Mountain Building 1. Deformation Zhao, W.-L., Davis, D. M., Dahlen, F. A. & Suppe, J. 1986. Origin of Convex
and Mechanical Energy Balance./o«/7ia/o/'Geop/iV.5ica//?«ea/T/;. 3906- Accretionary Wedges: Evidence From Barbados. Journal of Geophysi-
3922. cal Research,91, 10246-10258.
Davis. D., Suppe, J. & Dahlen, F. A. 1983. Mechanics of Fold-and-Thrust Zienkiewicz, O.C. & Godbole, P. N. 1974. Flow of Plastic and Visco-Plastic
Belts and Accretionary Wedges. Journal of Geophysical Research. 88, Solids with Special Reference to Extrusion and Forming Processes.
1153-1172. International Journal for Numerical Methods in Engineering, 8, 3-16.
Emerman, S. H. & Turcotte, D. L. 1983. A fluid model for the shape of 1977. The Finite Element Method. McGraw-Hill, New York, 851 p.
accretionary wedges. Earth and Planetary Science Letters, 63,379-384.
31
A developmental stage of a foreland belt
E. G. Bombolakis
Abstract: Active foreland-type belts indicate that recurring seismically related modes of
deformation played a fundamental role in ancient foreland belts. Earthquakes along ramps (?) or
listric segments of thrusts apparently can be larger in magnitude than along associated decollements.
When this situation exists, additional styles of fault slip must occur along decollements to generate
the larger earthquakes at the 'locked' inclined fault segments. Fault-slip styles in active tectonic
terranes include typical earthquakes, slow earthquakes, and recurring forms of fault creep. They
indicate that velocity-dependent damping is a key element in several faulting processes. Fault
models presented here take this parameter into account, and they are applied to the Hogsback sheet
in the Kemmerer region of the Wyoming Salient. They indicate that appreciable long-term fault
creep displacement cannot occur along decollements at average geological slip rates, with few
possible exceptions such as extensive layer-parallel shortening due to pressure solution. Data
analysis for Kemmerer indicates that the Hogsback sheet probably was emplaced by the more rapid
forms of recurring fault slip. The partitioning of elastic tectonic strain imposed by typical and slow
earthquakes accordingly may depend to a considerable extent on the relation between relaxation
times of deformational processes and the recurrence intervals of these fault-slip events.
nodal plane of the mainshock was the active fault, but a recent
Active continental tectonic terranes illustrate that upper crustal
deformation is strongly influenced by coseismic faulting and analysis of the three-dimensional velocity structure and
short-term fault creep events (e.g. Allen 1981; Savage 1983; seismicity provides very strong evidence that the mainshock
Schwartz & Coppersmith 1986; Wesson 1988; Sibson 1989; occurred along a 30° SW dipping portion of a low-angle blind
Stein & Yeats 1989). Consequently, the objectives here are thrust system at some 10 km depth (Eberhart-Phillips 1989).
(1) to present some key evidence of seismically related Relevant interpretations are that the mainshock occurred
deformation in foreland-type belts, (2) to quantify potential along an inclined fault segment of a fault-propagation fold
modes of dynamic fault slip along a decoUement aft of a (Stein & Yeats 1989), along a listric segment of an imbricate
thrust system beneath a major fold (Wentworth & Zoback
frontal ramp or listric segment of an imbricate, (3) to illustrate
applications with a field example from the Wyoming Salient, 1989), or along a frontal ramp associated with a fault-bend
and (4) to suggest how some of the elastic tectonic strains fold (Namson & Davis 1988a). The Coalinga anticline was
imposed by fault slip may be partitioned between brittle either uplifted or had its amplitude increased by half a metre
deformation and strain-rate dependent ductile deformation. or more (Stein 1987).
The Coalinga region is not unique. Pronounced seismicity
reportedly is related spatially to major ramps (or listric
Seismically related deformation in foreland belts segments of blind thrusts) associated with major folds in the
western Transverse Ranges (Namson & Davis 1988b) and in
Seismic moment tensor analyses indicate that moderate to the fold belt of the Los Angeles region where the Oct. 1,1987
large earthquakes account for 25 % to 70% of the upper crustal M|^ 5.9 Whittier Narrows mainshock occurred at about 14 km
fault slip occurring in several major continental zones where depth (Lin & Stein 1989). In both the Whitder Narrows and
active deformation is distributed over large distances from Coalinga mainshocks, the preferred nodal planes dip about
major plate boundaries (Ekstrom & England 1989). Evi- 30°, consistent with balanced cross sections of these regions
dence that seismically related modes of deformation are (Namson & Davis 1988a; Davis era/. 1989). In the Tell Atlas
important in foreland-type belts within these zones is illus- of Algeria, the axis of a well developed anticline in the
trated by the Himalayan foreland belt, the Shotori belt in hanging wall of a complex thrust sheet is parallel to the 30 km
eastern Iran, the Kopah Dagh belt northeast of Iran, the El surface rupture produced by the Ms 7.3 1980 El Asnam
Asnam thrust terrane in the Tell Atlas of Algeria, the western earthquake. Palaeoseismic studies indicate that two previous
Transverse Ranges of California, and by the Coalinga region large earthquakes had occurred along this zone with a re-
of the fold-and-thrust belt along the southern Coast Ranges currence interval of about 450 years (Swan 1988).
bordering the San Joaquin Valley. Coseismic faulting also appears to be associated with flats
The May 2, 1983 M 6.7 Coalinga mainshock and its and decollements. Low-angle nodal planes in the eastern half
aftershock sequence were associated principally with blind of the Himalayan foreland belt lie either on a basal decollement
thrusts and a major fold. Controversy has existed as to which or on flats associated with the Main Boundary thrust system
33
E . G . BOMBOLAKIS
(Baranowski etal. 1984). Examples are the Oct. 21,1964 M boundary between the Franciscan and Great Valley sequence
5.9 mainshock and the March 24, 1974 M 5.4 mainshock, strata (Eberhart-Phillips 1989), a potentially important
both of which had slip vectors plunging 2° or 3°. Similarly, stratigraphic type of barrier.
south of Coalinga and east of the North Kettleman Hills A fundamental phenomenon related to barriers is the
Dome, the M 5.5 Avenal earthquake is consistent with low- recurrence of earthquakes of similar size along specific
angle thrusting along a flat (Namson & Davis 1988a). segments of a fault zone (see Schwartz & Coppersmith 1986),
Most of the catalogued continental thrust-fault mainshocks as illustrated by the Oued Fodda thrust fault in the Tell Atlas
(M > 5) have preferred nodal planes dipping 20° to 60° (e.g. (Swan 1988). The 'characteristic' earthquakes apparendy
Ekstrom & England 1989). Instructive examples are the 1980 recur along the same fault segments until the geometry,
Ms 7.3 El Asnam mainshock on a complex fault dipping = 45° rheology, or boundary conditions of the fault zone are altered
NW at some 6 km depth in the Tell Atlas (Nabalek 1985; sufficiently to accommodate changes in the style of defor-
Yielding 1985), and the 1978 Ms 7.4 Tabas-e-Golshan mation.
mainshock nodal plane dipping 30° NE at about 10 km depth Instrumental monitoring of active faults has revealed that
in the Shotori belt (Berberian 1982). An important feature in the modes of fault slip include typical earthquakes of seconds
both cases is that detailed aftershock studies indicate that the duration, slow earthquakes of minutes to tens of minutes
major structures include listric imbricate faults that converge duration (Sacks et al. 1981; Beroza & Jordan 1989), acceler-
downward into low-angle decollements. And in both cases, ating-decelerating fault creep episodes each of an hour or so
like Coalinga, some of the smaller aftershock focal mecha- duration, and slower forms of fault creep (Wesson 1988).
nisms are indicative of low-angle slip onflatsor decollements. Potentially important mechanisms of long-term fault creep
include the LPS (layer-parallel shortening) thrusting described
by Geiser (1988), and thrusting along evaporite beds But
Observations for formulation of decoUement fault- despite more than 30 years of intensive study, 'long term'
slip models fault creep has been documented in relatively few active
continental locales; notably, along the central section of the
The analysis of seismic moment rates shows that a few large San Andreas fault zone, the Imperial fault to the south, one
earthquakes produce considerably more cumulative fault slip locale along the North Anatolian fault in Turkey (Wesson
along a fault than fairly numerous small earthquakes along 1988), the Nahan thrust in the Himalayas (Sinvhal et al.
the same fault (Brune 1968; Schwartz & Coppersmith 1986). 1973), and the Buena Vista thrust in California (Wilt 1958).
Seismic data, discussed above, indicate that earthquakes Attempts to discover long term fault creep in New Zealand
along ramps or inclined thrust segments can be larger than have been unsuccessful thus far (Scholz 1989).
along flats or decollements. Therefore, for this type of Velocity-dependent damping along faults accordingly
situation, additional styles of fault slip along a decollement, seems to be a key element in several faulting processes
such as fault creep, need to be considered; e.g. in order that (Bombolakis 1989b). For example, analyses in mechanical
sufficient elastic tectonic strains are stored at 'locked' listric engineering show that velocity-dependent damping along a
fault segments to generate the larger earthquakes at those slip surface controls the rate, amount, and time duration of
segments. slip (e.g. Den Hartog 1958; Timoshenko et al. 1974). The
In a review of the paradox of overthrust faulting. Price analyses incorporate various forms of velocity-dependent
(1988) emphasized that traditional models of fault slip, such damping in terms of a damping parameter 5 with respect to
as the Hubbert-Rubey model, cannot be applied to the entire inertial forces. The damping forces can be very complex, and
subhorizontal base of a thrust sheet, and that the fault so 5 is evaluated such that the dissipative energy associated
displacements are described more realistically by a Somigliana with 5 is equivalent to the dissipative energy produced by the
dislocation fault model. This model is one of the elastic actual damping forces (Timoshenko et al. 1974, p. 64,81 -88).
dislocation models frequently employed in geodetic analyses Consequently, relevant concepts from mechanical engineering
of faulting (e.g. Savage 1983; Lin & Stein 1989). However, should have an important bearing on faulting in the upper
these models do not take into account inertial forces of the continental crust.
fault blocks, nor inelastic behaviour along the faults. Con-
sequently, the models presented here are formulated to take
these parameters into account with respect to the following Fault slip models
observations in earthquake seismology and seismotectonics.
Active faults are associated with barriers defined as fea- Figure 1 provides a reference for discussion. Recurring
tures that interrupt or terminate rupture propagation (Aki fault slip occurs along length L of an active thrust-belt
1984). Examples of geometric barriers are a tear fault and an segment of unit width W, bounded by the topographic slope
imbricate zone that sequentially interrupted rupture propa- and vertical sections h^ and h^. The barrier shown in Figure
gation of the main thrust during the 1980 Ms 7.3 El Asnam 1 is a frontal ramp, but it alternatively could be represented by
earthquake (Yielding 1985). An example of a rheologic a listric segment of an imbricate fault. The other barrier in the
barrier is the thick evaporites in the Zagros (Jackson 1983). vicinity of a-b is not shown because various types of barriers
And in the case of the 1983 Coalinga mainshock, the upward are possible. In the case of a coseismic net slip event along
extent of the mainshock rupture ended at the approximate L, the slip equations of Bombolakis (1989a) indicate that fault
34
DEVELOPMENT STAGE OF A FORELAND BELT
k|-h ki (P - S)LW
/ ^ 1 - e' 2^' (cosh pt -I- sinh (3t)
x= k. + k-,
Y p'(T-tana + h 2 ) « L ' W (2) P (4)
is the fundamental frequency of the thrust-belt segment. Tan where p = V s/4 - (o„" . The importance of the fundamental
a is the topographic slope and p is average bulk density. Each frequencies of active fault blocks with respect to velocity-
composite stiffness constant is analagous to the spring con- dependent damping along faults is illustrated nicely by the
stant of a spring-mass system. The stiffer the spring, the case of 5 » c o ^ . For this case, 5/2 « p, causing the bracketed
larger the spring constant. Competent strata usually are expression to approach zero because of a mathematical
elastically stiffer than incompetent strata. The resulting identity between hyperbolic functions and exponential
fundamental frequency characterizes the simplest mode of functions. Therefore, X = 0 irrespective of time t, with the
motion that the active mass would have if it were not affected consequence that no significant fault creep can occur along L.
by dissipative forces. This frequency therefore is one of the This physical condition accordingly might be one of the basic
fundamental parameters inherent in the elastic rebound theory reasons why so few examples of iong term' fault creep have
of earthquake faulting. An example of how the elastic been documented in the upper continental crust since instru-
stiffnesses are calculated from seismic profiling data is illus- mental monitoring of active fault zones was initiated circa
trated in Table 2 of Bombolakis (1989a). 1960. And it is a physical condition that can be evaluated in
The time duration of a coseismic net slip in equation (1) is analyses of thrust-sheet emplacement, as illustrated in the
t = Tt/co^. For this t, equation (1) becomes identical in form following field example.
to the net slip equations of traditional crack-growth models
employed in earthquake seismology (see equation 26 of
Bombolakis 1989a). But like the traditional crack-growth Hogsback thrust sheet in the Kemmerer region,
models, equation (1) neglects velocity-dependent damping. Wyoming
If velocity-dependent damping is incorporated in the deri-
vation of equation (1), we obtain (Bombolakis 1989b) In general, damping includes both a constant component of
damping and a velocity-dependent component. S in equa-
tions (3) and (4) is the shear resistance associated with the
(P - S ) L W 6/2 constant component (Bombolakis 1989a). If we denote T_J as
X« 1 - e 2 (cos a)jt + -sin CO .t)
k, + ko
COH
(3) the average shear resistance associated with both components,
then (T^ - S) represents the average increment of shear resist-
ance associated with velocity-dependent damping along length
L of the decollement in Figure 1. For these conditions, the
approximate value of 5 is
(^d- S)
5=-
p • Xavg. • (-^ tan a + hj) (5)
where Xavg. is the average slip rate of a fault slip event along
L, and h^ is the thickness of the sedimentary package indi-
cated in Figure 1.
L »1 Basement The Hogsback thrust sheet illustrates the problem of long-
term fault creep of thrust sheets. Extensive deformation in its
frontal ramp region near Kemmerer is shown in Figure 2.
Figure 1. A developmental stage of a foreland belt. The temporally active
thrust-belt segment is bounded by L, h 1, hj the topographic slope, and unit (See Figs 1 & 2 in Delphia & Bombolakis 1988 for detailed
width W peipendicular to diagram. stratigraphic section and location in the Wyoming Salient).
The predominantly competent sequence encompassing the
35
E . G . BOMBOLAKIS
Ordovician Bighorn Dolomite (O^^) and the Triassic Nugget fault slip have occurred ? Consider again fault slip data from
sandstone (TR J had undergone considerable imbrication in active tectonic terranes, keeping in mind that equations (2)
the break back mode. In contrast, the same stratigraphic and (5) show that the fundamental frequency is more sensi-
sequence just above the basal decollement extends westward tive to length L of the active thrust-belt segment than the
from c-d for 30 km with no significant deformation apparent velocity-dependent damping parameter.
within the resolution of seismic reflection profiles and other The 'long term' fault creep along the central section of the
subsurface data of Dixon (1982) and Lamerson (1982), until San Andreas actually appears to be dominated by sequences
the trailing edge contact with a major ramp of the Absaroka of short-term fault creep events, each affecting a few or more
thrust is approached. km of fault length along the traces (Wesson 1988). Many of
Suppose that this section of the Hogsback plate is assumed the creep events are interpreted to be associated with yield
to have undergone long-term fault creep at an average geo- point phenomena (Nason & Weertman 1973; Bombolakis et
logical slip rate along L in Figure 1. The estimated rate for al. 1978; Wesson 1988), which is consistent with (P - S) of
thrusts in the Wyoming Salient (Wiltschko & Dorr 1983) is equations (3) and (4). In the case of thrust belts, field evidence
similar to the 10"^ to 10' cm/sec. rates calculated by Elliott of repeated cycles of work softening is reported for several
(1976) for external thrusts of the Canadian Rockies foreland low-angle thrust faults in the foreland zone of the Appalachi-
belt. L = 30 km for the Hogsback sheet, and h^ = 10 km with ans (Wojtal & Mitra 1986, 1988; Woodward etal. 1988).
a ~ 4° prior to post-thrusting erosion. Utilizing elastic Slip rates of fault creep events along the San Andreas
stiffness data for the Kemmerer region (Bombolakis 1989a), frequently range from 10' to 10* cm/sec. (Wesson 1988;
equation (2) indicates that the fundamental frequency of this Bilham 1989). Applying these data to the Hogsback sheet for
part of the Hogsback plate is less than 1 Hertz. A conservative length L = 5 - 10 km, equation (5) yields 8 in the range of 10
estimate of (T^ - S) is = 0.1 bar in equation (5) because esti- - 1O*" Hertz, whereas equation (2) yields a fundamental fre-
mated stress drops of fault creep along the central section of quency of only a few Hertz for the same length segments.
the San Andreas are in the range of 0.4 bar to several bars Consequently, for the more rapid fault creep events, more
(Wesson 1988). Consequently, for average geological slip cumulative fault slip is possible than for the slower fault creep
rates, equation (5) yields values of 5 in the range of 10'' to 10* events. For example, slip rates of slow earthquakes in the
Hertz. These values would increase by a factor of 10, for range of 0.1 - 1 cm/sec. yield values of 8 of about 10 ' to 10
example, if (i^ - S) = 1 bar, instead of 0.1 bar. Therefore - Hertz. Therefore, equations (2), (3), (4), and (5) indicate that
equation (4) stipulates that no significant fault creep displace- it is only slow earthquakes and relatively rapid fault creep
ment possibly could have occurred under these conditions, events, as well as typical earthquakes, that could have pro-
even if the shear resistance associated with velocity-dependent duced significant propagating fault slip between the trailing
damping were small. edge and the Hogsback frontal ramp. Since no significant
The net slip of this section of the Hogsback sheet exceeds fault displacement can result from recurring slow fault creep
12 km (Delphia & Bombolakis 1988). How, then, could the of work-hardening/work-softening material, a potentially
WEST EAST
? ??
- SL 0
/ \ PC / \ /
/ \ / \ / \ / \ / \ / \ / \ / \ / ^ / ^
Figure 2. Balanced cross-section of the Absaroka (A) and Hogsback (H) thrust sheets 15 km north of Kemmerer in the Wyoming Salient. The Lazeart
syncline (L) is demarcated by a solid-line curve drawn along contact between the Frontier formation and the Cretaceous Milliard clastic sequence (Kj^). Open
circles indicate locations of three deep wells close to the section. Three key marker beds within the predominantly competent stratigraphic sequence are
the Ordovician Bighorn Dolomite (O^^^), the Permo-Pennsylvanian Weber sandstone (PIP„), and the Triassic Nugget sandstone (TR^^). From Delphia &
Bombolakis (1988).
36
DEVELOPMENT STAGE OF A FORELAND BELT
37
E . G . BOMBOLAKIS
Partitioning Seismic Strains • be calculated from a steady-state creep flow law when appro-
priate. (P - S) therefore can be larger than the stress drop
1. Estimate x/O Gradient from Deformation Profile: associated with the initial phase of a fault-creep event. This
is one reason why the approximate sign is employed in
equations (3) and (4); other reasons are discussed elsewhere
(see discussion and relevant references in Bombolakis 1989a).
For very heavy damping, the fault displacement is very small,
regardless of whether (P - S) is larger than the stress drop
associated with the early phase of the motion. Consequently,
because of opposing damping and inertial effects, the relation
between 5 and the fundamental frequencies of active thrust-
2. Estimate Relaxation Times for Each Bed: block segments is one of the critical relations that determine
whether fault slip may occur as a typical earthquake, slow
(a) Layer-Parallel Shortening: earthquake, or a relatively rapid or slow fault creep event.
The analysis of the Hogsback thrust sheet in the Kemmerer
region illustrates potential modes of emplacement of external
thrust sheets in a foreland belt. For each style of fault slip
event analysed here, the increment of shear resistance asso-
T = 3n/E ciated with velocity-dependent fault damping is (x^ - S) < (P
- S). And for (t^ - S) as small as 0.1 bar, data analysis for the
(b) L a y e r - P a r a l l e l Shear: Kemmerer region demonstrates that the 12 or more km of
cumulative net slip could not have developed at average
T geological slip rates. Instead, the Hogsback sheet would have
had to be emplaced via more rapid types of recurring fault
slip. Potential exceptions permitting emplacement at average
geological slip rates include thrusting along salt and the LPS
thrusting described by Geiser (1988), which do not apply to
the Hogsback sheet (Delphia & Bombolakis 1988). Further-
T = n/G more, fault-rock studies of several foreland thrusts in the
Figure 4. Idealized example of partitioning layer-parallel shortening and Appalachians indicate that fault slip involved cycles of work
layer-parallel shear within a thrust sheet in terms of relaxation times T and hardening and work softening (Wojtal & Mitra 1986, 1988;
T '. See text. Woodward ^/a/. 1988). Hence, in view of the relatively rapid
forms of fault slip operating in active zones of the upper
barriers (Aki 1984). In thecaseoffaultcreepevents. Wesson continental crust, it is unlikely that the development of
(1988) finds that they can be viewed as arising from (1) stress foreland belts can be analysed exclusively in quasi-static
applied from external sources. (2) stress caused by geometry terms. The mechanical partitioning of elastic tectonic strains
and distribution of displacement on the fault, arising from the accordingly would need to include analyses based on rela-
elastic response of the surrounding medium to the tions between relaxation times and recurrence intervals.
displacements within the inelastic fault zone itself, and (3) An important observation in recent years is that syntectonic
constitutive relations and yield point phenomena that charac- extensional faulting is associated with compressional fault-
terize slip resistance along the fault. (Wesson's models ing in several thrust belts (e.g. Piatt & Leggett 1986; Wojtal
currently do not incorporate inertial forces). & Mitra 1986; Coward 1988). In Figure 1, the sedimentary
These parameters are incorporated here in prototype mod- package of the active thrust-belt segment along vertical
els that include inertial effects. In the case of a typical section h^ can undergo interludes of extension during dy-
earthquake, (P - S) in equation (3) is directly proportional to namic fault slip. However, it is not clear yet whether
stress drops calculated in earthquake seismology (Bombolakis extensional faulting would occur in the vicinity of h^, as was
1989a). The alteration of fault-zone material, however, can observed during the 1941 Hartford dyke detachment faulting
alter fault-slip style along specific segments of a decollement. event (Bombolakis 1981). Syntectonic extensional faulting
The constitutive relations for the segments probably are very in thrust belts therefore constitutes another potentially impor-
complex, and so it is more convenient to characterize several tant problem for future study.
effects of fault-rock behaviour by damping parameters.
Damping factor 5, for example, can be calculated directly
from basic field parameters in equation (5). In the case of Constructive suggestions for revision of the preliminary manuscript were
made by P. D'Onfro andW. Rizer of Conoco Inc., R. H. Manin III of New
episodic fault creep associated with yield point phenomena, England Research Inc., an anonymous reviewer, and the editor of this
P is the peak shear strength equal to or greater than the yield volume. Special thanks to Jeffrey Townsend for the arrangement and editing
point, depending on work hardening. The result is that (P - S) of a technically difficult manuscript.
is the stress drop associated with work softening, where S can
38
DEVELOPME.NT STAGE OF A FORELAND BELT
References
Aki, K. 1984. Asperities, barriers, characteristic earthquakes, and strong Lin, J. & Stein, R. S. 1989. Coseismic folding, earthquake recurrence, and the
motion prediction. Journal of Geophysical Research, 89, 5867-5872. 1987 source mechanism at Whittier Narrows, Los Angeles Basin,
Allen, C. R. 1981. The modem San Andreas Fault. In: Ernst, W. G. (ed.) The California. Journal of Geophysical Research. 94, 9614-9632.
Geotectonic Development ofCalifornia, Ruhey Volume 1. Prentice-Hall, Mitra, G., Yonkee, W. A. & Gentry, D. J. 1984. Solution cleavage and its
Englewood Cliffs. New Jersey, 223-249. relationship to major structures in the Idaho-Utah-Wyoming thmst belt.
Baranowski, J., Armbruster, J., Seeber, L. & Molnar, P. 1984. Focal depths Geology, 12, 354-358.
and fault plane solutions of earthquakes and active tectonics of the & Protzman, M. 1988. Strain trajectories and deformation profiles in
Himsdaya^s. Journal of Geophysical Research, 89, 6918-6928. thrust belts and extensional terranes. Geological Society of America
Berberian, M. 1982. Aftershock tectonics of the 1978 Tabas-e-Golshan Abstracts with Programs, 100, A107.
(Iran) earthquake sequence; a documented active' thin-and-thick skinned Nabalek, J. 1985. Geometry and mechanism of faulting of the 1980 El
tectonic' case. Geophysical Journal of ihe Royal Astronomical Society, Asnam, Algeria earthquake from inversion of teleseismic body waves
68, 499-530. and comparison with field observations. Journal of Geophysical Re-
Beroza, G. C. & Jordan, T. H. 1989. Comparison of free-oscillation excita- search, 90, 12713-12728.
tion events with global seismicity: evidence for silent earthquakes. Namson, J. & Davis, T. L. 1988a. Seismically active fold and thrust belt in
Transactions of the American Geophysical Union (EOS), 70, 397. the San Joaquin Valley, central Califomia. Bulletin of the Geological
Bilham, R. 1989. Surface slip subsequent to the 24 November 1987 Super- Society of America, 100, 257-273.
stition Hills, California, earthquake monitored by digital creepmeters. & Davis, T. L. 1988b. A structural transect of the western Transverse
Bulletin of the Seismological Society of America , 79, 424-450. Ranges, Califomia: implications for lithospheric kinematics and seismic
Bombolakis, E. G. 1981. Analysis of a horizontal catastrophic landslide./n: risk evaluation. Geology, 16, 675-679.
Carter, N. L., Friedman, M., Logan, J. M. & Steams, D. W. (eds) & Weertman, J. 1973. A dislocation theory analysis of fault creep
Mechanical Behaviour of Crustal Rocks, the Handin Volume. American events. Journal of Geophysical Research, 78, 7745-7751.
Geophysical Union. Geophysical Monograph, 24, 251-257. Philip, H. & Meghraoui. M. 1983. Structural analysis and interpretation of
1989a. Thrust-fault mechanics and dynamics during a developmental the surface deformations of the El Asnam earthquake of October 10,
stage of a foreland belt. Journal of Structural Geology, 11, 439-455. \9m. Tectonics, 2, 17-49.
1989b. Fault slip and partitioning of elastic tectonic strains during Piatt, J. P. & Leggett, J. K. 1986. Stratal extension in thrust footwalls, Makran
dynamic low-angle thrusting. Geological Society of America Abstracts accretionary prism: implications for thrust tectonics. American Asso-
with Programs, 20, A134-A135. ciation of Petroleum Geologists Bulletin, 70, 191-203.
, Hepbum, J. C. & Roy, D. C. 1978. Fault creep and stress drops in Price. R. A. 1988. The mechanical paradox of large overthrusts. Bulletin of
saturated silt-clay gouge./o«™a/o/'Gfop/!yx/ca//?fseair/i, 83,818-829. the Geological Society of America. 100, 1898-1908.
Brune, J.N.I 968. Seismic moment, seismicity, and rate of slip along major Sacks, I. S., Linda, A. T., Snoke, J. A. & Suyehiro, S. 1981. A slow
fault zones. Journal of Geophysical Research, 73, 777-784. earthquake sequence following the Izu-Oshima earthquake of 1978. In:
Coward, M. 1988. The Moine thrust and the Scottish Caledonides. In: Mitra, Simpson, D. W. & Richards, P. G. (eds) Earthquake Prediction. An In-
G. & Wojtal, S. (eds) Geometries and Mechanisms of Thrusting, with ternational Review. American Geophysical Union, Maurice Ewing Se-
Special Reference to the Appalachians. Geological Society of America ries, 4, 617-628.
Special Paper, 222, 1-16. Sanderson, D. J. 1982. Models of strain variation in nappes and thrust sheets:
Davis, T.L. , Namson, J. & Yerkes, R. F. 1989. A cross section of the Los a review. Tectonophysics, 88, 201-233.
Angeles area: seismically active fold and thrust belt, the 1987 Whittier Savage, J.C.I 983. Strain accumulation in the western United States. Annual
Narrows earthquake, and earthquake hazard. Journal of Geophysical Review of Earth and Planetary Sciences, 11, 11-43.
Research. 94, 9644-9664. Scholz, C. H. 1989. Mechanics of faulting. Annual Review of Earth and
Delphia, J. & Bombolakis, E. G. 1988. Sequential development of a frontal Planetary Sciences, 17, 309-334.
ramp, imbricates, and a major fold in the Kemmerer region of the Schwartz, D. P. & Coppersmith, K. J. 1986. Seismic hazards: new trends in
Wyoming thrust belt. In: Mitra, G. & Wojtal, S. (eds) Geometries and analysis using geologic data. In: Active Tectonics. National Academy of
Mechanisms of Thrusting, with Special Reference to the Appalachians. Sciences Press, Washington D.C., 215-230.
Geological Society of America Special Paper, 222, 207-222. Stein, R. S. & Yeats, R. S. 1989. Hidden earthquakes. Scientific American,
Den Hartog, J. P. 1956. Mechanical Vibrations. McGraw-Hill, New York. June, 48-57.
Dixon, J. S. 1982. Regional structural synthesis. Wyoming Salient of 1987. Contemporary plate motion and crustal deformation. American
western overthrust belt. American Association of Petroleum Geologists Geophysical Union, Reviews in Geophysics, 25, 855-863.
Bulletin, 66. \560-\5S0. Sibson, R. H. 1989. Earthquake faulting as a structural process. Journal of
Eberhart-Phillips, D. 1989. Active faulting and deformation of the Coalinga Structural Geology, 11, 1-14.
anticline as interpreted from three-dimensional velocity structure and Swan, F. H. 1988. Temporal clustering of Palaeoseismic events on the Oued
.seismicity. Journal of Geophysical Research. 94, 15565-15586. Fodda fauU, Algeria. Geology, 16, 1092-1095.
Ekstrom. G. & England, P. 1989. Seismic strain rates in regions of distributed Synvhal, H., Agrawal, P. N.. King, G.C.P. & Guar, V. K. 1973. Interpretation
continental deformation. Journal of Geophysical Research, 94, 10231 - of measured movement at a Himalayan (Nahan) thrust. Geophysical
10257. Journal of the Royal Astronomical Society. 34, 203-210.
Geiser, P. 1988. Mechanisms of thrust propagation: some examples and Timoshenko, S., Young, D. H. & Weaver, W. 1974. Vibration Problems in
implications for the analysis of overthrust xerranes.Journal ofStructural Engineering. John Wiley and Sons, New York.
Geology. 10, 829-845. Wentworth, C. M. & Zoback, M. D. 1989. The style of late Cenozoic
Hudleston, P. J. 1986. Extracting information from folds in rocks. Journal of deformation at the eastern front of the CalifomiaCoast Ranges. Tectonics.
Geological Education, 34, 237-245. 8, 237-246.
Jackson, J. A. 1983. The use of earthquake source studies in continental Wesson, R. L. 1988. Dynamics of fault creep. Journal of Geophysical Re-
tectonic geology. In: Kanamori, H. & Boschi, E. (eds) Earthquakes: wa/r/i, 93, 8929-8951.
Observation, Theory, and Interpretation, Course LXXXV. Proceedings Wilt, J. W. 1958. Measured movement along the surface trace of an active
of the International School of Physics "Enrico Fermi', Italian Physical thrust fault in the Buena Vista Hills, Kern County, Califomia. Bulletin of
Society. North-Holland, Amsterdam, 456-478. the Seismological Society of America, 48, 169-176.
Klinger, R. E. & Rockwell, T. K, 1989. Flexural-slip folding along the Wiltschko, D. V. & Dorr, J. A. Jr. 1983. Timing of deformation in overthrust
Elmore Ranch fault in the Superstition Hills earthquake sequence of belt and foreland of Idaho, Wyoming and Utah. American Association of
November 1987. Bulletin of the Seismological Society of America,19,297- Petroleum Geologists Bulletin, 67, 1304-1322.
303. Wojtal, S. & Mitra, G. 1986. Strain hardening and strain softening in fault
Lamerson, P. R. 1982. The Fossil Basin and its relationship to the Absaroka zones from foreland thrusts. Bulletin of the Geological Society of
thrust sy.stem, Wyoming and Utah. In: Powers, R. B. (ed.) Geologic America, 91. 61A-6S1.
Studies of the Cordilleran Thrust Belt, Volume 1. Rocky Mountain As- & 1988. Nature of deformation in some fault rocks from
sociation of Geologists, Denver, Colorado, 279-340. Appalachian thrusts. In: Mitra, G. & Wojtal, S. (eds) Geometries and
Mechanisms of Thrusting, with Special Reference to the Appalachians.
Geological Society of America. Special Paper, 222, 17-33.
39
E . G . BOMBOLAKIS
Woodward, N. B., Wojtal, S., Paul, J. B. & Zadins, Z. Z. 1988. Partitioning Yielding, G. 1985. Control of rupture by fault geometry during the 1980 El
of deformation within several external thrust zones of the Appalachian Asnam (Algeria) earthquake. Geophysical Journal of the Royal Astro-
otogen. Journal of Geology, 96, 351-361. nomical SocietyyHl, 641-670.
Yerkes, R. F., Ellsworth, W. L. & Tinsley, J. C. 1983. Triggered reverse fault
and earthquake due to crustal unloading, northwest Transverse Ranges,
California. Geology, 11, 287-291.
40
PART TWO
Physical modelling
One-dimensional models for plane and non-plane power-law flow in
shortening and elongating thrust zones
Steven Wojtal
Abstract: The movement of a thrust sheet, even in external parts of fold-thrust belts, occurs by
deforming rocks within a zone of finite thickness. A significant fraction of thrust-zone strains may
accrue slowly as viscous deformation. This contribution uses a simple one-dimensional model of
power-law flow to analyse the viscous deformation in thrust zones. The model generates profiles
of stress and velocity for plane and non-plane flows that shorten or elongate during shearing. These
stress and velocity profiles are, in essence,' snapshots' of stresses and velocities in real thrust zones
that sequentially shorten and elongate during thrust emplacement. A comparison of stress and
velocity solutions for flows with different stress exponents gives a way to examine structural
settings where a single weak layer develops and affects stresses throughout the sheet. Using data
on the orientations of stress principal directions during episodes of shortening and elongation
derived from the mesoscopic fault array in an external thrust zone from the southern Appalachian
fold-thrust belt, the plane-flow solution suggests that differential stresses in the thrust zone did not
exceed 20 MPa. Moreover, body forces due to the dipping upper surface of the sheet were
apparently not the primary source of the tractions during thrust zone shortening or elongation
episodes. The additional tractions that other portions of the thrust sheet exerted on this thrust-zone
segment had comparable magnitudes during shortening and elongation episodes. Comparing
plane and non-plane solutions indicates that (1) shearing may localize at shallower depths or occur
with different orientations for principal stresses relative to the thrust-zone boundaries, and (2)
shortening or elongation may occur at smaller magnitudes of in-transport compression or tension,
in non-plane structural settings like the growing tips of thrusts.
The traditional view of fold-thrust belts has each thrust sheet ematical model, based on field studies of several external
sliding over its footwall on a discrete fault surface (Hubbert fold-thrust belt thrust zones, that gives stresses and velocities
& Rubey 1959; Hsu 1969; Price 1988). This view does not in plane strain as functions of position below the surface of a
conform with the movement mechanisms known for thrust thrust sheet. The model provides insight into the forces that
sheets in the internal portions of fold-thrust belts (cf. Schmid drove the emplacement of an external thrust sheet in the
1975;Gilotti&Kumpulainen 1986). It is at odds increasingly southern Appalachian fold-thrust belt. It also guides infer-
with observations in the external portions of fold-thrust belts, ences on the effects of changes in material behaviour due to
where studies of well-exposed thrust zones indicate that initial differences in rock rheology or to the mesoscopic and
thrust sheets move by deforming rocks within a zone several microscopic structural changes that accompany strain accu-
metres to several tens of metres thick (Harris & Milici 1977; mulation. A second intent of this contribution is to extend this
Wojtal 1986; Piatt & Leggett 1986). Prevailing models for analysis to three-dimensions, considering non-plane flow in
thrust belt evolution, on the other hand, specify that wedges thrust sheets. Preliminary examination of the results of an
exhibiting uniform mechanical behaviour yield throughout added degree offreedom, even in the unsophisticated manner
while sliding above abasal detachment (Chappie 1978; Davis considered here, suggests that non-plane strain treatments are
etal. 1983; Emerman & Turcotte 1983). The intensities and warranted.
geometries of strains within foreland thrust sheets do not The model, derived from glacier flow analyses (Nye,
conform with the predictions of yielding-wedge models 1957; Paterson, 1981), considers the flow of a power-law
(Woodward 1988). Moreover, rocks at different positions sheet between parallel-sided boundaries. The formulation
within otogenic wedges and along distinct segments of thrust (described in greater detail below) does not specify a priori
zones within the wedges may exhibit distinct material prop- the stress exponent n for the sheet, so it is sufficiently general
erties (Piatt 1986; Wojtal & Mitra 1988). These factors may to treat linearly viscous sheets (« = 1) and perfectly plastic
limit the applicability of elegant analyses of otogenic de- sheets (« —^ oo). Stress and velocity distributions for flows
formation, and they make exact solutions for realistic otogenic with different stress exponents n provide some insight to the
wedges difficult to attain (Fletcher 1989). effects of different rheologies. More importantly, the model
A more tractable task is to examine limited segments of a provides explicit solutions for stresses and velocities in
thrust sheet where a single deformation mechanism prevails. laminar flows (i.e. heterogeneous simple shear flows) as well
Even then, one must tailor the analysis to fit the desired focus. as solutions for compressing and extending flows (flows that
One intent of this contribution is to present a simple math- shorten or elongate as they shear). This contribution uses
41
S. WOJTAL
those solutions to examine the elongation and shortening that moving relatively undeformed but fault-bounded blocks of
accompanies shearing in many fold-thrust belt thrust zones. rock past each other. From the standpoint of how the faults
in the arrays together accommodated the strains associated
with movement on major faults and from the standpoint of
Structure of foreland fold-thrust belt thrust zones how stresses were transmitted through the pervasively-faulted
rocks in the vicinity of major faults, it is therefore sensible to
Mesoscopic structure
take a larger view and examine this deformation at the scale
Immediately above and below many foreland fold-thrust belt of the whole thrust sheet.
thrust surfaces, pervasive arrays of mesoscopic minor faults Some foreland fold-thrust belt thrust zones possess thin
occur within thin (relative to the thrust sheet's thickness) layers of pervasively-cleaved rock along movement surfaces
layers (Harris & Milici 1978; Wojtal 1986; Piatt & Leggett {?\aXietal. 1987;Figs.6&7inWoodwardera/. 1988). The
1986; Mitra 1986). Faults in the arrays typically make either pervasive cleavage in these zones usually forms by localized
low (<45°) or high (>45°) angles to the thrust zone boundaries pressure solution in rocks with mixed clay-quartz or clay-
(Fig. 1). Faults at low angles to the thrust zone boundaries calcite compositions (Woodward et al. 1988), and it ac-
usually (1) have lineations whose pitches are close to 90°, (2) commodates thmst sheet movement by inhomogeneous ductile
also cut the enveloping bedding at low (<45°) angles, and (3) shearing in narrow layers like that envisioned by Kehle
exhibit reverse offsets of bedding. Faults at high angles to the (1970). This type of ductile shearing might not persist within
thrust zone boundaries have less consistent kinematics. an individual layer throughout the life of a thrust zone,
Whereas these faults generally also cut enveloping bedding at however. Continued shearing accommodated by diffusive
high (>45°) angles, the pitches of lineations on them may be mass transfer, particularly in thrust zones that exchange pore
close to 0°, close to 90°, or in between. Normal offsets of fluids with the surrounding rocks, could remove entirely the
bedding on high-angle faults are more common, but reverse relatively soluble phases. The resulting narrow layer of clay-
offsets occur. Where both low-angle and high-angle faults rich selvage might continue to accrue slip, or the localization
occur together, high-angle faults usually postdate low-angle of strain elsewhere in the thrust zone could generate another
faults. zone of pervasive cleavage. The occurrence of individual and
multiple selvage bands within several large displacement
thrust zones (e.g. Hunter Valley and Saltville thrusts in the
southern Appalachians and McConnell thrust in the Canadian
Rockies) suggests that inhomogeneous shearing of this type
may commonly contribute to thrust sheet movement.
Microscopic structure
In many fold-thrust belt thrust zones, cataclastic rocks (1)
Faults with . Faults without decorate individual fault surfaces and bed-parallel slip sur-
Slickensides Slickensides
faces, and (2) separate hangingwall rocks from footwall rocks
(Engelder 1974; House & Gray 1982; Wojtal & Mitra 1986;
Piatt et al. 1987). Thrust-zone deformation may generate
Figure 1. Downplunge projection (onto a vertical plane with a strike of either penetratively microfractured rocks with little bulk
110°) of a portion of the mesoscopically faulted rock in the Cumberland
Plateau thrust zone, southern Appalachians (after Harris & Milici 1977: strain or distinctive banded cataclasites and ultracataclasites
Wojtal 1986). The primary decollement in acoal layer lies parallel to the base that accommodated very high strains (Mitra 1984; Wojtal &
of the drawing, just below the edge of the drawing. Irregular line at base Mitra 1986). The magnitude of strain accommodated in
represents rubble that locally covers the detachment. Solid bold lines and cataclastically deformed rocks varies, of course, with rock
dashed bold lines show faults with slickensides preserved. Dash-dot bold
type and position within individual thrust zones, and from one
lines show faults with no slickensides. Fine lines show bedding traces. The
dashed fault traces are those faults with slickensides that do not conform with thrust zone to another. In thrust zones where temperatures
either of two well-defined palaeostress tensors (Wojtal & Pershing 1991). during deformation remained below about 200°C, siliciclastic
sandstones are penetratively fractured and/or pervasively
This pervasive mesoscopic faulting accommodates defor- faulted, but cataclasis does not normally account for large
mation that is continuous at the scale of a thrust sheet (Wojtal strains and banded cataclasites are absent. Banded carbon-
1989). The distribution of displacement among faults in these ate-matrix cataclasites or zones of pervasive cleavage in
arrays indicates that small numbers of faults account for most shales are common in these low temperature thrust zones,
of the slip in the zones of pervasive faulting (Wojtal & Mitra suggesting that shearing occurred in units other than
1986). Moreover, the mesoscopic faults that account for most siliciclastic sandstones (Wojtal & Mitra 1986; Woodward et
of the total slip constitute subsets of faults that conform with al. 1988; Arboleya 1989). Where temperatures during de-
well-defined palaeostress tensors (Fig. 2) (Wojtal & Pershing formation exceeded 200°C, banded quartz-rich cataclasites
1991). Bulk deformation in these pervasively-faulted zones are common in thrust zones (Mitra 1984; Wojtal & Mitra
was, then, 'block-boundary sliding' in the sense that bulk 1988). Regardless of thrust-zone temperature, banded
shortening, bulk elongation, and bulk shear occurred by cataclasites occur most often along primary thrust surfaces,
42
ONE-DIMENSIONAL MODELS FOR POWER-LAW FLOW IN THRUST ZONES
R=0.89
Nj-60
N,,=26
- « 20
(D ^ ^- CO (3)
• , ^ Ot C^
O (O T-
r- 1^ n
~ CM
Deviation Deviation
Figure 2. Stereographic projections giving the orientations of palaeostress principal directions calculated, using the attitudes of mesoscoplc faults and their
slickensides, for the portion of the thrust zone illustrated in Figure 1. This figure shows the two palaeostress tensors calculated for the Cumberland Plateau
thrust zone (Wojtal & Pershing 1991). The great circle in each siereographic projection gives attitude of the basal detachment in this thrust zone. The
designations 'first' and "second" for the stress tensors come from cross-cutting relationships among faults. Histograms beneath each stereographic projection
give the angular deviations of the fault/slickenside pairs used to calculate each tensor. R - (a^-a3)/(0| - a^). NU is the number of fault slickenside pairs
actually used to calculate each palaeostress tensor. NT is the total number of fault-slickenside pairs that could have constrained tensors during each program
iteration.
43
S. WOJTAL
50 Mm 50 Mm
Figure 4, (a) Plane polarized light photomicrograph of a ponion of Figure 3. (b) Approximately the same field of view as (a) under cathodo-luminescence.
The cataclasite matrix and vein material on one hand and wall rock and wall-rock fragments on the other hand exhibit significant differences in the intensity
and colour of their luminescence.
parallel to R and Y, support the inference that some luminesce, or vice versa. Where the cataclasite matrix and
intragranular fractures originated as intergranular slip sur- wall rocks luminesce differently, the cataclasite matrix and
faces. Mineral-filled intragranular fractures are more com- mineral-filled veins cutting the wall rocks usually exhibit
mon, however. Most mineral-filled intragranular fractures similar luminescence. Differences in calcite luminescence
resemble, in thickness and in the microstructure of their result from slight differences in concentrations of activators
mineral fillings, the late veins found elsewhere in the like Mn and suppressors like Fe (Marshall 1988). Different
cataclasites. Others have distinctive, thin mineral fillings and trace element chemistry in host rocks and veins, like different
orientations roughly appropriate fore microfractures (Fig. 11 c isotopic composition in hosts and veins (Rye & Bradbury
in Wojtal & Mitra 1986). These intragranular fractures may 1988; Wiltschko & Budai 1988), suggest that host rocks did
be extensional microcracks that never propagated past the not buffer solutions during mineral precipitation, perhaps
fragment boundaries because the matrix blunted the growing because fluid to rock ratios were high. Comparable trace
fracture tip. Such a situation requires the fine-grained matrix element chemistry for veins and cataclasite matrix suggests
to have been sufficiently ductile to dissipate fracture tip that both equilibrated with a single fluid phase. Matrix grains
stresses. The flow of fine-grained matrix into opening may, because of small diameters, have been highly reactive,
fractures (Figs 1 le,f& 12c in Wojtal & Mitra 1986) supports and their chemical equilibration with a fluid phase may have
the inference that the fine-grained matrix was, at times, occurred statically. This would not explain, however, why
highly ductile. fine-grained carbonates in the host rocks did not also equili-
Matrix ductility in cataclasites might result from either brate with that fluid or why no reacted carbonate rims occur
cataclasis or flow by diffusive mass transfer (Rutter 1984). on host fragments in the matrix. Dissolution and reprecipitation
Determining, if possible, which deformation mechanism of the fine-grained matrix during diffusion-accommodated
prevailed is important in order to know what are the mechani- flow, as Wojtal & Mitra (1986) proposed, could cause differ-
cal implications of banded cataclasites along thrusts. Cathodo- ences in trace element chemistry (Beach 1982) that are
luminescence provides one way to assess the contribution of restricted to the fine-grained matrix. Thus, cathodo-lumines-
diffusive mass transfer to deformation in some rocks. Under cence observations suggest that diffusive mass transfer was a
cathode-ray illumination, the luminescence of the fine-grained significant deformation mechanism in banded cataclasites
matrix in banded carbonate cataclasites is often qualitatively (Mitra 1984; Wojtal & Mitra 1986, 1988), even at tempera-
quite different from the luminescence of its presumed parent tures below 200°C. If diffusion creep accommodated matrix
rocks (Fig. 4). The matrix may luminesce intensely while tlow, cataclasites would have exhibited a linear stress-strain
wall rocks and wall-rock fragments in the cataciasites barely rate relation and strong grain size sensitivity.
44
ONE-DIMENSIONAL MODELS FOR POWER-LAW FLOW IN THRUST ZONES
The planar to smoothly-curved extensional veins and selvage relative to the displacement by slip on cataclasite-
shear surfaces occur in subparallel sets that cut fragments, selvage or cataclasite-cleaved rock contacts. Wall-rock de-
banding, and the distorted remnants of earlier subparallel formation does not provide unequivocal answers. O'Hara et
veins where all occur together. In some cases, these al. (1990), citing increased coal rank along mesoscopic
intergranular features abut or curve asymptotically into ma- faults, inferred that fault slip occurred in discrete, temporally
trix-rich bands aligned parallel to the thrust zone boundaries, abrupt, steps. Crack-seal fibres in mineral lineations on
suggesting that they formed during episodes of shear by mesoscopic faults also suggest episodic slip on individual
localized matrix flow in matrix-rich bands. Sets of veins and/ mesoscopic faults. These observations suggest that discrete,
or shear surfaces cutting through entire samples indicate that probably seismic, slip contributes to thrust sheet movement,
episodes of intergranular fracturing postdate matrix flow in an inference supported by analytical treatments (Bombolakis
some cataclasites. Most cataclasite samples possess both 1989; pers. comm.). Lineations on mesoscopic faults in many
undisiorted and uniformly-distorted intergranular features, thrust zones are, however, continuous fibres, suggesting that
suggesting that episodes ofintergranular fracturing repeatedly mesoscopic faults may move in part by continuous slip.
interrupted ductile flow in thrust zone cataclasites (Wojtal & Moreover, the microstructures in thrust zone cataclasites
Mitra 1986). Inasmuch as matrix flow accounts for nearly all indicate that a sizable fraction of the total shear strain in these
of the shear strain within these banded cataclasites, shear rocks occurred by continuous, i.e. viscous, flow. At present,
displacement apparently accumulated mainly by diffusive there are no general ways to ascertain/row rock structures
mass transfer. In thrust zones segments where banded alone whether viscous creep or discrete, seismic slip events
cataclasites constitute the only rocks along planar boundaries predominated. One contention of this contribution is that
between deformed hangingwall and footwall rocks (e.g. viscous deformation of rocks along external thrust faults is
segments of the southern Appalachian Copper Creek thrust significant, and that it provides important additional insight
described in Wojtal & Mitra 1986, or segments of the southern to the overall mechanics of thrust sheet movement.
Appalachian Saltville thrust described by Woodward et al.
1988), diffusive mass transfer accommodated the last size-
able increments of sheet movement. Since bulk flow by Interrelationship of mesoscopic and microscopic struc-
diffusive mass transfer probably exhibits a linear relationship tures
between differential stress and strain rate (Elliott 1973; Rutter Microstructures in cataclasite layers and mesoscopic struc-
1976), sheet movement on such a thrust zone would, in tures in the layers of pervasively faulted or cleaved rocks
essence, occur by shear of a viscous fluid {Kehle 1970). evolve concurrently. Using cross-cutting relationships in
In other thrust zone exposures, banded cataclasites occur pervasively faulted strata (Wojtal 1986; Piatt & Leggett
adjacent to pronounced selvage bands or pervasively cleaved 1986), cross-cutting vein patterns, or microfracture
shales (Fig. 5). Cataclasites in these settings possess the orientations in fragments from cataclasites (Woodward et al.
entire complement of microstructures indicating that ductile 1988), one may infer deformation histories for different
flow accommodated large shear strains. The concern most segments of foreland thrust zones. Thrust zone deformation
pertinent here is to assess the magnitude of displacement by histories commonly include shortening and elongation epi-
ductile flow within layers of cataclasite, cleaved rock, or sodes, with elongation following shortening (Piatt & Leggett
45
S. WOJTAL
1986; Woodward et al. 1988). The individual mesoscopic with planar and parallel top and bottom surfaces flows in
faults by which wall rocks shorten or elongate often extend response to surface forces that act at its trailing edge and body
directly into distinct intergranular slip surfaces in cataclasites, forces due to its dipping upper surface (Fig. 6). The driving
suggesting that both thrust zone components shortened or stresses in the treatment are those that operate in thrust
elongated coevally. This extension-then-contraction defor- systems (cf. Siddans 1984; Hudleston 1991), and the formu-
mation sequence could result from moving a sheet past an lation is general enough to consider compressing and extending
asperity or through an area of rough slip along its thrust zone flows.
(Piatt & Leggett 1986; Erickson & Wiltschko 1986). Consider that the sheet's upper surface dips in the direc-
Alternatively, this structural sequence may have evolved tion the sheet moves. Cartesian coordinates in the sheet have
during thrust surges (Coward 1982) akin to surging glaciers X| parallel to the upper boundary and pointed in the direction
(Sharper a/. 1988). Perhaps an analogy more precise than that of movement, x^ normal to the upper boundary and directed
with surging valley glaciers is the switching on and off of ice upwards, and x^ perpendicular to the Xi-x^ plane and directed
streams within the Antarctic ice-sheet (MacAyeal etal. 1988; in such a manner to make the coordinate frame right-handed
MacAyeal 1989a & b). Ice stream initiation occurs when a (Fig. 6). The Xi-x^ plane is a principal plane for stress and
portion of an ice-sheet becomes decoupled from its base, strain rate tensors in the sheet.
either due to increased subglacial water pressure or due to
weakening of subglacial till (Bentley 1987;MacAyeal 1989b).
The flow strengths of and stress distributions within external
thrust zones are not known precisely, but two observations
suggest that the basal layers of thrust sheets are weaker than
the sheets themselves. First, the localization of strain in
cataclasite layers along thrusts suggests that they became
weaker than the surrounding rocks even if they were not
originally weaker than them (Mitra 1984; Wojtal & Mitra
1986). Second, shortening and elongation episodes in perva-
sively faulted rock layers apparently occurred in response to
principal stresses aligned nearly parallel to thrust zone
boundaries (Wojtal & Pershing 1991). Principal stress direc-
tions aligned parallel to thrust zone boundaries rather than
oblique to them suggests that the sheets moved over zones Figure 6. Sketch showing the orientation of the coordinate frame used in
considerably weaker than the sheets themselves (cf. Chappie the mathematical model described in the text, and showing the general
geometry of the stresses acting on rocks in a thrust zone considered in the
1978, p. 1196; Stockmal 1983). In this respect, external thrust model.
zones apparently differ from internal thrust zones. In the
latter, thrust-parallel shear is distributed throughout signifi- The flow law at all points in the layer, written in terms of
cant volumes of sheets, suggesting that principal stresses the second invariants of the stress and strain rate tensors x and
were oblique to the thrust-zone boundaries near the thrust e,is
sheet base (Mitra & Elliott 1980; Ramsay era/. 1983;Murphy
1987). To be completely analogous to ice-stream flow, thrust e=r[T/ T*] (1)
surges would occur when a sheet had pre-existing, a suffi-
ciently deformable stratum composing its footwall, or if such where r is a unit strain rate, and T* is a material constant with
a deformable layer were generated during emplacement by the units of stress. The power law stress exponent, n, is a
strain accumulation or the influx of fluids. Inferring that dimensionless material constant with a single value through-
processes like these occur during thrusting is reasonable, but out the layer. When n = °°, strain rates are zero when x < T *
this inference cannot be tested directly. and finite when x = x * . Thus, x* is a plastic layer's yield
Mesoscopic and microscopic structures in external thrust strength.
zones indicate that sizeable fractions of the total deformation Individual strain rate components relate to deviatoric
accrued by quasi-viscous creep. It is, therefore, appropriate stress components (where i is the coordinate direction in
to examine analytically viscous deformation under conditions which a traction vector acts and j the direction of the outward
appropriate to thrust emplacement. Such an analysis will normal to the surface on which the traction vector acts)
provide important insight to the overall conditions by which according to
thrust sheets move.
e.. ={d\./dx. + dv./dx.)/2=Xa'..
y 1 J J 1 iJ (2)
Model for continuous flow in external thrust zones V. are velocity components, a', are deviatoric stress compo-
nents, and X = T x<"-'V(x*)"
To examine the viscous flow within external thrust sheets, In a plane-strain treatment of sheet movement (Nye 1957;
consider a model derived from the treatment of power-law Wojtal 1991), stress components in the layer, taken to vary
glacier flow by Nye (1957). In this treatment, a uniform sheet only with x^, are
46
ONE-DIMENSIONAL MODELS FOR POWER-LAW FLOW IN THRUST ZONES
= [T^-y^ (5)
47
S. WOJTAL
to the deformation in horses along thrusts, where thrust zone downward from the upper surface of the sheet, it is harder to
properties change with position along strike, or near active assess the effects of variations in layer rheology. Whereas
thrust tips. To treat sheets that spread or contract laterally and steep gradients, even discontinuities, may exist in velocity
thicken or thin subvertically as they move, consider that the fields (and thus in displacement fields), thrust zones probably
normal stress acting on the Xj-x^ plane does not equal (0,1 + cannot support steep gradients or discontinuities in their
o^^)/2. Take Oj, = pgx^ cosa - 2G, where G = A F and A is a stress distributions. Since the values of the stress components
constant. In this case (see Appendix), the distribution of when n = 1 and when « —> 00 limit the values of stress com-
stresses along a profile across a thrust zone is limited by ponents for other rheologies (Fig. 10), stresses in a thin layer
with a different rheology in a package of otherwise uniform
21 -(1/2)
[T^ - (y')'] (6) rheology could determine the stresses elsewhere in the col-
umn. If the rocks that compose external thrust sheets yiitially
Here again, T is the dimensionless second invariant of the exhibit highly non-linear flow and have an initial yield
stress tensor, y' is again a dimensionless distance below the strength of Fo, 0^ would essentially be given by the curve
upper surface defined using the value of the second invariant extending down from Fo in Figure 10 for a sheet undergoing
of the stress tensor at x^ = 0 (see Appendix). The general form compressing flow. If the thrust zone developed, as a result of
of the stress solution is identical to that for plane flow outlined microstructural and structural changes, a lower bulk strength
by Nye (1957), but the dimensionless distances are not (F|0) and a linear flow law, a,, would then be constrained by
identical (y' = [3/B2]<'«> y, where B- = 4A2 - 4A + 4). Well- the curves extending downward from F^o. In the vicinity of
defined stress solutions exist for all x,>/' ( = [3/B^]'"^'/). the thrust zone (near depth = y), a,, could become more
y' = y when A = 1/2 and Ojj = (a, j + '^^T^I'^- The value of the compressive as a result of the change in rheology. Since o^^
factor [3/B^]'"^'when A = 1/2, unity, is its minimum value. and a,, are unchanged, the orientation of the principal com-
When A is either greater than or less than 1/2, the factor [3/ pression would become more nearly parallel to the thrust
B^]'"^* is less than one. Thus, for perfectly plastic flows with zone boundaries. The shear stress at depth y would, in a layer
equal yield strengths, stress solutions for non-plane flows are with low viscosity, localize shear strains due to a larger value
well-defined for shallower depths than those for plane flows. of the velocity gradient dv.Jdx^. At shallower depths, 0,1
would become less compressive than the value required for
compressing flow, and nearly plug flow in the nearly plastic
Analyzing flow in external thrust zones sheet would prevail. The overall evolution of the thrust zone
would resemble, in this case, a surging glacier. A change in
Due to their simple formulation, the plane and non-plane flow the rheology of thrust-zone rocks may then have profound
solutions for stress and velocity presented here cannot model implications for the stress and velocity distributions through-
precisely how stresses and velocities vary throughout moving out the sheet.
thrust sheets. Flow in thrust sheets is rarely laminar, and
episodes of compressing or extending flow will alter the Stress
boundary conditions, which, by thickening or thinning the
sheet, alter in turn the driving forces, which further alters the
flow, etc. The solutions are, moreover, one-dimensional, i.e.
stresses and velocities vary only with distance below the
upper surface. One could, however, join solutions sequentially,
incorporating variations in boundary conditions into the
externally applied load parameter, F. Thus, compressing-
flow stresses and velocities, calculated using one functional
form for F, could model the flow at one instant within a
particular material column. Later in the history of the sheet,
Q.
extending-flow stresses and velocities, calculated using a <1>
Q
different functional form of F, could prevail in that same
material column. Viewed in this way, as predictors of
instantaneous stress and velocity profiles across a thrust zone,
the model provides some insight to thrust motion.
Consider first, since thrust sheets are rarely Theologically
uniform, the effects of differences in rheology on velocity and Figure 10. Plot showing a variation of stress components in compressing
displacement profiles. In laminar flow, stress components flow. Oijand ©22 vary onlywith depth, a,, depends on the value of the stress
vary with depth only, so differences in layer rheology will not exponent for the sheet, but it is must lie between the straight line (when n =
affect the values of stress components. The geometry of a 1) and quadrant of an ellipse (when « —^ oc) emanating downward from Fo
loose line in a sheet composed of layers with different stress (region shown by right-to-left diagonal ruling). Velocities in perfectly
exponents n will, however, depend on the layer's stress plastic sheets are well-defined for all depths shallower than y. If the strength
of the sheet drops toF,o, 0 | | will fall in the region outlined by the left-to-
exponent (Fig. 9). In compressing or extending flows where right diagonal ruling, and velocities in plastic sheets will be defined for all
stresses and velocities must be calculated by integrating depths shallower than y,. See text for more explanation.
48
ONE-DIMENSIONAL MODELS FOR POWER-LAW FLOW IN THRUST ZONES
^ t Jit
X
(I22C1
Figure 11. Mohr circles for stress for the Cumberland Plateau thrust zone, showing only the geometric relationships between Op, the parameter F from
the plane flow model, and the radius of the stress Mohr circle, Ra, during thrust-zone shortening episodes (a) and thrust zone elongation episodes (b). Compressive
stresses are negative. Palaeostress tensors calculated from mesoscopic fault surface and slickenside measurements fix the angle between the principal stress
directions and the thrust surface. As outlined in the text and in Wojtal (1991), one can calculate that Ra = 20 MPa by estimating overburden at 1.2 km and
estimating surface slope at 1°.
Combining field data on thrust zone stresses with this mately equal in shortening and elongation. Such strength
analysis provides an estimate of bulk sheet properties during isotropy for layered sedimentary rocks is, in the author's
thrust emplacement. At exposures of the base of the southern view, curious. The inferred sheet strength, 6 to 20 MPa, is
Appalachian Cumberland Plateau thrust sheet along the west roughly equal to the stress drops observed in shallow earth-
edge of the southern Appalachian fold-thrust belt, a zone of quakes (Sibson 1989).
faulted strata nearly 100 m thick composes the thrust zone If this thrust-zone deformation reflected 'far-field' stresses,
(Wojtal 1986). The thrust zone shortened and elongated and if those stresses varied smoothly across this thrust zone,
successively as the sheet moved. Principal stress directions stresses in strata beneath this local detachment should con-
during shortening and elongation episodes were essentially form generally with this simple plane flow solution for all
parallel to the thrust zone boundaries (Fig. 2). The pitch of points above a depth /, calculated using the 6 to 20 MPa yield
the stress principal directions in the a-c plane and the relation- stress here to be 6 to 12 km. Provided the plane flow solution
ships between 0| 1,0^2, and 0|, derived above, fix the relative held, detachments along subjacent weak horizons could oc-
magnitudes of F and the radius of a two dimensional stress cur, with stress principal directions in the vicinity of succes-
Mohr circle for the thrust zone (Fig. ll)(Wojtal 1991). Using sively lower detachments approaching 45° angles to thrust
conodont colour alteration index trends to infer an approxi- zone boundaries as the depths of detachments approach the
mate sheet thickness of 1 to 2 km and an approximate surface limiting depth, / = 6 to 12 km.
slope of 1° to 2°, these geometric relationships suggest that The southern Appalachian Hunter Valley thrust zone
at the base of this sheet (1) the parameter F had a magnitude originated at greater depths and accommodated larger net
of 3 to 10 MPa, and (2) the maximum differential stresses displacement than the Cumberland Plateau thrust zone, yet
were of the order of 6 to 20 MPa. The geometry of stresses principal stress directions during later shearing increments
indicates that the parameter F and the maximum differential were apparently also aligned nearly parallel to thrust zone
stresses had comparable magnitudes during shortening and boundaries (Wojtal & Pershing 1991). The deformation
elongation episodes (Fig. 11). during later movement increments in this, or any, large
This argument indicates that the magnitude of shear stress displacement thrust zone need not resemble that during the
resolved parallel to the base of this sheet was small relative to early history of the zone. Large thrust displacement may go
the parameter F. The pitches of stress principal directions hand in hand with the generation of a weak basal layer, like
change little with distance above the base of the sheet. Both a banded cataclasite, that allows stress principal directions to
observations suggest, not surprisingly, that the lateral move- rotate parallel to the thrust zone boundaries and allows
ment and attendant deformation of this sheet are not consist- resolved shear stresses across thrust zones to decrease.
ent with uniform sheet rheology. Shearing was localized in Moreover, erosional unroofing may reduce the overburden
a laterally-continuous coal bed, which is now the base of the and alter the position of a thrust zone relative to the upper
sheet, and thrust-related deformation accrued mainly in re- surface of the sheet. Only by examining the thrust-zone
sponse to stresses transmitted to this region by the surround- structures in the vicinity of more deeply-buried thrust tips,
ing sheet. These stresses include the 'push from behind' where total slip is small, can one determine whether external
responsible for the thrust-zone shortening episodes, and a thrusts ever are, like their higher temperature and pressure
'pull forward' responsible for the thrust-zone elongation counterparts, surfaces of high resolved shear stress.
episodes. The inferred strength of this sheet was approxi- The analysis of non-plane flow presented here suggests
49
S. WOJTAL
that non-plane flow may have significant consequences in the model generates profiles for stresses and velocities in
exactly those thrust zone settings where total slip is small. extending or compressing flows. Stress and velocity profiles
Consider, for example, thrust initiation in a package of plastic calculated with the model are, in essence, 'snapshots' of
sedimentary strata containing several weak layers. In the stress and velocity for flows that sequentially shorten and
centre of the incipient thrust mass, where stresses and strains elongate as they shear. These snapshots suggest that differ-
conform with plane flow, the pitches of stress principal ence in layer rheology may, in laminar flow, produce loose
directions within the a-c plane, will approach 45° angles to line geometries that change from layer to layer in a stratified
the layer boundaries and shear strain rates on planes parallel sheet. In compressing flow, changes in layer rheology due to
to layering will approach maximum values at depths of x, = microstructural or structural changes can alter subsequent
/. In the vicinity of thrust tips, where thrust-related shearing stress and velocity distributions in the thrust zone and the
departs from plane flow, the pitches of stress principal direc- sheet above it, generating in effect a surge-like behaviour in
tions within the a-c plane, will approach 45° angles to the the sheet.
layer boundaries and shear strain rates on planes parallel to Comparing calculated stress and velocity profiles data on
layering will approach maximum values at depths of x^ = /' - the stress orientations during discrete episodes of shortening
[3/B2]('«> /, where [3/B-l<"« < 1. Decollements might, then, or elongation episodes in the southern Appalachian
initiate at shallower depths where non-plane flow conditions Cumberland Plateau thrust zone suggests that differential
prevail. As a detachment grows in length and the thrust tip stresses in the thrust zone did not exceed 20 MPa. Moreover,
migrates away from a particular rock column, local flow body forces due to the dipping upper surface of the sheet were
conditions could evolve toward plane flow. Stress principal not the primary source of these tractions. Additional tractions
directions in the vicinity of the active detachments would that neighbouring portions of the thrust sheet exerted on this
rotate toward orientations parallel and perpendicular to the segment of the thrust zone were on the order of 10 MPa, with
active detachment during such an evolution. Lateral changes comparable magnitudes during shortening and elongation
in rock strength like those due to facies changes, even local episodes.
increases in rock strength, could cause then the maximum rate
The analysis further suggests that non-plane flow differs
of shearing parallel to layering to localize at a shallower level
from plane flow in significant ways, and that these differ-
within a particular stratigraphic column. If the strains asso-
ences may affect the distribution of stresses and velocities in
ciated with the early shear increments led to a weakening of
structural settings like the growing tips of thrusts. In highly
rock along the active detachment, an initially stronger rock
non-linear sheets, maximum rates of shear strain parallel to
mass would have led indirectly to a localization of shear strain
layering may occur at shallower depths in non-plane flows
and an eventual weakening now associated with the detach-
than in plane flows. If, as a thrust tip migrates past a rock
ment. A comparable sequence of changes in the pitches (in
column, a non-plane flow evolves into a plane flow and if
the a-c plane) of stress principal directions from 45° to the
early shear strain increments facilitate a weakening of the
thrust zone boundaries toward parallel to and perpendicular
layer experiencing maximum shear strain rates, the later
to the boundaries could occur as the active portion (i.e.
evolution of the detachment may be characterized by stress
slipped area) of a detachment grows in size during the surge.
principal directions in the a-c plane that are nearly parallel
Restricting analyses to plane flow may then, may severely
and perpendicular to the detachment. Both more data on the
limit our understanding of thrust zone evolution.
nature of stresses and strains in the vicinity of thrust zones and
more sophisticated three-dimensional analyses of thrust sys-
tem deformation are needed to more fully understand the
Conclusions nature of thrust zone evolution.
Thrust sheet movement, even in the external portions of fold- I thank Wendy for her patience and support in this project, Mike Bombolakis
thrust belts, occurs by deforming rocks within a zone of finite for kindly sharing his views on thrust zone mechanics, and Peter Hudleston
thickness, suggesting that analyses of viscous deformation in for giving me a preprint of his comparison of glaciers and thrust sheets.
zones with finite thickness will help elucidate the processes Thought-provoking criticism by Dave Wiltschko, Andy Schedl, Terry
Engelder, and Shankar Mitra helped me to organize this analysis. Comments
by which thrust sheets move. The simple one-dimensional by Rob Knipe and an anonymous reviewer helped me to clarify further the
model of power-law flow between parallel top and bottom manuscript. Finally, I thank Ken McClay both for his comments on this work
surfaces presented here provides insight to some deformation and his efforts in organizing the Thrust Tectonics 1990 Conference and
within thrust zones. With driving forces that include a publishing the conference proceedings.
dipping upper surface and supplemental horizontal stresses.
50
ONE-DIMENSIONAL MODELS FOR POWER-LAW FLOW IN THRUST ZONES
References
Arboleya, M. L. 1989. Fault rocks of the EslaThrust (Cantabrian Mountains, 1989b. Ice-shelf response to Ice-stream discharge fluctuations: III. The
N Spain - an example of foliated cataclasites. Annates Tectonicae, 3,99- effects of Ice-stream imbalance on the Ross Ice Shelf, Antarctica.
109. Journal ofGlaciology, 35, 38-42.
Beach, A. 1982. Deformation mechanisms in some cover thrust sheets from , Bindschadler, R. A., Jezek, K. C. & Shabtaie, S. 1988. Can relict
the external French Alps. Journal of Structural Geology, 4, 137-149. crevasse plumes on Antarctic ice shelves reveal a history of ice-stream
Bentley,C.R. 1987. Antarctic Ice streams: A review.you/'na/o/Geo/j/ivi/ca/ fluctuation? Annals glaciology, 11, 77-82.
Research, 92, 8843-8858. Marshall, D.J. 1988. Cathodolumlnescence of geological materials. Unwin
Bielenstein, H. U. 1969. The Rundle thru.st sheet, Banff, Alberta. Ph.D. Hyman, Boston.
Dissertation, Queen's University, Kingston, Ontario. Mitra, G. 1984. Brittle to ductile transition due to large strains along the
Bombolakis, E. G. 1989. Thrust fault mechanics and dynamics during a White Rock thrust. Wind River Mountains, Wyoming. Journal of
developmental stage of a foreland belt. Journal of Structural Geology, Structural Geology, 6,51-61.
11,439-456. & Elliott, D. 1980. Deformation of basement in the Blue Ridge and the
Chappie, W. M. 1978. Mechanics of thin-skinned fold and thrust belts. development of the South Mountain cleavage. In: Wones, D. R. (ed.) The
Geological Society of America Bulletin, 89, 1189-1198. Caledonides in the USA. Department of Geological Sciences Virginia
Coward, M. P. 1982. Surge zones in the Moine thrust zone of NW Scotland. Polytechnic Institute and State University Memoirs, 2, 307-311.
Journal of Structural Geology, 4, 247-256. Mitra, S. 1986. Duplex structures and imbricate thrust systems: Geometry,
Crossland, A. & Wojtal, S. 1989. Rheological implications of intragranular structural position, and hydrocarbon potential. Bulletin of the American
fractures in cataclasites from foreland fold-thrust belt thrusts (abst.). Association of Petroleum Geologists, 70, 1087-1112.
Program of the Geological Society of America, 21, A135. Mullenax, A. C. & Gray, D.R.I 984. Interaction of bed-parallel .stylolites and
Davis, D., Suppe, J. & Dahlen, F. A. 1983. Mechanics of fold and thrust belts extension veins in boudinage. Journal of Structural Geology, 6, 63-72.
and accretionary wedges. Journal of Geophysical Research, 88, 1153- Murphy, D. C. 1987. Suprastructure/infrastructure transition, east-central
1172. Cariboo Mountains, British Columbia: geometry, kinematics, and tec-
Elliott, D. 1973. Diffusion flow laws in metamorphic rocks. Geological Society tonic implications. Journal of Structural Geology, 9, 13-29.
of America Bulletin. 84, 2645-2664. Nye, J. F. 1957. The distribution of stress and velocity in glaciers and ice
1976a. The motion of thrust sheets. Journal of Geophysical Research, sheets. Proceedings of the Royal Society of London, A239, 113-133.
81, 949-963. O'Hara, K., Hower, J. C. & Rimmer, S. M. 1990. Constraints on the
1976b. The energy balance and deformation mechanisms of thrust emplacement and uplift history of the Pine Mountain thrust sheet, eastern
sheeta. Philosophical Transactions ofthe Royal Society ofLondon, A2S3, Kentucky: Evidence from coal rank trends. Journal of Geology, 98,43-
289-312. 51.
Emerman, S. H. & Turcotte, D. L. 1983. A fluid model for the shape of Paterson, W. S. B. 1981. The physics of glaciers. 2nd ed. Pergamon Press,
accretionary wedges. Earth and Planetary Science Letters, 63,379-384. Oxford.
Engelder, J. T. 1974. Cataciasis and the generation of fault gouge. Geological Piatt, J. P. 1986. Dynamics of orogenic wedges and the uplift of high-
Society of America Bulletin, 85, 1515-1522. pressure metamorphic rocks. Geo/o,?/ca/Sf«7>rvo//lmenfaBu//fr/n, 97,
Erickson, S. G. & Wiltschko, D. V. 1986. Analytical models of an a.sperity 10.37-1053.
onuthmstfiurface i'dbsL). Program ofthe Geological Society of America, & Leggett, J. K, 1986. Stratal extension in thrust footwalls, Makran
18, 595. accretionary prism: implications for thrust tectonics. Bulletin of the
Fletcher, R. C. 1989. Approximate analytical solutions for a cohesive fold- American Association of Petroleum Geologists, 70, 191-203.
and-thrust wedge: Some results for lateral variation in wedge properties , Leggett, J. K. & Alam, S. 1987. Slip vectors and fault mechanics in
and for fine wedge angle. Journal of Geophysical Research, 94, 10,347- the Makran accretionary wedge, southwest Pakistan. Journal of Geo-
10,354. physical Research, 93, 7955-7973.
Geiser, P. A. 1988. The role of kinematics in the construction and analysis Price, R. A. 1967. The tectonic significance of mesoscopic subfabrics in the
of geological cross sections in deformed terranes.//?.' Mitra, G. & Wojtal, southern Canadian Rockies of Alberta and British Columbia. Canadian
S. (eds) Geometries and mechanisms of thrusting, with special reference Journal of Earth Science, 4, 39-70.
to the Appaiach'ianfi. Special Paper of the Geological Society of America, 1988. The mechanical paradox of large overthrusts. Geological Society
222, 47-77. of America Bulletin, 100, \S9?,-\90».
Gilotti, J. & Kumpulainen, R. 1986. Strain-softening induced ductile flow in Ramsay, J. G., Casey, M. & Kligfield, R. 1983. Role of shear in the
the Sarv thrust sheet, Scandinavian Caledonides. Journal of Structural development of the Helvetic fold-and-thrust belt of Switzerland. Geol-
Geology, 8, 44-56. ogy, 11, 439-442.
Harris, L. D. & Milici, R. C. 1977. Characteristics of thin-skinned style of Rutter, E. H. 1976. The kinetics of rock deformation by pressure solution.
deformation in the southern Appalachians, and potential hydrocarbon Philosophical Transactions of the Royal Society of London, A283, 203-
traps. Professional Paper US Geological Survey 1018. 219.
House, W. M. & Gray, D. R. 1982. Cataclasites along the Saltville thrust, , Maddock, R. H., Hall, S. H. & White, S. H. 1986. Comparative
USA and their implications for thrust-sheet emplacement. Journal of microstructures of natural and experimentally produced clay-bearing
Structural Geology, 4, 257-269. fault gouges. Pure and Applied Geophysics, 124, 3-30.
Hsu, K. 1969. A preliminary analysis of the statics and kinetics of the Glarus Rye, D. M. & Bradbury, H. J. 1988. Fluid flow in the crust: An example from
overthrust. Eclogae Geologicae Helvetiae, 62, 143-154. a Pyreanean thrust ramp. American Journal of Science, 288, 197-235.
Hubbert, M. K. & Rubey, W. W. 1959. Role of fluid pressure in mechanics Schmid, S. M. 1975. The Glarus overthrust: Field evidence and mechanical
of overthrust faulting. Geological Society of America Bulletin, 70, 115- model. Eclogae Geologicae Helvetiae, 68, 247-280.
166. Sharp, M., Lawson, W. & Anderson, R. S. 1988 Tectonic processes in a
Hudleston, P. J. 1991. A comparison between glacial movement and thrust surge-type glacier. Journal of Structural Geology, 10, 499-516.
sheet or nappe emplacement and associated structures. In: Mitra, S. & Sibson, R. H. 1989. Earthquake faulting as a structural process. Journal of
Fisher, G. W. (eds) Structural Geology of Fold and Thrust Belts (Elliott Structural Geology, 11, 1-14.
Volume). Johns Hopkins University Press, Baltimore (in press). Siddans, A. W. B. 1984. Thrust tectonics, a mechanistic view from the West
Kehle, R. O. 1970. Analysis of gravity sliding and orogenic translation. and Central Alps. Tectonophysics, 104, 257-281.
Geological Society of America Bulletin, 81, 1641-1664. Stockmal, G. S. 1983. Modeling of large scale accretionary wedge deforma-
Logan, J. M., Higgs, N. G. & Friedman, M. 1981 Laboratory studies on tion. Journal of Geophysical Research, 88, 8271-8287.
natural gouge from the US Geological Survey Dry Lake Valley No. 1 Wilt.schko, D. V. & Budai, J. M. 1988. A model for fluid motion through
well, San Andreas fault zone. American Geophysical Union of Geo- layered rock: Evidence from the Idaho-Wyoming thrust belt (abst.).
physics Monograph, 24, 121 -134. EOS, Transactions of the American Geophysical Union, 69, 484.
MacAyeal, D. R. 1989a. Large-scale ice flow over a viscous basal sediment: Wojtal, S. 1986. Deformation within foreland thrust sheets by populations of
Theory and application to Ice Stream B, Antarctica. Journal of Geo- minor faults. Journal of Structural Geology, 8, 341-360.
physical Research, 94, 4071 -4087. 1989. Measuring displacement gradients and strains in faulted rocks.
Journal of Structural Geology, 11, 669-678.
51
S. WOJTAL
— 1991. Shortening and elongation of thrust zones within the Appala- & Pershing, J. 1991. Paleostresses associated with faults of large offset.
chian foreland fold-thrust belt. In: Mitra, S. & Fisher, G. W. (eds) Journal of Structural Geology , 13, 49-62.
Structural Geology of Fold and Thrust Belts (Elliott Volume). Johns Woodward, N. B., 1988. Geological applicability of critical-wedge thrust-
Hopkins University Press, Baltimore (in press). belt models. Geological Society of America Bulletin, 99, 827-832.
— & Mitra, G. 1986. Strain hardening and strain softening in fault zones , Wojtal,S.,Paul, J. B.&Zadins,Z.Z.1988.Partitioning of deformation
from foreland thrusts. Geological Society of America Bulletin, 97, 674- within several external thrust zones of the Appalachians. Journal of
687. Geology, 96, 35\-36\.
— & Mitra, G. 1988. Nature of deformation in fault rocks from Appala-
chian thrusts. In: Mitra, G. & Wojtal, S. (eds) Geometries and mecha-
nisms of thrusting, with special reference to the Appalachians. Special
Paper of the Geological Society of America, 222, 17-33.
Appendix
To narrow the considerable latitude in choices for the func- Differentiating with respect to Xj indicates that
tional form of F, choose dwjd\^=0. Invoking this condition
in equations (2) and combining them to eliminate X, Nye d^wjc)x^- = - B-Vj/Bx^Bx, (16)
(1957) showed that
and
>. = a/F = a/[x^ - (pgx,sina)2]"'2' (II) d/dx.[X(¥ + G)]/3 = a/ax, [ >i (2G - F)]/3 (17)
52
Centrifuge modelling of the propagation of thrust faults
John M. Dixon & Shumin Liu
In fold-thrust belts and accretionary prisms the horizontal widely to fold-thrust belts and accretionary prisms and be-
shortening and vertical thickening ofthe stratigraphic pile is came for a time the leading fold-thrust paradigm. In this
accommodated by layer-parallel shortening, folding and low- model, the thrust is clearly implied to develop first, and the
angle overthrust faulting. The relative importance of these fold is a product of passive bending. The model has the
mechanisms depends largely on the rheological properties of advantage of being easy to analyse in a rigorous geometric
the stratigraphic sequence, and the interference among them way, but it involves a large number of implicit assumptions
plays a major role in the structural evolution of fold-thrust (Ramsay 1990) that are valid in natural fold-thrust systems.
belts. The relationship between folds and thrusts has in Another genetic link between folds and thrusts is repre-
particular received a great deal of attention. For example, sented by the fault-propagation model (Williams & Chapman
Rogers & Rogers (1894), Heim (1878), Willis (1894), and 1983; Suppe & Medwedeff 1984; Suppe 1985). A fault-
other workers of that era viewed thrust faults as having propagation fold develops simultaneously with propagation
formed from pre-existing buckle folds caused by layer- of the ramp portion of a stepped thrust fault; the fold directly
parallel compression, the faults often propagating from breaks overlies the ramp and grows in amplitude as the thrust
through fold limbs in competent beds. In contrast, Buxtorf propagates across the bedding.
(1915) concluded that low-angle reverse faulting can produce A detachment fold (Jamison 1987) also develops at the
associated drag folds. termination of a propagating thrust, but differs from a fault-
Folds can also be formed passively by translation of a propagation fold in that it is not related to a ramp. This kind
thrust sheet over a ramp. Rich (1934) first developed the of fault-related fold represents localized shortening of the
'ramp-anticline' or 'fault-bend fold' theory to interpret the thrust plate above the tip of a bedding-parallel fault or zone
Powell Valley anticline (Virginia, USA) as a product of of decollement.
displacement of a thrust sheet on a fault with a trajectory that In all of these relationships between folding and thrusting,
is in part parallel to bedding and in part ramping upwards with the exception ofthe first, folding is either a consequence
across bedding. This mechanism has been analysed geo- of or concurrent with thrusting; the faulting (be it localized on
metrically (Boyer & Elliott 1982; Suppe 1983) and mechani- a discrete planar fracture or distributed in a zone of shear or
cally (Berger & Johnson 1980). It has been applied very decollement) is inferred to be the primary phenomenon, and
53
J. M. DIXON & S. Liu
Table 1. Model ratios applicable to models TH16, TH18, TH20, TH22, TH23 and TH24
folding a secondary consequence. One problem with this that early-formed faults continue to accumulate displacement
hierarchy is that it provides no obvious mechanical explana- even though later-formed faults have developed closer to the
tion for the localization of thrust ramps. foreland; thus the fold-thrust wedge deforms throughout
In the authors' opinion, the role of faulting has been over- while the toe is propagating.
emphasized. Furthermore, recent theoretical (e.g. Davis e? a/. Two previous papers (Liu & Dixon 1990, 1991) have
1983 and others) and analogue (e.g. Mulugeta & Koyi 1988) discussed the sequence of thrusting and along-strike struc-
model work has tended to focus on faulting and ignore the role tural variation (displacement transfer) in fold-thrust belts.
of folding because thrust belts have been assumed to consist This paper summarizes the observations of the three-dimen-
of a homogeneous material and bedding anisotropy has been sional interaction between folds and thrusts and the propaga-
neglected. On the basis of analogue model studies of the tion of thrust faults in duplex structures. The results from
interaction between buckling and faulting (and layer-parallel models TH16, TH18, TH20, TH22, TH23, and TH24 are
shortening), Dixon & Tirrul (1991) and Liu & Dixon (1990, discussed. The last four models have been discussed in the
1991) have proposed that early buckling play s a fundamental previous papers but here more details of the structural rela-
role in localizing thrust ramps and in guiding along-strike tionships in new three-dimensional diagrams are shown. The
propagation of thrusts, and thus may control the formation of stratigraphic sequences in models TH16 and TH18 differ in
duplex structures. In this regard, a return to the hierarchy detail from that of the other models, but the relationships are
proposed by Rogers & Rogers (1894), Heim (1878) and consistent.
others is proposed.
Folds and thrusts nucleate serially and propagate spatially
from hinterland to foreland, and the conventional view has Experimental procedure
been that early-formed structures lock up as younger ones
develop at the toe of a growing fold-thrust belt (e.g. Boyer & The centrifuge in the Experimental Tectonics Lab at Queen's
Elliott 1982). On the other hand, the critical Coulomb wedge University is capable of subjecting two specimens to
analysis of Davis et al. (1983) predicts that in order to main- accelerations as high as 20,000 g (see Dixon & Summers
tain a critical taper while new material is accreting to the toe (1985) for a detailed description of the machine). Readers are
of the fold-thrust belt, the wedge must continue to deform referred to Ramberg (1981) for a detailed discussion of the
internally (for example by slip on thrust faults throughout its theory of scale modelling and the centrifuge technique.
volume). The model studies of Liu & Dixon (1990) confirm The models are constructed of analogue materials. Plas-
ticine (Harbutt's Gold Medal Brand) and silicone putty (Dow
Coming dilatant compound 3179) have the appropriate
rheological properties to represent, respectively, competent
rocks such as limestone and sandstone, and incompetent
rocks such as shale, under natural conditions typical of the
upper part of the Earth's crust. The rheological similitude
achieved with these materials and the method of construction
of finely-laminated multilayers of the two materials have
been discussed in detail elsewhere (Dixon & Summers 1985;
Dixon & Tirrul 1991). It will suffice to repeat here that an
internally-laminated stratigraphic unit of a desired bulk com-
petency can be prepared by repeated rolling and stacking of
an initial couplet of layers of the two materials whose thickness
ratio corresponds to the thickness ratio desired for the unit.
Competent stratigraphic units have a higher proportion of
plasticine, whereas less-competent units contain more silicone
putty. Scale-model ratios chosen for the experiments described
here are listed in Table 1. Note that 1 mm in the model
represents 1 km in the prototype.
Figure 1. Dimensions and initial configuration of the models. The dimensions and initial geometric configuration of the
54
CENTRIFUGE MODELLING OF THE PROPAGATION OF THRUST FAULTS
models are illustrated in Figure 1, although the foreland The model stratigraphic units simulate rock units such as
stratigraphic sequence varies from one model to another. The limestone and shale with prototype thicknesses of 1670,1330
models all consist of a six-part stratigraphic sequence (Fig. or 1000 m, and 330 m, respectively. The units are internally
2), three competent and three incompetent units. For conven- laminated so that they simulate to some extent the bedding
ience these will be referred to as Units I—VI from top to anisotropy of natural sedimentary units. The model
bottom. Competent units I, III and V each contain internal stratigraphic pile represents a prototype which has 6-fold
laminae of plasticine without silicone putty interlayers. The stratigraphy with alternating competent and incompetent
competent units in model THl6 (Fig. 2a) are each 1.67 mm units. An example of such a sequence is provided by the
thick and contain 10 layers of blue and red plasticine, and in succession in the Appalachian Valley and Ridge Province in
model TH18 (Fig. 2b) they are each 1.33 mm thick and West Virginia, Virginia and Maryland (Parry 1978; Kulander
contain lOlayersofyellowandblueplasticine. Thecompetent & Dean 1986).
units in models TH22, TH23 and TH24 (Fig. 2c) contain four The models are subjected to horizontal compression from
laminae of plasticine of four different colours (yellow, red, one end by gravitational collapse and spreading of a 'hinter-
black and blue from top to bottom), while in model TH20 land wedge' of plasticine (see Fig. 1). The wedge begins to
(Fig. 2d) they contain 8 laminae, with a total unit thickness of fail when the acceleration in the centrifuge reaches about
1.0 mm. In all six models the incompetent units II, IV and VI 2500 g. As the wedge collapses, it loses gravitational poten-
each contain four laminae, two of silicone putty and two of tial; the acceleration must be gradually increased (to 4000 g
black plasticine in a 2:1 ratio of thickness. The total thickness in these experiments) to maintain the spreading. This in-
of the model stratigraphic pile is 6 mm in THl6; 5 mm in crease in acceleration causes a proportional (minor) change in
THl8; and 4 mm in models TH20, TH22, TH23 and TH24 the model scaling. The experiments are run in stages so that
(see Fig. 2). There is minor variance in strength between the progressive development of structures can be monitored.
plasticines of different colour, but no structural manifestations Some models are sectioned after each stage of deformation,
of this variance have been detected. parallel to the shortening direction, are photographed and
Figure 2. Stratigraphic sequences in the foreland portions of models TH16 (a), TH18 (b), H22, TH23, TH24 (c) and TH20 (d).
55
J. M. DIXON & S. Liu
Figure 3. Profile sections through model TH 16 at stage I through stage V. Thepositionsof sections relative to one edge of the model are indicated in inches
on the labels. The labels measure 10 mm \ 10 mm.
then reassembled for a subsequent stage of deformation. tapered slice of plasticine to its surface. This process is
Other models are run through several stages of deformation repeated after each stage until the whole deformation se-
without sectioning, and then sectioned horizontally or trans- quence is finished.
versely to the shortening direction at the final stage. For each An analysis of the magnitude of the stress applied to the
new stage the driving wedge is rejuvenated by addition of a edge of the model fold-thrust belt in similar models has been
56
CENTRIFUGE MODELLING OF THE PROPAGATION OF THRUST FAULTS
presented elsewhere (Dixon & Tirrul 1991). With a more competent units and localized groups of small folds in the
gently-tapered 'hinterland wedge' and with a 10 mm foreland incompetent units. The small folds in the incompetent units
stratigraphic pile the models commence to yield under hori- are localized beneath the anticlines in the competent units.
zontal compression estimated at about 2.3 x 10* Pa, which The folds nucleate serially, from hinterland towards foreland,
corresponds to a prototype stress of about 3.0 MPa. This and grow progressively in amplitude. By stage III, one thrust
value is below, but probably within one order of magnitude (labelled 'a' in Fig. 4) has ramped upward across the compe-
of, realistic levels of differential stress required to drive tent unit V in the position of the front limb of an earlier-stage
deformation of thrust belts (see Dixon & Tirrul (1991) for low-amplitude fold (fold 'a' in Fig. 4, stage II), and an
detailed discussion). The stress levels in the models de- incipient thrust ('b') has begun to propagate through the
scribed here are similar. lower part of unit V in the position of the front limb of another
fold ('b') that had formed in stage II. This suggests the serial
nucleation and progressive growth of thrust faults. By this
Nucleation and across-strike propagation of stage (III), in addition to the two thrusts, more low-amplitude
thrusts folds (folds 'c', 'd' and 'e') have formed in unit V in front of
the thrusts.
Sequence of thrust nucleation During stage IV (Fig. 4), the folds continue to increase
The progressive evolution of model TH16 was monitored by amplitude in unit I and unit III. New thrusts propagate across
cutting profile sections parallel to the shortening direction unit V in the front limbs of low-amplitude folds (folds 'c', 'd'
after each stage of the shortening. These are shown in and 'e' of stage III) on the forelandsideof those rampsformed
photographs in Figure 3 and in line drawings in Figure 4a. For at earlier stages, but the old thrusts continue to increase their
each section the Roman numeral refers to the model stage, displacements. Note that thrust 'e', like thrust 'b' at stage III,
and the Arabic number indicates the position of the line of is only an incipient thrust; this, as well, suggests the sequence
section relative to one edge of the model (distance in inches). of thrusting.
As demonstrated in other models (Dixon & Tirrul 1991; Liu At stage IV, Figure 3 shows that the groups of small folds
& Dixon 1990), the early stages of deformation (stages I and in the incompetent units are evenly spaced and are clearly
II) are characterized by harmonic buckle-fold trains in the localized beneath the anticlines of the overlying competent
units. Coincident with this (Figs 3 & 4a), the spacing of
thrusts in competent unit V is also constant. By following the
deformation sequence, it is easy to see that the thrusts have a
constant spacing because they are localized by the earlier
folding (Dixon & Tirrul 1991; Liu & Dixon 1990) which has
a buckling wavelength dependent on the material properties
and thicknesses of the mechanical units involved in the
folding process (e.g. Ramberg 1963).
Stage
rVI
n
TTT^
57
Figure 6. Photographs of the top (free) surface of model TH16 after each of stages I through V. showing the pattern of serial nucleation and foreland propagation of folds. The 'hinterland wedge' of plasticine (white
and t 7 s " r r i n mTrker "' '" ' ^ ' ''^'' "^"^ " ' ° ' ' " ' ' ' " " " ' " ' ^ ' ^ "*'''^"''^ ^'"""^ "^^ '^^^ ^^ ^"''^"^^ ^"^^^^ ^^''^f- ^""^ ^ " ^ ^ ^ "^ ^^e model is marked with a print of metric graph paper (cm. mm) for scale
Si.
:s'
CENTRIFUGE MODELLING OF THE PROPAGATION OF THRUST FAULTS
The competent units in model TH16 (Fig. 3) were con- In the models, fault ramps clearly propagate upward
structed with short strips of lighter-coloured (yellow) plasticine through the competent units (see Fig. 4, fault 'b' at stage III
embedded within them in the central portion of the model. and fault 'e' at stage IV). These ramps form at the tip of a
These strips extend across the width of the model (that is, bedding-parallel decollement zone in the underlying incom-
along strike) and serve as passive markers which can be used petent unit. This pattern of upward propagation, from the
to measure relative displacement between the competent incompetent unit into and through the overlying competent
units. At stage 1 the ends of the light-coloured markers in unit, is contrary to Eisenstadt & DePaor's (1987) suggestion
units I, III and V are situated in a vertical row. As shortening that ramps nucleate within competent units and propagate
proceeds, there is progressive shearing of the upper layers both upward and downward into neighbouring weak units.
towards the foreland due to drag against the rigid base of the
model. The differential displacement is accommodated by
decollement in the incompetent units II, IV and VI. Strain partitioning
The time of nucleation and the duration of displacement of Natural fold-thrust belts undergo shortening by three differ-
all the thrusts developed in unit V of model TH16 is docu- ent deformation mechanisms: layer-parallel shortening (LPS),
mented in Figure 4b. The data are obtained by careful buckling and faulting. We have previously analysed the
measurement of the displacement of competent unit V by partitioning of strain among these three mechanisms in the
thrusts after each stage. The plot shows that thrusts nucleate competent units of model TH24 (Liu & Dixon 1991). Using
successively, one by one, toward the foreland, but that an the same methods, it has been possible to reconstruct restored
older thrust does not die out or cease its displacement as a new
thrust forms in front of it. Model TH24 (Fig. 5) evolved in a
similar way (although it contains more thrusts), and shows the
same pattern of serial nucleation but long-lived displacement
of the thrusts. This is not in agreement with the simple
conception that imbricate thrusts develop serially with early-
II
formed thrusts becoming inactive as later-formed ones de-
velop towards the foreland. On the other hand, the continued 1.000
accumulation of displacement on early-formed faults within
the older, hinterland end of the fold-thrust belt is in agreement
with the critical Coulomb wedge model of Davis er a/. (1983).
The nucleation sequence of thrusts from hinterland to fore-
land is inherited from the sequence of development of folds
from hinterland to foreland. The propagation of folds (of
competent unit 1) across the top surface of model TH16 is
shown in Figure 6. 1.125
The evolution of a fold into a thrust can be observed in the
sections at stages II, III, IV and V of model TH24 (Fig. 7).
Note the structure in the location marked by '*' in the
sections. At stage II, the structure is a low-amplitude fold; it
could be classified as a 'detachment' fold (Jamison 1987) for
it overlies a zone of layer-parallel shear or decollement in the
underlying incompetent unit VI. At stage III, the lower part
of unit V is displaced by a propagating thrust ramp; the fold IV
is by now a 'fault-propagation' fold (Suppe & Medwedeff 1.250
1984; Suppe 1985) which grows in amplitude as the fault
propagates. The fault is shown as a solid line in unit V, but
it extends back towards the hinterland as a decollement
horizon within unit VI to the right of the ramp. By stage IV,
the thrust has propagated upward so as to ramp all the way
through unit V and into the overlying incompetent unit IV.
Now the fault follows a flat-ramp-flat trajectory and the
ramp-tlat geometry of the thrust begins to influence the shape
of the fold. By stage V the fold has evolved into a well- V
developed 'fault-bend' fold (Suppe 1983) or 'ramp anticline' 1.375
(Rich 1934; Dahlstrom 1970). Therefore, there is an evolu-
tion of the relationship between fold and thrust: the fold
nucleates first and localizes the thrust ramp in its foreland-
dipping limb; the fold and thrust dominate alternately during Figure 7. Line drawing from magnified photographs of sections through
model TH24 after the stages 0. Ill, IV and V. The structure marked by '*'
the deformation process. is discussed in detail in the text.
59
J. M. DIXON & S. Liu
\ / /
/ \ \ \
X
\ N •x \ 'l 0.625
\ \/ / 1
/ \ \ \ \
\ N N. \ ^ 1] 0.760
60
CENTRIFUGE MODELLING OF THE PROPAGATION OF THRUST FAULTS
Along-strike propagation
61
J. M. DIXON & S. Liu
Figure 11. Three-dimensional view of structures in unit-V duplex of model TH22, drawn from 22 transverse sections. The numbers and letters marking
individual structures (faults and folds) are referred in the text.
Figure 12. Three-dimensional view of structures in unit-V duplex of model TH 18, drawn from 22 transverse sections. The numbers marking individual
structures (faults and folds) are referred in the text.
62
CENTRIFUGE MODELLING OF THE PROPAGATION OF THRUST FAULTS
Figure 11. The section interval is 0.06 inches. The thrust that
can be seen to terminate near 'B' in Fig. 11 is marked '*' in
Fig. 14. The thrust exhibits decreasing displacement from
section 0.81 to 0.75. In sections 0.69 and 0.63 the thrust cuts
only the lower part of unit V and then dies into the core of a
small fold in section 0.56. This mechanism is also evident in
models constructed by Gardner & Spang (1973).
The phenomenon of transfer of displacement between
overlapping, en-echelon thrusts is also present in our models.
Figure 15 shows 8 sections cut through model TH39 (which
had a stratigraphic column similar to that of models TH22,
TH23 and TH24 (see Fig. 1)) after the final stage of deforma-
tion. While thrust 'a' increases its displacement along strike
from section A to section E, thrust 'b' decreases its displace-
ment in the same direction and dies into a fold in section E.
From section F to section H, the sense of displacement
Figure 13. Across-andalong-strike sections of unit Vin model TH18. The transfer reverses. Thrusts 'c' and 'd' have the same kind of
numbers at the right end of the transverse sections indicate the positions of relationship: from section A to section F, thrust 'c' decreases
these sections (in inches) relative to the far edge of the model. Note that some its displacement gradually, finally dying into a fold in section
ramps are arcuate (oblique relative to the right-to-left shortening direction).
See text for discussion.
F where thrust 'd' picks up its displacement.
63
1 J. M. DIXON & S. LIU ' J-
Figure 15. Transverse sections cut through model TH39 after the final stage of deformation. Only the bottom poriion of ihe model is shown, as a horizontal
section had been cut ihrough the level of the unit-V duplex. Thealong-sirike transfer of displacement between faults 'a' and "b'.and between faults 'c' and
'd' is discussed in the text.
menl of these thrusts than of thrusts in model TH22 (Liu & made by Elliott (1976) and in Liu & Dixon (1991): a large
Dixon 1991). However, one common factor emerges: none thrust may nucleate at one point and propagate in both
of the thrusts in either model keeps constant displacement directions along strike. Fortheotherthrusts, the displacement
along its full strike length. As was pointed out above, the profiles each show several maxima and minima. In this
faults have curvilinear traces; but it is apparent that variable model thrust 'c' has a particulariy large displacement of 8.7
displacement is not directly related to sinuousity of trace. mm along section 0.50, and its displacement decreases towards
Like those of thrusts 1 and 7 in model TH22 (Fig. 10 of Liu the other side of the model. In that portion of the model where
& Dixon 1991), the displacement profile of thrust d in model thrust 'c' has large displacement, thrusts 'b', 'd' and 'e' all
TH20 is also roughly bow-shaped with its maximum located have small displacements; these faults grow in the same
near the middle of the model. This fault confirms the point direction that 'c' decreases. This suggests displacement
64
CENTRIFUGE MODELLING OF THE PROPAGATION OF THRUST FAULTS
65
n'i'ij-i !•' jsti' f. J. M . D I X O N & S. Liu i-K-r/ :-r>'i-ij>|-i/.:?0
•-Ht
•"" • h-./i'ini"',
•'Ml : ••
•/ r - v . ,;;.,
•• - .•:: .; -Ji. t
'• f(:'t
66
CENTRIFUGE MODELLING OF THE PROPAGATION OF THRUST FAULTS
jfii' !'];-;
Qum-
0,875 1,375
.•;ri -y,-
1.000 1.500
1-125 1.625
»ii(i;'J
.rCI
1,250
1,750
Figure 19. Linedrawingsofenlargedphotographsof serial transverse sectionscutthrough centre part of model TH22 (Fig, 11), The numbers on the right
show the posilions of the sections {in inches) relative to the far side of the model. Fault 6 cuts the full thickness of unit V in section 0,87.5 but its displacement
decreases along sirike to seclion 1,375 and then increases to section 1,750 where it again cuts the full thickness of unit V, The fault is overlain by a fold
from section 1 ,(M) to seclion 1.625.
67
J. M. DIXON & S. Liu
pattern of folds. If such folds subsequently evolve into thrust faulting evolves through time: first, a 'detachment' fold
ramps, the ramps might also fail to merge but would form an forms above a zone of decollement; second, a fault ramp
overlap relationship with displacement transfer between them. propagates upward into the base of the competent unit at the
Another possibility is the merger of two nearly colinear position of foreland-dipping limb of the early fold, and the
ramps into a single fault with a sinuous, sigmoidal trace, or the fold becomes a 'fault-propagation' fold; third, when the fault
joining oftwo offset ramps by a transverse ramp. An example has propagated through the competent unit its trajectory
of this kind of relationship was documented in model TH23 bends into the overlying incompetent unit and with further
(Fig. 14 in Liu & Dixon (1991)). transport the hangingwall is modified by 'fault-bend' fold-
In summary, large thrusts may be the result of along-strike ing. Thus, folding and faulting alternate as the dominant
propagation and merger of smaller faults that nucleated at process, and bedding anisotropy is particularly important to
different points along the strike length of an earlier fold. The the localization of thrust ramps through its control on buckling
geometric relationship among nearly coplanar thrusts (inter- wavelength. In nature, low-amplitude (even incipient) folds
locking and connecting) depends on the relationship of the may produce sufficient stress concentration to trigger thrust
antecedent folds. faults in their foreland-dipping limbs.
Displacement transfer occurs between pairs of en-echelon
thrust faults, between pairs of imbricate thrusts, and between
Conclusions thrust faults and folds. The transfer accommodates structural
change along strike while the total transverse shortening
The analogue models described here contain a six-fold stays relatively constant.
stratigraphic succession, with units that alternate in bulk Some large thrusts exhibit along-strike variations of thrust
competence (weak at the base). Folds and faults nucleate and displacement with local maxima and minima. This pattern
propagate from hinterland to foreland but early-formed faults suggests that such thrusts may nucleate at different points as
at the hinterland end of the fold-thrust belt continue to coplanar faults, propagate along strike, and eventually join
accumulate displacement as the deformation progresses. Thus together because the nucleation points are all under an
the model fold-thrust belt undergoes continuous thickening anticline which formed by along-strike propagation and
in its hinterland portion while the toe or deformation front merger of colinear, earlier-formed periclinal folds. The
propagates towards the foreland. thrust displacement in such cases varies from smaller values
Horizontal shortening strain in the models is partitioned at linkage points to larger values at nucleation points. Dis-
among three deformation mechanisms: homogeneous, layer- placement transfer through transverse ramps and overlap-
parallel shortening; folding; and imbricate thrust faulting. ping thrusts may occur as a means of linking non-colinear
The relative amount of shortening accommodated by these propagating thrusts which nucleated beneath en-echelon
three mechanisms varies with depth at different stages of folds.
deformation, and varies in time at a given structural level.
Layer-parallel shortening is dominant; in some models it
increases throughout the experiment, but in others it declines Our analogue-modelling investigation of fold-thrust tectonics is supported
in importance while shortening by the other mechanisms in part by a grant from Exploration Research, Arco Oil and Gas Company,
accelerates. The models exhibit along-strike structural vari- The construction and operation of the Experimental Tectonics Laboratory
at Queen's University has been supported by grants from the Natural
ation and transfer of displacement between faults and folds. Sciences and Engineering Research Council of Canada to J. M. Dixon. S.
In the models it is clear that fault ramps nucleate at points Liu gratefully acknowledges scholarship support in the course of his Ph.D.
that are localized by earlier folding of competent units: a ramp studies and a travelling grant to the Thrust Tectonics 1990 Conference
granted by Queen's University.
propagates upward through the foreland-dipping limb of a
fold from a bedding-parallel decollement zone in an underly-
ing incompetent unit. The relationship between folding and
References
Berger, P. & Johnson, A. M. 1980. First-order analysis of deformation of a Davis, D., Suppe, J. & Dahlen, F. A. 1983. Mechanics of fold-and-thrust
thrust sheet moving over a ramp. Teclonophysics, 70, T9-T24. belts and accretionary wedges. Journal of Geophysical Research, 88,w
Boyer, S. E. & Elliott, D. 1982. Thrust systems. American Association of 1153-1172.
Petroleum Geologists Bulletin, 66, 1196-1230. Dixon, J. M. & Summers, J. M. 1985. Recent developments in centrifuge
Buxtorf, A. 1915. Prognosen und Befunde Beim Hauenstein Basis und modelling of tectonic processes: equipment, model construction tech-
Grenchenbergtunnel und die Bedeutung der Letzteren fur die Geologic niques and rheology of model materials. Journal of Structural Geology,
des Juragebirges. Basel: Verhandlungen des Naturforschunden die 7,83-102.
Gesellschafi, 27. & Tirrul, R. 1991. Centrifuge modelling of fold-thrust structures in a
Dahlstrom.C.D.A. l969.Balancedcrosssections.Canadian Journal ofEarth tripartite stratigraphic succession-. Journal ofStructural Geology, 13,3-
Sciences, 6, 743-757. 20.
1970. Structural geology in the eastern margin of the Canadian Rocky Eisenstadt, G. & DePaor, D. C. 1987. Alternative model of thrust-fault
Mountains. Canadian Society ofPetroleum Geologists Bulletin, 18,332- propagation. Geology, 15, 630-633,
406.
68
CENTRIFUGE MODELLING OF THE PROPAGATION OF THRUST FAULTS
Ellis, M. A. & Dunlap, W. J. 1988. Displacement variation along thrust Perry, W. J. Jr. 1978. Sequential deformation in the central Appalachians.
faults: implications for the development of large faults. Journal of American Journal of Science, 278, 518-542.
Structural Geology, 10, 183-192. Ramberg, H. 1963. Fluid dynamics of viscous buckling applicable to folding
Gardner, D.A.C. & Spang, J. H. 1973. Model studies of the displacement of layered rocks. American Association of Petroleum Geologists Bulle-
transfer associated with overthrust faulting. Canadian Society of Petro- tin, 47, 484-505.
leum Geologists Bulletin, 21, 534-552. 1982. Gravity, Deformation and the Earth's Crust. Second Edition.
Geiser, P. A. 1988. Mechanisms of thrust propagation: some examples and London: Academic Press.
implications for the analysis of overthrust tsnanef,. Journal of Structural Ramsay, J. G. 1990. Some geometrical problems of ramp-flat thrust models.
Geology, 10, 829-845. In: McClay, K. R. (ed.) Thrust Tectonics 1990, Programme with Ab-
Heim,A. \92\.UntersuchungenuberdenMechanismusderGehirgsbildung. stracts, Royal Holloway and Bedford New College, University of
Basel: Schwabe. London (4—7 April, 1990), 71.
House, W. M. & Gray, D. R. 1982. Displacement transfer at thrust termina- Rich, J. L. 1934. Mechanics of low-angle overthrust faulting as illustrated by
tions in Southern Appalachians - Saltville thrust as example. American Cumberland thrust block, Virginia, Kentucky, and Tenessee. American
Association of Petroleum of Geologists Bulletin. 66, 830-842. Association of Petroleum Geologists Bulletin, 1, 1584-1596.
Jamison, W. R. 1987. Geometric analysis of fold development in overthrust Rogers, W. B. & Rogers, H. D. 1894. On the physical structure of the
terranes. Journal of Structural Geology, 9, 207-219. Appalachian chain, as exemplifying the laws which have regulated the
Kulander, B. R. & Dean, S. L. 1986. Structure and tectonics of Central and elevation of great mountain chains generally. Association of American
Southern Appalachian Valley and Ridge Provinces, West Virginia and Geologists, Nature Reports. 474-531.
Virginia. American Association of Petroleum Geologists Bulletin, 70, Suppe, J. 1983. Geometry and kinematics of fault-bend folding. American
1674-1684. Journal of Science, 283, 684-721.
Liu, S. & Dixon, J. M. 1990. Centrifuge modelling of thrust faulting: strain 1985. Principles of Structural Geology. Englewood Cliffs: Prentice-
partitioning and sequence of thrusting in duplex structures, In: Knipe, R. Hall.
J. (ed.) Deformation Mechanisms, Rheology and Tectonics, Geological Suppe, J. & Medwedeff, D. A. 1984. Fault-propagation folding. Geological
Society of London Special Publication, 54, 431-434. Society of America Program with Abstracts. 16, 670.
& 1991. Centrifuge modelling of thrust faulting: Structural Williams, G. & Chapman, T., 1983. Strains developed in the hangingwalls of
variation along strike in fold-thrust belts. Tectonophysics, 188, 39- 62. thrusts due to their slip/propagation rate: a dislocation mode\. Journal of
Mulugeta, G. & Koyi, H. 1987. Three-dimensional geometry and kinematics Structural Geology, 5, 563-571.
of experimental piggyback thrusting. Geology, 15, 1052-1056. Willis, B. 1894. The mechanics of Appalachian structure. United States
O'Keefe, F. X. & Stearns, D. W. 1982. Characteristics of displacement Geological Survey Annual Report. Part II: 211-282.
transfer zones associated with thrust faults, in Powers, R.B. (ed.) Geo-
logic Studies of the Cordilleran Thrust Bell. Rocky Mountains Asso-
ciation of Geologists, 1, 219-233.
69
Physical models of thrust wedges
Liu Huiqi, K. R. McClay & D. Powell
Abstract: Scaled sandbox models have been used to simulate the growth and sequential
development of critical thrust wedges in isotropic cohesionless and anisotropic cohesionless
materials. Variations in the initial thickness of the layered sequence, the friction of the basal
detachment, and the anisotropy of the layered system have been systematically investigated.
Imbricate fans of dominantly foreland-vergent thrust systems are developed similar to those found
in accretionary prisms and in foreland fold and thrust belts. Critical taper wedges close to
theoretically predicted geometries are developed for intermediate values of basal friction Pj^ = 0.47
whereas for the lower value of basal friction low-taper wedges are formed with tapers less than
predicted by theory. Supra-critical wedges are formed when the basal friction equals or is greater
than the coefficient of friction in the wedge and the wedge has a high taper closer to the angle of
rest for the modelling material. The spacing/thickness ratio of foreland-vergent thrusts increases
as the layer thickness increases. The spacing of thrust faults increases with increased basal friction.
Higher basal friction or anisotropy within the layered systems favours displacement along
foreland-vergent thrusts and suppresses backthrusts.
Physical modelling of thrust structures dates from the earliest surface slope of the wedge topography (a) and the dip of the
days of structural geology. Cadell (1889) constructed the first basal decollement (p). The steady state wedge shape is
documented experiment in a squeeze box using clay material controlled by the relative magnitudes of the strength of the
to simulate the formation of structures found in the Northwest wedge material and the frictional strength of the basal de-
Scottish Highlands. This was closely followed by Willis tachment. Increasing the basal friction increases the critical
(1892) who modelled the structures of the Appalachians. taper, whereas increasing the internal strength of the wedge
Despite their limitations, physical models have provided material decreases the critical taper (Davis ef a/. 1983; Dahlen
many insights into the study of thrust tectonics (e.g. Hubbert etal. 1984). Other factors that affect the critical taper are pore
1937, 1951; Mulugeta & Koyi 1987; Mulugeta 1988a, b). fluid pressure, erosion and synchronous sedimentation at the
Recent physical modelling of thrust systems has been largely top of the wedge, and underplating of material at the base of
concerned with the development of critical wedges, with the wedge (Dahlen & Suppe 1988; Roure et al. 1990).
experiments performed under normal gravity (e.g. Davis et In the critical-tapered wedge theory the wedge material
al. 1983; Dahlen et al. 1984; Malavieille 1984) and under has been treated as homogeneous with an average rheology
higher gravitational loading in a centrifuge (e.g. Mulugeta & (Coulomb behaviour). Variations in cohesion, coefficients of
Koyi 1987; Mulugeta 1988 a, b). internal friction and basal friction have not been considered,
The critical-tapered, Coulomb wedge model proposed by nor have the roles of mechanical anisotropy, competency
Davis etal. (1983), Dahlen (1984) and Dahlen etal. (1984), contrasts, and variations in local stress fields (Price 1988).
together with the subsequent modifications and develop- Such parameters have not been systematically examined in
ments (e.g. Davis & Engelder 1985; Davis & von Huene previous model studies (Mulugeta 1988a, b; Mulugeta &
1987; Zhao et al. 1986; Dahlen & Barr 1989; Dahlen 1990) Koyi 1987). This paper presents the results of a detailed
provides a coherent explanation for the geometry and sequen- series of analogue experiments which investigate the influ-
tial propagation of thrust sheets in accretionary prisms and in ences of model thicknesses, basal friction variations and
foreland fold and thrust belts. According to this model, anisotropy within the critical-taper Coulomb wedge. The
thrusts propagate sequentially until an overall critical taper is experiments illustrate how increased basal friction dramati-
attained such that subsequent deformation involves transport cally affects the wedge geometry, how enhanced mechanical
of the whole wedge along the basal decollement. The state of strength due to increased thickness affects the final geometries
internal stress within such a critical wedge is such that of thrust wedges, as well as the spacing of the thrust sheets,
everywhere the wedge is on the verge of brittle failure and how anisotropy modifies the structural style of a critical-
according to the Coulomb failure criterion, and the basal taper Coulomb wedge.
shear stress is at a level for frictional sliding along the basal
detachment.
The critical taper of a steady-state geological wedge such Experimental method
as an accretionary prism or foreland fold and thrust belt, or
that of an appropriate analogue model, is defined by the Four series of experiments are described in this paper (Table
71
LIU HUIQI ET AL.
Table 1. Experimental Details. periments described in this paper, the base of the deformation
box was horizontal (P - 0°) and the initial lengths of the
Expt. Thickness J^b a+p l/h Shortening models were all 100 cm.
No. (cm)
Models were constructed by carefully sieving alternating
Homogeneous sand models
Series 1. Plastic sheet detachment layers of coloured and white sand (isotropic models) into the
C.56 • 3.0 0.37 2° 5.1 21.0% deformation box taking care to minimize local
C.63 2.5 0.37 2° 4.9 26.5% inhomogeneities in layer thickness and/or packing.
C.62 2.0 0.37 2° 4.2 32.0% Anisotropic models were constructed by sieving thin layers
C.5^ 1.5 0.37 2.5 - 3.5° 4.1 46.0%
of mica flakes (1 mm thick) between the alternating sand
C.59 1.0 0.37 3-4° 3.6 55.0%
Series 11. Drafting filmdetachment layers. The glass side-walls were cleaned using an anti-static
C.102 3.0 0.47 3.5-4° 5.1 37.5% cleaner to minimize the side-wall friction. Sequential defor-
C.103 2.5 0.47 4-5° 4.7 44.0% mation of the sedimentary succession was monitored and
C.I 04 2.0 0.47 5-6° 4.6 51.5% photographed at constant time intervals. Each experiment
C.105 1.5 0.47 6-8° 4.3 58.0% was run until a steady state was reached, i.e. an overall
C.106 1.0 0.47 8 - 10° 3.8 70.0%
critical-taper wedge was attained and visible internal defor-
Series 111. Sand paperdetachment
C.64 3.0 0.55 17-22^ 5.6 62.0% mation had ceased. At this stage the whole wedge either slid
C.73 2.5 0.55 20 - 22" 5.0 66.0% along the basal detachment {low basal friction experiments),
C.68 2.0 0.55 20 - 23'^ 4.7 66.5% or material was continuously underplated at the base and the
C.69 1.5 0.55 20 - 24'^ 4.2 71.5% wedge continued to grow with the surface slope approaching
C.71 1.0 0.55 20 - 24^^ 3.2 81.5%
the angle of rest of the modelling material (high basal friction
Anisotropic sand/mica models
Series IV. Consiani thickness experiments). The completed models were impregnated and
C.113 2.5 0.37 2 - 2.5° 4.4 31.0% serially sectioned in order to examine the resultant structures
C.112 2.5 0.47 8- !0° 4.5 53.0% in three dimensions. Side-wall friction generated thrusts
C.114 2.5 0.55 21-24" 4.5 67.0% which were curved in plan view but severe curvature was
|i^ = basal friction; 1 - slacing ot toreland-vergent thrust; h --layer thickness. limited to about 2 cm from the edges of the glass, and the
structures formed in the centre of the deformation box were
1). Series I-III used homogeneous sand models to investigate laterally consistent within the model. The deformation ob-
the effects of variations in both basal friction and initial model served through the glass side-walls was shown to be similar
thicknesses on the geometries and kinematics of the resultant to the intemal parts of the model. The serial sections shown
Coulomb wedge. Series IV experiments used sand-mica in this paper were taken from the centre of the models and are
models of a constant initial thickness (2.5 cm) to investigate representative of the deformation of the models as a whole.
the effects of variations in the basal friction on an anisotropic Each series of experiments have been duplicated and the
wedge system. results were found to be reproducible.
The apparatus used in this experimental programme was
a glass-sided, rectangular deformation box (Fig. 1) similar to
Physical properties of the modelling materials
that used by Davis etal. (1983). The motor, through the roller
system, pulls a sheet of detachment material at a constant rate Dry, cohesionless quartz sand has been commonly used as a
underneath the sand model, and as a result of the buttressing physical modelling material as it satisfactorily simulates the
force of the rigid 90° backstop, generates a Coulomb wedge brittle Coulomb behaviour of shallow crustal rocks in scaled
that simulates an accretionary prism or fold and thrust belt. laboratory experiments (Hubbert 1937; Horsfield 1977;
The internal dimensions of the box were 100-150 cm x 20 cm McClay& Ellis 1987a, b; Ellis & McClay 1988; Mandl 1988;
X 20 cm. The sheet of detachment material moved at a Mulugeta 1988a, b). Dry quartz sand has a Navier-Coulomb
constant displacement rate of 0.6 cm min '. In all of the ex- rheology and an angle of friction of 30° similar to that for
Moving
sheet
72
PHYSICAL MODELS OF THRUST WEDGES
I 400'
sliding friction \i^ = 0.47, whereas the sandpaper has a high
value of sliding friction \)i^ = 0.55 which is equal to the coef-
ficient of friction for quartz sand. The vertical backstop had
a coefficient of friction for sand sliding against it of |j. = 0.55.
200-
Results
- • — I — I — I — I I < I I I I I I I I I ' I
200 400 600 800 1000 1200 1400 1600 1800 2000
The results of four series of experiments (15 model s, Table 1)
Normal stress (Pa) are presented in this paper. (Duplicate models have produced
similar results (Liu unpubl. data)). In all models a foreland
propagating sequence of thrust slices formed a Coulomb
Figure 2. Mechanical properties of the modelling materials (sand and mica)
at failure. wedge-like geometry with a taper dependent upon both the
basal friction and, to a lesser degree, on the initial thickness
many sedimentary rocks (cf. McClay 1990). In terms of of the model (Table 1). In all of these models a rigid vertical
geometric similarity (Ramberg 1981), the length ratio of the backstop was used in order to produce wedges that closely
model experiments to natural examples is 10'^-10^i.e. a 1 cm approximate the theoretical geometries (cf. Davis et al. 1983;
sand layer corresponds to 100 m to 1 km of sedimentary strata Dahlen 1990). Further research is currently being carried out
in nature. However, the packing, distribution, and size of the using variable backstop geometries that may be more geo-
grains of the modelling material can not be accurately scaled logically realistic.
and consequently the fault structures in the analogue ex-
periments are not discrete failure planes but are dilatant
shear zones of varying width (Horsfield 1977; McClay 1990). Thrust sequences and thrust geometries
Dry quartz sand (grain size 200-300 |im, average density The progressive evolution of the thrust system in a typical
1.58 + 0.1 g cm 0 was used to simulate the brittle deformation
of isotropic sedimentary sequences (experiment series I-III,
Table 1) and thin layers of dry vermiculite mica (grain size 500-
500 \im (long axis), average density 1.0 g cm') were intro-
duced between the sand layers to provide a mechanical a. Sand paper: x = 71.1 + 0.55 o
anisotropy and hence simulate the deformation of anisotropic b. Drafting film: x = 23.7 + 0.47 o
sediments which deform by interlayer slip (experiment series 400
c. Plastic sheet: x = 19.4+ 0.37 0
IV, Table 1).
The mechanical properties of the sand and mica at very CD
low normal stresses (Fig. 2) have been investigated using a tW 300
similar apparatus to that described by Krantz (1991). Shear M
0)
i-
strength was measured for both the unfaulted and faulted
w
materials (after fault displacement of 5 mm) (Fig.2). Linear in 200
0)
Mohr envelopes were found for all materials with coefficients
CO
of friction ranging from \i = 0.55 for unfaulted sand to \l =
0.37 for unfaulted mica (Fig. 2). Low apparent cohesive
100
strengths were found for all materials (Xy= 166-189.5 Pa, Fig.
2). Faulted sand shows a strength reduction of approximately
10% (Fig. 2) whereas faulted mica gained strength (14%)
after 5 mm of fault displacement (Fig. 2). The latter feature 1 1 1 ' 1 1 1 1 1 1 1
is probably a result of reorientation of mica flakes so that they 0 100 200 300 400 500 600 700
obstruct the sliding plane at low displacements. However, at
Normal stress (Pa)
larger displacements, it is anticipated that entrainment of a
large number of mica grains into the shear plane would lower Figure 3. Frictional properties of the three basal detachment materials,
73
•^-i.-lHW L j y H U I Q I ET AL. '^•"''-"
,.0 |.
J* .^.ri
-•JO.:'
3
(0
-f!f;3 IV)
J-y.
Tl •j'J. uv 'j
(:^ ' . X! r! a.
• )0 •.f! f)
,)-. iq ~oot
*~ r
o
^-^'
.(V .•. -I
.L .'IT
Figure 4. Sequential development of Experiment C.l 19, with homogeneous sand, fJ^ = 0.47 and a 2.5 cm thick initial parallellayered sequence, (a) Initial
model 0% shortening; (b) 3% shortening; (c) 9% shortening; <d) 18% shortening: (e) 27% shortening; (f) 39% shortening; (g) 48% shortening; (h) 49%
shortening - end of the defoimation run with a stable wedge formed. Slight underplating of sand causes the wedge to grow vertically from (g) above.
74
PHYSICAL MODELS OF THRUST WEDGES
I
Another foreland-vergent thrust developed in the immediate
footwall of the first thrust and was followed by a second major
12-
foreland-vergent thrust approximately five cms in front of
thrust 1 (Fig. 4c). At 18% shortening thrust 2 has developed
a 10-
a small backthrust at the tip and thrust 3 had nucleated in front (0
of thrust 2 (Fig. 4d). The older, higher thrusts became o
a.
inactive and are passively back rotated once a new foreland- o
vergent thrust nucleated in the footwall. A similar pattern of Expt. C.119
nucleation and displacement on the most foreland-ward thrust I "~i
combined with back-rotation of the hinterland thrusts, is 10 20 30 40 50
20
Initial thickness variations c
o
In the experiments described in this paper a uniform parallel
sided sedimentary prism was used in order to invesigate the
effect of stratigraphic thickness on thrust spacing and wedge
II 10
75
Figure 6. Five sections of sand models with different thickness, all using intermediate basal friction, C.102, 3 cm model, C.I03, 2.5 cm model, C.104, 2 cm model, C.105, 1.5 cm model and C.106, 1 cm model.
PHYSICAL MODELS OF THRUST WEDGES
number of thrusts that form the final stable wedge; and Critical taper Coulomb wedges
(3) The critical taper of the wedge only decreases slightly
According to Coulomb wedge theory the critical taper (a -i- P)
with an increase in initial thickness from 9.5° for an initial
of a dry cohesionless wedge can be approximated by the
thickness of 1.0 cm to 7° for an initial thickness of 3.0 cms
equation (e.g. Dahlen 1990, eq. 16) -
(Fig. 6).
In all experiments except C.106 (initial thickness 1.0 cm) 1 - sin
an early backthrust commonly formed at the tip of each a + p = Hb (1)
1 + sin (()
foreland-vergent thrust (Fig. 6). In the experiments with
where a = topographic slope of the wedge, (3 = dip of the basal
thinner initial thicknesses more foreland-vergent thrusts
detachment, ^^^ = coefficient of friction of the basal detach-
formed with the result that the more hinterland thrusts have
ment, if - angle of internal friction of the material in the
undergone greater back rotation than for the wedges formed
wedge.
from greater initial thicknesses (Fig. 6).
For the analogue model experiments, substituting for ]x.^
and ([) gives theoretical critical wedge tapers of 7.4° for \\.^ =
Variation in basal friction 0.37 (plastic sheet), 9.4° critical taper for \)L^= 0.47 (drafting
film), and a critical taper of 11° for |j,^= 0.55 (sandpaper).
The effects of varying the basal friction on final wedge
Comparison of the theoretical values with actual values
geometry are shown in Figure 7. An increase in friction
(Table 1) shows that the best results were obtained for thin
produces a dramatic increase in the critical taper from 1.5° for
initial thicknesses (1.0-1.5cms) and for an intermediate basal
a basal friction of p.,^ = 0.37 (Fig. 7c), to a wedge taper of 7.0°
detachment friction of \y^ = 0.47 - the drafting film detach-
for a basal friction of |J,j^ = 0.47 (Fig. 7b), and to a wedge taper
ment. For low values of basal friction low-taper wedges
of 18.5° for a basal friction of \i^ = 0.55 (Fig. 7a). Concomi-
(taper less than that predicted by eq. 1) were formed because
tant with the increase in wedge taper there are corresponding
the deformation partitioned approximately equally between
increases in the lengths of thrusts in the wedge (Fig. 7). The
foreland-vergent thrusts and backthrusts (Figs 6 & 7) such
number of backthrusts associated with each foreland-vergent
that symmetric uplift occurred thus reducing the wedge taper.
thrust decreases with an increase in basal friction. For the
Wedges produced when the basal detachment had a frictional
lowest value of basal friction {\x.^ = 0.37) the deformation is
coefficient equal to that of the modelling material produced
characterized by asymmetric 'pop-up' structures that result in
critical tapers of 18-24°, greatly in excess to that expected
a low critical taper for the wedge (Fig. 7c).
from theory (11°) and unaffected by initial thickness of the
model (Table 1). In these cases a critical point is not
The ejfects of anisotropy reached - the wedge continues to build up as long as new
material is fed into the system and the wedge angle is closer
The introduction of anisotropy into the system by placing thin
to a dynamic angle of rest for the granular modelling material.
layers of mica flakes between the sand layers produces
Such wedges are supra-critical and the Coulomb wedge
critical tapers (Fig. 8) which are slightly increased when
theory is no longer applicable (it was an assumption in the
compared to the isotropic models (cf. Fig. 7). The styles of
theoretical analysis that the basal detachment is weaker than
deformation in the anisotropic models are very similar to the
the internal material). For a basal friction of \i.^^ = 0.37 there
isotropic examples (Figs 7 & 8) with possibly slightly more
is an initial thickness above which a Coulomb wedge is not
folding in the anisotropic models (the mica flakes facilitate
formed but rather the whole model slides on the basal detach-
layer parallel slip).
ment with no internal deformation. Such models are termed
supercritical and it has been found that the supercritical
thickness increases with an increase in the basal friction. In
Discussion
this case the model is significantly stronger than the basal
detachment and no internal deformation occurs.
Sandbox analogue models have successfully simulated the
formation of critical taper Coulomb wedges comparable to
those found in accretionary prisms and foreland fold-thrust Thrust geometries
belts (Davis et al. 1983; Dahlen 1990). In all models a
foreland nucleating and foreland-vergent thrust system de- In the analogue models (Figs 4,6-8) imbricate thrust fans are
veloped with varying degrees of backthrusting. Active produced which have similar forms to those found in accre-
foreland-vergent deformation was concentrated on the last- tionary prisms (e.g. Seely 1974; Westbrook 1982; Lewis et al.
formed foreland-vergent thrust such that the older, higher 1988; Westbrook et al. 1988) and in the frontal parts of
thrusts were passively back-rotated to steeper angles in a foreland fold and thrust belts (e.g. Bally etal. 1966; Dahlstrom
manner analogous to that found in many thrust belts (Bally et 1970; Lillie et al. 1987; Stanley 1990). In the models de-
al. 1966; Dahlstrom 1970; Boyer & Elliott 1982). In many scribed in this paper the spacing/thickness ratio of the thrust
models the tips of individual thrusts have an associated small sheets varies from 3.2-5.7 and increases with an increase in
backthrust (Figs 4, 6-8) similar to those found both in fold- basal friction in contrast to the findings of Mulugeta (1988b)
thrust belts (Alonso & Teixell 1991 - this volume) and in who found that this ratio increased with a decrease in basal
accretionary prisms (Lewis et al. 1988). friction. The length of individual thrust slices increased with
77
^ " - , ' " " • " :
Figure 7. Three representative sections of selected homogeneous sand models with the same layer thickness (2.5 cm); (a) High basal friction model, M-^ = 0.55; (b) Intermediate basai friction model, |j^, = 0,47; and (c)
Low basal friction model, ^l^^ - 0,37.
o
Figure 8. Three representative sections of sand/mica models with the same layer thickness (2.5 cm); (a) High basal friction model, |J|^ = 0.55; (b) Intermediate basal friction model, ji,, = 0.47; and (c) Low basal friction
model, |ij, ^ 0.37.
LIU HUIQI ET AL.
B
'Foreland' Foreland vergent imbricate fan Maximum uplift
Figure 9. Schematic diagram of a deformed wedge showing the three deformation zones and the general sequence of thrust faults. Note that the back thrusts
at the rear of the wedge are active at various stages of wedge growth.
an increase in initial model thickness (Fig. 6) and similar Similar slumping was also been reported by Mulugeta &
features are found in foreland fold and thrust belts e.g. in Koyi (1987) in their experiments, and can be compared to that
cross-sections of the Canadian Rocky Mountains (Price 1981) reported by Moore & Shipley (1988) in the Middle America
and of the Moine thrust zone (Elliott & Johnson 1981). The Trench accretionary prism. Whereas the experiments de-
analogue models show that a decrease in the basal friction scribed in this paper produce Coulomb wedges that show
increases the number of backthrusts in the system (Fig. 7) many similarities to foreland fold and thrust belts and to
with a more symmetric uplift and a low critical taper similar accretionary prisms, it must be remembered that the experi-
to that found in foreland fold and thrust belts with a weak ments use dry material and pore-fluid pressures have not been
basal decollement (e.g. Pakistan, Davis & Engelder (1985), simulated - a factor important in many accretionary prisms
Jadoon etal. (1991, this volume)). In the anisotropic models and in fold-thrust belts (Hubbert & Rubey 1959).
more shortening is required to achieve a stable wedge system
(Fig. 8) and deformation of the hangingwall above the fore-
land-vergent thrusts is accommodated by folding with sup- Conclusions
pression of backthrusts thus indicating that the anisotropic
models are mechanically weaker than the isotropic sequences. Physical models have been successfully constructed to simu-
The initial ramp angles of the foreland-vergent thrusts (25 - late the progressive deformation of brittle Coulomb wedges.
30°) and of the backthrusts (35-40°) do not show significant Imbricate fans of dominantly foreland-vergent thrust systems
variations with changes in basal friction thus indicating that are developed similar to those found in accretionary prisms
care must be exercised when attempting to deduce the coef- and in foreland fold and thrust belts. Critical taper wedges
ficient of basal friction from thrust fault angles (cf. Mulugeta close to theoretically predicted geometries are developed for
1988b). intermediate values of basal friction \i^ = 0.47 whereas for the
Figure 9 is a summary diagram illustrating the main lower value of basal friction low-taper wedges are formed
features of the Coulomb wedge models described in this with tapers below that predicted by theory. Supra-critical
paper. Zone A delimits an area adjacent to the back end wall wedges are formed when the basal friction equals or is greater
which has suffered maximum bulk shortening and thickening than the coefficient of friction in the wedge and the wedge has
and where foreland-vergent thrusts have suffered a maximum a high taper closer to the angle of rest for the modelling
amount of passive rotation. This section undergoes vertical material. In sandbox critical wedges, the spacing/thickness
uplift by synchronous movement on the foreland-vergent ratio of foreland-vergent thrusts increases as the layer thick-
detachment and on the backthrusts adjacent to the vertical ness increases, suggesting a process which is not geometri-
backstop. In zone B the foreland-vergent thrusts have the cally self-similar. The experimental data also indicate that
largest displacements, most regular spacing and form a the spacing of thrust faults increases with increased basal
foreland-vergent imbricate fan; zone C (the 'foreland') is friction. Higher basal friction or anisotropy within the
undeformed and has been passively transported on the basal layered systems favours displacement along foreland-vergent
detachment. The wedges have a concave upwards topogra- thrusts and suppresses backthrusts. In the anisotropic models
phy which suggests a strain hardening process at the rear of enhanced interlayer slip arising from the introduction of mica
the system (Zhao ef a/. 1986; Mulugeta 1988b). The slumped layers allows accommodation in the hangingwalls to be taken
material from each of the thrust fronts is later over-ridden by up by folds.
the foreland-vergent thrust faults which is similar to the
'erosional thrusts' mentioned by Burchfiel et al. (1982).
80
PHYSICAL MODELS OF THRUST WEDGES
Liu Huiqi gratefully acknowledges receipt of a research studentship from the RHBNC. Thanks are due to Brian Adams for constructing the deformation
Sino-British Friendship Scholarship Scheme administered by the British apparatus, to Martin Insley for helping with the shear tests and to Ian Davison
Council. This work was in part supported by the Industry Association of the and Tony Dahlen for helpful comments on an early version of the manu-
Geology Department, Royal Holloway and Bedford New College, Univer- script.
sity of London, and by the Structural Geology Modelling Laboratory,
References
Alonso, J. L. & Teixell A. 1991. Forelimb deformation in some natural Lillie, R. J., Johnson, G. D., Yousuf, M., Zamin, A.S.H. & Yeats, R. S. 1987.
examples of fault-propagation folds (this volume). Structural development within the Himalayan foreland fold-and-thrust
Bally, A. W., Gordy, P. L. & Stewart, G. A. 1966. Structure, seismic data, and belts of Pakistan. !n: Beaumont, C. & Tankard, A. J. (eds) Sedimentary
orogenic evolution of the southern Canadian Rocky Mountains. Bulletin Basins and Basin-Forming Mechanisms. Canadian Society of Petroleum
of Canadian Petroleum Geology, 14, 337-381. Gelogists Memoir, 12, 379-392.
Boyer, S. E. & Elliott, D. 1982. Thrust systems. Bulletin of the American Malavieille, J. 1984. Modelisation experimentale des chevauchements
Assocation of Petroleum Geologists, 66, 1196-1230. imbriques: Application aux chaines de montagnes. Geological Society of
Burchfiel, B. C , Wernicke, B., Willemin, J. H., Axen, G.J . & Cameron, C. France Bulletin,!, 129-138.
S. 1982. A new type of decollement thrusting. Nature, 300, 513-515. Mandl, G. 1988. Mechanics of Tectonic Faulting: Models and Basic Con-
Cadell, H. M. 1889. Experimental researches in mountain building. Trans- cepts. Developments in Structural Geology, 1, Elsevier, Amsterdam, 407p.
actions of Royal Society of Edinburgh, 35, 337-357. McClay, K.R. 1990. Deformation mechanics in analogue models of
Dahlen, F. A. 1984. Noncohesive critical Coulomb wedges: an exact extensional fault systems. In: Knipe, R. J. & Rutter, E. H. (eds). De-
so\\iiion. Journal of Geophysical Research, 89, 10125-10133. formation Mechanisms, Rheology and Tectonics. Geological Society of
1990. Critical taper model of fold-and-thrust belts and accretionary London Special Publication, 54,445-453.
wedges. Annual Review of Earth and Planetary Sciences, 18, 55-99. & Ellis, P. G. 1987a. Analogue models of extensional fault geometries.
& Barr, T. D. 1989. Brittle frictional mountain building, 1, Deformation In: Coward, M. P., Dewey, J. F. & Hancock, P. L. (eds) Continental
and mechanical energy budget. Journal of Geophysical Research, 94, Extensional Tectonics. Geological Society of London Special Publica-
3906-3922. tions, 28, 109-125.
& Suppe, J. 1988. Mechanics, growth, and erosion of mountain belts. & 1987b. Geometries of extensional fault systems developed in
In: Clark, S. P., Jr., Burchfiel, B. C. & Suppe, J. (eds) Processes in model experiments. Geology, 15, 341-344.
Continental Lithospheric Deformation. Geological Society of America Moore, G. F. & Shipley, T. H. 1988. Mechanics of sediment accretion in the
Special Paper, 218, 161-178. Middle America Trench off Mexico. Journal of Geophysical Research,
, , J. & Davis, D. M. 1984. Mechanics of fold-and-thrust belts and 93.8911-8927.
accretionary wedges: Cohesive Coulomb theory. Journal of Geophysi- Mulugeta, G. 1988a. Squeeze box in a centrifuge. Tectonophysics, 148,323-
cal Research,^, 10087-10101. 335.
Dahlstrom, C.D.A. 1970. Structural geology of the eastern margin of the 1988b. Modelling the geometry of Coulomb thrust wedges. Journal of
Canadian Rocky Mountains. Bulletin of Canadian Petroleum Geology, Structural Geology, 10, 847-859.
18, 332-406. & Koyi, H. 1987. Three-dimensional geometry and kinematics of
Davis, D. M. & Engelder, T. 1985. The role of salt in fold-and-thrust belts. experimental piggyback thrusting. Geology, 15, 1052-1056.
Tectonophysics, 119, 67-89. Price, R. A. 1981. The Cordilleran foreland thrust and fold belt in the
& von Huene, R. 1987. Inferences on sediment strength and fault southern Canadian Rocky Mountains. In: McClay, K. R. & Price, N. J.
friction from structures at the Aleutian Trench. Geology, 15, 517-522. (eds) Thrust and Nappe Tectonics,Geological Society of London Special
, Suppe, J. & Dahlen, F. A. 1983. Mechanics of fold-and-thrust belts and Publications, 9, 427-448.
accrelionary wedges. Journal of Geophysical Research,SS, 1153-1172. 1988. The mechanical paradox of large overthrusts. Geological Society
Elliott, D. & Johnson, M.R.W. 1980. Structural evolution in the northern part of America Bulletin, 100, 1898-1908.
of Moine thrust belt, NW Scotland. Transactions of Royal Society of Ramberg, H, 1981. Gravity, Deformation, and the Earth's Crust: In Theory,
Edinburgh (Earth Sciences), 71, 69-96. Experiments and Geological Application (2nd ed.). Academic Press,
Ellis, P. G. & McClay, K. R. 1988. Listric extensional fault systems - results London. 452p.
of analogue model experiments. Basin Research, 1, 55-70. Roure, F., Howell, D. G., Guellec, S. & Casero, P. 1990. Shallow structures
Horsfield, W. T. 1977. An experimental approach to basement controlled induced by deep-seated thrusting. In: Letouzey, J. (ed.) Petroleum and
faulting. Geologic en Mijinhouw, 56, 363-370. Tectonics in Mobile Belts, Editions Technip, Paris, 43, 15-30.
Hubbert, M. K. 1937. Theory of scale models as applied to the study of Seely, D. R., Vail, P. R. & Walton, G. G. 1974. Trench slope model, /n: Burk,
geologic structures. Geological Society of America Bulletin, 48, 1459- C. A. & Drake, D. L. (eds) The Geology of Continental Margins.
1520. Springer-Verlag, New York. 249-260.
1951. Mechanical basis for certain familiar geologic structures. Geo- Stanley, R. S. 1990. The evolution of mesoscopic imbricate thrust faults—
logical Society of America Bulletin, 62, 355-372. an example from the Vermont Foreland, USA. Journal of Structural
& Rubey, W. M. 1959. Role of fluid pressure in mechanics of thrust Geology, 12, 227-24].
faulting: I. Mechanics of fluid-filled porous solids and its application to Westbrook, G. K. 1982. The Barbados Ridge complex: Tectonics of a mature
overthrust faulting. Geological Society of America Bulletin, 70,115-166. forearc system. In: Leggett, J. K. (ed.) Trench-Forearc Geology. Geo-
Jadoon, I.A.K., Lawrence, R. D. & Lillie, R. J. 1991. Balanced and logical Society of London Special Publication, 10, 275-290.
retrodeformed geological cross-section from the frontal Sulaiman Lobe, , Ladd, J. W., Buhl, P., Bangs, N. & Tiley, G. J. 1988. Cross section of
Pakistan: Duplex development in thick strata along the western margin an accretionary wedge: Barbados Ridge complex. G^o/ogy, 16,631-635.
of the Indian Plate (this volume). Willis, B. 1892. The mechanics of Appalachian structure. US Geological
Krantz, R. W. 1991. Measurements of friction coefficients and cohesion for Survey 13th Annual Report, 271-281.
faulting and fault reactivation in laboratory models using sand and sand Zhao, W.-L., Davis, D. M., Dahlen, F. A. & Suppe, J. 1986. Origin of convex
mixtures. Tectonophysics, 188, 203-207. accretionary wedges: Evidence from Barbados. Journal of Geophysical
Lewis, S. D., Ladd, J. W. & Bruns, T. R. 1988. Structural development of an Research,n, 10246-10258.
accretionary prism by thrust and strike-slip faulting: Shumagin region,
Aleutian Trench. Geological Society ofAmerica Bulletin, 100,767-782.
81
Generation of curved fold-thrust belts: Insight from simple physical
and analytical models
S. Marshak, M. S. Wilkerson & A. T. Hsui
Abstract: Viscous-flow and sand-wedge models provide insight into factors that determine the
location, shape, and temporal evolution of fold-thrust belts whose traces are curved in plan view.
These models suggest that curves are localized by shear between segments of fold-thrust belts and
adjacent regions, and by lateral variations in sediment thickness. Furthermore, they indicate that
the shape of antitaxial (convex toward the foreland) belts may reflect the rheology and/or strength
of rock comprising a belt, as well as regional variations in sediment thickness, topography, and
backstop geometry. Finally, the models help distinguish between settings in which curvature is
secondary (i.e. curves are oroclines) and settings in which curvature is primary.
Many major orogens, and their component fold-thrust belts, curve geometry. To gain a clearer image of how obstacle
contain pronounced plan-view curves defined by lateral geometry controls flow trajectories, a physical model of
variation in structural trends. For example, the Appalachians viscous flow that used cooled glycerine as the deforming
contain two major antitaxial curves (convex in the direction medium was constructed. The need to have more quantitative
of vergence) and two major syntaxial curves (concave in the constraints on the shape of deforming lobes led the authors to
direction of vergence). Work during the past few decades construct a finite-difference model describing crustal defor-
(e.g. Carey 1955; Ries & Shackleton 1976; Beutner 1977; mation. Finally, to gain insight into the kinematic evolution
Burchfiel 1980; Eldredge et al. 1985; Lowrie & Hirt 1986; of curved fold-thrust belts, three types of sand-wedge models
Marshak & Tabor 1989) has clarified our understanding of that allowed simulation of collision with obstacles, impact of
structural and palaeomagnetic patterns within curved belts, an indenter, and superposition of strike-slip shear were con-
and has emphasized the persistence of three outstanding structed.
problems: (1) Why do curves form? (2) What factors control
the shape of individual curves? and, (3) Do curves initiate
Analytical model of fluid flow
with their present shape, or do they represent plan-view
bending and tangential extension of pre-existing straight Model setup. This model uses the evolution of a passive
structures? marker line on the surface of a fluid sheet to describe curve
The purpose of this paper is to describe results of viscous- development accompanying interaction of a growing fold-
flow and sand-wedge modelling experiments which help thrust belt with obstacles in the foreland. The fluid sheet
constrain solutions to these problems. Curved fold-thrust represents a thrust wedge, and the marker line represents the
belts form in a variety of different tectonic settings, but this structural trend as a function of location. All points on the
paper examines only curves formed as a consequence of fold- marker line are constrained to flow along parallel lines. To
thrust belt interaction with obstacles in the foreland, curves simulate interaction with obstacles, the endpoints of the
formed in advance of an indenter, and curves formed adjacent marker line are assigned a velocity of zero. Initially, the
to strike-slip faults. In this context, the term obstacle refers marker line is straight and is perpendicular to the flow
to any feature that restrains thrust-sheet movement at its direction, but as flow progresses, it arches into a parabola, as
lateral margins - physically, obstacles may be basement described by:
massifs, pinch-outs of stratigraphic glide horizons, or strike-
slip faults. An indenter is a rigid mass or backstop of finite x = {(h/2)"*' — [y-(h/2)]"*'} / F(n+1) (1)
length (measured perpendicular to the transport direction)
that pushes into the deforming medium. In this equation, which is similar to the equation defining
one-dimensional channel flow, x defines the distance be-
tween the initial and final positions of a point on the marker
Modelling procedures and observations line as measured in the direction of flow, y defines the
distance between a point and the left endpoint of the marker
Several different types of models were used to investigate the line prior to flow, and h is the distance between endpoints of
generation of curved fold-thrust belts. The first was an the marker line measured along the y-axis (Fig. 1; Wilkerson
analytical model of viscous flow describing the interaction of etal. 1988; modified from Turcotte & Schubert 1982, p. 316-
a fluid layer with obstacles to test the effects of rheology on 318, and Hsui & Youngquist 1985; see also Johnson 1970).
83
S. MARSHAK ETAL.
s • ^ ^
t ' ^ ^ ,
.rrt-.<c«oQ:orcr-
(b)
.L»J>>)»>>>>>>>>>>>>)>>>>>>>>>>rr>;^.
Non-Newtonian (b)
<WCWWWMWMW«W(|
y^ \
N••, ,f>
Himalayas 'm^
84
GENERATION OF CURVED FOLD-THRUST BELTS
85
S. MARSHAK EJ AL.
200
150
100
50
•/
200 200
\] 1) 1 ^^^ (d)
150 150
^
f\
100
*0 1 /
100
Differential
50 <J C Large Critical Angle 50 Obstacle Geometry
Figure 5. Form-line contour maps showing the shape of flow fronts generated by the finite-difference model, (a) Shape of the flow front after about 3 million
years for a material whose behaviour is described by a critical angle of 0°. Contour interval is 333 m, lowest contour is 533 m. Note the flat-crested trace
of the contours; (b) Shape of the flow front after about 3 million years for a material whose behaviour is described by a critical angle oi5°. Contour interval
is 160 m, lowest contour is 360 m. Note the parabolic trace of the contours, (c) Shape of the flow front after about 3 million years for a model where the
regional flow is oblique to a base line drawn between the two obstacles, (d) Shape of the flow front after about 3 million years for flow where one obstacle
(upper) is larger than the other.
can also be used to show that flow geometry is related to flow cm', internal friction angle of about 30°) that behaves as a
direction or obstacle geometry; an asymmetric flow front is Coulomb material. Sand is spread in a smooth 2 cm-thick
readily generated by the model either by making the regional layer across the base of a 75 x 120 x 30 cm wooden deforma-
flow oblique to a baseline drawn between the obstacles (Fig. tion box, so the dimensional scale factor is between 10^ and
5c), or by making one obstacle larger than the other (Fig. 5d). 10"', similar to values used by Horsfield (1977) and McClay
& Ellis (1987). Slightly dampening the sand made fault
traces developed during shortening visually more distinct,
Sand-wedge models though their shape was identical to those formed in dry sand.
Model setup. Sand wedges formed by compression have The capillary effect of the water films on the sand grains
proven to be successful tools in developing an intuitive effectively creates negative pore pressure (see Lambe &
understanding of fold-thrust belt generation (e.g. Davis et al. Whitman 1979), a factor that does not affect the slope of the
1983; Mulugeta & Koyi 1987; Mulugeta 1988). This paper failure envelope. In order to maintain constant sand damp-
focuses on models that illustrate development of fold-thrust ness during the runs, sand wedges were generated at a strain
belt geometries in map view (see also Burg et al. 1987; Davy rate of about 0.5 cm/s. Though this rate is significantly faster
& Cobbold 1988). The models employ commercial silica than rigorous scaling dictates, studies suggest that the yield
sand (0.3 - 0.6 mm diameter grains, bulk density of 1.57 g/ envelope of Coulomb materials is practically independent of
86
GENERATION OF CURVED FOLD-THRUST BELTS
Indenter Configuration
Obstacle Configuration
87
S. MARSHAK HI AL.
Observations
Effect of cross-strike shear. The conveyor bell configuration
allowed examination of the effects of imposing cross-strike
shear on a fold-lhrust belt. Movement of the wide mylar
conveyor, which simulates imposition of shear during thrust-
ing, results in the development of a ridge-like fold-thrust belt
bordered on the hinterland by a lowland, a pattern that is
characteristic of some real fold-thrust belts (Royden &
Burchfiel 1989). The ridge grows asymmetrically (Fig.
9a,b), with the foreland slope approaching T and the hinter-
land slope approaching 30°. As a consequence of shear
between the sand and the box walls (right-lateral on one side
and left-lateral on the other), faults on the foreland side
initiate with a syntaxial curvature (Fig. 9a). Faults extend
across the width of the deformation box and each slips almost
simultaneously along its entire length. Movement of the Hgure 10. Photosof sand wedgcsdevelopedinadvanceof an indenter. The
narrow mylar conveyor beneath a linear sand wedge simu- indenter is the dark face near the lop of each photo. Scale is 15 cm long. All
photographs are map view, (a) Early stage in wedge development. Note that
lates imposition of shear on a pre-existing fold-thrust belt. the early formed thrusts (indicated by steps in ihe sand surface) are elliptical,
After movement, the straight fault traces were bent into (b) Intermediate stage in wedge development. Note that ihe younger thrusts
antitaxial curves, one above each edge of the mylar conveyor. are less eccentric than the older thrusts, (c) Late stage in wedge development.
The curve adjacent to the left-lateral shear zone is similar in Note the overall paraboMc shape of ihe deformation front. Oldest faults
(adjacent to the indentor face) are flat-crested. The grid roughly shows strain
appearance to the curve of the Makran fold-thrust belt where
distribution in the map plane, (d) Close up of the strain grid along the side
it intersects the Chaman fault in Pakistan (see Fig. 15 in of the wedge.
GENERATION OF CURVED FOLD-THRUST BELTS
u
the thrust wedge immediately after initiation of the first fault
indicates that the two parameters are directly proportional to
(c) one another (Fig. I la,b) as defined empirically by the rela-
tion: W = 2.75H. This proportionality suggests that the
location of a curve is affected by lateral variations of sand
Figure 11. Relationship between sand thickness and 'width' (the distance thickness in the foreland and that, once formed, the evolution
between the indenler face and the trace of the first formed thrust), (a) of a curve is controlled in part by lateral variations in the
Schematic map showing the definition of width. On the right side of the
thickness of the wedge itself.
diagram is an elliptical fault trace in front of a straight indenter face. Length
is the distance between endpoints. The movement direction is indicated by To physically illustrate how lateral variation in sand
the arrow; (b) Graph of width as a function of sediment thickness for six thickness localizes curves, sand was pushed with an indenter
experiments; (c) Schematic cross section emphasizing why the distance that had matching notches removed from both sides along its
between the toe of the wedge (marked by 'v') and the indenter face depends
base (Fig. 8). The full thickness of sand was displaced,
on sand thickness. Shaded area is sand; the thick black line is the indenter,
which moves in the direction indicated by the arrow. The slope of both therefore, only adjacent to the central part of the indenter,
wedges is the same. where the indenter contacted the box floor; adjacent to the
notches, only three-quarters of the sand thickness was dis-
Marshak 1988). placed. When the indenter moves, the sand wedge formed
Incienter-generated curves. Pushing into the sand with an adjacent to the notches is relatively narrow (as measured
indenter leads to the development of an antitaxial fold-thrust
wedge, and allowed observation of how curved structures
initiate and evolve. Following the first increment of move-
ment, a fault develops in advance of the indenter (Fig. I Oa).
The shape of the fault trace depends on indenter length
(measured perpendicular to movement direction) relative to
sand thickness. In experiments with a 2 cm-thick sand layer,
the first fault formed in front of the 18 cm-long indenter
initiates as an elliptical arc (spoon-shaped in three dimen-
sions; cf. Davy & Cobbold 1988), whereas the first fault
formed in front of the 31 cm-long indenter was curved near Figure 12. Photo of a sand lobe formed in front of the notched indenter
the ends of the indenter, but was parallel to the indenter face sketched in Figure 8. This photo is taken looking vertically down on the
away from the ends. With progressive displacement into the surface, and movement is from the top to the bottom of the photo. The lobe,
which is 15 cm-long measured parallel lotheindenterface,forms where the
foreland, the width (measured parallel to the movement sand being pushed by the indenter is thicker. Note that the spacing of fauhs
direction) of the thrust wedge increases and its interior depends on wedge width.
89
S. MARSHAK ETAL.
both scales.
The relationship between sediment thickness and thrust-
wedge width may explain the frequently referred to associa-
tion between sedimentary thickness in passive-margin or
foreland basins and the position of fold-thrust belt salients
(Rankin 1976; Thomas 1977; Lillie& Davis 1990). There is
a direct proportionality between the width of the wedge after
formation of the first fault and the thickness of the sand being
deformed (Fig. lla,b). Thus, if the critical taper (Davis c^tv/.
1983) of a sand wedge is to be maintained, a thrust wedge
must be wider where the sand layer being deformed is thicker
(Fig. 1 lc;cf.Breen 1989). For a given indenter displacement,
more sand must be moved where the sand is thicker than
Figure 13. Photo showing the sand lobe developed on the foreland side of
where it is thin. In order to accommodate for extra sand in the
two glass obstacles. The curvature of originally straight faults to the wedge, and at the same time maintain the critical taper,
hinterland of the obstacles was imposed after collision, and the curvature of faulting propagates further into the foreland (Fig. He).
the faults to the foreland of the obstacles is primary. Movement is from the Therefore, an antitaxial lobe develops over the region where
top to the bottom of the photo. sediment was originally thicker.
90
GENERATION OF CURVED FOLD-THRUST BELTS
What controls the shape of curves? caused by restraints on movement (e.g. obstacles, strike-slip
faults).
Deformation fronts of antitaxial curves range in shape from
The shape of the deformation front of an antitaxial curve
parabolic (e.g. the Himalayas) to flat-crested (e.g. the
changes as the fold-thrust belt grows into the foreland. All of
Carpathians) in plan. Furthermore, within a given belt, the
the models (e.g. Figs 3 & 10) show that the interlimb angle of
foreland fault traces tend to be more parabolic than hinterland
the curves decreases as the deformation front propagates into
fault traces (Figs 6, 10c, 14b). Elastic models suggest that
the foreland. The sand model, in addition, suggests that
such a pattern reflects the geometry of the instantaneous
during early stages in the formation of antitaxial curves, curve
stress field (Fig. 14). The analytical parallel-flow model
geometry is controlled by the ratio between the length of the
suggests that the shape of a curve (parabolic vs. flat-crested)
indenter and the thickness of the sediment being deformed.
may also depend on the rheology of the material being
Qualitatively, if the ratio is very small, then the curve initiates
deformed (cf. Johnson 1970) which could reflect the defor-
as a parabola, if it is intermediate, then the curve initiates as
mation mechanism contributing to strain development in a
an arc of an ellipse, and if the ratio is large, the fault is curved
belt (e.g. pressure-solution mechanisms in shallow belts,
near the ends of the indenter but is parallel to the indenter face
dislocation glide/climb in deeper belts; Spratt 1983).
away from the ends. Initially flat-crested deformation fronts
A major difficulty with the analytical model is the require-
evolve into arcs of ellipses as the thrust wedge grows. Each
ment that displacement trajectories of all points on the marker
new fault at the toe initiates with progressively lower eccen-
line be straight and mutually parallel. Since the work of Seuss
tricity, until ultimately the deformation front resembles a
(1904) and Argand (1924; Fig. 15), it has been recognized
parabola. Once formed, a thrust sheet is subjected to pure
that movement trajectories in real orogens may be divergent.
shear as new thrust sheets add to the toe of the wedge.
Modem kinematic data from real orogens suggest that move-
Therefore, the curvature of a given fault decreases with time.
ment at any given point along the curve is typically perpen-
This effect coupled with change in fault shape as a function
dicular to the curve; for example, the P-axes of fault-plane
of distance from the indenter face means that curves towards
solutions for earthquakes in the Himalaya are perpendicular
the hinterland are flatter than curves toward the foreland, a
to the range front (Seeber 1984) and slip lineations on thrust
phenomenon that is observed in many antitaxial fold-thrust
faults in the Wyoming salient are typically perpendicular to
belts (e.g. the Verkhoyansk and Jura Ranges; Figs 6 & 14b).
the regional map trace of the faults (Crosby 1969). Also,
The finite-difference model also predicts such variation in
considering the structural complexity that is locally charac-
cui^ve geometry (Fig. 5) as a function of position in the wedge.
teristic of fold-thrust belts (e.g. Coward & Potts 1983; Marshak
1986), motion of discrete blocks within a belt may be some-
what chaotic, thereby leading to non-parallel movement Do faults initiate with curved traces, or are curves
especially near the endpoints of curves. secondary?
Finite-difference models suggest that the contrast be-
One of the principal debates concerning curved fold-thrust
tween parabolic and flat-crested curves need not reflect
belts concerns the question of whether curves are primary
rheology, but alternatively may reflect the overall strength of
(i.e. initiate with their present structural trends) or are second-
the fold-thrust belt (perhaps controlled by lithology, degree
ary (i.e. are oroclines). The models in this paper emphasize
of fracturing, or deformation conditions). It was found that
that true oroclines can form due to superposition of cross-
curve shape changes as a function of the critical angle, a term strike shear on an originally straight fold-thrust belt, and due
representing strength in the model. As the critical angle to collision with obstacles in the foreland (see Carey 1955 and
increases, the flow front becomes more parabolic. This Marshak 1988 for discussion of other settings in which
model suggests that antitaxial curve shape reflects interaction oroclines form). In the case of collision with obstacles, only
between a regional topographic potential driving the orogen structures that had formed prior to collision are bent. Primary
into the foreland (e.g. caused by plate convergence or curves can form in advance of an indenter, to the foreland of
overthickening of crust) and local topographic potentials obstacles in coUisional zones, or as a consequence of along-
strike variations of sediment thickness in the foreland. Curves
like the Pennsylvania salient, for example, which are spa-
tially associated with thicker strata deposited in a continental
re-entrant, may have initiated with a curved trajectory and
need not reflect significant oroclinal bending. Palaeomagnetic
data so far have not demonstrated large amounts of bending
in this salient, and structural data have not shown major
along-strike extension (Schwartz & Van der Voo 1984;
Miller & Kent 1986). The initial movement on a curved fault
at the toe of the thrust wedge results in some tangential
extension and some rotation, but the amount is less than
would accompany the development of a curved fault trace
Figure 15. Reproduction of a sketch from Argand (1924) showing his from an originally straight fault and may explain the lack of
concept of a 'double virgation of the first type' (i.e. divergent flow in a curved reported tangential extension in many natural orogenic curves.
orogen). The arrows indicate assumed flow lines.
91
S. MARSHAK ET AL.
Partial support for this work was provided by NASA grant NAGS-1312 to importance of sediment thickness to us, C. Hedlund for help with the sand-
Hsui, NSF grant EAR-84-07785 to Marshak, and an NSF graduate fellow- wedge modelling, and G. P. Salisbury for suggestions concerning curve
ship to Wilkerson. We wish to thank P. R. Cobbold and K. R. McClay for nomenclature.
helpful reviews of the manuscript, R. D. Hatcher, Jr. for emphasizing the
References
Argand, E. 1924 (1977 translation by A.V. Carozzi). Tectonics of Asia. Hafner Marshak, S. 1986. Structure and tectonics of the Hudson Valley fold-thrust
Press, New York, 218p. belt, eastern New York State. Geological Society of America Bulletin, 97,
Beutner, E. C. 1977. Causes and consequences of curvature in the Sevier 354-368.
erogenic belt, Utah to Montana. In: Wyoming Geological Association 1988. Kinematics of orocline and arc formation in thin-skinned
Twenty-Ninth Annual Field Conference Guidebook. Wyoming Geo- orogens. Tectonics, 1, 73-86.
logical Association, 353-365. & Tabor, J. R. 1989. Structure of the Kingston orocline in the
Breen, N. A. 1989. Structural effect of Magdalena fan deposition on the Appalachian fold-thrust belt. New York. Geological Society of America
northern Columbia convergent margin. Geology, 17, 34-37. Bulletin. 101,6^3-101.
Burchfiel, B. C. 1980. Eastern European Alpine system and the Carpathian McClay, K. R. & Ellis, P. G. 1987. Analogue models of extensional fault
orocline as an example of collision tectonics. Tectonophysics, 63,31 -61. geometries. In: Coward, M. P., Dewey, J. F. & Hancock, P. L. (eds)
Burg, J. P., Bale, P., Brun, J. P. & Girardeau, J. 1987. Stretching lineation and Continental Extensional Tectonics. Geological Society Special Publi-
transport direction in the Ibero-Armorican arc during the Siluro-Devonian cation, 109-125.
collision. Geodinamica Acta, 1, 71-87. Miller, J. D. & Kent, D. V. 1986. Paleomagnetism of the Upper Devonian
Carey, S. W. 1955. The orocline concept in geotectonics. Papers and Pro- Catskill Formation from the southern limb of the Pennsylvania salient:
ceedings of the Royal Society of Tasmania, 89, 255-288. Possib\eevidenceo{oroclinalrotation.Geophysical Research Letters,13,
Coward, M. P. & Potts, G. L. 1983. Complex strain patterns developed at the 1173-1176.
frontal and lateral tips to shear zones and thrust zones. Journal of Mulugeta, G. 1988. Modelling the geometry of Coulomb thrust wedges.
Structural Geology, 5, 383-400. Journal of Structural Geology, 10, 847-859.
Crosby, G. W. 1969. Radial movements in the western Wyoming salient of & Koyi, H. 1987. Three-dimensional geometry and kinematics of
the Cordilleran overthrust belt. Geological Society of America Bulletin. experimental piggyback thrusting. Geology, IS, 1052-1056.
80, 1061-1078. Nalivkin,D.V. (editor-in-chief) X'ibi.Geological map of the Union ofSoviet
Davis, D., Suppe, J. & Dahlen, F. A. 1983. Mechanics of fold-and-thrust belts Socialist Republics, sheet 7. Ministry of Geology of the USSR,
and accretionary wedges. Journal of Geophysical Research, 88, 1153- 1:2,500,000.
1172. Rankin, D. W. 1976. Appalachian salients and recesses: Late Precambrian
Davy, P. & Cobbold, P. R. 1988. Indention tectonics in nature and experi- continental breakup and the opening of the lapetus Ocean. Journal of
ment. 1. Experiments scaled for gravity. Bulletin of the Geological In- Geophysical Research, 81, 281-288.
stitute of the University of Uppsala, N.S., 14, 129-141. Ries, A. C. & Shackleton, R. M. 1976. Pattemsof strain variation in arcuate
Eldredge, S., Bachtadse, V. & Van der Voo, R. 1985. Paleomagnetism and fold belts. Philosophical Transactions of the Royal Society ofLondon A,
the orocline hypothesis. Tectonophysics, 119, 153-179. 282,281-288.
Horsfield, W. T. 1977. An experimental approach to basement controlled Royden, L. & Burchfiel, B. C. 1989. Are systematic variations in thrust belt
faulting. Geologic en Mijnbouw, 56, 363-370. style related to plate boundary processes? Tectonics, 8, 51-62.
Hsui, A. T., Wilkerson, M. S. & Marshak, S. 1990a. Generation of non- Schwartz, S. Y. & Van der Voo, R. 1984. Palaeomagnetic evaluation of the
Newtonian-like mountain belt curvatures within a Newtonian crust orocline hypothesis in the central and southern Appalachians. Geophysical
(abst.). Transactions of the American Geophysical Union (EOS), 71,637. Research Letters, 10, 505-508.
, & 1990b. Topographically driven crustal flow and the Seeber, L. 1984. Large components of rotation and/or longitudinal stretching
generation of curved fold-thrust belts. Geophysical Research Letters, (in are predicted for the Himalayan accretionary wedge (abst.). International
press). Association of Seismology and Physics of the Earth's Interior, Regional
& Youngquist, S. 1985. A dynamic model of the curvature of the Assembly.
Mariana Trench. Nature, 318, 455-457. Seuss, E. 1904. The Face of the Earth. Clarendon Press, 604p.
Jaeger, J. C. & Cook, N.G.W. 1979. Fundamentals of rock mechanics. Spratt, D. A. 1983. Deformation mechanisms associated with the initiation
Chapman & Hall, London, 593p. and propagation of thrust faults (abst.). Transactions of the American
Johnson, A. M. 1970. Physical processes in geology. Freeman, Cooper & Geophysical Union (EOS), 64, 318.
Company, San Francisco, 577p. Thomas, W. A. 1977. Evolution of Appalachian-Ouachita salients and
Lambe,T. W.& Whitman, R.V. l979.Soil mechanics,SI version.3o\\nV^i\ey recesses from reentrants and promontories in the continental margin.
& Sons, New York, 553p. American Journal of Science, 277, 1233-1278.
Laubscher, H. P. 1972. Some overall aspects of Jura dynamics. American Turcotte, D. L. & Schubert, G. 1982. Geodynamics applications of con-
Journal of Science, 272, 293-304. tinuum physics to geological problems. John Wiley & Sons, New York,
Lillie, R. J. & Davis, D. M. 1990. Structure and mechanics of the foldbelts 450p.
of Pakistan. In: McClay, K. R. (ed.) Thrust Tectonics 1990, Programme Wilkerson, M. S., Marshak, S. & Hsui, A. T. 1988. Dynamic modelling of
with Abstracts, Royal Holloway and Bedford New College, University an orocline in the Idaho-Wyoming fold-thrust belt: Implications for
of London, (4-7 April, 1990), 69. modelling the rheology of the upper crust. Geological Society ofAmerica
Lowrie, W., & Hirt, A. M. 1986. Paleomagnetism in arcuate mountain belts. Abstracts with Programs, 20, A58.
In: Wezel, F.-C. (ed.) The Origin ofArcs, Developments in Geotectonics
27. Elsevier, 141-158.
92
PART THREE
Thrust geometries and thrust systems
Thrust faults in inverted extensional basins
K. R. McClay & P. G. Buchanan
Abstract: Thrust faults in inverted extensional basins may not exhibit the simple low-angle, ramp-
flat thrust trajectories that are characteristic of many foreland fold and thrust belts, but are expected
to show more complex geometries controlled by the architecture of the earlier extensional fault
system. The results of sandbox analogue models are reviewed as guides to possible thrust fault
architectures in contracted extensional basins. High-angle thrust faults and thrust faults with
convex-upwards shapes are characteristic of inverted extensional fault systems. Footwall-vergent
shortcut thrusts and hangingwall-vergent backthrusts are characteristic of the inversion models.
Fault sequences are complex and do not follow simple 'footwall-nucleating - footwall-vergent'
rules. The architecture of the pre-existing extensional fault system exerts a strong control on the
character of thrust faults in inverted basins and must be taken into account when analysing such
regions.
Inversion tectonics involves a switch in tectonic mode from Thrust systems in inverted basins
extension to compression such that extensional basins are
contracted and become regions of positive structural relief. Thrust systems in inverted basins vary from small, high-angle
This process is generally accepted to involve the reactivation thrust faults in basins which have undergone only small
of pre-existing extensional faults such that they undergo amounts of contraction to large, low-angle thrust systems
reverse slip and eventually may become thrust faults (Cooper which override and obliterate the pre-existing extensional
& Williams 1989; Williams et al. 1989). Pre-existing architecture (Fig. 2). Extensional fault systems which have
extensional basins have been incorporated into many fold and undergone low amounts of contraction (inversion) commonly
thrust belts - in the Alps (Butler 1989; de Graciansky et al. show reverse movement on only a few of the faults that could
1989; and others), in the Pyrenees (Puigdefabregas et al. possibly have undergone reactivation (Fig. 2a). For example
1986; Berastegui et al. 1990; Martinez et al. 1989) and the the inversion structures across the South Hewett fault zone in
Canadian Cordillera (McClay ef a/. 1989). The architecture the southern North Sea show a growth anticlinal structure
of thrust systems in inverted extensional basins will be with the development of a number of small footwall shortcut
influenced and in some cases strongly controlled by the pre- thrusts and only localized reactivation of extensional faults in
existing extensional fault geometries (cf. Cooper & Williams the main domino array within the basin (Fig. 2a). Growth
1989; Coward et al. 1991; McClay et al. 1989; Hayward & folding (i.e. the fold develops in strata that are being depos-
Graham 1989). In such situations thrustfaults have geometries, ited at the same time as the fold grows, e.g. Suppe etal. (1991,
trajectories and sequences that are significantly different to this volume)) is typically developed with thinned growth
those found in many foreland fold thrust belts which deform strata (Suppe et al. 1991, this volume) deposited above the
platformal strata (e.g. Dahlstrom 1970; Boyer & Elliott reactivated basin margin fault and thicker stratigraphic inter-
1982). vals deposited on the flanks of the structure (Fig. 2a). Within
Inversion geometries have been commonly modelled as the inverted domino fault array the thrust faults are typically
involving complete reactivation of the basin forming small, high-angle faults which show net extensional dis-
extensional faults such that the extensional sedimentary placement at depth. In examples where listric extensional
wedge is pushed up - 'inverted' above the pre-extensional faults have been reactivated during inversion, hangingwall
'regional elevation' (Fig. 1, Bally 1984). Such conceptual anticlines are developed and 'pop-up' structures in the region
models of inversion ignore the mechanics of the inversion of the crestal collapse graben (McClay 1990) are character-
process and do not necessarily resemble natural inversion istic (Fig. 2b). At greater values of contraction reactivated
geometries (Fig. 2). It is clear that the range in structural extensional faults propagate into the post-rift and syn-inver-
styles and geometries resulting from the inversion and con- sion sequences, flattening upwards as they do so (Fig. 2c).
traction of previously distended crust is likely to be far more Complex 'arrowhead' or 'harpoon' geometries result from
complex than the relatively simple imbricate thrust geometries the wedge of syn-rift sediments being elevated above regional.
in areas of layer cake stratigraphy (Boyer & Elliott 1982). Where extensional half-grabens have been subjected to greater
This paper aims to illustrate thrust geometries in inverted contraction, thrust faults may either transect across the half-
extensional basin systems at various scales and to demonstrate, graben thus dissecting it (Fig. 2d) or new thrust faults will
using sandbox analogue models, the geometries and mechanics develop out of the half-graben (Fig. 2d). In cases of strong
of thrusting during inversion. contraction pre-existing extensional faults may also be rotated
93
K. R. MCCLAY & P. G. BUCHANAN
Experimental method
Two types of sandbox experiment are briefly reviewed in this
paper. The first type, the simple listric and the ramp-flat listric
fault shapes, consisted of a fixed footwall geometry (wooden
block) over which a deformable hangingwall was first extended
and then contracted by horizontal recompression (see also
Buchanan & McClay 1991). The second type of model
experiment was designed to simulate a series of rigid domino
Post Inversion blocks that firstly undergo extension and then are inverted or
pushed back to their pre-extensional configuration.
T7A =v- y Basement
The simple listric and ramp-flat listric experiments were
carried out in a glass walled deformation rig (Fig.3a) 150 cm
Figure 1. Conceptual model of an inverted half-graben system (modified long, 20 cm wide and up to 20 cm deep. Initial model
after Bally 1984). (a) Extension; (b) Partial inversion: (c) Total inversion. dimensions were typically 30 cm long, 20 cm wide and 10 cm
deep. The sand models were constructed between the end
and re-utilized as thrust faults (Fig. 2d). Basement blocks walls of the deformation rig and consisted of alternating
may be incorporated into the thrust system via footwall layers of white and coloured sand (average grain size 300
'shortcut' thrusts thus creating exotic horses within the thrust |im). Dry cohesionless quartz sand has an angle of friction of
complex (Fig. 2e). Extensional faults which are unfavour- 31° (Liu et al. this volume) and successfully simulates the
ably oriented for reactivation remain bypassed by the brittle deformation of sedimentary rocks in the upper crust
contractional tectonics (Fig. 2e). In areas of very strong (McClay 1990; Horsfield 1977). Faulted sand has a lower
thrusting such as the Alps (Butler 1989; de Graciansky et al. strength than unfaulted sand (Liu etal. 1991, this volume) and
1989) the thrust faults appear to bypass pre-existing extensional as such one would expect fault zones formed in extension to
faults which display little or no reactivation. The thrust at least partly reactivate during inversion. The models are
trajectories are, however, strongly controlled by the broadly scaled such that 1 cm in the model represents between
extensional fault architecture with thrust ramps located at 100 m and 1 km (McClay 1990).
pre-existing extensional fault steps. Deformation was achieved by moving one of the end walls
In the examples shown in Figure 2 it is obvious that the at a constant displacement rate of 4.16 x 10"^ cm sec'. A
architecture of the extensional basin fault system exerts a plastic sheet between the sand model and the footwall block
strong influence on the thrust systems developed during / base of the deformation rig simulates a low friction (^^ = 0)
inversion and thrust faulting. If the thrust faults nucleate or basal detachment. For extension the hangingwali is pulled
grow from reactivated extensional faults they may be shal- down and along the basal detachment, whereas for inversion
low-dipping in their upper sections but steepen downwards, the hangingwall is pushed back up the detachment. During
thus not following a more conventional ramp-flat trajectory. extension syn-rift layers were incrementally added to the
The volume problem caused by contracting an extensional model in order to simulate syn-rift sedimentation. At the end
half-graben may generate new thrusts which nucleate and of the extensional phase, a thin post-rift sequence was added
propagate out of the 'half-graben' not necessarily in a footwall to the models and during the contraction, syn-inversion layers
propagating sequence. The stratigraphic sequences under- were introduced in order to prevent the formation of unstable
going thrusting are not 'layer cake' but exhibit significant surface slopes as thrust faults propagated to the surface of the
thickness changes with the resultant implications for thrust model.
trajectories. Hangingwall anticlines in extensional basin The second type of experiment was designed to simulate
94
THRUST FAULTS IN INVERTED EXTENSIONAL BASINS
2.0-^ ^ , ^ \ ^ • ^ ^ • ^ ^ ^ \ ^ v ' ^ \ ^ ^ v ^ \ ^ ^ ^ ^ ^ ^ ^ ^ • ^ ^ ^ • ^ ^ ^ \ ^ ^ \ ^ x \ ^ ^ ^ ^ ^ \ ^ ^ \ ^ ^ • ^ ^
seconds Basement Pre-rift Post-rift
•*. \ V
Pre-Triassic TrJassic Lower Cretaceous
Syn-lnversion Post-Inversion 5 km
i^ Late-Cretaceous-Miocene Pliocene-Recent
1 km
2 km -
^^^^^^^m^m
\ ^ X \ \ •s s
3 km
SW NW
km
1 km
h » J Basement Ijjjjjjjji;! Syn-rifl | | Post-rift | % ^ Syn-lnversion
Figure 2. Examples of thrust faults in inverted basin systems, (a) Geoseismic section across the South Hcwett fault zone, southern North Sea. The inverted
domino fault array shows a major growth anticline at the left hand end of the section and high angle thrust faults have produced arrowhead or 'harpoon'
structures at the right hand end of the section; (b) Cross-section through the Eakring oilfield, England (modified after Eraser ef a/, 1990). The extensional
half-graben system has undergone partial inversion with contractional displacement on the main extension fault and a 'pop-up' at the location of the Caunton
oil field; (c) Cross-section through pan of the Sunda Arc (modified after Letouzey 1990) showing inversion of half-graben fault systems to produce
characteristic arrowhead and harpoon geometries.
95
-*''^' K. R. MCCLAY & P. G. BUCHANAN
a 1A •••. T - r n . i o .
SW "Re-utilised extensional fault" "out of the half-graben thrusts" NE
•iTfc^' \ \ \ \ \ \
- ^ V V V V ^ .' ^ .' ^ y ^ "transecting thrust
^ y^^^'^^^'^j-^'y^y'^^'^-'^^^-'^^^y^ cutting down section in^^''.^''^
'-^''•^^y'^y''-y^y'^y''-y^y'^y^^^y'^^^y''the transport direction"^^''^''^
^ • ^ ^ ^ ^ ^ ^ ^ . ^ ^ > . ^ • ^ ^ i ^ . s •. , , ^ , . . ^ % s \ \ \
Basement •
kmO-
HORIZONTAL = VERTICAL
Syn-Rlft Syn-Rift Post-Rift 5 km
Silurian-Lower Devonian Upper-Devonian - Visean ^ Namurian
Figure 2. (d) Strongly inverted haif-graben system from the Kechika Trough, northern Rocky Mountains, Canada (adapted after McClay ei al. 1989). Pre-
existing extensional faults have been rotated and re-utilised. 'Out of the half-graben' thrusts occur together with a through going thrust that cuts down
stratigraphy in the direction of transport; (e) Inversion structures in the South Wales coalfield (modified after Powell 1989). Note the development of short-
cut faults in the footwall to the reactivated Benton fault system.
domino extensional fault blocks and utilized the glass-sided are given in McClay (1990) and Buchanan & McClay {1991).
deformation rig ( Fig. 3b, Buchanan 1991). This apparatus
consists of four rigid domino shaped plates linked by a trellis
Limitations of the analogue models • ~ ' •
system such that they can be simultaneously extended by
moving the end wall to the right and inverted by moving the The limitations of analogue models of inversion tectonics are
end wall back (Fig. 3b). In this series of experiments a 3 cm discussed in detail by Buchanan & McClay (1991) and are
thick pre-rift sequence was placed in the apparatus and synrift briefly summarized here. The most fundamental limitation of
sediments were added incrementally during extension. A 1.5 the first .series of models (the simple listric fault and the ramp-
cm thick post-rift sequence was added at the end of extension flat listric fault models) is that they incorporate a rigid
and then the models were inverted by pushing the end wall footwall fault block (Fig. 3a) and any possible footwall
back such that the domino fault blocks slid back to their pre- shortcut thrusts are confined to the post-rift and syn-inversion
extensional configuration. strata. In addition, the rigid nature of the footwall block in
In both series of experiments, progressive deformation these experiments generates a marked mechanical contrast
during both the extension and inversion phases was moni- between the sandpack and the footwall such that strong
tored using time lapse photography. The completed models buttressing effects are to be expected (see below). The
were impregnated and serially sectioned to study the final sandbox modelling also cannot simulate compaction, ther-
deformation geometry. Details of the experimental methods mal and pore fluid pressure effects, all of which may affect the
96
THRUST FAULTS IN INVERTED EXTENSIONAL BASINS
97
K. R. MCCLAY & P. G. BUCHANAN
Figure 4. Simple listric fault - Experiment 18. (a) Model at the end of 6 cm extension. Top of the pre-rift sequence is marked by the upper grey-black-
grey sequence. Syn-rift grey and white layers were added incrementally up to the top of the footwall block. A 2 cm thick, horizontal post-rift sequence has
been added prior to inversion: (b) Serial section of the model at the end of 7 cm inversion. See text for details.
are generated by the reactivated main detachment, by the out by a hangingwall syncline related to the ramp in the detach-
of graben thrusts and by the hangingwall-vergent backthrust ment, from a larger roll-over with superimposed graben
system (Fig. 4b). structures associated with the lower listric segment of the
For the simple listric fault system the maximum uplift as detachment.
indicated by the upper syn-rift, lower post-rift sequences, is Inversion was achieved in a similar manner to that for the
listric fault model and the magnitude of contraction exceeded
focused above the upper part of the original listric extensional
detachment, and forms the crest of a large growth anticline. extension (by 12 cm). Reactivation of the lower listric
The sequence of thrusting is complex with reactivation of the segment of the basal detachment resulted in the propagation
basal detachment throughout the inversion. Footwall shortcut of a major, footwall-vergent thrust fault upwards and by-
thrusts form a break-back sequence early in the inversion passed the higher and smaller of the two crestal collapse
history whereas forward-breaking backthrusts dominate the graben (Fig. 5b). This footwall-vergent thrust becomes a
latter stages of inversion (see also Buchanan & McClay growth fault with significant thickness variations in the syn-
1991). inversion sequence. Minor thrusts parallel to the major thrust
in its hangingwall were also generated. The geometry of this
footwall-vergent thrust is relatively planar with almost no
Ramp-flat listric faults deformation in its footwall (Fig. 5b). Thus the upper exten-
The structural architecture which evolves above an exten- sional roll-over structure is preserved as an exotic, fault-
sional ramp-flat detachment geometry has been discussed in bounded, lozenge-shaped block unaffected by the contrac-
detail by Ellis & McClay (1988) and McClay & Scott (1991). tion. As in the case of the listric fault model (Fig. 4b), the
The basic structural elements are similar to those seen above curvature of the basal detachment at the lower listric section
the listric fault described above. An upper roll-over and results in the development of a series of 'forward-breaking',
associated crestal collapse graben develops above the upper hangingwall-vergent 'backthrusts'. These have a convex-
listric portion of the detachment (Fig. 5a). This is separated upwards profile geometry caused by the presence of the non-
98
THRUST FAULTS IN INVERTED EXTENSIONAL BASINS
POST-RIFT
Figures. Ramp-flat lislric fault - Experiment I 12. (a) Model at the end of 6 cm extension. Themodelconsistsof alternating layers of coloured sand with
thin mica layers between them in order to facilitate layer parallel slip (McClay & Scott 1991). The top of the pre-rift sequence is marked by the upper grey-
black-grey sequence. Syn-rift grey and white layers were added incremenlally up to the top of the footwall block. A 2 cm ihick, horizontal post-rift sequence
has been added prior to inversion; (b) Serial section of the model at the end of 18 cm inversion. See text for details.
POST-RIFT
Figure 6. Domino fault system, (a) Model at the end of 13 cm extension. Top of the pre-rift sequence is marked by the upper grey-black sequence. Syn-
rift grey and while layers were added incrementally up to the top of the fooiwail block. A 1.5 cm ihick, horizontal post-rift sequence has been added prior
to inversion; (b) Serial section of the model at the end of 13 cm of inversion. See text for details. No syn-inversion strata have been added lothis model,
99
K. R. MCCLAY & P. G. BUCHANAN
deflectable end wall, i.e. this is an end effect reflecting the whilst below, the layering remains in net extension. As a
limitation of the experimental apparatus. These forward- result, null points where net displacement across the fault is
breaking backthrusts, in conjunction with the major footwall- zero and classic arrowhead or harpoon structures are formed.
vergent thrust, produce a characteristic broad 'pop-up' struc- Reactivation of the antithetic faults during inversion does not
ture centred above the soling-out point of the lower listric generally occur in these experiments (Buchanan 1991). The
section of the original extensional detachment. The maximum style of the contractional deformation is strongly asymmet-
uplift in the hangingwall is located above the top of the ramp ric, all the thrust faults verging in the same direction (i.e.
section of the detachment between the upper flat and the footwall-vergent) up the dip of the reactivated domino faults.
lower listric segment. As a consequence of the relatively The sequence of fault reactivation generally occurs from right
planar nature of the major footwall-vergent thrust, little to left in the models as the right hand end wall is pushed back
rotation is observed in the hangingwall which is simply to its pre-extensional position.
translated up and over the undeformed footwall. The footwall-
vergent thrust becomes 'thin-skinned' in this region where
the hangingwall beds of the pre-rift are sub-parallel to the Discussion
fault surface (Fig. 5b). The sequence of thrusting is that of
reactivation of the lower part of the basal detachment; gen- This paper has briefly illustrated some of the thrust geometries
eration of the major footwall-vergent thrust (that bypasses the found in both experimental and natural examples of inverted
upper part of the extensional detachment); formation of sedimentary basins. Clearly, such structures have particular
minor hanging wall-vergent thrusts that nucleate from the tips characteristics that are markedly different from the classic
of the extensional faults bounding the lower crestal collapse 'ramp - flat' thrust systems that deform the platformal regions
graben; and finally the development of major forward-break- of continental margins (e.g. Rocky Mountains and the Ap-
ing backthrusts as a result of buttressing against the lower palachians, Bally era/. 1966;Dahlstrom 1970; Boyer& Elliott
listric part of basal detachment surface. 1982).
The experimental models of inverted extensional fault
systems illustrate the sequential development of reactivation
Domino faults
of pre-existing extensional faults as thrust faults, the formation
The extensional architecture produced above a series of of growth structures and the nucleation of new thrust faults.
domino faults is illustrated in Figure 6a and is described in
detail in Buchanan (1991). The structure is similar for each
Reactivation of pre-existing extensional faults as thrust
domino fault and consists of a half graben filled with a syn-
faults
rift wedge in which the layering is planar. The final profile
geometry of the domino faults has a slight listric shape In the experimental models pre-existing extensional faults
acquired during extension as a result of the interaction between were reactivated as high-angle thrust faults whose architec-
sedimentation and rotation (Vendeville & Cobbold 1988). ture was controlled by the shape of the extensional detachment
Minor planar antithetic faults form in the hangingwall to each system (Fig. 7). Simple listric faults inverted such that the
domino fault and root at the top of the rigid basement. These fault takes on a concave-up geometry and the thrust fault
faults increase in size away from the static margin fault on the steepened upwards into the post-rift and syn-inversion strata
left-hand side of the model. (Fig. 7a). In ramp-flat listric fault systems the lower listric
Inversion of the half graben is achieved by pushing the part of the extensional detachment reactivated with a new
individual fault blocks back to their pre-extensional configu- footwall-vergent thrust system nucleating upward from the
ration such that the top of the rigid basement is returned to ramp such that the upper part of the listric extensional
horizontal. Contractional strain in the sand pack above is detachment was bypassed entirely (Fig. 7b). In the domino
accommodated by reactivation of the domino faults and models the extensional faults can reactivate and became
along several footwall shortcut faults (Fig. 6b). Displace- convex upwards in the upper part of the syn-rift and post-rift
ment is generally transferred from the domino faults in the strata (Fig. 7c). The inversion produces characteristic wedge
basement onto the shortcut faults in the cover. These shortcuts shaped syn-rift geometries that are termed 'harpoon' or
are planar to slightly convex upwards in profile and define arrowhead structures (Badley etal. 1989; Buchanan & McClay
downward tapering wedge-shaped blocks at the margin of the 1991). In many cases the thrust faults are typically steep and
individual half grabens. The contractional strain associated their shapes strongly controlled by the pre-existing extensional
with each reactivated domino fault is concentrated in the fault architecture.
immediate footwall. Reactivation of the entire length of the
domino faults causes steeply dipping thrust faults to develop
in the post-rift sequence (Fig. 6b). All the extensional and Reactivation of crestal collapse graben faults
contractional faulting in the cover relating to each domino Crestal collapse graben faults (Figs 4, 5, 7a & b) show only
fault, links into a single master fault in the rigid basement partial reactivation with new thrust faults nucleating from the
below. Reactivation of individual domino faults causes post- upper tips of both the antithetic and synthetic crestal collapse
rift and upper syn-rift layers to be put into net contraction graben faults. Extensional architecture is preserved at depth
100
THRUST FAULTS IN INVERTED EXTENSIONAL BASINS
Figure 7. Summary diagrams of inverted extensional fault systems, (a) Inverted listric extensional fault showing the rigid basement footwall block, pre-
rift layers, syn-rift layers added incrementally during extension, post-rift layers and syn-inversion strata, (b) Inverted ramp-flat listric extensional fault
system showing the rigid basement footwall block, pre-rift layers, syn-rift layers added incrementally during extension, post-rift layers and syn-inversion
strata. Note the development of a prominent hangingwall bypass thrust that leaves the upper roll-over unaffected by the inversion. Major backthrusts lead
to the development of a triangular 'pop-up' structure, (c) Diagram for the inversion of a domino fault array showing reactivation of the domino faults and
the development of footwall shortcut thrusts in the pre-rift and syn-rift strata.
(Figs 7a & b). The thrust faults typically have a convex bounding extensional faults (Figs 2a, 4b & 7a). They are
upward trajectory flattening into the post-rift sequence. fault-propagation growth folds, typically asymmetric with
onlap sequences in the syn-inversion strata. The growth
Growth structures asymmetry is footwall-vergent (Figs 2a & 4b).
101
K. R. MCCLAY & P. G. BUCHANAN
102
THRUST FAULTS IN INVERTED EXTENSIONAL BASINS
Regional Regional
Regional Regional
Regional
•jp^fi• ^ • ^ • ^ • ^ •^
Regional
• y
•
\ \ \ \ \ \
y
y
y
y
y y y
• y y y y y y y y
• y y y y y y y y y y
• y y y y y y y y y y y y
f ^ y y y y y y y y y y y y y y y y y y y y y y y y y y y y y
Figure 8. (a- j) Conceptual models for thrust faults developed by the dip-slip inversion of a listric fault system. See text for detailed discussion. Note the
regional at the top of the syn-rift strata in (a).
103
K. R. MCCLAY & P. G. BUCHANAN
References
Badley, M. E., Price, J. D. & Blackshall, L. C. 1989. Inversion, reactivated inversion. Bulletin American Association of Petroleum Geologists, 69,
faults and related structures - seismic examples from the southern North 582-600.
Sea. In: Cooper, M. A. & Williams, G. D, (eds). Inversion Tectonics. Hayward, A. B. & Graham, R. H. 1989. Some possible characteristics of
Geological Society of London Special Publication, 44, 201-219. inversion. In: Cooper, M. A. & Williams, G. D. (eds) Inversion Tecton-
Bally, A. W. 1984. Tectogenese et sismique reflexion. Bulletin Societe ics. Geological Society of London Special Publication, 44, 17-39.
Geologique France, 7, 279-285. Horsfield, W.T. 1977. An experimental approach to basement controlled
Bally, A.W., Gordey, P.L. & Stewart, G.A. 1966. Structure, seismic data, and faulting. Geologic en Mijnbouw, 56, 363 - 370.
orogenic evolution of the southern Canadian Rocky Mountains. Bulletin Letouzey, J. 1990. Fault reactivation, inversion and fold-thrust belt. In:
of Canadian Petroleum Geology, 14, 337 - 381. Letouzey, J. (ed.) Petroleum and Tectonics in Mobile Belts. IFP Editions
Berastegui, X., Garcia-Senez, J. M. & Losantos, M. 1990. Tectono-sedimen- Technip, Paris, 101-128.
tary evolution of the Organya extensional basin (central south Pyrenean Liu, Huiqui, McClay, K. R. & Powell, D. P. 1991. Physical models of thrust
unit, Spain) during the Lower Cretaceous. Bulletin Societe Geologique wedges (this volume).
France, »,25\ -264. McClay, K. R. 1990. Extensional fault systems in sedimentary basins. A
Boyer, S. M. & Elliot, D. W. 1982. Thrust systems. Bulletin American As- review of analogue model studies. Marine and Petroleum Geology, 7,206
sociation of Petroleum Geologists, 66, 1196-1230. -233
Buchanan, P. G. 1991. Geometries and kinematic analysis of inversion McClay, K. R., Insley, M. W. & Anderton, R. 1989. Inversion of the Kechika
tectonics from analogue model studies. Ph.D thesis (unpubl.). University Trough, Northeastern British Columbia, Canada. In: Cooper, M. A. &
of London. Williams, G. D. (eds) Inversion Tectonics. Geological Society of London
Buchanan, P. G. & McClay, K. R. 1991. Sandbox experiments of inverted Special Publication, 44, 235-257.
listric and planar faults systems. /«: P. R. Cobbold (ed.) Experimental and McClay, K. R. & Scott, A. D. 1991. Experimental models of hangingwall
Numerical Modelling of Continental Deformation. Tectonophysics, 188, deformation in ramp-flat hstric extensional fauh systems. /«: P. R. Cobbold
97-115. (ed.) Experimental and Numerical Modelling of Continental Deforma-
Butler, R.W.H. 1989. The influence of pre-existing basin structure on thrust tion. Tectonophysics, 188, 85-96.
system evolution in the Western Alps. In: Cooper, M. A. & Williams, G. Martinez, A., Verges, J., Clavell, E. & Kennedy, J. 1989. Stratigraphic
D. (eds) Inversion Tectonics, Geological Society of London Special framework of the thrust geometry and structural inversion in the south-
Publication, A4, 105-122. eastern Pyrenees: La Garrotxa area. Geodinamica Acta, 3, 185-194.
Cooper, M. A. & Williams, G. D.(eds) 1989. Inversion Tectonics. Geological Powell, C. M. 1989. Structural controls on Palaeozoic basin evolution and
Society of London Special Publication, 44, 376p. inversion in southwest Wales. Journal of the Geological Society of
Dahlstrom, C.D.A. 1970. Structural geology in the eastern margin of the London, 146, 439-446.
Canadian Rocky Mountains. Bulletin of Canadian Petroleum Geology, Puigdefabregas, C , Munoz, J. A. & Marzo, M. 1986. Thrust belt develop-
18, 332-406. ment in the eastern Pyrenees and related depositional sequences in the
Ellis, P. G. & McClay, K. R. 1988. Listric extensional fault systems: results southern foreland basin, /n; Allen, P. A. & Homewood, P. (eds) Foreland
from analogue model experiments. Basin Research 1, 55-71. Basins. International Association of Sedimentologists Special Publica-
Etheridge, M. A. 1986. On the reactivation of extensional fault systems. tion, 8, 229-246.
PhilosophicalTransactions ofthe Royal Society ofLondon,Series A,3n, Sibson, R. H. 1985. A note on fault reactivation. Journal of Structural Ge-
179-194. ology,1,151-15^.
Fraser, A. J., Nash, D. F., Steele, R. P. & Ebdon, C. C , 1990. A regional Suppe, J., Chou, G. T. & Hook, S. C. 1991. Rates of folding and faulting
assessment of the intra-Carboniferous play of Northern England. In: determined from growth strata (this volume).
Brooks, J. (ed.) Classic Petroleum Provinces. Geological Society of London Vendeville, B. & Cobbold, P. R. 1988. How normal faulting and sedimen-
Special Publication, 50, 417-440. tation interact to produce listric fault profiles and stratigraphic wedges.
de Graciansky, P. C , Dardeau, G., Lemoine, M. & Tricart, P. 1989. The Journal of Structural Geology, 10, 649-659.
inverted margin of the French Alps and foreland basin inversion. In: Williams, G. D., Powell, C. M. & Cooper, M. A. 1989. Geometry and
Cooper, M. A. & Williams, G. D. (eds) Inversion Tectonics. Geological kinematics of inversion tectonics. In: Cooper, M. A. & Williams, G. D.
Society of London Special Publication, 44, 41-59. (eds) Inversion Tectonics. Geological Society of London Special Publi-
Harding, T.P. 1985. Seismic characteristics and identification of negative cation, 44, 3.-15.
flower structures, positive flower structures, and positive structural
104
Rates of folding and faulting
determined from growth strata
John Suppe', George T. Chou^ & Stephen C. Hook^
Abstract: Many upper crustal folds on the scale of 1 -10 km in compressive mountain belts grow
by kink-band migration as a result of fault-bend, fault-propagation, or box folding. One or both
kink-band boundaries sweep through the rock as the kink bands widen during fold growth. The
kink bands typically have a constant width in pre growth strata, which are strata that existed before
deformation, whereas we predict - and observe on seismic lines - an upward decrease in kink-band
width within the stratigraphic sequence deposited during fault slip and associated fold growth -
here called growth strata. In fact this growth stratigraphic sequence provides a complete,
decipherable record of the kinematics of deformation, much in the same way that sea-floor
magnetic anomalies provide a decipherable record of plate kinematics. It is the continual addition
of material that provides the detailed record of motion in both cases. In the simplest folds the kink
band has a constant width within the pregrowth strata, narrows upward through the growth strata,
and finally has a zero width at the top of the growth stratigraphy.
The rate of folding can be described geometrically as the rate of kink-band migration divided by
the sedimentation rate, which is dimensionless. Furthermore, if key beds can be dated, absolute
fault-slip rates can be computed. Examples are presented from the Gulf of Mexico, the Philippines,
Venezuela, California, and Oklahoma. Folding rates in many of these examples are observed to
be 1-2 mm/y lasting for 1-8 Ma. Translation of folding rates into fault-slip rates requires
knowledge of the fault shape and its relationship to fold shape. Nevertheless, even if fault geometry
is poorly constrained, fault slip rates generally can be estimated to within narrow limits based on
rates of kink-band migration, a fact that makes growth structures valuable for assessing earthquake
hazards.
It should be noted that we are not arguing that constant-dip kink-band migration is the only important folding mecha-
nism in the upper crust, rather that the predictions of these theories of fault-related folding are in fact borne out in a
number of specific cases. Some upper crustal folding does involve progressive limb rotation. For example, some folds
appear to lock up and then amplify at a later stage in deformation (e.g. Suppe & Medwedeff 1990, Fig. 1). The case of
progressive limb rotation is beyond the scope of this paper.
105
J. SUPPE ET AL.
a. C.
Active Axial Fixed Axial,
Surface Surface/
Kink-Band Fixed Axial Active Axial
Migration Surface Surface
Figure 1. The five possible kinematic classes of constant-dip kink-band migration. The motion of the axial surfaces relative to the material is indicated.
All five folds appear the same because they are composed of layers that predate the deformation, that is pregrowth strata. Only layers deposited during
deformation, that is growth strata, record the deformational history (Fig. 2).
a. b. C.
Kink-Band
Migration /
Z
T
r
Active Axial Fixed Axial Active Axial
Surface Surface Surface
e. f.
Figure 2. The five possible classes of constant-dip kink-band migration with strata deposited during deformation, called growth strata. The five folds only
differ within the growth strata, which record the kinematics of deformation. It is assumed in these models that sedimentation rate is constant relative to fold
growth rate and that sedimentation keeps ahead of deformation such that there is never bathymetric or topographic expression of the fold. The motion of
the axial surfaces is indicated relative to the material.
106
RATES OF FOLDING AND FAULTING DETERMINED FROM GROWTH STRATA
km
I 1-1.1 J J • • •
107
J. SUPPE EJ AL.
Figure S. The fixed axial surface within the growth strata is called a growth
axial surface, which is the locus of particles that were deposited along the Tb^ Tu+U (2)
active axial surface. The growth axial surface does not bisect the angle
between the fold limbs because there is a change in thickness that reflects a
In general the sedimentation and uplift rates could be continu-
primary change in sedimentation rate across the the active axial surface
(eq. 1). ously varying; therefore, in the general case the orientation of
a growth axial surface will be variable and its shape will be
predicted by the standard fault-related folding mechanisms curved or bent. However, in the seismic examples (Figs 3 &
(although case 2e is unusual, it is known to exist in normal 4) the growth axial surfaces are relatively straight overall
faulting). All five kinematic possibilities have been observed indicating a relatively constant ratio of deformation rate to
at map-scale in nature. Figure 3 is a seismic example similar sedimentation rate. Examples are given later in this paper that
to Figure 2c in which the right-hand axial surface is active and record major variation in this ratio.
the left-hand axial surface is inactive. Figure 4 is a seismic
example similar to Figure 2b.
It should be noted in both the models and the seismic Measurement of kink-band migration rates
examples (Figs 2b, 2c, 3 & 4) that the beginning of deforma-
tion - that is, the pregrowth-growth boundary - can easily be Growth axial surfaces and growth triangles provide a con-
identified as the top of the constant-width kink-band, for tinuous quantitative record of the history of deformation. A
which the two axial surfaces are parallel. Similarly the end of simple measure of the kink-band migration during a time
growth is marked by the apex of the upward narrowing kink interval t is the limb length L divided by the thickness of
band or growth triangle. Also note that the fixed axial surface sediment deposited T^ (Fig. 5), which is here called the
within the growth strata - here called the growth axial surface dimensionless growth rate G.
(Fig. 5) - does not bisect the angle between the fold limbs
km
Figure 6. East-west seismic profile of the east flank of the Allegria anticline, western Cebu, Philippines. The beginning of growth is marked by the inflection
and reversal of dip on the right-hand axial surface, indicating a kink band of the class shown in Figure 2b. Approximately 900 m of growth strata are preserved
which record 600 m of limb lengthening; therefore the dimensionless growth rate G is about 0.7. The average vertical exaggeration in the upper 1 s is about
0.8.
108
RATES OF FOLDING AND FAULTING DETERMINED FROM GROWTH STRATA
• \ 1
109
J. SUPPE ET AL.
Dimensionless Growth
b. c.
Rate G= L/T = 0.6
G= L/T = 0.6
e.
G = 1JT = 8
Angular
1^ Disconformity
Figure 9. Models showing the effect of reducing the sedimentation rate in a growth structure. The upper models are for one active axial surface moving
to the right. The lower models are for two active axial surfaces moving outward relative to the material. Angular disconformities develop when the boundary
between deposition and non-deposition, which is the edge of topographic or bathymetric relief, lies along an active axial surface.
km
l-ls
Figure 10. An east-west seismic profile of a growth structure similar to Figure 9e from Antelope Hills, San Joaquin basin, California. Vertical and horizontal
scales are approximately equal.
110
RATES OF FOLDING AND FAULTING DETERMINED FROM GROWTH STRATA
km
i-1 s
:,: .#ii«angi-;::<ff— p
,_ _-«——"•".'• : iiiiiiiii"*'" ..J..''.
- .-5??$!:3|
Figure 11. Seismic profile showing a short-term decrease in sedimentation rate relative to the rate of limb lengthening at a depth of about 0.9 s; Santa Barbara
Channel, California. The average vertical exaggeration in the upper 1.5 s is about 1.7. The apex of the growth triangle is obscured by horizontal multiples
at depths less than 0.7 s. Seismic profile provided by Texaco USA, BP Exploration Inc., and Sun Operating LP.
orders of magnitude but are expected to be commonly one or complexities are largely ignored in this paper.
two orders of magnitude less than deformation rates, leading Relatively short term fluctuations in sedimentation rate -
to dimensionless growth rates G = LIT^^ of 10-100 or greater for example caused by climactic variation - are expected to
(Fig. 8). A fast deformation rate is expected to be on the order be reflected in the fine structure of the growth axial surfaces.
of centimetres per year whereas fast sedimentation rates are For example the seismic line in Figure 11 shows a local
on the order of millimetres per year. fluctuation in dimensionless growth rate at a depth of about
Growth structures appear drastically different with lower 0.9 s whereas the long-term dimensionless growth rate is
sedimentation rates; for example Figure 9 shows that even a quite constant. It is of course possible that these changes in
modest one-half reduction of sedimentation rate makes the dip of the growth axial surface could reflect short-term
uplifting block non-depositional in the models of Figure 2. fluctuations in deformation rate, but sedimentation rate is a
This gives rise to time-transgressive, cross-cutting surfaces more likely cause given known climatic fluctuations.
of nondeposition, here called angular disconformities. An
example of an onlapping time-transgressive disconformity
similar to Figure 9e is shown in Figure 10. If the uplifting Growth fault-bend folding
block undergoes erosion, then time-trangressive angular
unconformities are predicted, examples of which are shown So far we have considered growth structure only from the
in later sections. perspective of kink-band migration, ignoring the fundamen-
The simple models at high sedimentation rate in Figures 2 tal causes of the motion of active axial surfaces. This
and 9 assume that sedimentation is independent of deformation. approach is useful because growth structure is observed
In contrast we may expect that the bathymetric or topographic commonly at shallow depths on seismic profiles, whereas the
relief that is produced at low sedimentation rates will have a deep structure causing the axial surfaces to move commonly
direct impact on the process of sedimentation. Both sedi- is not imaged or is poorly imaged. Insight into these deep
mentary facies and sedimentation rates will be affected by structural causes is obtained most rapidly by constructing
structure - for example pelagic drape, downslope transport, models of the growth structures that are predicted for the
and shoreline facies - which give rise to more complex fold known common mechanisms of fault-related folding. Fault-
and reflector geometry within the growth strata. The differ- bend folding is discussed in this section and fault-propaga-
ences between the model in Figure 9e and the data in Figure tion folding in the next. The growth structure associated with
10 are thought to be caused by heterogeneous sedimentary normal faulting is presented in detail elsewhere (Xiao &
facies and associated heterogeneous compaction. These Suppe 1990), but is essentially analogous to compressive
111
J. SUPPE ErAL.
A' A B' B
deposition
112
RATES OF FOLDING AND FAULTING DETERMINED FROM GROWTH STRATA
2 km
Figure IS. Seismic example of an erosive growth fault-bend fold analogous to the front flank and crest of the model fault-bend fold in Figure 14; eastern
Venezuelan basin. Notice that the synclinal axial surface terminates at the unconformity above and the fault below. The steep axial surface connecting two
segments of unconformity in the front flank of the anticline records an interval of higher sedimentation rate relative fold growth rate. The poorly imaged
nearly flat fault is the Pirital thrust. The average vertical exaggeration in the upper 1.0 s of syncline is approximately 0.8.
back kink band are active and are analogous to the axial thickness of strata deposited over the ramp decreases and the
surfaces in the kink-band model of Figure 2f. dip of the growth axial surfaces decreases. In the limit of
It is important to realize that in fault-bend folding the nondeposition over the ramp (Fig. 13b) the growth axial
particles are moving everywhere parallel to the local orienta- surface on the back limb is horizontal and is an angular
tion of the fault. Uplift takes place instantaneously only over disconformity whereas the growth axial surface on the front
the ramp, therefore thin sedimentation takes place over the limb is parallel to the front-limb dip and is a surface of
ramp in the zone bounded by the two active axial surfaces (A disconformable onlap. For any horizon within the growth
and B in Fig. 12a and B and B" in Figs 12b & 12c). The sequence, the region in which a bed is missing records the
sedimentation rate over the ramp is equal to the sedimentation region of non-deposition over the ramp. If we move into the
rate in the basin minus the uplift rate over the ramp. Therefore erosive regime we observe down cutting into the pregrowth
the swath of thin beds running diagonally upwards from left strata (Fig. 13c). The magnitude oferosion is proportional to
to right in Figure 12c are the beds that were deposited over the the time spent over the ramp; for example note the progres-
ramp and the growth axial surfaces bounding the thin sediments sively deeper erosion updip on the front flank of Figure 13c.
record the locus of particles deposited on the active axial Figure 14 is a model of a fault-bend fold that has slipped
surfaces that bound the region of thin sedimentation over the beyond the top of the ramp in the erosive regime. Figure 15
ramp. is a seismic example from eastern Venezuela that is analo-
If we lower the sedimentation rate in our models the gous to the front flank and crest of the model. Notice the
progressive down cutting of the pregrowth strata and the
angular
progressive onlap of the growth strata. Also notice in the
disconformity
model that the unconformity has been folded on the crest such
Growth that the unconformity is dipping but the beds above and below
Pre- are horizontal. This relationship is observed at the right edge
Growth of the seismic example and in other lines along strike. This
same model has been applied to Lost Hills anticline, Califor-
nia (Medwedeff & Suppe 1986a, 1986b; Medwedeff 1989).
Figure 16. Model of growth fault-bend fold with sedimentation rate less If the uplifting block does not erode then substantial
than uplift rate and non-deposition over bathymetric or topographic relief. topography or bathymetry can develop, as shown in Figure
The structurally highest part of the fold is along the back anticlinal axial
16. Notice that once the fault slips beyond the top of the ramp
surface where the angular disconformity is being folded. Note that growth
strata are missing at the front anticlinal axial surface. A similar fold from the the angular disconformity is folded such that the
Perdido foldbelt of the Gulf of Mexico is modelled in detail by Mount er al. bathymetrically highest point on the fold is along the back
(1990). anticlinal axial surface whereas growth strata are missing
113
J. SUPPE ET AL.
km
Figure 18. Seismic profile of the back limb of a growth fault-propagation fold showing an upward widening triangle of gently dipping beds similar to Figure
17, Oak Ridge anticline, Ventura basin, California. A well section of this structure along strike is given by Suppe & Medwedeff (1990), which shows it to
be a fauU-propagation fold less than a million years old.
114
RATES OF FOLDING AND FAULTING DETERMINED FROM GROWTH STRATA
115
J. SUPPE ET AL.
unconformities were documented by Tomlinson (1952) in his where s^ is the fault slip before the bend and s is the slip be-
classic paper 'Odd Geologic Structures in Southern Okla- yond the bend (Suppe 1983, eq. 16). Note - there is a
homa'. difference in sign convention between the published theories
Dibblee (1966,1986) mapped a structure analogous to the of fault-bend folding (Suppe 1983) and fault-propagation
predicted front-limb structure of Figures 21b & 23 in the folding (Suppe & Medwedeff 1990): Equation (6) is written
Transverse Ranges of California as shown in Figure 24. in the sign convention for fault-propagation folding. P| is the
Notice that on the left-hand side of Figure 24 - below point A hangingwall cutoff of the folded beds (see Suppe 1983).
- the Juncal Formation lies conformably on Cretaceous shale.
At point A, where the anticlinal axial surface disappears, the Fault-Bend Folding Fault-Propagation
contact suddenly makes a right-angle bend to become a nearly Folding
orthogonal angular unconformity (compare point A in Fig. e, =- ei= +
21). The angular unconformity extends to the right to point
B where the anticlinal axial surface reappears, similar to point Pi=- Pi= +
B in Figure 21. To the right of point B the Juncal Formation
Yi = + Yl= +
lies on the Espada Formation - roughly parallel but with
several kilometres of section cut out between A and B. These (()= + i, = +
116
RATES OF FOLDING AND FAULTING DETERMINED FROM GROWTH STRATA
2 km
is^i s
#-2
Figure 22. Seismic profile of the back limb and crest of a growth fault-propagation fold from southern Oklahoma. The growth axial surface terminates
in the middle of the back limb at a time-transgressive disconformity similar to the model of Figure 21a. Average vertical exaggeration in the upper 1.5 s
is about 0.9. Seismic profile provided by Professional Geophysics Inc. (PGI).
2 km
Figure 23. Seismic profile of a growth fault-propagation fold from southern Oklahoma, which is similar to the model of Figure 21 b. The structure grew
under erosive conditions at an early stage, after which it stopped growing or grew slowly relative to the sedimentation. The structure was reactivated late
in the depositional sequence. The strong angular unconformity that is predicted for the front limb (Fig. 21 b) has been encountered in drilling but is not imaged
on the seismic profile. A similar unconformity is shown in Figure 24. Average vertical exaggeration in the upper 1.5 s is about 1.5. Seismic profile provided
by Halliburton Geophysical Services, Inc.
117
J. SUPPE ET AL.
118
RATES OF FOLDING AND FAULTING DETERMINED FROM GROWTH STRATA
'sin(Ye-62)" and
Wf = h
sinYeSin02
(15)
back-limb length L^
Combining (14) and (15) we obtain L^ slip s
[cotYe* -cotyi]
Wb = Lb = h [cotYe*-coty i ]
[ sin02 sin(Ye+yi-92)-l L sinyi J
sinyj /sin Ye "
J sinYi / sin Ye "] r s i n ( y i + 9 2 ) 1 Lsini
+ h n92 sin( y e + y i - 9 2 ) -
.sin92 sin(Ye+Yi~92)- )J L sinyi J
(17) (21)
Fault slip s in a fixed-axis fault-propagation fold is The above equations give relationships between fault slip
and limb length and width for fault-propagation folds. Inas-
J- sinYi / s i n y^
much as back-limb growth is most easily observed, Figure 28
sine2 sin( Ye+yi-92)
(18) gives a graph of the relationship between back limb dip ?>^ =
2(7t-y,) and LJs with lines of constant fault bend <[) or
Combining (18) with (15), (16), and (17) we obtain the hangingwall cutoff angle 0^ based on the above equations and
desired relationships between slip and limb lengths and numerical solution of the fundamental equations of fault-
widths propagation folding of Suppe & Medwedeff (1990). This
graph is discussed below.
'sin(Ye-92)"
front-limb width Wf sinYeSin92
slip s sinYi / s i n y^
sin 92 sin( Ye+yi-92)
(19) .^
Figure 26. Graphs of the relationship between limb dip 5 and limb length
Figure 25. Fault-bend fold showing angular relationships for derivation of LI slip for synclinal fault-bend folds with lines of constant fault bend $ or
relationship between limb length and fault slip. hangingwall cutoff angle Gj.
119
J. SUPPE ET AL.
0.8
0.7 -
-1—»
0.6
C
0)
0.5
E
0.4
Q.
"a5
0.3
0.2
0^ 20° 40^ 60^
back dip 5
Figure 28. Graph of the relationship between limb dip 8 and slip IL (back limb length) for constant-thickness fault-propagation folds with lines of constant
fault bend (j) and hangingwall cutoff angle 91.
Figure 29. Models of a kink band with a 30° limb dip showing the range of fault shapes that can produce the same fold. Note the variation in fault slip
with constant fold geometry.
120
RATES OF FOLDING AND FAULTING DETERMINED FROM GROWTH STRATA
Discussion of relationships between limb length and slip 1990). Therefore fault slip commonly can be closely con-
strained from fold geometry.
We would like to estimate fault-slip rates from observation of
shallow growth structure and to assign uncertainties based
upon our uncertainties in the underlying fault geometry. For
Conclusions
example Figure 27 shows schematically how the same kink
band might form by slip on a variety of faults. The equations
The well known processes of fault-related folding such as
and associated graphs obtained in the previous sections allow
fault-bend and fault-propagation folding specifically predict
us to assign these uncertainties in fault-slip rate based on our
fold growth by constant-dip kink-band migration. This
uncertainty in fault geometry. This discussion assumes a two-
implies specific fold geometries within the stratigraphic
dimensional geometry - essentially dip-slip motion. If the
sequences deposited during fold growth, which are observed
folding is produced by oblique slip through a fault bend, such
as shown by seismic examples. Specifically, kink bands
that there is a strike-slip component, then the kink band decrease in width upward through the growth strata. Rates of
records the dip-slip component. kink-band migration are observed in a few structures to be on
In the schematic example of Figure 29 a limb dip of 30 the order of 1-2 mm/y. Straightforward relationships are
degrees is displayed. If based on growth structure we believe derived between limb length and fault slip which commonly
the kink band is produced by a synclinal fault bend then allow fault slip to be determined within rather narrow limits
Figure 26 shows that the ratio of limb length to fault slip L/s from shallow growth structure, even if fault geometry is
ranges between 1.0 and about 2.0. If we can constrain the poorly constrained - a fact that makes growth structures
problem further, for example regional geological relation- valuable for assessing earthquake hazards.
ships might indicate that the upper fault dip 0^ must be less
than about 50°, then we can further limit Us to the range 1.0 Suppe is grateful to the Guggenheim Foundation for support during a
to 1.25. Thus with only rather modest information on fault sabbatical leave in 1978-79 when the initial theory of growth fault-bend
geometry we can constrain fault slip rather closely from folding was developed (Fig. 12). However most of the theory of growth
growth structure. structure was developed during Fall of 1985 when Chou and Hook were
Similar conclusions are reached for fault-propagation visiting professors at Princeton, supported by Texaco USA. We are all
grateful to Texaco for the continued support and encouragement of this work
folds (Fig. 28); namely that the relationship between limb and to the many sources of the seismic examples, including Texaco USA;
length and fault slip does not vary rapidly with fault geometry Professional Geophysics Inc. (PGI); BP Exploration Inc.; Sun Operating
at a constant limb dip. For example with a back-limb dip of Limited Partnership; Nippon Western US Company, Ltd; Halliburton Geo-
15-25°, which is quite common, the fault slip ranges between physical Services, Inc.; R. Prieto, R. del Pilar, and A. Saldivar-Sali. We
0.4 and 0.6 times the back-limb length for almost all fault thank Steve Boyer and Shankar Mitra for reviews of the manuscript. The
work on the relationship between fold growth and fault slip and on the
geometries. Fault geometry of fault-propagation folds com- examples from California were supported by USGS NEHRA grant 14-08-
monly can be constrained rather closely from shallow geo- 0001-G1699,
logical data such as ratios of limb widths (Suppe & Medwedeff
References
Dibblee, T. W. Jr. 1966. Geology of the central Santa Ynez Mountains, Santa , Suppe, J. & Hook, S.C. 1990. A forward modelling strategy for
Barbara County, California. California Division of Mines and Geology balanced cross sections. American Association of Petroleum Geologists
Bulletin, 186, 1-99. Bulletin, 14, 52\-53\.
1986. Geologic map of the Hildreth Peak Quadrangle, Santa Barbara Rich, J. L. 1934. Mechanics of low-angle overthrust faulting as illustrated by
County, California. Thomas W. Dihblee Jr. Foundation Map, DF-07. Cumberland thrust block, Virginia, Kentucky, and Tennessee. American
Medwedeff, D. 1989. Growth fault-bend folding at southeast Lost Hills, San Association of Petroleum Geologists Bulletin, 18, 1584-1596.
Joaquin Valley, California. American Association of Petroleum Geolo- Suppe, J. 1983. Geometry and kinematics of fault-bend folding. American
gists Bulletin,13, 5A-tl. Journal of Science, 283, 684-721,
& Suppe, J. 1986a. Growth fault-bend folding - precise determination 1985. Principles of Structural Geology. Prentice-Hall Inc., Englewood
of kinematics, timing, and rates of folding and faulting from syntectonic Cliffs, New Jersey, 1-537.
sediments. Geological Society ofAmerica Abstracts with Programs, 18, & Chang, Y. L. 1983. Kink method applied to structural interpretation
692. of seismic sections, western Taiwan. Petroleum Geology of Taiwan, 19,
& 1986b. Kinematics, timing, and rates of folding and faulting 29-49.
from syntectonic sediment geometry. EOS Transactions ofthe American & Medwedeff, D. A. 1984. Fault-propagation folding. Geological Society
Geophysical Union, 68, 1223. of America Abstracts with Programs, 16, 670.
Mosar, J. & Suppe, J. 1991. Role of shear in fault-propagation folding (this & 1990. Geometry and kinematics of fault-propagation folding.
volume). Eclogue Geologicae Helvetiae, 83 (Laubscher volume), (in press).
Mount, V.S. & Suppe, J. 1990. Seismic structural analysis of the deep-water Tomlinson, C. W. 1952. Odd geologic structures in Southern Oklahoma.
Perdido foldbelt, Aliminos Canyon, Northwest Gulf of Mexico. In: Future American Association of Petroleum Geologists Bulletin, 36, 1820-1840.
Petroleum Provinces of North America. American Association of Pe- Xiao, H. B. & Suppe, J. 1990. Origin of rollover. American Association of
troleum Geologists Memoir (in press). Petroleum Geologists Bulletin (submitted).
121
Role of shear in fault-propagation folding
Jon Mosar' & John Suppe
Abstract: The effect of layer-parallel shear on the shapes of fault-propagation folds is explored
for the two theories of fault-propagation folding of Suppe & Medwedeff (1990): (1) constant layer
thickness and (2) variable front-limb layer thickness; the range of possible fold shapes is
significantly expanded relative to the case of no shear. In this analysis, a homogeneous differential
layer-parallel shear is applied to the beds that are cut by the thrust fault in the cores fault-
propagation folds. This shear may be applied in three ways: (1) self-similar fold growth during
fault propagation with constant shear applied instantaneously along the bed of the fault tip, (2)
progressively increasing shear during fault propagation, and (3) shear after the fault is locked
modifying the existing fault-propagation fold. The final shape of the fold is independent of the
history of shearing in relation to fault slip. The fold shapes are largely governed by the fault
steepness and the amount of imposed shear. Strong differential shear profiles and shallow faults
produce overturned or thickened front-limbs. Little differential shear and steep faults result in
upright or thinned front-limbs.
'Present address: Musee Geologique, Universite de Lausanne - BFSH2, CH-1015 Lausanne, Switzerland
123
J. MOSAR & J. SuPPE
contrast, the deformational history of a final fold shape Figure 2. Self-similar growth of a simple-step fault-propagation fold. The
becomes clearly non-unique in the case of non-zero shear. In differential layer parallel shear (angular shear a) is determined by the
general the shear may be applied in three ways: (1) self- original fold geometry. Once this geometry is acquired, a stays the same as
similar fold growth during fault propagation with constant the fauh propagates from (a) through (d), only the width D of the shear profile
increases. (Fault ramp angle = 30°; anticlinal interlimb angle in fold core =
shear applied instantaneously along the bed of the fault tip, 60°; interlimb angle of syncline at fault tip = 90°; a = 31.8°).
which is shown in Figure 2, (2) progressively increasing shear
applied in various possible ways during fault propagation and occuring within specific stratigraphically controlled weak
fold growth, and (3) shear applied after the fault is locked horizons may be important.
modifying the existing fault-propagation fold. The final The conditions for conservation of line-length and area
gross shape of the fold in pre-existing strata is independent of can be expressed quantitatively in a mathematical formula-
the history of shearing in relation to fault slip, but the fold tion using only angles describing the fold shape (Fig. 3).
shape in growth strata will record the deformational history Exact solutions involving shear have been derived for the
(Suppe era/. 1990). general case, not involving a decollement (Suppe &
The fold develops kinematically by consuming layer par- Medwedeff 1990). This phase will only be concerned with
allel slip and the overall deformation is one of simple shear. simple step-up fault-propagation folds, that is, when the fault
The amount of layer parallel shear across the fold is determined steps up at a given angle 0^ from a bedding plane decollement
by the fold geometry (interlimb angles as well as angle and (Fig. 3). The results indicate that in this simple step case, the
height of the ramping thrust plane). Thus some fold geometries fold shape (viz. the various internal axial angles) is a function
may require a differential bedding parallel shear to be of two independent variables only: S , the shear associated
transmitted through the structure (see also Mitra 1990, p. with layers in the fold core; and 0^. i^Q fault step-up angle.
925). A displacement profile of single layers through the flat Two different fault-propagation folding theories have
lying beds in the trailing part of the fault-propagation fold, been developed (for details on the quantitative derivations see
may then be a straight line perpendicular to bedding or a more Suppe & Medwedeff 1990).
complex line if excess layer parallel shear is implied by a
different fold geometry. Excess layer parallel shear can, for
example, be induced by bending of beds along a supplemen- Constant-thickness theory
tary axial surface (Suppe 1985, p. 318). Although excess In the constant-thickness model, bed length and layer thick-
layer parallel shear appears to be minor in many examples of ness are conserved throughout the fold. The quantitative
fault-propagation folding (Suppe & Medwedeff 1990), shear geometric relationships are shown in equation (1). Note that
124
ROLE OF SHEAR IN FAULT-PROPAGATION FOLDING
CONSTANT THICKNESS THEORY overturned. The front-limb dip is always parallel to the back-
A B limb axial surfaces!
iK
(a) '"''"'' ~'''''
% ,
,!
Kinematic aspects
The kinematic aspects and characteristics of fault-propaga-
tion folding can be addressed in several different ways, only
a few of which are briefly described here. Once the fold is
FIXED AXIAL SURFACE THEORY initiated, it grows self-similarly (except for later possible
imposition of additional shear, as discussed above). Struc-
A B tural relief increases during propagation and the flat anticlinal
top progressively disappears (Fig. 1).
': I ••: 1
Icos 02
Sp = Icoty^ + (2)
sin 02
125
J. MOSAR & J. SUPPE
Ilk f
^•:t^%'
Figure 5. Kinematic propertie.s of diffcreni simple step fault-propagation fold models (detail see text), (a) constant-thickness theor>. shallow fault, low
83 =V5°;7*=21.6*';Y= 29.1°, {b)conslant-thicknesstheory, steep fault, high 62 = 45°;Y*=5l.2'';Y=73.7°,(c)fixed axial thickness model with front limb
thickening: TJ J. = 0.35. shallow fault, low 6^ = 10°; 7^*= 18.6°; y^ = 19.7°. (d) fixed axial thickness model with front limb thinning: T^/ T^ = 1.37 steep
fault, high 02 = 40°; yg*=41.2°; y ^ ^ 67.5°. Arrows indicated sense of material moving through active axial surfaces.
07 = 55' 9^ = 55'
\ Te/Ti=^l37
Figure 6. Influence of fault-
ramp angle 6^ on fold shape in
simple step fault-propagation
folds with conslanl ramp height.
09 = 45= 0 2 = 40° Shear is zero; step-up angle
changes from (a) through (d).
The total displacement at the
fold back end of case (d) in the
Te/Ti = 0.69 fixed axial surface theory could
not be represented accurately.
02 = 30° 02 = 20°
Te/Ti = 0.35
09=15° 02=10°
126
ROLE OF SHEAR IN FAULT-PROPAGATION FOLDING
127
J. MOSAR & J. SUPPE
TelTi = 0.55
(a) L!.4.^fc^.>fe;.7i.>l.:...l:^2sr>^SirIi?.-^!Z..'.^r:.r.
Te/Ti = 0.71
(b) kpMiptppnMiiiMMi
(d)
Sp = - 0.41 a = - 22.5^ Sp = - 0.86 a = - 40.6°
Figure 7. Influence of shear in fold core layers in simple-step fault-propagation folds with constant ramp height. Siiear decreases from (a) througti (d);
step-up angle for all four cases is 30°. Fold core angles y * respectively y^' are: (a) 25°, (b) 30°, (c) 40°, (c) 50°.
are geometrically consistent with fault-propagation folding. and lift-off folds because displacement goes to zero at the
Three types of solutions can be found (Fig. 10): basal layer, rather than a finite value as in detachment folds).
(a) first, a set of situations where geometries (sets of fold In this case no thrust surface exists and no slip has occurred
internal angles) are consistent with the fault-propagation fold other than layer parallel slip. The axial plane C of the fold
models. That is, displacement occurs along a fault surface - core intersects the back-limb axial surface B' at the basal
a decollement portion and a ramp portion in simple step-up horizon of shear (Fig. 10b);
models - which implies that the axial surface C of the fold core (c) third, a set of geometries which are inconsistent with
intersects the ramping fault segment between the fault tip and fault-propagation folding by thrust faulting. The displace-
the point where the fault steps up from the decollement (Fig. ment sense along the ramp segment of the fault appears to be
10a); a normal fault and the axial plane C in the fold core intersects
(b) second, a set of solutions in which simple parallel-kink the back-limb axial surface B' (Fig. 10c).
folding without faulting is produced (simple shear folding, The geometries expected by the third type of solutions are
not to be confused with detachment folds such as box folds not considered here. They fall within the upper right of the
128
ROLE OF SHEAR IN FAULT-PROPAGATION FOLDING
s 2.5
S
-a
J 0
C/3
-1
-2.5
Fault angle O2
Figure 8. Inter-limb half angle 7* as a function of fault-ramp angle Oj and shear S for constant-thickness fault-propagation fold models (solutions from
eq.(l)).
129
J. MOSAR & J. SuPPE
130
ROLE OF SHEAR IN FAULT-PROPAGATION FOLDING
References
Boyer, S. E. 1986. Styles of folding within thrast sheets: examples from the & Namson, J. 1989. Equal-area-balancing. American Journal of Sci-
Appalachian and Rocky Mountains of the US A and Canada. Journal of ence, 289, 563-599.
Structural Geology 8, 325-339. Mosar, J. 1989. Deformation interne dans les Pr6alpes Medianes (Suisse).
Coates, J. 1945. The construction of geological sections. Quarterly Journal Eclogae Geologicae Helvetiae, 82, 765-793.
of the Geological, Mineralogical and Metallurgical Society of India, 17, Mount, V.S., Suppe, J. & Hook, S. C. 1990. A forward modelling strategy
1-11. for balancing cross sections. American Association ofPetroleum Geolo-
Dahlstr6m,C.D.A. 1969. Balanced cross sections. Canadia«yourna/o/£arr/! g««, 74,521-531.
Science, 6,743-757. Namson, J. S. 1981. Structure of the western foothills belt, Miaoli-Hsinchu
1990. Geometric constraints derived from the law of conservation of area, Taiwan; I southern part. Petroleum Geology of Taiwan, 18,31 -51.
volume and applied to evolutionary models for detachment folding. Pfiffner, O. A. 1990. Kinematics and intra-bed strain in mesoscopically
American Association of Petroleum Geologists Bulletin, 74, 336-344. folded limestone layers: examples from the Jura and the Helvetic zone
Faill, R. T. 1973. Kink-band folding. Valley and Ridge Province, Pennsyl- of the Alps. In: 1. ann. meeting Swiss Tectonic Studies Group, Basel,
vania. Geological Society of America Bulletin, 84, 1289-1341. Abstract vol.
Goguel, J. 1952. Traite de tectonique. Masson, Paris, 383p. Srivastava, D. C. & Engelder, T. 1990. Crack-propagation sequence and
Jamison, W.R. 1987. Geometric analysis of fold development in overthrust pore-fluid conditions during fault-bend folding in the Appalachian
terranes. Journal of Structural Geology 9, 207-219. Valley and Ridge, central Pennsylvania. Geological Society of America
Kilsdonk, B. & Wiltschko, D. V. 1988. Deformation mechanisms in the Bulletin, 102, 116-128
southeastern ramp region of the Pine Mountain block, Tennessee. Suppe, J. 1983. Geometry and kinematics of fault bend folding. American
Geological Society of America Bulletin, 100, 653-664. Journal of Science, 283, 648-721.
Laubscher, H. P. 1962. Die Zwei-phasenhypothese der Jurafaltung. Eclogae 1985. Principles of Structural Geology. Prentice-Hall, Inc., New-
Geologicae Helvetiae, 55, 1-22. Jersey, 537p.
1976. Geometrical adjustments during rotation of a Jura fold limb. , Chou, G. T., & Hook, S. C. 1991. Rates of folding and faulting de-
Tectonophysics, 36, 347-365. termined from growth strata (this volume)
1977. Fold development in the Jura. Tectonophysics, 37, 337-362. & Medwedeff, D. A. 1984. Fault propagation folding. Geological Society
Marshak, S. & Mitra, G. 1988. Basic methods of structural geology, ed. Prentice of America Abstracts with Program, 16, 670.
Hall, 446p. & Medwedeff, D. A. 1990. Geometry and kinematics of fault-
Mitra, G. 1988. Area balanced models of fault propagation folds. Geological propagation folding. Eclogae Geologicae Helvetiae, 83 (Laubscher
Society of America Abstracts with Program, A56. volume, in press).
1990. Fault-propagation folds: geometry, kinematic evolution, and
hydrocarbon traps. American Association of Petroleum Geologists, 74,
921-945.
Appendix
Although various angles, and hence fault-propagation fold smaller than the length xz of the ramping fault segment (Fig.
shapes are consistent with the mathematical formulation 11). Expressed differently, the angles y* (constant-thickness
derived from volume conservation, not all solutions are theory) and y* (fixed axial surface theory) have to be in a
geometrically feasible. range such that the axial surface of the fold core always
Several boundary conditions impose limiting constraints intersects the ramping fault segment. Conversely, the axial
on the variety of fault-propagation fold shapes (Suppe & surface of the fold core cannot intersect the back limb axial
Medwedeff 1990). (1) Internal fold angles cannot exceed surface associated with the fault bend at the intersection of the
certain upper limits. The step-up angle 9^ has to be less than decollement and the ramping fault segment (Fig. 10c).
The consistency conditions for the general (not only
simple step-up from a decollement) fault-propagation fold
model is given by : xy < xz. Using trigonometric relation-
ships, xy and xz can than be expressed as:
sin 62 (4)
and
sin Y[
xy = (5)
Figure 11. Outline of parameters involved in deriving the general consist- sin P2 •S'" 7e
ency conditions. General case fault-propagation fold, fixed axial surface
model.
From internal angular relationships (Suppe & Medwedeff
1990) we know that:
60°, hence y, and (y^* + y*) have to be smaller than 90° and
larger than 60°. (2) To be consistent with the fault-propaga-
tion fold model, the distance xy between the fault tip and the P2 = 180 - (7* + r^ +Pi (6)
intersection of the axial surface of the fold core has to be
131
J. MOSAR & J. SUPPE
Pi = 02- 5\,= 02 + 2/1 - 180 (7) For the constant-thickness fault-propagation fold model,
where (y.*+ y^*) = 2"^ we thus will have the following limiting
Replacing and substituting we then obtain the following condition:
relationship:
sin 02 < sin (if - 02 + 5b) (11)
^ > '^'"^' ti, (8) In a simple-step fault-propagation fold with a fixed front
sin02 sinYs sin \(yi+ Ye) " P)\ anticlinal axial surface, we have 9^ = 8^ Thus the former
equation simplifies to:
(10)
132
2-D reconstruction of thrust evolution using the fault-bend fold method
Reini Zoetemeijer* & William Sassi
Abstract: This paper presents a computer-aided method for analysing geological cross sections
in compressive terranes. The deformation of thrust sheets is simulated using the geometrical model
of fault-bend folding. The modelling focuses on complex thrust configurations like multiple
imbrications and duplexes. This paper discusses the geometrical constraints of the fault-bend fold
model to highlight the applicability and methodology of the kinematic reconstruction method. An
application to a section from the Jura Mountains shows that complex thrust geometries can be
analysed. Improvements on earlier work concern the treatment of transported piggy-back thrust
sheets and the introduction of sedimentation and erosion processes. With the current computer
method it is possible to investigate the stratigraphic patterns that develop in the syn-orogenic
basins. This can provide insight in understanding the kinematic evolution of thrust systems.
The conceptual model of fault-bend fold deformation may be esses in the simulation.
used for structural interpretation of geological cross-sections The main limitations of the geometric model of deforma-
as a method of restoration and may serve as such, as an aid to tion are discussed in the paper. These concern the treatment
seismic interpretation. With fault-bend fold deformation one of 'high-angle fault-ramp geometry' and the fusion of two
assumes conservation of bed length and orthogonal thickness kink-axes during deformation. Earlier assumptions intro-
throughout deformation and formation of kink-band type of duced to simplify passive deformation are abandoned and it
folding. This model belongs to a class of geometric models is demonstrated that passively transported thrust sheets can
of finite deformation, like the vertical simple shear which is be deformed without production of back shear deformation.
also used in section balancing (Moretti & Larrere 1989). The Finally, sedimentation and erosion processes are introduced
model of fault-bend folding was described in detail by Suppe so that the stratigraphic patterns that develop in the sedimen-
(1983), who demonstrated its applicability for the Pine tary overburden during thrust displacement can be studied.
Mountain thrust sheet in the southern Appalachians, and the
fold- and thrust belt of western Taiwan. Suppe & Medwedeff
(1984) and more recently Medwedeff (1989) have further Geometric method
extended the analysis describing the kinematic relationships
which may prevail between fault-related folds and contempo- The computer method (Endignoux & Mugnier 1990) is based
raneous sedimentation processes. on the equations of Suppe (1983). The solutions of these
The fault-bend fold model may also be used to reconstruct equations strictly satisfy the conservation of bed-thickness
the forward evolution of thrust emplacement (Endignoux and bed-length, and therefore volume remains constant dur-
1989). Although a number of forward computer methods ing the deformation. The mathematical problem is described
making use of this model have been proposed (Usdansky & by equations (1) and (2) and involves four parameters:
Groshong 1989; Chou & Suppe 1987; Charlesworth
&McLellan 1986; Endignoux 1989), very few case history -sin(7^e)f sin(2"i^9)- sinOI
tan 9 = cos(Y-0)[ sin(2Y-9)- sinOJ-siny
studies have been reported in the literature (Endignoux et al. (1)
1989). This paper attempts to demonstrate that such a
modelling approach may be applied successfully to rather P= 9 —(p + (ji —2y) (2)
complex thrust geometries. An example of modelling the
progressive evolution of thrusting is presented using a geo- 9 is the cut-off angle of the fault with stratigraphic layering,
logical section taken from the Jura Mountains, western Alps, (p is the change in fault dip, P is the cut-off angle after bending
studied by Guellec et al. (1990). and 7 is the axial angle, between kink-axis and layering (Fig.
The work has been done using a computer program that 1). The preservation of layer thickness is guaranteed when
was developed by Endignoux (1989) and described in detail taking as the kink-axis, the bisectrix between the two bedding
by Endignoux & Mugnier (1990). This program has been surfaces, before and after bending. In the case where two
modified and extended in order to enable analyses of complex solutions of yexist (mode I and mode II of Suppe (1983)), one
fault geometries (e.g. to allow faults to be activated in any has to chose one of the two possibilities. Figure 1 shows two
order) and to incorporate sedimentation and erosion proc- schematic kink folds, one for an increase of fault dip (anti-
'Present address: Department of Earth Sciences, Vrije Universiteit, PO Box 7161,1007 MC Amsterdam, The
Netherlands
133
R. ZOETEMEIJER & W . SASSI
Figure 1. Fault-bend fold method showing (a) anticlinal fold in case of Figure 2. High-angle fault-ramp geometry. For this situation, no kink-
decrease in fault dip and (b) synclinal fold in case of increase in fault dip. angle y exists for which bed-length and thickness remain conserved, when
Parameters: 6 is the cut-off angle of the fault relative to bedding attitude, (p trying to bend the toe towards the surface.
is the change in fault dip, (3 is the cut-off angle after bending and y is the axial
angle, between kink and layering.
134
2 - D RECONSTRUCTION OF THRUST EVOLUTION
Finally, in order to satisfy the local lack of material that is sediments in a synclinal shape, in such a way that kink-axes
necessary for the further deformation of high-angle fault- intersect. The kink-axes will undergo fusion and, if one treats
ramp geometry, a certain amount of extra bed-parallel shear the problem in the mathematically correct way, another kink-
is introduced to ensure preservation of overall volume (Suppe axis has to split up elsewhere in the model on the same shear-
1983; Jamison 1987), (Fig. 3a). Due to length preservation, level. The problem of kink-axis fusion and splitting up is
the change in shape at the rear boundary or trailing edge of the difficult to solve. The solution is no longer unique; several
model represents the amount of required layer-parallel shear kink-axes can be considered for splitting up. However, the
(also called 'negative' back shear). In the proposed approach volume of material lost in the reconstruction is usually very
the value of the kink-angle y is determined such that the amount small compared to the volume of material needed to allow
of extra bed-parallel shear is minimized (see also Endignoux folding in the case of high-angle fault-ramp geometry. In this
& Mugnier 1990; Moretti et al. 1990). method the loss of material during the fusion of kink-axes is
not compensated, and as a result, this will produce a (positive)
backshear towards the hinterland of the models.
Fusion of kink-axes
The deviation from a vertical rear boundary (positive or
Back shear may also occur when the constructed kink-axes negative back shear) measures the amount of material that is
come to intersection during the kinematics of the deforma- needed (high-angle fault-ramp geometry) or is in excess
tion. An important increase in fault dip can bend the overlying (kink-axis fusion) because the assumption of volume preser-
NW SE
MOLASSE BASIN
EXTERNAL BORNES ARAVIS ULTRAHELVETIC
SALEVE Parmelan Plateau Merdassler
desGlieres LaClusaz Flumet _,<, 2
Cruseilles Thorens Gli<
^km
Restoration
NW SE
F~~] ?.
EI34
EZ15
~\ 7
Okm
Figure 4. Balanced cross section along the ECORS profile with restoration before Cenozoic deformation. 1: Urgonian limestone; 2: Subalpine Tithonian
limestone and Upper Jurassic platform carbonates; 3: Middle Jurassic limestone; 4: Basement of Jura Mountains and External Belledonne; 5: Basement of
Internal Belledonne; 6: Thrusts and faults; 7: Potential thrusts. I: Cenozoic of Bresse graben and Molasse basin; II Oligocene flysch; III, IV and V: Cretaceous
(III Bresse graben, Jura Mountains and Molasse basin, IV: Presubalpine domain, V: Subalpine domain); VI and VII: Jurassic (VI: Bresse graben. Jura
Mountains and Molasse basin, VII: Presubalpine and Subalpine domain); VIII: Basement. After Guellec et al. (1990). The frame indicates the position of
the area tested by numerical modelling in Figure 5.
135
/= R. ZOETEMEUER & W . SASSI
••'••••: -•? r'\ ; :;-'K ; ; > f i j f - ' : ' : h ; > . ; /w.n-- ^ •-!;• •;"> • ' r . . - - . , ••ir-
Figure 5. Numerical modelling of the external zone of the Jura Mountains, (a) Restored section of the external zone of the Jura Mountains from Figure
4. (b) Starting template for the forward modelling (Faults labelled 1-9). (c) Forward model after 2 km of displacement, (d) Forward model after 4 km
displacement, (e) Forward model after 6 km of displacement, (f) Forward model,6km of displacementwitherosion. (g) Comparative geological section
(from Fig. 4). . " .• •
vation can not be satisfied locally in these cases. Thus, in seismic interpretation of the Jura Mountains (Guellec et al.
contrast to treatments where extra bed-parallel shear is a free 1990) (Fig. 4). Due to the complexity of the area, the study
parameter (e.g. fault-propagation folding (Suppe 1990a)), in was restricted lo the extreme external zone of the Jura,
the present approach extra bed-parallel shear is only calcu- between Coligny & Oyonnax (Fig. 4). The seismic section
lated to increase the range of solutions for the couples (p-y), cuts the mountain belt well along strike and images thrusts
in case of high-angle fault-ramp geometry and lo simplify the emplaced between 10 Ma (Tortonian) and 4 Ma (Lower
treatment of kink-axis fusion. The influence of the extra bed- Pliocene) (Mugniere/a/. 1990). Triassicevaporites(Keuper
parallel shear has to be minimized in order to support the and Muschelkalk) are considered to form the decollement
coherence of the modelling results. level during Neogene compressional events in the fold and
The following example of the Jura Mountains illustrates a thrust belt. The situation before the compressional events is
solution with minimum bed-parallel shear. described in the restoration of this section (Fig. 5a). It is
obtained using a section balancing technique assuming con-
servation of length. Due to the extensive erosion three
Jura Mountains reference horizons are chosen: Urgonian, Upper Jurassic and
Dogger limestones.
The code was tested on a real case example taken from a Layer-thicknesses remain constant all along the section,
136
2 - D RECONSTRUCTION OF THRUST EVOLUTION
137
R . ZOETEMEIJER & W . S A S S I
Hi... m
Figure 7, Reproductionof duplex imbrications. 4duplexes with various initial fault spacing, respectively 15, 10,5 and 3 units, (a) results of initial model
with passive deformation assuming constant cut-off angles, and (b) same model as in a, but with adjustment of cut-off angles according to the fault-bend
fold method.
Erosion and sedimentation processes contemporaneous
with thrusting are simulated by numerically following the
evolution of time-lines which may be defined at any given
stage of deformation. For kinematic purposes one may
control independently thrust velocities and rate of sedimen-
tation and erosion. The continual addition of 'material' on
top of the model will fossilize a stratigraphic pattern in the
newly formed basin.
As mentioned above, the geometric description of imbri-
cate fault-bend folding cannot distinguish the order of fault-
ing in the deformation history. However, basin stratigraphic
analyses of the sedimentary overburden may reveal informa-
tion on the kinematics of deformation, which should be used
to constrain the reconstruction (Fig. 8).
Figure 8. Synthetic model of imbricate fault-hend folding. The order in An example of how sediments can register thrust evolu-
which thrusts are activated is not visible from the duplex geometry, but the tion is described by Roure etal. (1990), in their study of the
overlying sediments have recorded the evolution and the different situations southern Apennines accretionary wedge. The observed piggy-
are therefore clearly distinguishable.
back basins, formed by accumulation of syn-tectonic depos-
STEP 1
- ^ ^ ACTIVE ^ 9 1 ACTIVE
CTiTD o TILTED PIGGY-BACK TILTED PIGGY-BACK , t t ) , ROTATED FAULT CTCD o TILTED IPIGGY-BACK
^"=^^2 I ^tDACTIVE STEP 2 , / ^ b i t K ^ - tpiFOLDED
' ^-<»'~;'--.' CD, ACTIVE
'pJ nf-Ti„e F— ~-—^ I to ACTIVE
Figure 9. Development of geometry in piggy-back basins in relation to deep-seated structure, (a) depocenlers migrate away from the front of the underlying
fault when the displacement along the thrust is sufficiently slow, (b) The same general geometry of deposits is observed when the thrust is being tilted by
younger thrusts, (c) Depocenters migrate toward the front of the underlying fault when a previously flat portion of thrust becomes folded, {after Roure el
al. 1990).
138
2-D RECONSTRUCTION OF THRUST EVOLUTION
Figure 11. Two stages in the simulation of double ramp thrust structure in which time-lines record the evolution of thrusting. The velocity of the rear
boundary is taken constant. The simulated sedimentation has a constant velocity equal to the Initial vertical growth of the deepest fold {v = v . sin (p). However,
displacement along the fault changes as a function of the slip ratio and the kinematics of the deep fold are recorded in the basins, not only close to the fold
but also more towards the foreland.
139
R. ZOETEMEIJER & W . SASSI
References
Charlesworth, H.A.K. & G. C. McLellan. 1986. Refold; Fortran 77 program , S. Triboulet, & L. Endignoux. 1990. Some remarks on the geometrical
to construct model block diagrams of multiply folded rocks. Computers modeling of geological deformations. In: Letouzey, J. (ed.) Petroleum
and Geosciences, 12, 349-360. and tectonics in mobile belts. Editions Technip, Paris, 155-162.
Chou, T. G. & J. Suppe. 1987. Balanced computer-modeling of complex Mugnier, J. L., S. Guellec, G. Menard, F. Roure, M. Tardy, & P. Vialon.
thrust structures. Eos abs., 68, 44, 1451. 1990. Crustal balanced cross-sections through the external Alps deduced
Endignoux, L. 1989. Une modelisation numerique bidimensionelle de from the ECORS profile. In: Roure, F., Heizman, P. & Polino, R. (eds)
revolution cinematique et thermique des structures chevauchantes. Deep structures of the Alps. Memoirs de la Societe. Geologique de
Ph.D., Univ. Joseph Fourier, Grenoble, 243p. France 156, Suisse 1, Italiana 1 (in press).
Endignoux, L. & J. L. Mugnier. 1990. The use of a forward kinematical Roure, F., D. Howell, S. Guellec, & P. Casero. 1990. Shallow structures
model in the construction of a balanced cross-section. Tectonics, 9,1249- induced by deep-seated thrusting. In: Letouzey, J. (ed.) Petroleum and
1262. tectonics in mobile belts. Editions Technip, Paris, 15-30.
, I. Moretti & F. Roure. 1989. Forward modelling of the Southern Suppe, J. 1983. Geometry and kinematics of fault-bend folding, American
Apennines. Tectonics, 8, 1095-1104. Journal of Science, 283, 684-721.
Guellec, S., D. Lajat, A. Mascle, F. Roure & M. Tardy. 1990. Deep seismic 1990a. Geometry and kinematics of fault-propagation folding. Eclogae
profiling and petroleum potential in the Western Alps: Constraints from Geologicae Helveticae, 83 (in press).
ECORS data, balanced cross-sections and Hydrocarbon maturation 1990b. Rates of folding and faulting determined from growth strata. In:
modeling. In: Pinet, B. & Bois, Ch. (eds) The potential of deep seismic McClay, K. R. (ed.) Thrust Tectonics 1990, Programme with Abstracts,
profilingfor hydrocarbon exploration, Editions Technip, Paris, 425-437. Royal Holloway and Bedford New College, University of London, (4-7
Jamison, J. W. 1987. Geometrical analysis of fold development in overthrust April, 1990), 16.
terranes. Journal of Structural Geology, 9, 2, 207-219. & D. A. Medwedeff. 1984. Fault-propagation folding. Geological Society
Medwedeff, D. A. 1989. Growth Fault-bend folding at Southeast Lost Hills, of America Program with Abstracts, 16, 670.
San Joaquin Valley, California. American Association of Petroleum Usdansky,S.I.,&R. H.Groshong. 1989.Thrustrampland2,/?ocjbvare/«(;'.,
Geologists Bulletin, 73,1, 54-67. Geological Software Catalog, Spring, 65-66.
Moretti, I. & M. Larrfere. 1989. LOCACE: Computer-aided construction of
balanced Geological Cross-section. Geohyte, 16-24.
140
Kinematic models of deformation at an oblique ramp
Theodore G. Apotria'*, William T. Snedden^,
John H. SpangS David V. Wiltschko'
Abstract: Kinematic models for the deformation of hangingwall material moving over a footwall
oblique ramp are developed by considering two end members of assumed mechanical behaviour,
vertical shear and layer-parallel shear. In the former case, material is sheared vertically and
displacements remain within the tectonic transport plane; the deformation is accommodated by
thinning of the hangingwall over the ramp. In the later case, material is deflected out of the
transport plane such that the pitch angle of the hangingwall particle path in the plane of the oblique
ramp is equal to the angle between the transport direction and the strike of the oblique ramp. As
a result, shear strains above the oblique ramp are non-zero in both the transport and transport-
normal planes. The deflection and transport-normal shear strains are a minimum for the special
cases of pure frontal and lateral ramps, and maximum at an intermediate orientation, depending
on oblique ramp dip. Fault-bend folds are similar in most respects for both vertical shear and layer-
parallel shear mechanisms. At frontal ramp - oblique ramp intersections, synformal or antiformal
multiple bends in the footwall generate, respectively, second order hangingwall synclines or
anticlines, which terminate along strike into simple fault-bend folds. For the layer-parallel shear
mechanism along the pure oblique ramp, deflected hangingwall material passes through the
transport plane, conserving area and volume. At the rearward intersection zone (concave toward
the transport direction), hangingwall material diverges resulting in local strike-parallel extension.
This extension may be a mechanism for the generation of transverse faults (or 'tear faults') in the
hangingwall. At the forward intersection zone (convex toward the transport direction), displace-
ment paths converge resulting in local strike-parallel shortening. The attitude of the oblique ramp
and the amount of displacement significantly affect the map geometry and magnitude of lateral
strains.
A characteristic of thin-skinned thrust faults is that they are Jamison 1987; Groshong & Udansky 1988, in thrust terranes;
parallel to the stratigraphy along 'flats' and cut across the Verral 1981;Gibbs 1983;Whiteefa/., 1986;Groshong, 1988,
stratigraphic section along 'ramps'. Ramps are termed fron- in extensional terranes). Mechanical models have addressed
tal, oblique (or transverse), or lateral, depending on their ramp deformation ofisotropic half-spaces (Berger& Johnson
orientation with respect to the regional transport direction 1982; Kilsdonk & Fletcher 1989) as well as isotropic and
(Fig. 1). Similar ramp - flat geometries have been docu- anisotropic finite layers (Wiltschko 1979, 1981). Here, a
mented in extensional terranes as well (e.g. Gibbs 1984; kinematic model is developed bounded by two end members
Bosworth 1985; Wernicke & Burchfiel 1982), and these are of assumed mechanical behaviour, vertical shear and layer-
termed normal fault ramps. Ramps are a special case of fault- parallel shear. The two end members bound a range of
bends. In general, a fault-bend is a change in fault surface behaviour where a simple shear is imposed at an arbitrary
attitude. angle to bedding, a range also investigated in two dimensions
The purpose of this paper is to establish a criterion for the by White et al. (1986). However, this discussion is focused
kinematic development of a hangingwall moving over a on vertical shear and layer-parallel shear only. The three-
footwall oblique ramp. As the hangingwall negotiates an dimensional displacement and strain behaviour of these two
oblique ramp, do material particles remain entirely within the mechanisms due to motion over an oblique ramp are outlined,
tectonic transport plane (xz, Fig. 1), or are they deflected and discussed with respect to the implications on the devel-
resulting in out-of-plane deformation? opment of fault-bend folds. Here, the analysis is restricted to
Two-dimensional kinematic models of hangingwall de- the motion of hangingwall material by developing simple
formation over fault-bends include those with a preferred kinematic models, with no regard for the state of stress due to
directionofsimpleshear(e.g. Sanderson 1982;Suppe 1983; irregular footwall topography. The mechanics of this process
* Present address: Shell Development Company, PO Box 481, Houston, Texas 77001, USA
141
T. G. APOTRIA ET AL.
142
KINEMATIC MODELS OF DEFORMATION AT AN OBLIQUE RAMP
based on kinematic end members within which observed thickness of the hangingwall is reduced over the ramp, and
behaviour can be evaluated, and unobserved behaviour can the boundary between sheared and unsheared material (axial
be predicted. surface) is vertical. The magnitude of the shear strain in the
forelimb and backlimb in the xz plane is a function of the
ramp angle, 6 (Sanderson, 1982):
End members of mechanical behaviour
,w,z
.dj/ aw
£ x z - f.
u,x
-yz \ (;?f + ^ ) Figure 3. Ifthe increment of strain at
a ramp-flat hinge is infinitesimal, ex-
pressions for tensorial shear strain in
Vertical Shear the xz and yz plane can be used to
illustrate how displacements are related
to shear strains for vertical and layer-
parallel shear, u, v, and w are respec-
tively small displacements in the x, y,
and z direction. The x-y plane of the
B reference frame indicated corresponds
to bedding in the folded layer. The
dashed lines outline the undeformed
cube, the solid lines represents the
deformed cube, (a) For vertical shear,
displacements are parallel to z with
zero out-of-plane displacement, (b)
Layer-parallel shear admits
displacements in the x and y directions.
Deformed cube consists of components
Layer-Parallel Shear resolved parallel to x and y.
143
T. G. A P O T R I A ET At.
net strain. This mechanical behaviour is likely to occur if the Thinning of Hanging Wall Over
hangingwall is isotropic with respect to mechanical layering, Ramp (Vertical Shear)
if the rocks are vertically jointed, if the deformation occurs
under conditions where ductile mechanisms predominate, or
perhaps where rocks are unlithified or unconsolidated.
Layer-parallel shear
Alternatively, shear may occur parallel to bedding such that
bedding-normal thickness remains constant throughout the
deformation, analogous to parallel folding (Fig. 2b). This
mechanism is also called flexural slip if the shear occurs
along bedding planes, or flexural flow if the shear is otherwise
distributed through the rock (i.e. at the grain scale). To
maintain constant bed thickness, the axial surface bisects the
angle between the flat and the ramp. The magnitude of the
shear strain at each ramp - flat intersection is (Sanderson
1982):
Yx7. = 2tan
(f) (4)
% Thinning
Figure 5. Folding over a ramp is accommodated by thinning of the
Using the same analogy with tensorial shear strain, for hangingwall for the case of vertical shear. Thickness is measured perpen-
plane deformation e = 0 and in the case of layer-parallel dicular to bedding before (t) and after (t') passing through a ramp-flat hinge
shear, 3w/3x = 0 in equation (2). Infinitesimal displacements (Fig. 2). % thinning = (1- t'/t) • 100. The percentage of thinning in the xz
are parallel to bedding in the xz plane. If the upper and lower plane increases with both a and b, since each corresponds to an increase of
the component of ramp dip in that plane.
footwall flats are parallel, the shear strains at the lower and
upper hinges are opposite in sign, resulting in zero net shear
strain. The layer-parallel shear mechanism is likely to occur Application to oblique ramps
in the external, 'colder' levels of layered hangingwall sheets
To extend the analysis to three dimensions, initially consider
where interbed slip due to mechanical layering is prominent.
the path of a particle passing over a footwall oblique ramp
This mechanism provided the basis of folding models devel-
(Fig. 4). For the case of a frontal ramp, where two-dimensional
oped by Suppe (1983), Suppe & Medwedeff (1984), and
plane deformation is assumed, a particle moves in the direction
Jamison (1987).
of tectonic transport (x), moves up the ramp in a direction
normal to its strike, and continues in the transport direction on
the upper footwall flat (path 2, Fig. 4). The particle path along
the oblique ramp may not be quite as simple, and will be
distinctly different for the two end-members described above.
The oblique ramp, in general, will be a plane along which
Rearward Intersection
oblique slip occurs.
Zone In the case in which a vertical shear of the hangingwall
rocks is imposed at the flat-ramp hinge, material is sheared in
the xz and yz planes so that particles in the hangingwall move
vertically (Fig. 3a). Assuming an infinitesimal strain incre-
ment to illustrate the displacements, the tensor shear strain
components reduce to:
_i/aw\
(5)
'xz ~2V9x/
Figure 4. Footwall geometry including the salient footwall parameters,
a is the angle between the transport direction and the strike of the oblique
ramp. P is the dip of the oblique ramp. 5 is the dip of the frontal ramp. 9 is = i/aw\ (6)
the pitch of the particle path in the plane of the oblique ramp. If the strike
of the frontal ramp is perpendicular to the transport direction, then the
angle between the oblique ramp and the frontal ramp is 90-a. LQ is the where 9u/9z and 3v/3z , the horizontal displacement terms,
initial distance between two adjacent particles on the lower flat, L, is the
are zero. Therefore, there are only vertical displacements in
final distance on the upper flat.
the xz and yz plane in three dimensions. The shear strains are
144
KINEMATIC MODELS OF DEFORMATION AT AN OBLIQUE RAMP
..HI .
For small values of oblique ramp dip (p), the pitch of the
particle path (9), is nearly equal to the transport direction-
oblique ramp angle (a). This is also true at low and high
values for a, where the oblique ramp approaches a simple
frontal or lateral ramp, respectively.
The case corresponding to an end member model in which
layer-parallel shear is imposed at the flat-ramp intersection
results in a slight modification of equations (2) and (3) above
(Fig. 3b) such that:
TXdzJ (10)
V ^
- a (12)
Figure 6. The relation 9 - a for the case of layer-parallel shear can be The condition that 9=a for layer-parallel shear is a result
demotistraied using a simple physical model. The rigid footwall is repre-
sented by a wooden block and the hangingwall by a thickness of paper sheets, of the condition that there is no deformation in the plane of
which is a useful analogue for folding by layer-parallel shear. Here, all but
one sheet of paper has been removed and marked with a line representing the
transport direction. As the sheet is moved up the oblique ramp (Fig. 6b), the
particle path represented by the line is deflected out of the transport plane
such that 9 = a. In general, the panicle paths on the lower and upper flats are
parallel, but no longer contained in the same vertical plane. (Camera is
directed parallel to the transport direction).
where p^ and P are apparent dips of the oblique ramp in the PLAN VIEW
X and y directions, respectively (see also Fig. 7). There is no
displacement of the particle out of the transport (xz) plane for
the case of vertical shear. The two components of shear strain
are accommodated by thinning of the hangingwall over the
oblique ramp (Fig. 5). If there is no displacement of a
hangingwall particle path out of the transport plane, its path Figure 7. Parameters used to calculate the out-of-plane deflection for the
may be specified in the plane of the oblique ramp. In this case, layer-parallel shearmechanism. (a) p^ is the apparent dip ofthe oblique ramp
in the displacement direction, p is the angle between the deflected path and
an expression is derived for the relationship between a and the strike of the ramp measured in ihe horizontal plane, z, d, x. y, p, w, q are
the angle q, where q is the pitch of the particle path in the plane leg lengths of right triangles, h is the height of the ramps, (b) In plan view,
of the oblique ramp (see Fig. AI - Appendix 3): ihe deflection is a-p. Other symbols are listed in Appendix 1.
145
T. G. APOTRIA ET AL.
Deflection Out of Transport Given the displacement path of particles in the hangingwall
Plane (Layer-Parallel Shear) for the layer-parallel shear mechanism (eq. 12), the resultant
90 shear strain can be calculated, analogous to that in equation
(4), using the apparent dip of the oblique ramp in the direction
1c^r^.
\
i\\
of displacement (see P^ in Fig. 7). The shear strain in the
displacement direction can then be resolved into x and y
components.
This analysis has utilized two kinematic end member
\
models to describe the displacement paths and shear strains
](» \
<•
of hangingwall material at an oblique ramp. These displace-
a *if
t
'30
4(>/5C [) / 7ft
ment paths can be applied to determine fault-bend fold forms
over oblique ramps by making a transition from particle paths
/ /U to three dimensional volumes of rock.
^•8 0
1 / >
ii The fold forms that develop over oblique ramps differ for
each of the two end members of assumed mechanical behav-
iour, vertical and layer-parallel shear. For layer-parallel
0 Deflection 50 shear, the two-dimensional analytic solution of Suppe (1983)
is used, while a graphic solution is utilized for the vertical
(oc-p) shear mechanism. The purpose of this paper is not to present
Figure 8. Variation in the amount of deflection for a spectrum of ramp a complete three-dimensional theory of fault-bend folding,
geometries. The magnitude of the deflection increases with the dip of the but to perform the analysis in two dimensions, initially
oblique ramp (P). For any value of p, the deflection is zero at a=0° and a=90°, assuming plane strain, and constructing serial cross sections
and reaches a maximum at an intermediate value. In other words, there is no across two frontal ramp - oblique ramp intersections (Fig. 9).
out-of-plane deflection for pure lateral and frontal ramps. The location of the
deflection maximum increases from a= 45° to 90° as P increases from 0° to
It is then suggested as to how the theory must be modified to
90°. account for three dimensional geometry and kinematics.
cos 3 sin g
sin p : (13)
(1-(sin a sin P)2)l/2
146
KINEMATIC MODELS OF DEFORMATION AT AN OBLIQUE RAMP
Vertical shear
Fault-bend folds produced by this mechanism were con-
structed graphically by displacing the hangingwall up the D
ramp, maintaining constant vertical thickness (Fig. 10). Like
parallel fault-bend folding (Suppe 1983), two axial surfaces
are generated at each fault-bend (dashed lines), which unlike
parallel fault-bend folding, are vertical. One remains fixed to
147
T. G. APOTRIA ET AL.
A'
A' c c PP"
\>—
^
^ m g \ ^
2=
C c' B R' C C'
A \
fe^^^^^
' ^ :
i^
^ ^
A- V C- B A' B
c c ' R'
' 3 ^
i ^ H ^ ^ , , 1.11. ii|.iii.
A' C C'
? B-
-f-^
5;^.
\ JL i_ I
"^1
A A- C C'
A'
B B' B'
^
(V-
:2:
Figure 12. Detailed serial cross sections across ramp intersection zones for the layer-parallel shear mechanism. These sections are labelled a, b, c, d, e,
f, g, h, i,j, k, I, and have a displacements indicated by the distance along the ramp between axial surfaces A and A' in section a. The horizontal erosion
level in section a is the datum for the map in Figure 13. The location of these cross sections is given in Figure 9. See text for explanation.
148
KINEMATIC MODELS OF DEFORMATION AT AN OBLIQUE RAMP
deformation in the plane of bedding so that bedding line At the synformal footwall bend near the upper flat (section
length is conserved. Due to slip parallel to bedding, slip along g, Fig. 12), axial surface C is at the upper bend, localizing
the fault is lost to folding through each bend (Suppe 1983, eq. surfaces D-D'. The orientation of the D-D' and the dip of
16). Interbed slip and resultant angular shear also results in beds within is determined from the cutoff angle of beds
a forelimb dip that is steeper, in general, than in the case of between C-C, rather than the cutoff angle which determined
vertical shear (Fig. 11). Axial surfaces bisect bedding in the the dip of axial surface B-B' (horizontal beds between C and
hangingwall, thereby maintaining constant bed thickness. B'). Because this new cutoff angle is less, the dip of axial
Sections A and B are folds over frontal ramps dipping 30°, surfaces D-D' is correspondingly less than B-B' (see Suppe
resulting in forelimb dips of 60°. In section C, an apparent dip 1983, Fig. 7). Section k shows similar interference structures.
of 22° yields a forelimb which dips about 23°. Clearly, as From these serial cross sections, a map of the various dip
ramp dip decreases, the amount of interbed slip decreases. domains bounded by axial surfaces was generated at a speci-
Therefore, at small values of ramp dip, the forelimb dip fied horizontal erosion level (Fig. 13). In general, the
approaches the same dip as the ramp. As in the case of vertical hangingwall geometry is simple over pure frontal and oblique
shear, the shallower, oblique ramp apparent dip results in a ramps with two pairs of axial surfaces corresponding to each
difference in structural relief between frontal ramp anticlines fault-bend. Complexities arise at the intersection zones due
and the oblique ramp anticline which changes with displace- to multiple fault-bends and interfering axial surfaces. Within
ment. The antiformal bend in section B produces a second the intersection zones, multiply oriented dipping panels of
order anticline in the hangingwall. Similarly, there is a rock contribute an additional component of three-way clo-
second order syncline in the hangingwall of section D. Note sure between frontal and oblique ramp anticlines. Domain
that ramp anticlines produced by layer-parallel shear are dips in the intersection zones are the sum of two components:
broader than those produced by vertical shear. (1) dip in the plane of the cross section predicted by fault-bend
149
T. G. APOTRIA ET AL.
f
The cross sections and map generated in the previous section
assumed two-dimensional deformation, resulting in a first B ey=-.58
order approximation of the geometry and kinematics. Ex-
tending the results to three dimensions requires additional
considerations and assumptions. With regard to geometry, the \
use of two-dimensional fault-bend fold theory to generate
folds over oblique footwall ramps, in cases where cross
sections are oblique to the strike of ramp segments, does not
take into account the required apparent thickening in the p = 70O
hangingwall. Slight modifications in the orientation of axial
surfaces and in apparent fault offset should accompany the 6 = 30O
apparent thicknesses in obliquely oriented cross sections.
Additional kinematic assumptions are required as well. Figure 14. Two different ramp geometries are depicted in map view. In both
Frontal ramp fault-bend folds outside the ramp intersections cases, the frontal ramp dip (5) is 30°, and the angle between the transport
are assumed to deform by plane strain. That is, no material direction and the strike of the transverse ramp (a) is 45°. The deflected path
of hangingwall material is shown by the flow lines, as derived from equation
moves laterally from the frontal ramp domain into the oblique
(12) and Figure 8. (a) Oblique ramp dip (b) = 30°. (b) p = 70°. See text for
ramp domain. Displacement within oblique ramp fault-bend further explanation.
folds between the ramp intersection zones is specified by
equation (12), where an equal amount of material moves into separated by a distance L|. In case (a), the oblique ramp dip
the transport plane as moves out, thereby conserving area and
(P) is 30° and the deflection (a-p) is 4°. At the front inter-
volume. As we have shown earlier, material in the hangingwall
section zone, the resultant elongation across the entire inter-
is deflected out of the transport plane at the oblique ramp
section zone (AL/L) is -0.25. At the rear zone, the elongation
(Figs 6-8). Given the above assumptions, this deflection
is 0.33. In case (b), p=70° and the deflection angle is 26°. The
results in flow paths in the hangingwall which converge at the
elongation at the front hinge is -0.58, and at the rear hinge is
forward frontal ramp - oblique ramp intersection zone, and
1.4. With a and 5 constant, an increase in the oblique ramp
diverge at the rearward intersection zone. Convergence
dip increases the longitudinal strain. Although the magnitude
within this zone results in local lateral shortening and thick-
of the displacement path deflection (Fig. 8) is relatively small
ening, divergence results in local lateral extension and thin-
for oblique, dips under 40°, they should result in longitudinal
ning. Therefore, it appears then that at frontal ramp - oblique
ramp intersections, layer-parallel shear alone cannot accom- strains at ramp intersections that may be measurable in the
modate the deformation. field.
Figure 15 is a schematic block diagram depicting how the
In Figure 14, two different ramp geometries are depicted lateral strains might be manifest in the field. At the rearward
in map view. The deflected path of hangingwall material is intersection zone, rocks originally on the lower flat will be
shown by the flow lines, as derived from equation (13) and strained (y ) at the lower ramp hinge. Above the line of
Figure 8. A calculation of the longitudinal elongation (e ) at intersection of the two ramps, two shear strains are imposed
each ramp intersection has been made graphically by meas- (y^^,-Y^) as well as a lateral extension due to divergent
uring the line length change between two particles, originally deflection (e ). The extension may be manifest as fracturing
a distance L^ apart (see Fig. 4). One particle undergoes pure or normal faulting in the hangingwall. If the extension is
frontal ramp deformation, while the adjacent particle encoun- accommodated by a normal fault, hinterland dipping rocks of
ters the oblique ramp and is deflected out of the transport the backlimb will show an apparent left-lateral strike-slip
plane. On the upper footwall flat, the two particles are now offset in map view. Deflection and lateral extension may be
150
KINEMATIC MODELS OF DEFORMATION AT AN OBLIQUE RAMP
Figure 15. Schematic block diagram depicting how the lateral strains may be manifest in the field. Each cube of rock represents a small element of the
hangingwall moving up the frontal ramp-oblique ramp intersection zone. See text for further explanation.
a mechanism for 'tear faulting' in the hangingwall, which, in tonic transport plane; the deformation is accommodated by
this case, is normal faulting with an oblique slip component. thinning of the hangingwall over the ramp. In the later case,
At the forward intersection zone, a component of lateral material is deflected out of the transport plane such that the
shortening (e ) should result in folds, faults, or cleavage at a pitch angle of the particle path in the plane of the oblique ramp
high angle the regional transport direction. is equal to the initial angle between the transport direction and
the strike of the oblique ramp. This deflection results in shear
strains imposed at ramp - flat hinges. The deflection and out-
Discussion of-plane shear strain are zero for the special cases of pure
frontal and lateral ramps, and maximum at an intermediate
On the basis of previous work, oblique ramps are suspected oblique orientation, depending on ramp dip. Fault-bend folds
to be structures where the common assumption of plane strain are grossly similar for both vertical shear and layer-parallel
is violated. They may be the locus of oblique folds, faults, shear mechanisms. At frontal ramp - oblique ramp intersec-
cleavages, and fractures, and tend to be avoided when con- tions, synformal or antiformal multiple bends in the footwall
structing balanced cross sections. One component in under- generate second order hangingwall synclines or anticlines.
standing the geometry and kinematics of oblique ramp defor- These terminate along strike into simple fault-bend folds. For
mation requires a three-dimensional modelling approach. the layer-parallel shear mechanism, the deformation is more
Kinematic models for the deformation of hangingwall complex than vertical shear. Along the simple oblique ramp,
material moving over a footwall oblique ramp are developed deflected hangingwall material passes through the transport
by considering two end members of assumed mechanical plane, conserving area and volume. Within the intersection
behaviour, vertical shear and layer-parallel shear. The ob- zones, axial surface interference generates additional panels
served behaviour may be expected to lie between the two of dipping beds At the rearward intersection zone (concave
idealized end members, depending on the intrinsic and extrin- toward the transport direction), hangingwall material di-
sic conditions during thrusting. In the former case, material is verges resulting in strike-parallel extension, or the develop-
sheared vertically and displacements remain within the tec- ment of a 'gap'. This extension may be a mechanism for the
151
T. G. APOTRIA ET AL.
generation of transverse faults (or 'tear faults') in the and/or erosion ofa thrust fault (e.g. Woodward, 1987b). The
hangingwall which are, in fact, oblique slip normal faults. At detection of out-of-plane strains may eliminate cases (a) and
the forward intersection zone (convex toward the transport (b) where subsurface control of the thrust fault is lacking.
direction), displacement paths converge resulting in strike- Furthermore, the models are a prerequisite to developing
parallel shortening, or 'overlap'. The attitude of the oblique techniques of three-dimensional cross section construction
ramp and the amount of displacement significantly affect the and restoration. Depending on the end member model and
map geometry and magnitude of lateral strains. fault-bend geometry, the out-of-plane displacements can be
The models have several important implications. First, specified, allowing oblique structures to be palinspastically
deflection of material and accompanying longitudinal strains restored. For example, over simple oblique ramp segments,
suggests that the layer-parallel shear mechanism alone is area and volume are balanced regardless of out-of-plane
insufficient at ramp intersections given the assumptions of flow. At ramp intersections, extensional or shortening longi-
the model. To maintain compatibility at the intersection, tudinal strains, perhaps manifest as folding or faulting, will
oblique folds, faults, and fabrics are expected, the sense and need to be incorporated into balanced sections in addition to
magnitude of which are predicted. Secondly, oblique ramp penetrative strains.
anticlines are not, in general, an indication of the local The models presented here may be an alternative approach
transport direction. In both vertical shear and layer-parallel to transpression and differential transport models (Sanderson
shearmodels, material flow in the hangingwall is not coplanar & Marchini 1984; Sanderson 1982; Coward & Kim 1981;
with the principal shortening direction inferred from fault- Coward & Potts 1983) which employ a strain factorization
bend fold axes or other hangingwall fabrics, neither of which, method. A common aspect of these models is that differential
in general, remain in the transport plane. This conclusion has displacement along the basal detachment, presumably occur-
been substantiated independently by mechanical modelling ring in the vicinity of a lateral thrust tip or oblique ramp,
of oblique fault surfaces using a continuum mechanics ap- imposes a shear strain in the plane of the detachment (Y^ ). The
proach (Apotria 1988, 1990). out-of-plane strain component (y ) due to deflection may
Some authors have suggested that motion over an oblique give similar results.
ramp requires that the hangingwall undergo regional strike- In extensional terranes, Gibbs (1984) suggests that trans-
parallel extension (as opposed to local, over the intersection verse faults at high angles to the regional trend are fundamen-
zones) as the hangingwall moves from the lower footwall flat tal structures associated with normal faults. The analogy with
through the oblique ramp (e.g. Butler 1982). The supposition oblique ramps in thrust belts is proposed by Gibbs, in which
that regional strike-parallel extension is required is based on complex rotational, synthetic dip, and strike slip components
conservation of line length in a direction parallel to the strike allow extension to transfer style and displacement along a
of the frontal ramp, but is dependent on the bulk deformation graben. Similarly, both shortening and extensional structures
mechanism of the hangingwall. For example, in the case of can develop depending on the relative orientations of the
layer-parallel shear, the apparent extension is accommodated transverse fault and normal faults. The models proposed
by layer-parallel slip and out-of-plane shear strains. If one herein may accommodate regional extension by simply re-
relaxes the assumption that deformation is plane over pure versing the transport direction in Figure 1. For the layer-
frontal ramp segments, out-of-plane displacement of material parallel shear model, equation (12) holds for extensional
from the frontal ramp domain into the oblique ramp domain oblique ramps. More appropriate models in extensional
may accommodate the apparent regional extension suggested terranes may employ inclined simple shear, rather than ver-
by Butler, and perhaps distribute the local lateral strains due tical or layer-parallel shear. Models of this type are in
to deflection over a greater area. A component of strike- progress.
parallel extension could occur if the hangingwall deformed as
an isotropic bending beam, with extensional strains in the
outer arc and shortening strains in the inner arc.
Preparation of this manuscript benefited from discussions with D. Goff, S.
The model may be useful in deciphering oblique footwall G. Erickson, D. DePaor, and D. Medwedeff. Financial support was contrib-
ramping from other along-strike variations. In map view, a uted by Texaco E&P Technology Division, ARCO Oil and Gas Co., Shell
Western E&P Inc., Chevron USA Inc., Unocal Science and Technology
change in stratigraphic separation along the trace of a thrust Division, Geological Society of America and Sigma Xi grants-in-aid of
is often interpreted as an oblique ramp. However, nearly research. We appreciate the constructive reviews by D. J. Sanderson, K. R,
identical map geometries can result from: (a) a change in McClay, and an anonymous reviewer.
hangingwall displacement along strike, and (b) uplift, plunge.
References
Apotria, T. G. 1988. A 3-D solution for hangingwall motion over a wavy fault Berger, P. & Johnson, A. M. 1980. First order analysis of deformation ofa
surface. Geological Society ofAmerica Abstracts with Programs, 20: 7, thrust sheet moving over a ramp. Tectonophysics, 70, T9-T24.
A57. Bosworth, W. 1985. Geometry of propagating continental rifts.Nature, 316,
1990. The Kinematics and Mechanics of Oblique Ramp Deformation 625-627.
within Fold and Thrust Belts. Ph.D. dissertation, Texas A & M Univer- Boyer, S. E. 1985. Hydrocarbon trap styles in fold and thrust belts and related
sity. terranes. Ojfshore Technology Conference, OTC 4873, 297-305.
152
KINEMATIC MODELS OF DEFORMATION AT AN OBLIQUE RAMP
Boyer, S. & Elliott, D. 1982. Thrust systems. American Association of Pe- Petrini, H. & Wiltschko, D. V. 1986. Some simple geometrical models for
troleum Geologists Bulletin, 66, 1196-1230. strains due to motion over lateral ramps. Geological Society of America,
Butler, R.W.H. 1982. Hangingwall strain: a function of duplex shape and Abstracts with Programs, 18, 717.
footwall topography. Tectonophysics, 88, 235-246. Pfiffner, O. A. 1981. Fold and thrust tectonics in the Helvetic Nappes (E.
Couples, G. D. & Lewis, H. 1988. 'Thrust belt' structures with 'foreland' Switzerland). In: McClay, K. R. & Price, N. J. (eds). Thrust and Nappe
influence: the interaction of tectonic styles. In: Schmidt, C. J. & Perry, Tectonics. Geological Society of London Special Publication, 9,319-328.
W.J. (eds). Interaction of the Rocky Mountain Foreland and the Cordilleran Sanderson, D. J. 1982. Models of strain variation in nappes and thrust sheets:
Thrust Belt. Geological Society of America Memoir, 171, 99-110. a review. In: G. D. Williams (ed.). Strain within Thrust Belts,
Coward, M. P. & Kim, J. H. 1981. Strain within thrust sheets. In: McClay, Tectonophysics, 88, 201- 233.
K. R. & Price, N. J. (eds). Thrust and Nappe Tectonics. Geological So- & Marchini, W.R.D. 1984. Transpression. Journal ofStructural Ge-
ciety of London Special Publication, 9, 275-292. ology, 6, 449-458.
& Potts, G. J. 1983. Complex strain patterns developed at the frontal Schmidt, C. J., O'Neil, J. M. & Brandon, W. C. 1988. Influence of Rocky
and lateral tips to shear zones and thrust zones. Journal of Structural Mountain foreland uplifts on the development of the frontal fold and
Geology. 5, 383-395. thrust belt, southwestern Montana. In: Schmidt, C. J. & Perry, W.J. (eds).
Gardener, D.A.C. & Spang, J. H. 1973. Model studies of the displacement Interaction of the Rocky Mountain Foreland and the Cordilleran Thrust
transfer associated with overthrust faulting. Bulletin of Canadian Pe- Belt, Geological Society of America Memoir 171, 171-202.
troleum Geologists, 21, 534-552. & O'Neil, J. M. 1982. Structural evolution of the southwest Montana
Gibbs, A. D. 1983. Balanced cross section construction from seismic transverse zone. /«.• R. B. Powers (ed.), Geologic Studies ofthe Cordilleran
sections in areas of extensional tectonics.yoM/-«a/ ofStructural Geology, r/iz-uirBe/r.vo/./. Rocky Mountain Association of Geologists, 193-218.
5, 153-160. Suppe, J. 1983. Geometry and kinematics of fault-bend folding. American
1984. Structural evolution of extensional basin margins. Journal of the Journal of Science, 283, 684-721.
Geological Society of London, 141, 609-620. & Medwedeff, D. A. 1984. Fault-propagation folding. Geological Society
Goldburg, B.L.I 984, Displacement transfer between thrust faults near the of America, Abstracts with Programs, 16, 670.
Sun River in the Sawtooth Range, northwestern Montana. Montana Verral, P. 1981. Structural interpretation with applications to North Sea
Geological Society 1984 Field Conference. 211-220. \>roh\emi: Joint Association of Petroleum Exploration Courses, London,
Groshong, R. H. 1988. Half-graben structures: balanced models of extension Course Notes No. 3.
fault-bend folds. Geological Society of America Bulletin, 101, 96-105. Wernicke, B. & Burchfiel, B. C. 1982. Modes of extensional tectonics.
& Udansky, S. I. 1988. Kinematic models of plane-roofed duplex Journal of Structural Geology, 4, 104-115.
styles. In: Mitra, G. & Wojtal, S. (eds). Geometries and Mechanisms of Wheeler, R. L. 1980. Cross-strike structural discontinuities: possible explo-
Thrusting, with special reference to the Appalachians. Geological So- ration tool for natural gas in the Appalachian overthrust belt. American
ciety of America Special Paper 222, 197-206. Association of Petroleum Geologists Bulletin, 64, 2166-2178.
Jamison, W. R. 1987. Geometric analysis of fold development in overthrust White, N. J., Jackson, J. A. & McKenzie, D. P. 1986. The relationship
terranes. Journal of Structural Geology. 9, 207-219. between the geometry of normal faults and that of the sedimentary layers
Kilsdonk, B. & Fletcher, R. C. 1989. An analytical model of hangingwall and in their hangingwalls. Journal of Structural Geology. 8, 897-909.
footwall deformation atramps on normal and thrust faults. Tfcwnop/iyi/ti, Wiltschko, D. V. 1979. A mechanical model for thrust sheet deformation at
163, 153-168. a ramp. Journal of Geophysical Research. 84, 1091-1104.
Lageson, D. R. 1984. Structural geology of the Stewart Peak Culmination, 1981. Thrust sheet deformation at a ramp: summary and extensions of
Idaho-Wyoming thrust belt. American Association of Petroleum Geolo- an earlier model. /«; McClay, K. R. & Price, N.J. (eds). Thrust and Nappe
gists Bulletin, 6S,40\-416. Tectonics. Geological Society of London Special Publication, 9, 55-63.
Mitra, G. & Yonkee, W. A. 1985, Relationship of spaced cleavage to folds Woodward, N.B.I 987a. Primary and secondary basement controls on thrust
and thrusts in the Idaho-Utah-Wyoming thrust bt\\.Journal ofStructural sheet geometries. In: Schmidt, C. J. & Perry, W. J. (eds). Interaction of
Geology,7,361-313. the Rocky Mountain Foreland and the Cordilleran Thrust Belt. Geo-
Mitra, S. 1988. Three-dimensional geometry and kinematic evolution of the logical Society of America, Memoir 171. 353-366.
Pine Mountain thrust system, southern Appalachians. Geological So- 1987b. Stratigraphic separation diagrams and thrust belt structural
ciety of America Bulletin. 100, 72-95. analysis. Wyoming Geological Association Guidebook, Thirty-Eighth
Field Conference, 69-77.
Appendices
Appendix 1: List of symbols Appendix 2: Derivation of r for case of layer-parallel shear (Fig.
7)
a: angle between transport direction and strike of oblique ramp
measured in horizontal plane. Ramp height h is assumed to be 1. p, d, z, and w are arbitrary
5: dip of frontal ramp triangle leg lengths to be eliminated leaving a final expression
(3: dip of oblique ramp relating angles.
Pj! apparent dip of oblique ramp in the direction of displacement
for the layer-parallel shear mechanism (A2-1)
P^: apparent dip of oblique ramp in x-direction
p = sin P
P : apparent dip of oblique ramp in y-direction
d= P (A2-2)
6: pitch of hangingwall particle path in the plane of the oblique sin a
ramp h
sin Pd = (A2-3)
p: angle between particle path in the plane of the oblique ramp 'd
and the strike of the oblique ramp measured in the horizontal h
plane z = tan Pd (A2-4)
Yjj^: shear strain in the xz (transport) plane
J : shear strain in the yz plane
1h
(A2-5)
Y^: shear strain in the plane containing the displacement path for
w =" tan p
the layer-parallel shear mechanism w
sin P = (A2-6)
z
153
T. G. APOTRIA ET AL.
By back substitution:
^ tan a (A3-7)
COS P
154
stress controls on fold thrust style
William R. Jamison*
Abstract: Fold-fault relationships in overthrust terranes can usually be placed in the spectrum of
fold/thrust styles ranging from fault-bend to fault-propagation to detachment folds. The specific
fold/thrust styles of three major folds in the Wyoming-Idaho-Utah thrust belt have been assessed
geometrically. Whitney Canyon-Carter Creek anticline is a fault-bend fold. Haystack Peak
anticline is a detachment fold, and Hunter Creek anticline is probably a fault-propagation fold. All
three structures involve essentially the same mechanical stratigraphy. It is postulated that the
preferential development of a specific fold/thrust style reflects a fundamental competition between
buckling and faulting in the layered rock package. Both deformational processes may be
represented by instability envelopes in three-dimensional stress space. The form and location of
these two envelopes is a function of the mechanical stratigraphy. The fold/thrust style that
develops depends upon which instability surface is intersected first by the stress path. Depth of
burial and regional tectonics are the major factors determining the stress path. In a thrust belt
setting, the buckling instability surface is likely to be the initial intersect of the stress path only at
relatively shallow depths of burial. At greater depths, the faulting instability surface is the initial
intersect. This suggests that detachment folds develop most readily in the shallow subsurface,
whereas fault-bend folds dominate in the deeper subsurface. Consequently, outcropping struc-
tures may not always be appropriate analogues for the structures occurring at depth.
Geologists working in overthrust terranes have recognized and the choice of fold/thrust style, appears to be a function of
for over a century that the folds they observed were inherently the state of stress at the time of fold initiation as well as the
linked to thrusts, either bedding-parallel thrusts (decollements) mechanical stratigraphy.
or thrust ramp segments (e.g. Willis 1890; Rich 1934). The
specific nature of the fold/thrust interaction can usually be
categorized as one of three basic styles (Fig. 1), viz. fault- W y o m i n g thrust belt e x a m p l e s
bend folding (e.g. Rich 1934; Suppe 1983), fault-propagation
folding (e.g. Faill 1973; Williams & Chapman 1983; Suppe The Wyoming-Idaho-Utah thrust belt (Fig. 2a) has a gentle
& Medwedeff 1984), and detachment folding (e.g. Willis but persistent southerly plunge. Numerous major hangingwall
1890; Laubscher 1977). All three styles exist, and are anticlines are well exposed in the northem portion of the
common in nature. The focus of this paper is on some of the thrust belt. To the south, the correlative structures, many of
mechanical factors that influence the selective development which are highly constrained by development drilling, reside
of these various fold/thrust styles. in the subsurface. The detailed cross-sectional geometries of
The mechanical characteristics of the rock package, i.e. three of these folds have been determined, and are used to
the mechanical stratigraphy, are a major factor affecting the infer fold/thrust style. Hunter Creek and Haystack Peak
macroscale deformational characteristics of layered rock. It anticlines are both surface exposed structures, located in the
is an important factor in determining fold/thrust style, but it Prospect and Absaroka thrust sheets respectively, whereas
is not the dictating influence. To support this contention, Whitney Canyon-Carter Creek anticline is a subsurface struc-
examples of different fold/thrust styles (all from the Wyo- ture in the Absaroka thrust sheet (Fig. 2a).
ming-Idaho-Utah thrust belt) that have developed in the same
rock package are presented.
Mechanical stratigraphy
The three end-member styles of fold/thrust interaction
(Fig. 1) are categories of fold/ramp timing. In the fault-bend As these are Laramide-age structures, they may have in-
fold, ramping precedes folding; in the detachment fold, volved the entire Mesozoic and Palaeozoic section, a sequence
folding precedes ramping; and in the fault-propagation fold, of sedimentary rocks ~5 km thick (Fig. 2b). The mechanical
ramping and folding are synchronous events. The different response of this sequence was likely dominated by units
fold/thrust styles thus reflect the relative dominance of within the Palaeozoic and lower Mesozoic section. The
buckling vs. faulting (folding vs. ramping). A stress space major competent struts in this stratigraphic sequence are the
approach is used to investigate the competition between these Bighorn, Madison, and Thaynes carbonates, each 150 m to
two processes. The choice between buckling and faulting, 300 m thick. The Weber-Wells and Nugget Formations, both
'Present address: Centre for Earth Resources Research, Department of Earth Sciences, Memorial University of Newfound-
land, St. John's, Newfoundland, Canada A l B 3X5
155
W. R. JAMISON
Haystack Peak
w>i_Hunter Creek
Anticline ->t Anticline
Montana
Wyoming
Colorado
Figure 2 (a) Reference map for the Wyoming-Idaho-Utah thrust belt showing the location of the thrust belt in the western USA, the surface traces of the
major thrusts, and the locations of the three anticlines discussed in this paper, (b) Generalized stratigraphy of the Wyoming-Idaho-Utah thrust belt. Standard
lithological symbols used. Units not shown to relative scale.
156
STRESS CONTROLS ON FOLD/THRUST STYLE
Fault-F*ropagation Folding
,25%' .
.-•'- / r i 4 4 .
-50% i \ !
.i00%L-i-'^^-^^y^9nJ
, . . , . . / : M M ! iI
10 4 - / ' - ' ^ ' : - ; , , - - ' .,200% ~ - - i - i - - r - r - - r - p y
^ forelimb / .| J
thickening/i • ; • i • i :
0 —I
30 1
60 '—\—•—^H—'—^-r
90 120 150 180
Figure 3. (a) Surface map of major structural features along the Strawberry fold interlimb angle (y)
Creek drainage, Salt River Range, Wyoming. Modified from Rubey (1973).
Rock units depicted by age groups: Q-Quatemary; T-Tertiary; K-Creta-
ceous; J-Jurassic; JTr-Jurassic plus Triassic; P-Permian plus Pennsylvanian;
M-Mississippian; DO-Devonian plus Ordovician; DOC-Devonian plus
Ordovician plus Cambrian; C-Cambrian. Schematic cross-section interpre-
tation of Haystack Peak and McDougal Pass anticlines along the Strawberry
Creek drainage as either (b) fault-propagation folds or (c) detachment folds.
157
W. R. JAMISON
ture, say 30° (e.g. Serra 1977; Beyer & Elliott 1982), the fault-
Detachment Folding (a/f=3) propagation fold model with 7^= 42° still has a forelimb thin-
ning of over 60%.
50% x:50% 100% : :
For analysis of the detachment fold interpretation, the
value of a/f must be specified, where a is the fold height and
f is the thickness of the ductile unit filling the core of the
structure (Jamison 1987). The fold height is 1000 m. The
Cambrian shales are the likely material to act as the ductile
core unit. The Cambrian shales have a maximum thickness
200% i
in the area of about 350 m (C. Bartberger pers. commun.
1984). Assuming that the decollement is at the base of these
shales, the a/f value is 1000 m/350 m, or about 3. This is a
t no solution minimum value; if the decollement is higher in the Cambrian
shales, f is smaller and a/f is larger. With an a/f of 3, an
interiimb angle of 42° and a backlimb dip of 57°, the geomet-
;//;',';/Torelimb _
JV;,';- thickening ric analysis indicates a forelimb thinning of 20% (Fig. 6).
The field measurements of Haystack Peak anticline agree
well with the geometric models of detachment folding, but
Oi 1 '"'•"""i ' r are not compatible with the model of fault-propagation fold-
0 30 60 90 120 150 180 ing. This geometric analysis, along with the absence of any
fold interlimb angle (y) direct evidence of a large ramp associated with this fold,
strongly favours the interpretation of Haystack Peak anticline
as a detachment fold. Though McDougal Pass anticline has
not been studied in comparable detail, its strong similarity in
form and setting to the Haystack Peak structure suggests it is
stiff unit
also a detachment fold.
f ductile unit
West East
S/L
Carter PA.
Creek — — 2 km s/sea
__DQCl
^ Absaroka Thrust Absaroka Thrust
Whitney
Canyon
S/L
t:: — 2 km s/sea
Absaroka Thrust
d ^-^^s^-^'^'
Figure 7. Cross-section imterpretations of the Whitney Canyon-Carter Creek oil field structure. Modified from Weir (1983). TC-Twin Creek Fm.; NA-
Nugget plus Ankareh Fms.; TDW-Thaynes plus Dinwoody plus Woodside Fms.; PA-Phosphoria plus Weber & Amsden Fms.; M-Madison Fm.; DOC-
Devonian plus Ordovician plus Cambrian units. Angular measurements made using top of the Dinwoody Fm. Wells penetrating to the top of the footwall
of the Absaroka thrust beneath this structure record only Cretaceous rocks.
158
STRESS CONTROLS ON FOLD/THRUST STYLE
159
W. R. JAMISON
of about 1000 m.
Hunter Creek and its companion anticline have cumula-
tive fold heights (2300 m) comparable to the stratigraphic
thickness of the hangingwall truncation (1600 m to 2000 m).
They can thus potentially be either fault-bend or fault-propa-
gation folds associated with the Prospect thrust. If Hunter
Creek anticline is alternatively interpreted as either a fault-
propagation fold associated with the Shepard thrust or a
detachment fold, this hangingwall truncation is only partially
accounted for. These latter interpretations would also require
an abnormally thick Cambrian section or involvement of the
crystalline basement in the core of Hunter Creek anticline.
These conditions are not compatible with the regional
stratigraphy or the structural style of the thrust belt. Thus, it
is more probable that Hunter Creek anticline is either a fault-
Hunter Creek bend or fault-propagation fold associated with the Prospect
Anticline
thrust.
Because displacement along the Prospect thrust has been
several kilometres (Royse et al. 1977; Dixon 1982), Hunter
Creek anticline must have been transported well to the
foreland of its associated footwall ramp. The geometry of a
transported fault-propagation fold is very similar to a mode 11
fault-bend fold (Jamison 1987). Geometric evaluation of
Figure 9. (a) Surface map of the major structural features between Shepard both interpretations for the Hunter Creek anticline can, there-
Creek and the Prospect thrust trace, Clause Peak quadrangle, Wyoming. fore, be made using fault-bend fold relationships (Fig. 11).
Modified from Schroeder (1973). Rock units depicted by age groups: QT- The exposure of Hunter Creek anticline on the south side
Quatemary plus Tertiary; K-Cretaceous; JTr-Jurassic plus Triassic; Mz- of the Shepard Creek drainage has been surveyed and pro-
Mesozoic; P-Permian plus Pennsylvanian; M-Mississippian; PM-Permian
jected into a cross-sectional plane normal to the fold axis
plus Pennsylvanian plus Mississippian; DO-Devonian plus Ordovician. (b)
Schematic cross-section of major structures along east-west line from (354704°). The Madison section is 310 m thick in the central
Shepard Creek to the Prospect thrust trace. forelimb, which is close to the normal stratigraphic thickness
180
160
STRESS CONTROLS ON FOLD/THRUST STYLE
of the Madison Formation in this region. The fold interlimb principal stress axes relative to the material layering is
angle (y) determined using the mid-forelimb dip is 94° (Fig. important in determining the response of this stratified mate-
10). No forelimb thickness change and y=94° suggests that rial to a given stress state. In order to address the layering
Hunter Creek anticline is either a mode II fault-bend fold or effects, a very specialized stress condition is assumed for the
a transported fault-propagation fold with a ramp angle (a) of following discussions. Specifically, it is assumed that the
27° (Fig. 11). There is no direct control on a, but this inferred principal stress axes, indicated as (5^^, o^^, and a^, are parallel
a=27° is very compatible with the generalized observations and perpendicular to layering (Fig. 13). There is no restric-
of Serra (1977) and Boyer & Elliott (1982). In the lower tion on the relative or absolute magnitudes of the principal
Madison Formation, right in the hinge region, ^=69° and the stresses, only their orientation relative to bedding. This
forelimb has locally thinned about 20% (Fig. 10). Fora=27° special set of conditions is referred to as the 'working stress
andy=69°, a thinning of 25% is indicated (Fig. 11), providing state'.
further support for the inferred a= 27° and the fault-bend/
fault-propagation fold interpretation. Working Stress Space
The geometric analysis does not provide a direct method
to distinguish between the mode II fault-bend fold and the
transported fault-propagation fold. From kinematic consid-
erations, though, it may be argued that a mode II fault-bend
fold should exhibit substantial deformation in the forelimb
(Jamison 1987). The absence of such deformation suggests
that this structure is probably a fault-propagation fold.
Figure 13. The 'Working Stress Space' is the special case of principal stress
axes oriented parallel and perpendicular to bedding, referenced as shown.
Stress controls
Faulting instability envelope
Haystack Peak anticline is definitely a detachment fold.
Whitney Canyon-Carter Creek anticline is definitely a fault- The stress conditions for the initiation of brittle faulting
bend fold. Hunter Creek anticline is probably a fault-propa- depend on both the mean stress and the differential stress as
gation fold. All three of these structures involve the same described, for example, by the Mohr-Coulomb or Drucker-
fundamental mechanical stratigraphy. This inconsistency in Prager failure criteria. In stress space, the condition of zero
the nature of the fold/thrust interaction in the same rock differential stress is described by the hydrostat (H in Fig. 14),
package argues that some factor beyond mechanical the line defined by the condition o =0 =0 . (E is the projec-
stratigraphy influences variations in fold/thrust style. In the tion of H onto the o^^-<3^^^ plane.) A given state of stress plots
following, the potential control by the state of stress at the as a point in stress space, and any point not on H indicates a
inception of fold development is explored. The underlying non-zero differential or deviatoric stress. The farther the
theme is that the observed variations in fold/thrust style point is from H, the larger the differential stress. The faulting
reflect the competition between folding and faulting, and that instability envelope for an isotropic material is symmetric
this competition can be assessed by examining the form of the about H. The conical faulting instability surface shown in
instability envelopes for buckling and faulting in terms of Figure 14a is defined by the Drucker-Prager failure criteria
stress. (Drucker&Prager 1952):
STRESS SPACE
Of = P J I + V J 7 (1)
where a^ is the failure stress, J^ and J^' are the first stress in-
variant and the second invariant of deviatoric stress, respec-
Figure 12. Stress space is a three- tively, and P is a coefficient (>0 for compression positive) that
dimensional reference system with axes may vary as a function of J j. As long as the state of stress plots
representing the principal stress magnitudes.
as a point internal to this cone the material is stable, but when
the stress state moves to contact with this conical failure
The working stress space
surface the material will fault of fracture (under conditions of
Stress space (Fig. 12) is a three-dimensional reference space brittle deformation).
whose axes are the principal stress values (e.g. Jaeger & Laboratory tests of rock samples with definitive layering
Cook 1967). It is a non-physical space; only the principal have found that the failure stresses are (1) relatively reduced
stress magnitudes are recorded. There is no reference to any in value for a, oriented less than 45° to layering at low
orientation data in stress space, including the principal stress confining pressures and (2) independent of principal stress
orientations. This is not of particular concern when dealing orientations at high confining pressures (e.g. McGill & Raney
with a mechanically isotropic material. However, the folds 1970; Donath 1961). In the working stress state, these
discussed above involve a distinctly stratified rock sequence, anisotropic effects can be incorporated, at least qualitatively.
which is mechanically anisotropic. The orientation of the The layering-related anisotropy will move the faulting instability
161
W. R. JAMISON
the hydrostat (H). E is the projection of H onto the a^, j -a|^2 P'sne- A sectional by an equivalent amount (Timoshenko 1936), i.e. o ,=0^1+01,2"
view in the a^-E plane (b) cuts through the centre of the conical failure a^. In the C5h|-o,,2 plane, the buckling instability envelope is
surface. The region internal to the cone is the area of stability for faulting, simply the straight line intersecting both of these axes at o^^
and the region external to the cone is the area of instability for faulting. The
solid curved lines depict the faulting instability envelope for an isotropic
(Fig. 15b). Putting the instability surface together from the
material. The dashed curved line is the suggested modification of the faulting Oj^i-a^ and CJ^|-0|,2 planes yields a buckling instability surface
instability envelope for an anisotropic material in the working stress space. for the working stress space of cylindrical form (Fig. 15c).
Buckling Instability
envelope towards H along the a^^ and o^^, axes (as suggested
by the dashed line in Fig. 14b). This shift becomes less
pronounced as the stress values increase. The net effect is to CTh2=0
make the faulting instability surface for layered rock a some- region of stability
what lopsided cone.
region of instability
162
STRESS CONTROLS ON FOLD/THRUST STYLE
163
W. R. JAMISON
necessitated for continued fold growth. This fault develop- occurrence of different fold/thrust styles is a function of both
ment during the incipient stages of fold growth may yield a the mechanical stratigraphy and the stress path of the rock
fault-propagation fold. package. The mechanical stratigraphy determines the form
The development of Hunter Creek anticline as a fault- and location of both instability surfaces. The stress path is
propagation fold may be rationalized using either explana- dependent on the depth of burial and the regional tectonics.
tion. There is no Cambrian strata exposed in the core of this In terms of determining fold/thrust style in a thrust belt
fold. It is, thus, possible that the Prospect thrust beneath with a fairly uniform stratigraphy, such as the Wyoming-
Hunter Creek anticline carries little or none of the Cambrian Idaho-Utah thrust belt, depth of burial becomes a critical
shales, the only lithology that would serve as the critical core- factor. In general, detachment folding is likely to develop
filling material for detachment folding. Fault-propagation only at relatively shallow depths, and fault-bend folding will
folding would then evolve as per the second scenario above. become the dominant fold/thrust style at greater depths.
Alternatively, Hunter Creek anticline may have developed at Because fault-bend folds can have larger interlimb angles
a burial depth intermediate between Haystack Peak and than fault-propagation folds and most detachment folds
Whitney Canyon-Carter Creek anticlines. Although roughly (Jamison 1987), folds in thrust belts might be expected to be
the same units outcrop at Hunter Creek anticline as at Hay- more open structures at depth than in outcrop or in the shallow
stack Peak anticline, the former structure has probably been subsurface. An important consequence of this concept is that
transported several kilometres foreland and over a kilometre the structures observed in outcrop may not always be appro-
upsection from its developing location (Dixon 1982). The priate analogues for subsurface structures.
stress conditions at the inception of the Hunter Creek fold The concepts of stress controls on fold/thrust style pre-
may have been located near the intersection of the two sented here have assumed (1) a very special set of principal
instability envelopes. stress orientations and (2) the deformation is appropriately
treated via brittle and elastic descriptions. However, these
assumptions must be dropped for the generalized application
Conclusions of stress control and deformational style. If the principal
stress axes deviate from the working stress space conditions,
Geometric analysis provides a methodology for distinguish- both the configuration and the location of the instability
ing different fold/thrust styles. The application of this proce- envelopes, especially for buckling, will change, but the basic
dure to three folds in the Wyoming-Idaho-Utah thrust belt concepts and conclusions presented here should remain in-
indicates that fault-bend, detachment, and, possibly, fault- tact. As conditions of deformation shift into the semi-brittle
propagation folds can all occur in the same stratigraphic and ductile regimes, deformation becomes progressively
sequence. To explain the implication that fold/thrust style is more strain-rate dependent. Consequently, the use of discrete
at least partially independent of mechanical stratigraphy, the instability surfaces in stress space may not be not appropriate
concept of competing buckling and faulting instability sur- for conditions of ductile deformation.
faces in stress space has been developed. In this approach, the
References
Boyer, S.E. & Elliott, D. 1982. Thrust systems. American Association of America Bulletin, 81, 2949-2958.
Petroleum Geologists Bulletin, 66, 1196-1230. Rich, J.L. 1934. Mechanics of low-angle overthrust faulting as illustrated by
Couples, G.D., Weir, G.M. & Jamison, W.R. 1987. Structural development Cumberland thrust block, Virginia, Kentucky, Tennessee. Bulletin of the
and hydrocarbon entrapment at Whitney Canyon and Yellow Creek American Association of Petroleum Geologists, 18, 1584-1596.
Fields, Wyoming overthrust belt. Wyoming Geological Association Rubey, W.W. 1973. Geologic map of the Afton quadrangle and part of the
Guidebook, 38th Field Conference, 275-285. Big Piney quadrangle, Lincoln and Sublette Counties, Wyoming. 1:62,500.
Dixon, J.S. 1982. Regional structural synthesis, Wyoming salient of western United States Geological Survey, Map 1-686.
overthrust belt. American Association ofPetroleum Geologists Bulletin, Schroeder, M.L. 1973. Geologic map of the Clause Peak quadrangle,
66, 1560-1580. Lincoln, Sublette and Teton Counties, Wyoming. \:24,Q00. United States
Donath, F.A. 1961. Experimental study of shear failure in anisotropic rocks. Geological Survey, Map GQ-1092.
Geological Society of America Bulletin, 76, 985-999. Serra, S. 1977. Styles of deformation in the ramp regions of overthrust faults.
Drucker, D.C. & Prager, W. 1952. Soil mechanics and plastic anaylsis for Wyoming Geological Association Guidebook, 29th FieldConference, 487-
limit design. Quarterly of Applied Mathematics, 10, 157-165. 498.
Faill, R.T. 1973. Kink-band folding. Valley and Ridge Province, Pennsylva- Suppe, J. 1983. Geometry and kinematics of fault-bend folding. American
nia. Geological Society of America Bulletin, S4, 1289-1314. Journal of Science, 283, 684-721.
Jaeger, J.C. & Cook, N.G.W. 1969. Fundamentals of rock mechanics. & Medwedeff, D.A. 1984. Fault-propagation folding. Geological So-
Chapman and Hall Ltd., London, 515p. ciety of America Annual Meeting Program with Abstracts, 16, 670.
Jamison, W.R. 1987. Geometric analysis of fold development in overthrust Timoshenko, S. 1936. Theory of elastic stability. McGraw-Hill, New York,
terranes. Journal of Structural Geology, 9, 207-219. 518.
Johnson, A.M. & Page, B.M. 1976. A theory of concentric, kink and Weir, G. 1983. Structural analysis of Whitney Canyon-Carter Creek field,
sinusoidal folding and of monoclinal flexuring of compressible, elastic Uinta and Linclon Counties, Wyoming. Amoco Production Company
multilayers. VII. Development of folds within Huasna syncline, San Luis Report WN-22-83R.
Obispo County, California. Tectonophysics, 33, 97-143. Williams, G. & Chapman, T. 1983. Strain developed in the hangingwalls of
Laubscher, H.P. 1977. Fold development in the Jura. Tectonophysics, 37,337- thrust sheets due to their slip/propagation rate: a dislocation model.
362. Journal of Structural Geology, 5, 563-571.
McGill, G.E. & Raney, J.A. 1970. Experimental study of faulting in an Willis, B. 1890. Mechanics of Appalachian structure. United States Geo-
anisotropic, inhomogeneous dolomitic limestone. Geological Society of logical Survey 13th Annual Report, pt. II, 211-281.
164
Kinematics of large-scale asymmetric buckle folds in overthrust shear:
an example from the Helvetic nappes
Mark G. Rowan* & Roy Kligfield
Abstract: Kinematic analysis of asymmetric detachment folds from the Wildhom nappe, central
Helvetics, Switzerland, supports models of buckle fold formation in overthrust shear between
nappe boundaries. The overall geometry shows a 2500 m thick multilayer of competent Jurassic
limestones deflected into super- and subjacent less competent marls and shales by buckling above
the basal decollement. Detailed investigation and measurement of the distribution, orientation,
and relative timing of mesoscale structures, primarily dilatant veins and solution cleavage, allow
documentation and quantification of the following kinematic development: (1) the nappe
boundaries originated at a low angle (ca. 10°) to undeformed bedding; (2) local stress reorientation
and buckling instabilities initiated symmetric buckle folds possibly characterized by tangential
longitudinal strain; (3) the far-field stress directions imposed by the shear zone became dominant,
and mesoscale structures developed asymmetrically on backlimbs and forelimbs; and (4) buckling
ceased when fold interlimb angles reached 90-100°, and further shearing caused the fold limbs and
axial planes to rotate into their current orientations. The total shear strain required is 2.9, equivalent
to ca. 8 km of displacement of the overlying Cretaceous with respect to the base of the nappe. It
is suggested that only an overthrust shear model can explain all the observations, and that other
models, such as gravitational sliding and buttressing by normal faults, are incompatible with the
data.
Large-scale asymmetric folds are commonly associated with variation in finite and incremental strains, and the observed
thrust faults. Several popular kinematic models have been metamorphic gradients.
advanced to explain their origins: the fault-bend fold model Simple shear has been invoked as the dominant factor in
relates fold development to movement over thrust ramps fold development in other areas as well. Sanderson (1979)
(Rich 1934; Suppe 1983), and the fault-propagation fold proposed that variations in fold geometries in southwest
model links fold amplification to growth and propagation of England, from open, upright folds to tight, asymmetrical,
blind thrust faults (Suppe & Medwedeff 1984; Suppe 1985). recumbent folds, were caused by increasing amounts of
Many folds, however, belong in neither of these categories, simple shear. Similar explanations have been applied to
and are best described as detachment folds, cored by rela- account for spatially varying fold styles from other thrust
tively incompetent lithologies, that develop due to shortening belts (Bruhn 1979; Kligfield etal. 19SI; Bosworth & Vollmer
and/or shearing of a multilayer above a basal decollement. 1981; Tanner & Macdonald 1982; Gibson & Gray 1985).
The Helvetic nappes of Switzerland offer a classic ex- While the proposed role of simple shear offers an attractive
ample of detachment folds. Some kinematic models suggest explanation for the observed deformations, it remains a
these folds formed due to either gravitational sliding (Lugeon model based solely on geometries and, occasionally, strain
1943;TrUmpy 1969,1973) or gravitational spreading (Milnes patterns; evidence for the kinematic development of a single
& Pfiffner 1977; Merle 1986, 1989). Another proposes a fold in overthrust shear has not been provided. Furthermore,
major component of pure-shear compression due to buttress- there is some question as to the mechanism of fold initiation:
ing againstpre-existing normal faults during nappe movement while some models (Ghosh 1966; Manz & Wickham 1978;
(Lemoineefa/. 1986;Gillcristefa/. 1987), and a fourth model Sanderson 1979; Gibson & Gray 1985) require only simple
is based on Rocky Mountain-style thin-skinned thrust tecton-
shear, Ramsay et al. (1983) proposed early ramp anticlines,
ics (Boyer & Elliott 1982; Butler 1985). Finally, in the model
Casey & Huggenberger (1985) invoked a pre-existing gentle
supported by the evidence presented here, the folds devel-
warping of bedding, and Gillcrist et al. (1987) suggested
oped in simple shear between thrust faults bounding the
compression due to buttressing.
nappes (Laubscher 1983; Ramsay et al. 1983; Casey &
In this paper, data are presented which document the
Huggenberger 1985; Dietrich & Casey 1989). The latter
kinematic history of large-scale asymmetric detachment folds
authors stressed that only this model is capable of explaining
from an area of the central Helvetic nappes in the Bernese
the wedge-shaped nappe geometries, the orientation and
Oberland, Switzerland (Fig. 1). It is shown that overthrust
'Present Address: Alastair Beach Associates, 11 Royal Exchange Square, Glasgow G l 3AJ, Scotland
165
M. G. ROWAN & R. KLIGFIELD
Geological setting • r
166
LARGE SCALE ASYMMETRIC BUCKLE FOLDS, HELVETIC NAPPES
Meiringen
J— Thrust fault
Other fault
Anticline
Syncline
• Sample site
Cross section
Figure 3. Simplified structural map of study area. The northwestern boundary is the Brienzer See, and the southeasteni boundary is the basal detachment
of the Wildhom Nappe. Crosses refer to Swiss national map grid coordinates.
the veins and cleavage, form the foundation for documenting (Shanley & Mahtab 1975) and field observations. Mean
the proposed role of simple shear in fold initiation, amplifi- orientations were determined by eigenvector analysis. Each
cation, and modification. set is defined relative to site bedding, and does not necessarily
Vein and cleavage orientations were measured at nineteen have the same orientation as ihe equivalent set at another site.
representative sites (Fig. 3, Table 1). The vein data for each The relative timings of the vein and solution cleavage popu-
site were divided into sets (Fig. 5) using cluster-analysis lations (Table 1) were determined by examining cross-cut-
BG fold
1-3000
2000
- 1000
*-0
Figure 4. Simplified structural cross section across center of study area (line of section shown in Fig. 3). Thick lines are faults, thin lines are fold hinge lines,
straligraphic patterns as in Figure 2. Topographic elevations in vicinity of cross section range from 700 m to 2500 m above sea level.
167
M. G. ROWAN & R. KLIGFIELD
ting relationships. Although veins oriented normal to the fold the array, and are presumably caused by spatially or tempo-
axes are common in all sites, they play no role in the kinematic rally varying stress fields (Pollard era/. 1982). Furthermore,
model and are not considered further. individual veins are demonstrably not shear veins, in that any
In a given structural domain, the number and orientations fibres are oriented perpendicular to vein walls. The vast
of the different sets with respect to bedding, and their relative majority of veins, therefore, appear to be tensile, mode-1
timing, are fairly consistent between separate sites and even fractures.
varying lithologies (Table 1). Backlimbs (Fig. 5a) are domi-
nated by coeval bedding-parallel cleavage and bedding-
Model
normal veins (set B) and a late phase of more steeply-dipping
veins (set D). The low-dip domains have an approximately Both experimental models (Ghosh 1966; Manz & Wickham
axial-planar cleavage and rare bedding-parallel (set A) and 1978) and theoretical calculations (Treagus 1973) indicate
north-dipping (set C) veins. Forelimbs (Fig. 5b) are charac- that simple shear oriented obliquely to an undeformed layer
terized by early bedding-parallel veins (set A) and bedding- embedded in less competent matrix forms symmetric buckle
perpendicular cleavage, an intermediate phase of steeply- folds (Fig. 6a,b). Whereas Treagus (1973) suggested that
dipping veins (set C), and a late phase of bedding-normal folds may become asymmetric only if competency contrasts
veins (set B) and bedding-parallel cleavage. are low. Price (1967) and Ramsay & Huber (1983, p.27)
Kinematic interpretation of veins is dependent upon their argued that asymmetric buckles form whenever the principal
origin. Although the majority of veins in a given set (mean compressive stress is inclined to bedding. Sanderson (1979),
of 83%) are nonsystematically spaced, the remainder are on the other hand, proposed that asymmetric folds form by
found in en-echelon arrays. Several factors suggest most of passive rotation of earlier-formed folds in progressive simple
these arrays do not represent shear zones (cf. Ramsay & shear (Fig. 6b,c). He described fold geometries that range
Graham 1970; Beach 1975): (1) most veins with similar from open and upright to tight and recumbent over a distance
orientations are isolated and tabular; (2) less than 1% are of less than 20 km, and attributed the differences to varying
sigmoidal; (3) of hundreds of arrays examined, only three amounts of simple shear: as shear increases, fold limbs and
showed evidence of displacement parallel to the array axial planes rotate as passive markers until, at very high shear
boundaries; (4) volume loss within the arrays is no greater strains, isoclinal folds form with both limbs and axial planes
than outside the arrays (unpubl. data; cf. Beach 1974); and (5) at a low angle to the shear direction. Asymmetric folds have
vein-to-array angles are generally very low (10-30°). Most similarly been produced by numerical modelling of symmet-
arrays are interpreted to have formed according to the model ric folds in simple shear (Skjemaa 1980; Ramsay etal. 1983;
of Olson & Pollard (1988), in which randomly spaced Casey & Huggenberger 1985), and the model has been
microfractures develop into en-echelon vein arrays during applied to explain fold geometries from several areas (Bos-
growth due to mechanical interaction around vein tips. Others worth & Vollmer 1981; Tanner & Macdonald 1982; Gibson
can be seen to merge into a primary vein oriented parallel to & Gray 1985).
Table 1. Mean orientations and relative timing of veins sets and cleavage
Domain Site Bedding A* B* C* D* PSlt PS2:i: Timing§
Backlimb Jq. NLa 054,198 069,72N (24) 043,275 b,psl/B,PSl
Jq.NLb 092,105 080,74N (2) 050,52N (9) 050,888(15) 043,178 B,c/C,P81,bA3
Jhe.NLa 080,248 068,7IN (69) 090,828 (25) 059,235 B/D
Jhe.NLb 052,118 098,74N(10) 088,828 (22) 091,03N B/D
Jhe.NLc 090,248 076,66N (20) 110,878(21) 5o - par. B,psl/P51,b/D
Jhs.NLa 076,265 080,74N(6I) 057,30N (6) 044,90(102) 5o - par. B,c/C,b,d/D
Jhs.NLb 069,265 085,67N (36) 120,48N(9) 068,88N(15) 079,375 B,c/C,b,d/D,c
Crest JhcNLFa 046,12N 048,13N(5) 024,84N(11) 030,105
Jhe.NLFb 046,12N 050,06N (2) 060,648(1)
Jhs.NLF 053,27N 038,465(1) 039,445
Hinge Jhe.NLN 023,39N 058,64N (6) 032,415(19)
Jhe.OTV 035,90 073,81N(1) 094,39N (27) 002,355 (5)
Forelimb Jq.OTa 059,468 060,458 (5) 044,72N (64) 028,685 070,36N A,PS2/C/PS1
Jq.OTb 071,458 073,438 (3) 050,788 (28) 5o - par. A/C/P51
Jq.OTc 033,598 095,43N (20) 098,82N (20) 105,10N(15) 046,535 8o • per. P52/C/P51
Jhe.OTa 050,368 059,448 (32) 060,45N (6) 039,88N (23) 066,508 A/C,b/B,P81,c
Jhe.OTb 50,525 055,588 (4) 050,36N (9) 070,73N(12) 042,558 A/C/B
Jhs.OTa 062,448 055,40N (79) 065,80N (3) 064,508
Jhs.OTb 052,558 052,558 (2) 062,35N (7) 066,73N (7) So - par. So - per. A,P82/C/B,P51
Note: axial-perpendicular veins not listed.
* A, bedding-parallel veins; B, bedding-perpendicular veins; C, flexural slip/flow veins consistent with hinge pin; D, flexural slip/flow veins with
opposite sense; number of veins measured in parentheses.
t Early or only cleavage; 8o-par. is bedding-parallel.
i Second, later cleavage; So-per. is bedding-perpendicular.
§ Timing relationships, with different phases separated by slashes (oldest on left, youngest on right). Commas separate mutually cross-cutting sets;
capital letters designate dominant or only occurrence of a set; lower-case letters designate minor occurrence of a set.
168
LARGE SCALE ASYMMETRIC BUCKLE FOLDS, HELVETIC NAPPES
Backlimb Forelimb
site Jhe.NLa site Jhe.OTa
n = 102 n = 99
Figure 5. Equal-area, lower-hemisphere plots of dilatant veins from representative sites (both in Echinoderm Member): (a) backlimb, (b) forelimb. Dots
- poles to veins, diamond - pole to site bedding, squares - poles to mean vein orientation for each set. Capital letters refer to different vein sets: A - bedding-
parallel, B - bedding-perpendicular, C - veins related to flexural slip/flow during folding, D - late flexural slip/flow veins, E - axial extension veins, n - number
of veins.
169
M. G. ROWAN & R. KLIGFIELD
170
LARGE SCALE ASYMMETRIC BUCKLE FOLDS, HELVETIC NAPPES
171
M. G. ROWAN & R. KLIGFIELD
rotation into the shear direction (Skjemaa 1980); this is dinal strain (Ramsay 1967). Small parasitic folds sometimes
consistent with field observations which indicate no such show radial dilatant veins and solution cleavage on the
rotation. Taking the thickness of this portion of the Wildhom outside and inside, respectively, of the folds. Cores of the
nappe to be 2500-3000 m, the displacement of the top of the major folds can also show more extensive fracturing, although
Jurassic multilayer with respect to the base of the nappe is this is not developed everywhere. Hinge areas of the early
about 8 km. This distributed displacement is entirely within symmetric folds are broad and poorly defined, so that the role
the Wildhom nappe, and is separate from the discrete dis- of tangential longitudinal strain in the initial buckling is
placements along the boundary thrusts. difficult to determine. It is probable, but not proven, that
initial buckling was manifested primarily by tangential lon-
gitudinal strain, and that later rotation was accompanied
Discussion principally by flexural slip and flow.
Brittle and chemical processes were important compo-
Through analysis of fold geometry, dilatant veins, and solu- nents of the Wildhom nappe deformation (see also Groshong
tion cleavage, the kinematic history of the asymmetric folds et al. 1984), and were primarily responses to the regional and
from the Wildhom nappe has been examined. It has been local stresses imposed by shearing along the nappe bounda-
demonstrated that the evidence is compatible with fold initia- ries. Vein formation accommodated both limb lengthening
tion and passive rotation in simple shear parallel to the nappe and thickening, depending on orientation, as well as axial
boundaries; furthermore, it is suggested that only this sce- extension, with total volume gains of up to 10% (unpubl.
nario can fully explain all the observed features. The kin- data). Dissolution, however, was the major factor in creating
ematic analysis corroborates earlier arguments invoking the thickness variations apparent in Figure 4. Although
overthmst shear in the Helvetic nappes (Laubscher 1983; maximum volume losses in both the Malm limestone and
Ramsay et al. 1983; Casey & Huggenberger 1985; Ramsay Echinoderm Member are less than 10%, limestones of the
1989; Dietrich & Casey 1989), and supports proposed expla- Schwarzhom Beds, where thinning is most pronounced on
nations for asymmetric fold development in other areas overturned limbs, have volume losses of up to 35% (unpubl.
(Sanderson 1979; Bruhn 1979; Kligfield et al. 1981; Bos- data). Dissolution in the interbedded marls may be even
worth & VoUmer 1981; Tanner & Macdonald 1982; Gibson greater, as marls make up 45% of the Schwarzhom Beds
& Gray 1985). thickness on backlimbs, but only 15% on forelimbs. This
It could be argued that the early symmetric folds were appears contradictory at first, as backlimbs have always been
caused not by simple shear at an angle to bedding, but instead properly oriented for limb thinning, whereas forelimbs did
by buttressing against pre-existing normal faults, as in the not reach appropriate dips until late in the folding history. It
models of Lemoine et al. (1986) and Gillcrist et al. (1987). may be that intense shear associated with flexural slip/flow in
These authors clearly demonstrated the effects of buttressing the weak marls of the overturned limbs facilitated the disso-
in the western Alps, and suggested that similar deformation lution process.
styles, modified by intense simple shear, can explain the In closing, it is important to emphasize the differences in
geometry and strain of the Morcles Nappe. Dietrich & Casey the fold kinematics documented here versus the fault-bend
(1989), however, pointed out the inadequacies of this model: and fault-propagation models which dominate the recent
first, that two of the three Helvetic nappe boundaries are literature. Widely varying stmctural styles exist in the high-
marked by clearly visible isoclinal synclines rather than by level, low-temperature portions of separate, and even indi-
normal faults; and second, that buttressing implies a signifi- vidual, fold-and-thrust belts, and different explanations are
cant pure shear component and corresponding steep axial required. For example, models appropriate to the Rocky
planes at the front of the nappe, features not characteristic of Mountains should not necessarily be applied to Alpine nap-
the Helvetic nappes. Furthermore, the present study has pes. The variable styles are caused by many factors. The most
shown that early mesoscale stmctures are not symmetric on critical include: tectonic setting (collision vs. subduction),
both limbs, as would be expected if buttressing were a presence and orientation of older faults, depth of burial and
significant factor. geothermal gradient, strain rate, and relative thicknesses and
Although the driving mechanism for fold development competencies of the sedimentary layers. It is the unique
was overthmst shear, the actual deformation processes that combination of such elements in each fold-and-thrust belt
accommodated buckling and rotation were varied. Certainly that determines the response to regional compression through
flexural slip played a major role during forelimb rotation, as folding, faulting, or more penetrative strain.
indicated by the abundant bedding-plane fibres. A compo-
nent of flexural flow is also suggested, both by the en-echelon We wish to thank R. Ratliff, J. Cosgrove, and an anonymous reviewer for
constructive comments, and K. McClay and A. Scott for their tireless efforts
veins on the forelimbs and by finite strain patterns (unpubl.
in organizing the conference and putting together this pubhcation. The
data). But folding was not by flexural mechanisms alone, as research was supported by NSF grant EAR 86-16640 to R. Kligfield.
some evidence exists for deformation by tangential longitu-
172
LARGE SCALE ASYMMETRIC BUCKLE FOLDS, HELVETIC NAPPES
References
Beach, A. 1974. A geochemical investigation of pressure solution and the Olson, J. & Pollard, D. D. 1988. Interpretation of joint sets based on fracture
formation of veins in a deformed greywacke. Contributions to Miner- mechanics (abst). Geological Society of America Abstracts with Pro-
alogy and Petrology, 46, 61-68. grams, 20, A31S-A3\9.
1975. The geometry of en-echelon vein arrays. Tectonophysics, 28,245- Pollard, D. D., Segall, P. & Delaney, P. T. 1982. Formation and interpretation
263. of dilatant echelon cracks. Geological Society of America Bulletin, 93,
Bosworth, W. & Vollmer, F. W. 1981. Structures of the Medial Ordovician 1291-1303.
Flysch of eastern New York: deformation of synorogenic deposits in an Price, N. J. 1967. The initiation and development of asymmetrical buckle
overthrust environment. Journal of Geology, 89, 551-568. folds in non-metamorphosed competent sediments. Tectonophysics, 4,
Boyer, S. & Elliott, D. 1982. Thrust systems. American Association of Pe- 173-201.
troleum Geologists Bulletin, 66, 1196-1230. Ramsay, J. G. 1967. Folding and Fracturing of Rocks. McGraw-Hill, New
Bruhn, R. L. 1979. Rock structures formed during back-arc basin deforma- York.
tion in the Andes of Tierra del Fuego. Geological Society of America 1981. Tectonics ofthe Helvetic nappes./«.McClay,K. R. & Price, N.
Bulletin, 90, 998-1012. J. (eds) Thrust and Nappe Tectonics, Geological Society of London Special
Burkhard, M. 1988. L'Helvetique de la bordure occidentale du massif de Publication, 9, 293-309.
I'Aar (evolution tectonique et metamorphique). Eclogae geologicae 1982. Rock ductility and its influence on the development of tectonic
Helvetiae, SI, (,i-ll4. structures in mountain belts. In: Hsii, K. (ed) Mountain Building Proc-
Butler, R.W.H. 1985. The restoration of thrust systems and displacement esses, 111-127. Academic Press, London.
continuity around the Mont Blanc massif, NW external Alpine thrust 1989. Fold and fault geometry in the western Helvetic nappes of
belt. Journal of Structural Geology, 1, 569-582. Switzerland and France and its implication for the evolution ofthe arc of
Casey, M. & Huggenberger, P. 1985. Numerical modelling of finite-ampli- the western Alps. In: Coward, M. P., Dietrich, D. & Park, R. G. (eds)
tude similar folds developing under general deformation histories. Journal Alpine Tectonics, Geological Society of London Special Publication, 45,
of Structural Geology, 7, 103-114. 33-45.
Dietrich, D. & Casey, M. 1989. A new tectonic model for the Helvetic , Casey, M. & Kligfield, R. 1983. Role of shear in development of the
nappes. In: Coward, M. P., Dietrich, D. & Park, R. G. (eds) Alpine Helvetic fold-thrust belt of Switzerland. Geology, 11, 439-442.
Tectonics, Geological Society of London Special Publication, 45,47-63. & Graham, R. H. 1970. Strain variation in shear belts. Canadian
Ghosh, S. K. 1966. Experimental tests of buckling folds in relation to strain Journal of Earth Sciences, 7, 786-813.
ellipsoid in simple shear deformations. Tectonophysics, 3, 169-185. & Huber, M. 1983. The Techniques ofModern Structural Geology, Vol.
Gibson, R, G. & Gray, D.R. 1985. Ductile-to-brittle transition in shear during I, Strain Analysis. Academic Press, London.
thrust sheet emplacement. Southern Appalachian thrust belt. Journal of & 1987. The Techniques of Modern Structural Geology, Vol. 2,
Structural Geology, 1, 513-525. Folds and Fractures. Academic Press, London.
Gillcrist, R., Coward, M. P. & Mugnier, J.-L. 1987. Structural inversion and Rich, J. L. 1934. Mechanics of low-angle overthrust faulting as illustrated by
its controls: examples from the Alpine foreland and the French Alps. the Cumberland thrust block, Virginia, Kentucky and Tennessee.
Geodinamica Acta, 1, 5-34. American Association of Petroleum Geologists Bulletin, 18,1584-1596.
Groshong, R. H. Jr., Pfiffner, O. A. & Pringle, L. R. 1984, Strain partitioning Sanderson, D. J. 1979. The transition from upright to recumbent folding in
in the Helvetic thrust belt of eastern Switzerland from the leading edge the Variscan fold belt of wouthwest England: a model based on the
to the internal zone. Journal of Structural Geology, 6, 5-18. kinematics of simple shediT. Journal of Structural Geology, I, 171-180.
Giinzler-Seiffert, H. 1924. Der geologische Bau der ostlichen Faulhomgru- Skjemaa, L. 1980. Rotation and deformation of randomly oriented planar
ppe im Berner Oberland. Sonderabdruck aus Eclogae geologicae and linear structures in progressive simple shear. Journal of Structural
Helvetiae, 19. Geology, 2, 101-109,
Heim, A. 1919. Das helvetische Deckengebirge. Sonderabdruck aus Ge- Shanley, R. J. & Mahtab, M. A. 1975. FRACTAN: a computer code for
ologic der Schweiz, Bd. II. Tauchnitz, Leipzig. analysis of clusters defined on the unit hemisphere. US Bureau of Mines
Kligfield, R., Carmignani, L. & Owens, W. 1981. Strain analysis of a Information Circular, 8671.
northern Apennine shear zone using deformed marble breccias. Journal Stauffer, H. 1920. Geologische Untersuchung derSchitthorngruppe im Berner
of Structural Geology, 3, 421-436. Oberland. Unpublished Ph.D thesis, Universitat Bern.
Laubscher, H. P. 1983. Detachment, shear, and compression in the central Suppe, J. 1983. Geometry and kinematics of fault bend folding. American
Alps. Geological Society of America Memoir, 158, 191-211. Journal of Science, 283, 648-721.
Lemoine, M., Bas, T., Amaud-Vanneau, A., Amaud, H., Dumont, T., Gidon, 1985. Principles of Structural Geology. Prentice-Hall, New Jersey.
M., Bourbon, M., De Graciansky, P.-C, Rudkiewicz, J.-L., Megard- & Medwedeff, D. A. 1984. Fault-propagation folding. Geological Society
Galli, J. & Tricart, P. 1986. The continental margin of the Mesozoic of America Abstracts with Programs, 16, 670.
Tethys in the western Alps. Marine and Petroleum Geology, 3,179-199. Tanner, P. W. G. & Macdonald, D.I.M. 1982. Models for the deposition and
Lugeon, M. 1943. Une nouvelle hypothese tectonique: la diverticulation. simple shear deformation of a turbidite sequence in the South Georgia
Bulletin de la Societe Vaudoise des Sciences Naturelles, 42, 301-303. portion of the southern Andes back-arc basin. Journal ofthe Geological
Manz, R. & Wickham, J. 1978. Experimental analysis of folding in simple Society. London, 139, 739-754.
shear. Tectonophysics, 44, 79-90. Treagus. S. H. 1973. Buckling stability of a viscous single-layer system,
Merle, O. 1986. Patterns of stretch trajectories and strain rates within oblique to the principal compression. Tectonophysics, 19, 271-289.
spreading-gliding nappes. Tectonophysics, 124, 211-222. Triimpy. R. 1969. Die helvetischen Decken der Ostschweiz: Versuch einer
1989. Strain models within spreading nappes. Tectonophysics, 165,57- palinspastischen Korrelation und Ansatze zu einer kinematischen Ana-
71. lyse. Eclogae geologicae Helvetiae, 62, 105-142.
Milnes, A. G. & Pfiffner, O. A. 1977. Structural development of the 1973. The timing of otogenic events in the central Alps../«.- De Jong,
Infrahelvetic complex,eastern Switzerland. EclogaegelogicaeHelvetiae, K. A. & Scholten, R. (eds) Gravity and Tectonics, Wiley, New York,
70, 83-95. 229-251.
173
Forelimb deformation in some natural examples of fault-propagation
folds
Juan Luis Alonso' & Antonio TeixelF
Abstract: An analysis of two natural examples of fault-propagation folds from the Pyrenees and
the Cantabrian zone (N Spain) is presented. These folds possess common features despite their
different geological settings, i.e. related to thrust tectonics in the Pyrenean example and to strike-
slip faulting in the Cantabrian zone. Transition from fault-bend to fault-propagation folding along
individual thrusts is related to changes in lithology. The fault-propagation folds studied are tight
and markedly asymmetric with respect to limb thickness and shape. The forelimbs are curvilinear
and display progressive thinning of beds towards the fault surface, which can be interpreted in
terms of a characteristic heterogeneous strain superimposed onto flexural shear, and restricted to
them. This strain has been factorized by means of grid construction, and is found to be a
combination of thrust-parallel heterogeneous simple shear and pure shear, attributed here to a shear
zone with diverging walls and extrusion. In contrast with previous models, it is shown that thrust-
parallel simple shear is not necessarily distributed above the entire fault surface but may instead
be restricted to regions where motion was inhibited such as tip zones or ramps.
Anticline-syncline fold pairs which accommodate fault dis- decollement located at the base of the sandstones (Larra floor
placement ahead of tip lines are commonly referred to as thrust, Teixell 1990). This thrust system is folded into
fault-propagation folds (Suppe & Medwedeff 1984). These overturned folds with local cleavage development (S^ in Fig.
folds are often broken as a result of further tip propagation. A Ic), related to the emplacement of lower basement thrusts.
kinematic model for fault-propagation folding was presented The Tejerina fault-propagation fold, located within the
by Suppe (1985), in which displacement was accommodated Cantabrian zone, is related to a strike-slip fault developed in
by kink folds developed as a result of layer-parallel shear. the northern, subvertical limb of the earlier Tejerina syncline
Other models to account for displacement variations along (Figs 3 & 4a). The stratigraphic succession involved consists
faults, and the differences in thrust-parallel shortening between of clastic rocks of Carboniferous age which unconformably
hangingwall and footwall have been both theoretically in- overlie the Esla nappe (Alonso 1985,1987). This subvertical
voked (Elliott 1976; Berger & Johnson 1980; Sanderson limb was folded into a broad, open fold in which various
1982; Williams & Chapman 1983; Coward & Potts 1983), structures were formed depending on the lithology and
and documented from nature (Geiser 1988; Hyett 1990). thickness of the beds. In the northern, inner arc of the fold,
In this paper an analysis of two natural examples of fault- where thick-bedded alternations of quartzitic and carbonate
propagation folds from non-metamorphic terranes of both the conglomerates predominate, fault-propagation folds were
Pyrenees and the Cantabrian zone of N Spain is presented. formed. By contrast, in the south, where a thin-bedded
These folds display common geometric features even though carbonate conglomerate and shale multilayer exists, only
they are from different geological settings (related to thrust smaller, conjugate kink folds and faults occur, involving less
shortening as would be expected in the outer arc of the fold
tectonics in the Pyrenees and to strike-slip faulting in the
(Fig. 3b).
Cantabrian zone). This paper shows that the deformation
occurring in the fold forelimbs can be factorized into a
flexural shear component, and a superimposed heterogene-
ous strain, both accommodating displacement variations along Geometry
the faults.
The Aragues and Tejerina structures are tight, fault-related
folds of hectometric scale. Their detailed geometry is shown
Geological setting in Figures 2 and 4. In Aragiies, the folds are inclined, with
subhorizontal axes. In Tejerina, the fold axes are vertical, and
The Pyrenean examples of fault-propagation folds presented therefore the map pattern constitutes a true fold profile. In the
belong to an imbricate fan developed in Upper Cretaceous Aragiies examples, the hangingwall anticlines, well-exposed,
sandstones and Palaeogene carbonates located just south of have been chosen for analysis. In Tejerina, the vertical
the Axial zone of the Central Pyrenees (Figs 1 & 2), in the syncline occurring in the north wall of the strike-slip fault is
Aragiies valley. The individual thrusts sole into a tighter than the south wall anticline, also better exposed, and
175
J. L. ALONSO & A. TEIXELL
A / T - K ^ ^ FRANCE
^ ^ - ^ 4 _ ^ 0 lOOKm
SPAIN
>, KEY
° [:::::| T u r b i d i t e s and m a r l s
^ j j |_jf^g5|Q[.^g5 Qp(j dolomites
=^ FOLD IN Fig.2
Figure 1. (a) and (b) Location maps for the fault-propagation folds in the Aragues valley (south central Pyrenees), (c) Simplified geological cross-section.
See (b) for location.
ARAGUES E
1
t« ^ 3
4
— - ^ -^—r^j^. QHWIO^
H = V
Figure 2. Detailed cross-sections of the same structure on both sides of the Aragues valley, showing the geometry of thrust-related folds and backthrusts
(see Fig. 1 for location). YgL- backlimb axial angle; 0 ^ ^ , hangingwall cutoff angle; Bj^, footwall cutoff angle; Key: 1, well-bedded limestones; 2, massive
limestones; 3, fine-grained limestones and dolomites; 4, dolomites; 5, quartz sandstones (Marbore formation).
176
FORELIMB DEFORMATION IN FAULT-PROPAGATION FOLDS
CANTA8RIAN /
A*ZONE II TEJERINA SYNCLINE
^ ' r //Fig. 3B
r-
Figure 3. (a) and (b). Location maps for the Tejerina fault-propagation fold, in the Cantabrian zone. Key: 1, shales; 2, calcareous conglomerates; 3, quartz
conglomerates.
was thus selected. The tighter nature of the north wall fold is fault-propagation folds of Suppe (1985).
probably related to the existence of a thick shaly packet, In Tejerina, the fold axial plane of the northern fault block
easily deformed by contact strain, immediately north of the originates from the trailing flat to ramp transition (Fig. 4). In
fault-propagation fold. contrast, in the Aragues example the axial plane originates
There are several similarities between the Aragiies and from the upper part of the ramp, near the boundary between
Tejerina fault-propagation folds. Both examples consist of the quartz sandstones and the overlying carbonates, leaving
asymmetric folds with a straight backlimb in which the bed an unfolded section of sandstones in the lower part of the
thicknesses remain unchanged, except for the hinge zones. In ramp (Fig. 2). This feature can be related to a lithologically-
contrast, the forelimbs display a characteristic curvilinear controlled upward change in slip/propagation rate, being
form, which results in markedly asymmetric axial angles lower within the sandstones than in the carbonates, where
(angles between the fold limbs and the axial plane). In fault-propagation folds develop. Angular values and other
addition, the forelimb regions show a progressive thinning parameters of the structures are presented in Figures 2 and 4,
towards the fault plane, where the beds show quasi-asymp- and in Table 1.
totic attitudes (Figs 2 & 4). This deformation took place
mainly by pressure solution mechanisms, evidenced by the
occurrence of stylolites and rough cleavage in Aragiies, and Strain factorization
pitted pebbles in Tejerina. Small fractures and veins are also
present. In order to explain the curvilinearity and thickness variations
In the Tejerina example, the fold shape changes, in the in the forelimbs of the fault-propagation folds studied, con-
direction of displacement reduction, from chevron-like sider a model in which a symmetric fold shape that would
through rounded to box forms. This transition is similar to result from the common folding mechanisms (layer-parallel
that occurring in the kink folds of the theoretical model for shear and tangential longitudinal strain) is modified by a
28°
Shales Calcareous
conglomerates
Quartz conglomerates
5 0 0 m.
Figure 4. (a). Detailed geological map constituting the profile of the Tejerina fault-propagation fold, yg^^, backlimb axial angle; 9,,^^, cutoff angle in north
wall of the fault (see Fig. 3 for location), (b) Restoration of the Tejerina structure.
177
J. L. ALONSO & A. TEIXELL
^North South
Wall Wall
TEJERINA 157° 20-90° 0° 32° 133° 0.86 170
(Top Ist.Conglom.)
Table 1. Angular values and other parameters of the fault-propagation folds analysed in this study.
deformation restricted to the forelimb. This restriction is structed hangingwall cutoff angle (Fig. 6). If an orthogonal
supported by the constant thickness and the absence of grid is drawn onto the forelimb of the symmetric fold (Figs 6a
significant strain which characterizes the straight backlimbs. & c), and if it is subsequently overlain on the actual fold shape
The procedure is used solely for convenience in the factori- with a displacement d = 1^,-1 j (axial plane displacement), a
zation analysis, and a particular deformation sequence is not deformed grid can be constructed using the distortion of
implied. bedding (Figs 6b & d). The superimposition of both the
If an ideal symmetric fold is constructed using the backlimb undeformed and deformed grids is shown in Figure 7. This
axial angles measured in the folds studied, keeping the length figure illustrates a shear zone with diverging walls and
of the axial plane constant, the forelimb obtained has the same heterogeneous strain, a model which is comparable to the
cross-sectional area as that of the actual fold (Fig. 5). This extrusion effect of Ramsay & Huber (1987, Fig. 26.35) and
suggests plane strain deformation with the intermediate strain Dietrich & Casey (1989). In the deformation zone shown
axis perpendicular to the fold profile. here, one of the diverging walls (axial plane) is translated
Furthermore, it can be observed that the cutoff lengths of parallel to itself along the fault slip direction, and the other
the individual layers in the actual fold forelimb (1, in Figs 6b wall (fault plane) constitutes a discontinuity in the displace-
& d) are shorter than the orthogonal thicknesses measured in ment field which experiences shortening. The resulting
the backlimb, implying that thrust-parallel shortening oc- extrusion accounts for the characteristic curvilinearity of the
curred in the forelimb. The along-thrust stretch which modi- forelimbs studied (Figs 6b, d & 7). The strain distribution
fied the ideal symmetric fold into the present shape is defined within the deformation zone (Fig. 7) indicates the occurrence
as of thrust-parallel stretch (pure shear) combined with hetero-
geneous simple shear to maintain strain compatibility.
s =iA (1)
where It is possible that strain extrusion models similar to the one
l„ = tysine„^ (2) proposed here also apply to true hangingwall ramps of fault-
bend folds, as a result of the higher resistance to motion in
where t^ is the original bed thickness and 6'^^ is the recon- ramp regions than in flats. In the hangingwall ramp of the
ARAGUES W TEJERINA
Figure 5. Reconstruction of symmetric folds from the backlimb axial angles measured in the Aragiies W and Tejerina structures. The actual fold profile
is represented by continuous lines, whilst the dashed lines represent the bedding traces of the reconstructed symmetric folds. In the Aragues example the
reconstruction has been made taking into account the mean thickness of the entire backlimb.
178
FORELIMB DEFORMATION IN FAULT-PROPAGATION FOLDS
ARAGUES W
TEJERINA
Figure 6. Modification of the symmetric folds into the actual fold forms, (a) and (c) Reconstructed symmetric folds, with undeformed orthogonal grids,
(b) and (d) Presently observed fold forms showing the superimposed heterogeneous deformation by means of the deformed grid, l,,, initial cutoff length;
1,, present cutoff length; d, displacement.
Eslanappe (Cantabrian zone), some incompetent stratigraphic angle can be explained by the fact that the folds in Tejerina are
units show progressive thinning towards the thrust surface, rounded, implying thus less shortening than the kink folds of
with cutoff lengths much smaller than the original orthogonal Suppe's model. The different geometries of the fault trajec-
thicknesses; by contrast, in the hangingwall flat, thrust- tories shown in Figure 4b and in the south wall ramp in Figure
parallel shortening is negligible (Alonso 1985). Hence het- 4a may be attributed to the deformation of the fault by
erogeneous simple shear, supported by a well-developed fold heterogeneous layer-parallel shear related to the develop-
train in the hangingwall ramp, would not necessarily affect ment of the south wall fold. In this southern wall, the cutoff
the whole thrust sheet but instead would occur only in the angles are now around 80°. To the north, the fault branches
frontal culmination wall. This differs from previous models into two splays, the northwestemmost splay shows a higher
in which simple shear superimposed on fault-bend folds is ramp angle than the southern part of the ramp. This can be
envisaged to affect the whole thrust sheet, i.e. both hangingwall
ramp and flat regions (Elliot 1976; Ramsay et al. 1983; Suppe
1983). In our model, simple shear is a result of heterogeneous
thrust-parallel stretch, necessary for the compatibility of
deformation.
179
J. L. ALONSO & A. TEIXELL
related to the subsequent propagation of the fault along the of fault-propagation folds has led to the following conclu-
axial plane of the anticline located in the northern part of sions:
Figure 4, after the fault-propagation fold developed, in agree- - The fault-propagation folds studied display common
ment with the model of Suppe (1985, p. 350 & Fig. 9-47) and geometric features despite differing geological settings (i.e.
also Mitra (1990). Since the fold facing is towards the south, thrusting and strike-slip faulting).
this anticline plays the same role as the syncline which - The lithology of the stratigraphic units involved is a
develops ahead of the tip of the fault in the model of Suppe major controlling factor on the slip/propagation rate, thus
(1985). governing the existence or absence of fault-propagation
In Aragues, the footwall ramp of the fault-propagation folds, as well as their geometry. Carbonate rocks appear to
fold studied is not exposed, but a restoration of the footwall offer greater resistance to fault propagation than siliciclastic
ramp of a higher thrust, removing the effect of the associated rocks.
folds, gives a value of Q^ around 30°. This angle is compa- - The fault-propagation folds studied are asymmetric and
rable to that expected from Suppe's model for an value of y characterized by a straight backlimb with constant bed thick-
equal to 40°, which is the observed in the constructed sym- ness, and a curvilinear forelimb, cutoff against the fault,
metric fold of Aragues. which shows progressive thinning of beds towards the fault.
In the examples studied, variations in displacement along - This characteristic geometry is inferred to be produced
the faults can only be observed in Tejerina, where both the by a heterogeneous deformation, which is superimposed onto
north wall and south wall cutoffs of the fault are fairly well that resulting from flexural mechanisms and is restricted to
exposed. Here, the fault-parallel shortening which only the forelimb regions. The nature of this deformation has been
occurred in the northern block as a result of the above- studied by means of the construction of orthogonal grids on
mentioned superimposed heterogeneous deformation, is re- ideal symmetric folds reconstructed from the backlimb axial
sponsible for 48 m of displacement variation (d in Fig. 6). angles. It consists of a combination of heterogeneous thrust-
Additional displacement variation may be related to the parallel simple shear and pure shear, occurring in a shear zone
differences in cutoff lengths caused by flexural shear proc- with diverging walls and extrusion.
esses as in the models of Suppe (1985) and De Paor (1987). - The folds obtained by removing the effect of the super-
Comparable geometries to those of the forelimbs of the imposed heterogeneous deformation display angular rela-
folds studied, with progressive bed thinning towards the tionships (ramp angles and axial angles) which are in good
fault, have been obtained experimentally by Mulugeta & agreement with previous theoretical models of fault-propa-
Koyi (1987) and Lui et al. (this volume). These experimental gation folds.
models also contain backfolds, whose geometry is strikingly
similar to the hangingwall box fold of Aragues E (Fig. 2a).
We appreciate the reviews by Ken McClay and two anonymous referees
which substantially helped to improve the manuscript. The comments by the
Conclusions colleagues of the Structural Geology Group at the University of Oviedo are
also acknowledged.
The geometric and kinematic analysis of two field examples
References
Alonso, J. L. 1985. Estructura y evolucion tectonoestratigrafica de la Region Hyett, A. J. 1990. Deformation around a thrust tip in Carboniferous lime-
del Manto del Esia (Zona Cantabrica, NW de Espafia). Institucion Fray stone at Tutt Head, near Swansea, South Wales. Journal of Structural
Bemardi no de Sahagun. Diputacion Provincial de Leon, 276p. Geology, 12, 47-58.
1987. Sequences ofthrusts and displacement transfer in the superposed Mitra, S. 1990. Fault-Propagation Folds: Geometry, Kinematic evolution,
duplexes of the Esla Nappe Region (Cantabrian Zone, NW Spain). and Hydrocarbon traps. American Association of Petroleum Geologists
Journal of Structural Geology, 9, 969-983. BuZ/cfm, 74, 921-945.
Berger, P. & Johnson, A. M. 1980. First-order analysis of deformation of a Mulugeta, G. & Koyi, H. 1987. Three-dimensional geometry and kinematics
thrust sheet moving over a ramp. Tectonophysics, 70, T9-T24. of experimental piggy-back thrusting. Geology, 15, 1052-1056.
Coward, M. P. & Potts, G. J. 1983. Complex strain patterns developed at the Ramsay, J. G., Casey, M. & Kligfield, R. 1983. Role of shear in development
frontal and lateral tips to shear zones and thrust zones. Journal of of the Helvetic fold-thrust belt of Switzerland. Geology, 11, 439-442.
Structural Geology, 5, 383-399. & Huber, M. I. 1987. The techniques of modern Structural Geology.
De Paor, D. G. 1987. Stretch in shear zones: implications for section Volume 2: Folds and Fractures. Academic ftess, London, 309-700.
balancing. Journal of Structural Geology, 9, 893-895. Sanderson, D. J. 1982. Models of strain variation in nappes and thrust sheets:
Dietrich, D. & Casey, M. 1989. A new tectonic model for the Helvetic a review. Tectonophysics, 88, 201-233.
nappes. In: Coward, M. P., Dietrich, D. & Park. R. G. (eds), Alpine Suppe, J. 1985. Principles ofStructural Geology. Prentice-Hall, New Jersey.
Tectonics. Geological Society of London Special Publication, 45,47-63. & Medwedeff, D.A.I 984. Fault-propagation folding. Geological Society
Elliot, D. 1976. The energy balance and deformation mechanisms of thrust of America Abstracts with Programs, 16, 670.
sheets. Philosophical Transactions ofthe Royal Society ofLondon, A283, Teixell, A. 1990. Alpine thrusts at the western termination of the Pyrenean
289-312. Axial Zone. Bulletin de la Societe Geologique de France, 8 (VI), 241-
Geiser, P. A. 1988. Mechanisms of thrust propagation: some examples and 249.
implications for the analysis of overthrust terranes. Journal ofStructural Williams, G. & Chapman, T. 1983. Strains developed in the hangingwalls of
Geology, 10, 829-845. thrusts due to their slip/propagation rate: a dislocation mode\. Journal of
Structural Geology, 5, 563-571.
180
The geometric evolution of foreland thrust systems
M. P. Fischer' & N. B. Woodward^
Abstract: Thrust system models which consider the observed variation in structural geometry
from minimum displacement regions to maximum displacement regions to be equivalent to the
temporal variation in structural geometry for any cross section through the thrust sheet are
incompatible with observations of lateral variations in ramp angles and fold geometry in well-
exposed thrust systems in Wyoming and Tennessee. The self-similar model of thrust system
evolution assumes ductile bead strain is always accommodated by tip-line folding, and suggests
that variations in structural geometry observed anywhere in the thrust system are the product of
the continued deformation of currently observed lateral tip structures. These variations, however,
are interpreted to reflect the influence of the intrinsic physical properties of the thrust system on
thrust propagation. Rather than representing the progressive deformation of tip folds, the structural
geometty of many thrust systems is most likely indicative of spatial and/or temporal variations in
these intrinsic properties.
Since direct observation of changes in the geometry of an (i.e. developing thrust system geometry). Likewise, since
evolving thrust system is not possible, assumptions must be maximum total thrust displacement is suggested to occur
made when reconstructing thrust system kinematics. The beneath the internal portions of thrust sheets (Elliott 1976),
assumption that presently observed spatial variations in de- structures in these regions should represent the final stages of
formation (i.e. thrust system geometry) are equivalent to deformation (i.e. thrust system geometry). It follows from the
inferred temporal variations in deformation (i.e. geometry) is self-similar model that many of the fundamental geometric
widely employed in geological modelling (see Means 1976, characteristics of a thrust system are determined at the earliest
p. 27-30 for a discussion of this technique; Dickenson & stages of thrusting, and that some record of this early defor-
Snyder 1979; Delaney etal. 1986; Wojtal 1986) and has been mation (i.e. geometry) should be observed in older parts of the
called an 'Eulerian' approach (Wojtal & Mitra 1988). This system (e.g. 'background strain' of Williams & Chapman
approach can yield perfectly viable kinematic models provided 1983; overprinting of deformation described by Wojtal 1986;
there is sound evidence of sequential overprinting of defor- Mitra 1987; Blenkinsop & Drury 1988; Wojtal & Mitra
mation (Wojtal & Mitra 1988). Because such overprinting 1988). To examine the factors which might influence initial
relationships are often difficult to discern in macroscopic thrust geometry, it is therefore logical to study the deforma-
studies of the evolution of thrust systems, however, this paper tion (i.e. structural geometry) at lateral thrust tip lines.
discusses an approach wherein the deformation path (i.e. Faults are commonly modelled as individual large-scale
temporal geometric evolution) of individual portions of the fractures (Segall & Pollard 1980; Aydin & Nur 1982; Aydin
thrust system is inferred based on an understanding of how 1988; Rybicki 1989), which in turn are equated with displace-
the local physical properties of system constituents affect ment discontinuities or dislocations (Bilby & Eshelby 1968).
fault kinematics.
Thrust fault surfaces are often conceptualized as Somigliana
dislocations (Eshelby 1973) where displacement is zero at the
edges of the dislocation (i.e. thrust tip lines; Elliott 1976;
Self-similar modelling techniques Williams & Chapman 1983; Chapman & Williams 1985;
Pfiffner 1985; Price 1988). Elliott (1976) recognized the
A commonly used, but rarely explicitly outlined technique strain inherent at the edges of such dislocations and suggested
(hereafter called a self-similar model) used to describe the this strain was accommodated by a 'ductile bead' of defor-
geometric and kinematic evolution of thrust systems is based mation propagating just ahead of the moving thrust tip line in
on the observation that over geological time spans the dis- much the same way deformation in a crack tip process zone
placement at any particular location on a single thrust surface accommodates the strain developed at fracture terminations
is proportional to the duration of thrusting in that region (Broek 1978).
(Elliott 1976). Hence, since total thrust displacement (i.e. The consistent observation of tip folds (asymmetric, blind
displacement of the lowest hangingwall cutoff) goes to zero fold complexes, e.g. Elliott 1976; Boyer & Elliott 1982) at
at lateral thrust tip lines, the structures observed in these thrust terminations (Buxtorf 1916; Spang & Brown 1981;
regions should represent the earliest stages of deformation Thompson 1981; Berger & Johnson 1982) has led to the
181
M. P. FISCHER & N . B . WOODWARD
Figure 1. Conceptual model ofan ideal self-similarthrust sheet. Portion of thrust fault {lined surface) between teeth is exposed at surface. Deformed strata
{shaded layers) have been reconstructed above an ideal horizontal erosion surface.
hypothesis that thrust faults normally die out into these Case studies
structures (Elliott 1976; Laubscher 1976; Williams 1980;
Williams & Chapman 1983; Boyer 1986), and implies that Since along-strike changes in the geometry of a self-similarly
deformation within the ductile bead is consistently mani- propagating thrust sheet represent the geometry of the system
fested as folding in front of a thrust tip. Figure 1 is a at different times during the kinematic evolution, the practi-
conceptual model of a simple thrust system consisting of a cal applicability of the self-similar model can be tested using
single thrust sheet that evolved according to the self-similar relatively simple thrust systems in which the three-dimen-
model. In this system a thrust fault deforms strata during sional geometry and/or kinematics is well understood.
forward and lateral propagation, and dies out laterally into tip
folds. Along strike towards the centre of the thrust sheet these
Baldy thrust system
anticlines are truncated and increasingly offset, preserving a
syncline in the footwall and an asymmetric, fold-first (e.g. The Baldy thrust system is located in the northern portion of
Dahlstrom 1970) anticline in the hanging wall. the Idaho-Wyoming thrust belt in the Snake River Range and
Figure 2. (a) Maps depicting the principal faults of the Idaho-Wyoming thrust belt and the Snake River Range {shaded box). Baldy thrust system is shaded
on large-scale map. A= Absaroka thrust, C= Crawford thrust, D= Darby thrust, JP= Jackson-Prospect thrust, M= Meade thrust, P - Paris thrust, T= Tunp
thrust (Modified from Woodward, 1986).
182
G E O M E T R I C EVOLUTION O F FORELAND THRUST SYSTEMS
Figure 3. Map of the surface trace of the Baldy thrust combined with a hangingwall cutoff map and structure contour drawn on the thmst surface. Contours
are in metres. Schematic cross sections A—A" through C—C" are intended to show generalized hanging wall and footwall geometries only. Sections are
not to scale, but relative thicknesses of units are correct. Note the decrease in thrust displacement from section A—A' to section C-—C', the increase in fault
dip towards the southeast and the along-strike variation of thrust geometry in Devonian-Mississippian strata (black layer). (Modifiedfrom Woodward, 1986).
183
M. P. FISCHER & N . B . WOODWARD • Vj ,;i.
36*30'
36" 15'
36'00'
35°45'
84°00' 83 •45'
Figure 4. Mapsdepictingprincipalfaultsof the central east Tennessee Valley and Ridge province. Symbols: BVT= Beaver Valley thrust, CCT=Copper
Creek thrust, DCT= Dunn Creek thrust, DVZ= Dumplin Valley fault zone, GCT^ Guess Creek thrust, GST= Great Smoky thrust, HVT= Hunter Valley
thrust, KVT= Knoxville thrust, MST^ Mill Spring thrust, P= Pulaski thrust, S VT= Saltville thrust, TKT^ Town Knobs thrust. WVT= Wallen Valley thrust.
Rocky Valley thrust system is shaded on larger-scale map. (Modified from Woodward et. al 1988). '"" ' '"
section represents an earlier stage of deformation (relative tonear 100°, but may be as small as 50°. The Rocky Valley
the other sections), and that all 'later' (i.e. more internal) anticline, well developed throughout the thrust sheet, and the
geometries are the product of progressive deformation of a leading-edge anticline, well developed only in the centre of
structure like that depicted in section C-C. In this case, an the sheet, merge northwestward to form one fold. The Blue
early-formed footwall ramp through Devonian-Mississip- Spring Creek anticline is interpreted as the product of later,
pian and Mississippian strata (sections C-C and B-B') must local out-of-sequence thrusting which deforms the south-
change cutoff angle through time and eventually be trans- western portion of the Rocky Valley thrust sheet (Fischer
formed into a footwall flat (section A-A'). Similar aged 1989).
footwall strata must be folded and unfolded during evolution The Rocky Valley anticline overlies the crest of a footwall
of the thrust system. ramp through Cambrian-Ordovician strata of the Conasauga
and Knox Groups, and exhibits varying geometry both along
strike and at different stratigraphic levels (Figs 5c & 6). In the
Rocky Valley thrust system
southwestern portion of the field area the fold is chevron-
The Rocky Valley thrust system is located in the central east shaped and exposed in Knox and uppermost Conasauga
Tennessee Valley and Ridge between the Saitville thrust and Group strata (Fig. 6a). In the centre of the study area the
Dumplin Valley fault zone (Fig. 4). The Rocky Valley thrust anticline is only exposed in Conasauga Group strata and is
transports Cambrian-Ordovician strata northwestward over geometrically similar to a fault-bend fold (Suppe 1983; Fig.
strata as young as lower Middle Ordovician, extends for 6b). In the northeast the fold is exposed throughout nearly the
roughly 50 km along-strike, and dies out within a few kilome- entire stratigraphic section, but exhibits a transition in geom-
tres of the study area (Figs 4, 5a & b). Displacement on the etry near the Knox-Conasauga Group contact. Conasauga
fault as calculated from balanced cross sections varies from strata maintain a fault-bend fold geometry while Knox Group
3.4 km in the southwest, to 6.6 km in the centre, to 5.3 km in strata are folded into a single-hinged chevron anticline. A
the northeast. The thrust sheet is deformed into leading-edge, roughly 150 m thick shale near the top of the Conasauga
intrasheet (Blue Spring Creek), and trailing-edge (Rocky Group serves as a detachment between the fold styles. Geo-
Valley) anticlines (non-genetic use of Boyer 1986 terminology; metric and kinematic modelling by Fischer (1989) suggests
Fig. 5c). The Rocky Valley anticline is generally an upright, this stratigraphic variation in fold form is most likely the
asymmetric, north west-verging fold in which the forelimb is result of differing fault-fold relations in each section of the
much shorter than the backlimb (terminology of Dahlstrom, stratigraphic column such that Knox strata deform by a fold-
1970). Interlimb angles of this fold vary from 160° in the first (Dahlstrom 1970), fault-propagation fold (Suppe 1985)
northeast to 80° in the southwest, where the anticline is mechanism, while Conasauga strata concurrently deform by
locally overturned. The leading-edge fold is generally an a fauh-first {Dahlstrom 1970), fault-bend fold mechanism.
asymmetric, north west-verging fold in which the forelimb is This model explains the transition in exposed fold geometry
overturned to the south. Interlimb angles of this fold average between the central and southwest portions of the study area
184
GEOMETRIC EVOLUTION OF FORELAND THRUST SYSTEMS
'.'^\
^ W Chickamauga Group
r (430 m
nj
o
•>
n
T)
O Knox Group
(890 m)
(b)
r
cd
o
(- Conasauga Group
E'
(fl (700 m
U
Rome Fm.
(basal detachment)
:•('••
•'O!
Blue Spring Creek ~1.5
anticline
Rocky Valley
anticline
' • •'• - y ' •• :• . . : • , . : • _ : \ ' • • ' H i • • ; - , _ ... ' . ; : • ; - -.^ . i.,(= • -. '"T . ' • ! •
" - •• .' • .' - T • •'< • ' : ' b r v • ' V f : . M i • ^ ' . - ' , - : . ' . • • • . ; I ; ,• •••
• : • " • . . • " ' • ' • . . . y ' . ^ . . - , • • • : . ' • • ' • . ; \ ^ ' J • . ' , • • . ; ; . ! ; , • ' • , • • . - . ' ^ , . : , ,
Figure 5. (a) Generalized geological map of the Rocky Valley thrust syslem in the vicinily of New Market, Tennessee. Rocky Valley thrust shown with
teeth on hangingwall. Other faults in the area are moderate to low angle reverse faults of lesserdisplacement. (b)Stratigraphic column depicting units exposed
in the Rocky Valley thrust system. Thicknesses of unils are averages. Standard llthological symbols are used. Note that the enlire Chitkamauga Group
is not exposed in the field area, and that the Rome Formation is never exposed in the field area. Groups correspond roughly to incompeleni (Conasauga and
Chickamauga Groups) and competent (Knox Group) structural lithic units (Woodward & Rutherford. 1989). (c)Block diagram ofihe Rocky Valley anticline
showing along-strike variations in fold geometry. Vertical scale is in kilometres below sea level (SL). Horizontal scale along section lines equal to vertical
scale. Distances between seclions are not to scale. Black layer is lowermost massive dolostone in Knox Group, stippled layer is thick limestone unit in
Conasauga Group (Fig. 5b). See Figure 5a for locations of section lines. MS- Mill Spring thrust, RVT- Rocky Valley thrust, SD- Saltville detachment.
185
M. P. FISCHER & N . B . WOODWARD ,--i,">
186
GEOMETRIC EVOLUTION OF FORELAND THRUST SYSTEMS
folding precedes thrusting. This paper points out, however, (Coward & Potts 1983). Fold-first geometries should de-
that although tip folds are often presumed to be the most velop when significant plastic deformation occurs before
common physical manifestation of ductile bead strain (Elliott failure, whereas fault-first geometries develop in regions
1976; Williams 1980; Williams & Chapman 1983; Boyer where failure occurs before significant plastic deformation
1986), other types of deformation may also accommodate accumulates. Williams & Chapman (1983) quantitatively
this strain (e.g. distributed internal strain, localized shear described this relation in terms of a fault slip versus propaga-
zones; Elliott 1976; Chapman & Williams 1985; vertical tion ratio (S/P). If a thrust sheet is transported a long distance
strike-slip faulting and fracturing; King & Yielding 1984). over a fault which has propagated a much shorter relative
The observation that tip folds are not the only structures distance, S/P is large and the theoretical strain gradient at the
which form to accommodate ductile bead strain suggests a fault tip is similarly quite large. Areas with large S/P values
ductile bead may not propagate self-similarly, and that many are more likely to form fold-first geometries because more
thrust system geometries may not be explained by the self- strain can be accommodated by plastic deformation.
similar model. If a ductile bead does not propagate in self- In the Rocky Valley thrust system, although different
similar fashion, structures currently preserved at lateral thrust structural geometries are observed throughout the system,
terminations only represent the ephemeral last stage of thrust- different structural lithic units (Woodward & Rutherford
ing (Knipe 1989; Woodward 1990), and should have no 1989) exhibit characteristic fold geometries in various loca-
kinematic relation to structures in the more internal parts of tions in the thrust sheet. Macroscopic folds in competent
the system. structural lithic units consistently exhibit fold-firstgeometries
A structural geometry observed anywhere in a thrust while folds in incompetent units exhibit fault-first geometries
system is most likely the product of the kinematics of thrust- (e.g. Coward & Potts 1983; Boyer 1986). These observations
ing in that portion of the system. Thrust kinematics are in turn suggest competent structural lithic units in the area may cause
controlled by the extrinsic (e.g. temperature, pressure) and a locally increased S/P ratio, and thereby inhibit fault propa-
intrinsic (e.g. pore fluid pressure, stratigraphy, structural gation while intensifying deformation around the thrust tip.
position) characteristics of the system. In most foreland In contrast, the S/P ratio in incompetent units may be small,
thrust systems, where extrinsic properties do not vary signifi- such that the thrust tip advances relatively rapidly, and there
cantly within the system, stratigraphy is generally accepted as is little time for strain to accumulate at the fault tip. Other
the controlling influence on macroscopic thrust geometry field studies (Morley 1987) have similarly shown an in-
(Rich 1934). Furthermore, thrust sheet geometry is largely creased tendency for thrusts to die out in competent sections
controlled by cutoff angle, displacement and overall thrust of a stratigraphic column, and further suggest these strata may
geometry (Suppe 1983; 1985; Jamison 1987), and is therefore cause a locally increased S/P ratio. This relationship inferred
also largely a function of stratigraphy. However, while from field studies is also in agreement with the resuhs of
numerous researchers have noted the influence of stratigraphy numerous theoretical, quantitative studies of the influence of
on both thrust and thrust sheet geometry (Berger & Johnson material competency on fracture propagation.
1982; Tippet era/. 1985; Woodward 1990),exactly how this In a situation analogous to the local interaction between a
influence is exerted in determining which of the two end thrust fault tip and strata of different mechanical properties,
member thrust sheet geometries (i.e. fault- and fold-first) Dundurs & Mura (1964) and Dundurs (1967) reported that
develops in a particular area is unknown. both screw and edge dislocations were 'repelled' by inclu-
In a simplistic sense, the relative rate of brittle versus sions of a higher rigidity modulus than the medium which
plastic deformation around a thrust tip is the limiting factor in contains the dislocation. Rybicki (1989) examined the in-
determining which of the two geometric forms develops teraction of two colinear cracks with an intervening high
Figure 7. Hangingwall cutoff map for the Rocky Valley thrust. Dots show locations of 218 shallow drill holes used to constrain the map. Cutoff widths
increase from southwest to northeasi by an average of 200%. Shaded pattern shows the cutoff for the Maynardville Limestone portion (uppermost formation)
of the Conasauga Group. Cutoffs; a- Rome Fm.-Conasauga Group, b= Conasauga Group-Knox Group.
187
M. P. FISCHER & N. B. WOODWARD
rigidity inclusion and showed that the mere presence of the in the overall structural style of a thrust sheet, the spatio-
inclusion inhibited the fractures from propagating toward one temporal variation in the style and geometry of ductile bead
another. In addition, the stress induced in the inclusion by the structures suggests there is little value in explaining the
cracks is higher than the corresponding stress in a homogene- geometry of a thrust sheet by the progressive deformation of
ous medium, and this induced stress increases with increasing currently observed lateral tip structures. Consequently, this
rigidity of the inclusion. The fracture mechanics study of paper questions the usefulness of the self-similar model, and
Lemiszki & Landes (1989) reported that rocks of higher to a degree, the 'Eulerian' approach upon which it is based,
competency will shorten (by folding) significantly before a as an untested general assumption which has not been ad-
mode III fracture will propagate through the hinge of the fold, equately documented. At one scale of frontier investigation
while in less competent strata, the fracture will propagate the model was a useful advance over all previous (generally
before significant fold shortening occurs. These studies non-viable, inadmissible and unbalanceable) approaches.
further suggest competent strata may cause a locally in- However, as more data become available, modem investiga-
creased S/P ratio and an increased tendency to form fold-first tions of thrust kinematics cannot accept a priori the self-
geometries. similar approach without much greater scrutiny.
Although the spatial variation in stratigraphy appears to Thrust system geometry is most likely determined by the
have been the dominant influence on the geometric evolution kinematics of thrusting which occurred throughout the re-
of the Rocky Valley thrust system, other intrinsic properties gion. The geometry of the Rocky Valley thrust system is
certainly played a role. Because the self-similar model explained by a stratigraphic variation in the mode of thrust
pre-determines many fundamental characteristics of thrust propagation and a lateral variation in cutoff angle. Thrust
system geometry and ignores any relation between the intrin- kinematics in this system appear to have been most strongly
sic properties of the system and thrust kinematics, interpreta- influenced by local stratigraphic (facies) and structural
tions based on the model are severely limited. Instead of the (competency) variations. Spatial variations in thrust system
self-similar model, we believe it is more beneficial to examine geometry do not necessarily reflect the evolution of the
the variation in the intrinsic properties of a thrust system and system, but may instead reflect a spatial or spatio-temporal
how these variations affect thrust kinematics. variation in the intrinsic properties of the system. It is this
variation of intrinsic properties which ultimately determines
thrust kinematics of thrusting and hence the geometric evo-
Conclusions lution of thrust systems.
References
Aydin, A. 1988. Discontinuities along thrustfaults and thecleavage duplexes. Blenkinsop, T. G. & Drury, M. R. 1988. Stress estimates and fault history
In: Mitra, G.&Wojtal,S.{sds)Geometriesand mechanisms of thrusting, from quartz microstructures.yourna/o/S/rMcmra/Geo/o^j, 10,673-684.
with special reference to the Appalachians. Geological Society of Boyer, S. E. 1986. Styles of folding within thrust sheets: examples from the
America Special Paper, 222, 223-232. Appalachian and Rocky Mountains of the USA and Canada. Journal of
& Nur., A. 1982. Evolution of pull-apart basins and their scale Structural Geology, 3/4, 325-339.
independence. Tectonics, 1, 91-105. & Elliott, D. \9i2.'Y\\ruslSysl£ms,. American Association of Petroleum
Bartholomew, M. J. 1987. Structural evolution of the Pulaski thrust system, Geologists Bulletin, 86, 1196-1230.
southwestern Virginia. Geological Society ofAmerica Bulletin, 99,491 - Broek, D. 1978. Elementary Engineering Fracture Mechanics. Alphen ann
510. den Rijn, The Netherlands, Sijthoff and Noordhoff International Pub-
Berger, P. & Johnson, A. M. 1982. Folding of passive layers and forms of lishers, 437p.
minor structures near terminations of blind thrust faults - application to Buxtorf, A. 1916. Prognosen und Befunde beim Hauensteinbasis und
the central Appalachian blind thrust./oMr«a/o/Sfrac(«ra/G£o/og>', 4,343- Grenchenberg tunnel und die Bedeutung der letzteren fUr die Geologic
354. des Juragebirges. Verhandlungen des Naturforschunden die Gesellschaft
Bilby, B. A, & Eshelby, J. D. 1968. Dislocations and the theory of fracture. Basel, 27, 185-254.
In: Leibowitz, H. (ed) Fracture, an Advanced Treatise, 1, 99-182.
188
GEOMETRIC EVOLUTION OF FORELAND THRUST SYSTEMS
Chapman, T. J. & Williams, G. D. 1985. Strains developed in the hanging Mitra, S. 1988. Three dimensional geometry and kinematic evolution of the
walls of thrusts due to their slip-propagation rate: a dislocation model; Pine Mountain thrust system, southern Appalachians. Geological Soci-
Reply. Journal of Structural Geology, 7, 759-762. ety of America Bulletin, 100, 72-95.
Coward, M. P. & Potts, G. J. 1983. Complex strain patterns developed at the 1987. Regional variations in deformation mechanisms and structural
frontal and lateral tips to shear zones and thrust zones. Journal of styles in the central Appalachian orogenic belt. Geological Society of
Structural Geology, 5, 383-399. America Bulletin, 98, 569-590.
Dahlstrom, C.D.A. 1970. Structural geology in the eastern margin of the Morley, C. K. 1987. Lateral and vertical changes of deformation style in the
Canadian Rocky Mountains. Bulletin of Canadian Petroleum Geology, Osen-R0a thrust sheet, Oslo region./oMrna/o/Sfrucfura/Geo/ogj', 9,331-
18, 332-406. 343.
Delaney, P.T., Pollard, D. D., Ziony, J. T., & McKee, E. H. 1986. Field Pfiffner, O. A. 1985. Displacements along thrust faults. Eclogae Geologae
relations between dikes and joints; Emplacement processes and paleostress Helvetiae,78,313-333.
analysis. Journal of Geophysical Research, 91, 4,920-4,938. Price, R. A. 1988.The mechanical paradox of large overthrusts. Geological
Dickenson, W. R. & Snyder, W. S. 1979. Geometry of triple junctions related Society of America Bulletin, 100, 1098-1908.
to San Andreas transform. yoMrna/o/Gcop/iyiica/Zfeiearcft, 84,561-572. Rich, J. L. 1934. Mechanics of low-angle overthrust faulting illustrated by
Diegel, F. A. 1986. Topological constraints on imbricate thrust networks, Cumberland thrust block, Virginia, Kentucky and Tennessee. American
examples from the Mountain City window, Tennessee, USA. Journal of Association of Petroleum Geologists Bulletin, 18, 1584-1596.
Structural Geology, 8, 269-280. Rodgers, J. 1953. Geologic map of east Tennessee with explanatory text.
Dundurs, J. 1967. On the interaction of a screw dislocation with Tennessee Department ofConservation, Division ofGeology Bulletin, 58,
inhomogeneities. In: Eringen, A.C. (ed) Recent Advances in Engineer- parts I and II, I68p.
ing Sciences, 2, 223-233. Rybicki, K. R. 1989. On fault slip zones in the external shear-field separated
& Mura, T. 1964. Interaction between an edge dislocation and a circular by a high rigidity asperity (barrier) - inplane and antiplane strain models.
inclusion. Journal of the Mechanics and Physics of Solids, 12,177-189. Geophysical Journal, 96, 101-115.
Elliott, D. 1976. The energy balance and deformation mechanisms of thrust Segall, P. & Pollard, D. D. 1980. Mechanics of discontinuous faults, yowr/ia/
sheets. Philosophical Transactions of the Royal Society of London, 283, of Geophysical Research, 85, 4,337-4,350.
289-312. Spang, J. H. & Brown, S. P. 1981. Dynamic analysis of a small imbricate
Eshelby, J. D. 1973. Dislocation theory for geophysical applications. thrust and related structures. Front Ranges, southern Canadian Rocky
PhilosophicalTransactionsofthe Royal Society ofLondon,274,3'i\-338. Mountains. In: McClay, K. R. & Price, N. J. (eds) Thrust and Nappe
Evans, M. 1989. The structural geometry and evolution of foreland thrust Tectonics. Geological Society of London Special Publication, 9, 143-
sysl£n\s,northeTn'ViTgmia.Geological Society of America Bulletin, 101, 149.
339-354. Suppe, J. 1985. Principles of Structural Geology: Englewood Cliffs, New
Fischer, M. P. 1989. Structural geometry of the Rocky Valley thrust system Jersey, Prentice-Hall Inc., 537p.
in the vicinity of New Market, Tennessee (MS thesis); Knoxville, 1983. Geometry and kinematics of fault-bend folding. American
University of Tennessee, 345p. Journal of Science, 283, 684-721.
Harris, L. D. 1970. Details of thin-skinned tectonics in parts of Valley and Tippett, C. R., Jones, P. B. & Frey, F. R. 1985. Strains developed in the
Ridge and Cumberland Plateau provinces of the southern Appalachians. hangingwalls of thrusts due to their slip/propagation rate: A dislocation
In: Fisher, G. W., Pettijohn, F. J., Reed, J. C , Jr. & Weaver, K. N. (eds) model: Discussion. Journal of Structural Geology, 7, 755-758.
Studies of Appalachian Geology: Central and Southern. New York, 161- Thompson, R. I. 1981. The nature and significance of large 'blind' thrusts
173. within the northern Rocky Mountains of Canada. In: McClay, K. R. &
Jacobeen, F. & Kanes, W. H. 1975. Structure of Broadtop synclinorium. Price, N. J. (eds) Thrust and Nappe Tectonics. Geological Society of
Wills Mountain anticlinorium, and Alleghany frontal zone. American London Special Publication, 9, 449-462.
Association of Petroleum Geologists Bulletin, 59, 1136-1150. Williams, G. & Chapman, T. 1983. Strains developed in the hanging walls
& 1974. Structure of Broadtop synclinorium and its implications of thrusts due to their slip-propagation rate: A dislocation model.Journal
for Appalachian structural style. American Association of Petroleum of Structural Geology, 5, 563-571.
Geologists Bulletin, 58, 362-375. Williams, R. E. 1980. Creation and geometry of the thrust fracture. Geological
Jamison, W. R. 1987. Geometric analysis of fold development in overthrust Society of America Abstracts with programs, 12, 549.
terranes. Journal of Structural Geology, 9, 207-219. Wojtal, S. 1986. Deformation within foreland thrust sheets by populations of
King, G.C.P. & Yielding, G. 1984. The evolution of a thrust fault system: minor faults. Journal of Structural Geology, 8, 341-360.
Processes of rupture initiation, propagation and termination in the 1980 & Mitra, G. 1988. Nature of deformation in some fault rocks from
El Asnam (Algeria) earthquake. Geophysical Journal of the Royal As- Appalachian thrusts. In: Mitra, G. & Wojtal S. (eds) Geometries and
tronomical Society, 77,915-933. mechanisms of thrusting, with special reference to the Appalachians.
Knipe, R. J. 1989. Deformation mechanisms - recognition from natural Geological Society of America Special Paper, 222, 17-34.
lectonites. Journal of Structural Geology, 11, 127-146. Woodward, N.B.I 990. Deformation styles and geometric evolution of some
Kulander, B. R. & Dean, S. L. 1986. Structure and tectonics of the central and Idaho-Wyoming thrust belt structures. David Elliott Memorial Volume,
southern Appalachian Valley and Ridge and Plateau provinces, West The Johns Hopkins University, (in press).
Virginia and Virginia. American Association of Petroleum Geologists 1986. Thrust fault geometry of the Snake River Range, Idaho and
Bulletin, 70, 1674-1684. Wyoming. Geological Society of America Bulletin, 97, 178-193.
Laubscher, H. P. 1976. Fold development in the Jura. Tectonophysics, 37,337- & Rutherford, E. Jr. 1989. Structural lithic units in extemal orogenic
362. zones. Tectonophysics, 158, 247-267.
Lemiszki, P. J. & Landes, J. D. 1989. Mixed-mode fracture propagation , Walker, K. R. & Lutz, C. T. 1988. Relationships between eariy
induced by folding ahead of a fault: Analogy to fault-propagation folds. Paleozoic facies patterns and structural trends in the Saltville thrust
Geological Society of America Abstracts with programs, 21, 136. family, Tennessee Valley and Ridge, southern Appalachians. Geologi-
Means, W.D. 1976. Stress and Strain. New York, Springer-Verlag, 339p. cal Society of America Bulletin, 100, 1758,1769.
189
Some geometric problems of ramp-flat thrust models
John G. Ramsay
Abstract: Various classic models used to relate fold geometry to thrust fault geometry are
critically discussed and it is shown how modifications might be made which could provide better
fits to geological data. A region on the South coast of Dorset, UK, where well developed small
scale thrust faults are developed, is described and the geometric features of the fold and fault
geometry related to the previously proposed modified models.
The Appalachian 'Rocky Mountain' thrust model, originally least needs some justification in the particular geological
described by Rich (1934) and refined to a high level of region under investigation. It does seem improbable that rock
sophistication by many subsequent workers (Boyer & Elliott Theology, which is generally considered to control the initial
1982; Suppe 1983), seems to have become a ruling guide to locations of ramps (where thrusts cut across more competent
much recent section construction in orogenic zones. In the members of the succession) or flats (generally sub-parallel to
author's opinion these fault bend- and fault propagation- the incompetent layers), should play no further role in con-
models are being applied rather indiscriminately, and to trolling the subsequent development of the overall structure.
situations where they are geologically inappropriate. Thus, In the classic region of the Helvetic nappes of Switzerland
published sections of highly water impregnated sediments at rock lithologies do have a considerable influence of the
the toe of an accretionary wedge (Davis et al. 1983; Knipe & development of the structures in the nappes and fold develop
Needham 1986) are often geometrically indistinguishable which are not just the result of passive layer bending above a
from those of foreland thrust belts developed in lithified rocks variably inclined thrust surface. Because thrust surfaces cut
of varying competence (Price 1981), and almost identical upwards across the stratigraphy in the overthrust sense, shear
reconstructions have been applied to highly deformed, folded distortions in the transported nappe block inevitably lead to
and regionally metamorphosed rocks from the central parts of layer parallel shortening (Ramsay etal. 1983; Ramsay & Huber
orogens and to crystalline basement deformed at deep crustal 1987, p. "ill-2)1%). This shortening sets up buckle folds in the
levels (Coward 1983; Butler 1983). more competent layers with initial wavelengths controlled by
Although today there is available an enormous amount of layer thickness and competence contrast. The amplitudes of
excellent seismic data from fault zones and this provides very these folds depends upon the extent of the internal shear and
sound guide lines to certain aspects of thrust and fold geometry, also upon the inclination of the bedding surfaces to the thrust
perhaps there is a tendency to fit these data too readily to sheet boundaries (Fig. 1).
previously selected geometric models. The fold geometry of In the classic model of fault bend folding the folds are only
the hangingwalls of thrust faults is usually clear and relatively developed in the hangingwall of the structure (Fig. 2). In
easy to interpret, the data from footwall sectors is often much natural examples of such structures which can be observed in
less clear and should be analysed with circumspection. In the surface exposures folds may also be found in the footwall.
sections below a number of modified kinematic models for One interpretation of these footwall folds is that they result
the interrelationships of fold and fault geometry will be from the development of other fault bend folds beneath the
discussed. observed structure (Suppe 1983). In such situations Suppe
shows how such features can be used to extrapolate down-
wards so as to predict the locations of the ramps and flats of
The relationships of folds and faults in thrust the underlying thrusts. However, this explanation might not
zones offer a universal model, especially where a well developed
footwall synform is present. Several alternative models
Two models are usually proposed to account for the interac- should be considered. Heim (1921) developed a model based
tive development of folds and faults in thrust systems; the on many years of field work in the Helvetic nappes that
fault bend- and fault propagation-models. In both, folding is suggested that thrust surfaces developed as a secondary
essentially a consequence of the progressive development of feature of folds, and were the result of the shearing out of the
the fault. The individual layers are deflected and folded as the middle limbs of antiform-synform pairs (Fig. 3). Although
overlying thrust sheet passes over irregularities in the thrust this theory is not in fashion today, it should be considered as
surface or develop as a fault tip propagates through unfaulted a possibility in regions of strongly developed regular folds,
rock. The Theological properties of individual layers play no especially where the form of footwall synforms mirrors that
guiding role in controlling the fold geometry. This assump- of hangingwall antiforms.
tion, even though convenient for the purposes of computing Another well known interpretation to account for folds in
the interlimb geometry of the folds, is either unrealistic, or at the footwall of a fault is that localized shear displacements
191
J. G. RAMSAY
B.
192
GEOMETRICAL PROBLEMS WITH RAMP-FLAT THRUST MODELS
, /! r! M f W to
'ni.-;i hrtii u:i
:i' •!•* ' , : . ;1t i •-••
iv
'-:;KJ- r!t ^ ><•!
J- H i j.^.f .
o
'..I:ji '/ !V'
.J
•ii; r /f
y . • f"; ,-
.iiiCi
(! ? ^
,--»l. •-;»>
• .•^•>t;r;ji ;
l'.' .•-V , (
'..MvS.iK
Figure 4. Tlirust structures seen in a quarry face of Hmestones and marls al Holderbank {30 km WNW Ziirich. Switzerland) with passive hangingwall and
deformed footwall {cf Fig. 2B, Model 2).
ment. However, if the rocks above and below the thrust had Model 1, deformed hanging wall (Boyer and Elliott, 1982)
more or less similar rheologies, and if the ground surface was
fairly high above the thrust plane then geometrically there
might be other solutions to the displacement field than that
shown in Figure 2A. For example it would be possible to take
up the same overall shortening between the two sides of the
thrust with a passive hangingwall and a folded footwall (Fig.
2B), or it would be possible to take up shortening with mirror
like deflections on both sides of the thrust (Fig. 2C). Al-
though these possibilities might shock the advocates of a
unique solution to the fault-bend model, nature provides
examples of both. Figure 4 shows a 50 m high quarry face in
Jurassic limestones (competent) and marls (incompetent) in
the folded Jura of Switzerland which seems in perfect accord
with the geometry of Figure 2B, and in the later discussion of
thrusting at Kimmeridge Bay the model of Figure 2C seems Model 2, deformed hanging and foot walls
to be predominant (Fig. 13). Why should downward i ' / f " l ' - J . ' . : , • • . -,.• • , . - :'
193
J. G. RAMSAY
/m/m/\
doing, it becomes unstrained (Elliott 1976; Ramsay & Ruber
8 1987, p. 538). At present little data exist to back up this
model, although small scale structural features of cleavage
and pressure shadows have been described in the Appala-
chian fold and thrust belt which fit this plan quite well
(Beutner ef a/. 1988). The strain plan ofthe model of Figure
7B is quite different from that of the kink model and should
show closer correspondence with contact strain buckle models
Figure 6. Typical active buckle fold developed by shortening of a competent
layer (black) in an incompetent rock showing progressively deamplifying
(Ramberg 1963; Dieterich 1970; Ramsay 1966, 1976). Fu-
folds in the zone of contact strain of the buckle. ture research might be effectively oriented to obtain more
detail ofthe fold geometry, finite strain and incremental strain
especially on a centimetric or decimetric scale. Although history of well exposed hangingwall anticlines.
large kink folds are frequently mentioned in the literature, the Special problems arise when investigating the geometric
fold geometry is rarely analysed in detail to see if it is really features of thrust zones located in the deeper, more internal
in accord with this model. In fact perfect kink band geometry parts of orogenic belts. In such regions the thin-skin concept
on a large scale is extremely rare (one excellent example has of an easily deformable cover overlying a rigid, undeformable
been described by Collomb & Donzeau 1974). Kink folds basement are untenable. The basement rocks are ductile or
develop only in terranes where the rocks are very uniformly
bedded with regular thickness periodic alternations of com- Kink band model -constant layer thickness (Suppe 1983)
petent and incompetent layers, or where there is a rather Incompetent layer
uniform anisotropy, such as is seen in slate belts. It is clear
that such conditions do not generally pertain in most average
rock sediment sequences, and this means that the kink model
is unlikely to provide a best fit to the fold geometry. With the
kink model the fold geometry shows a uniform forni and is
unconstrained upwards away from the generating ramp and
flat as far as the Earth's surface. Such geometry requires a
considerable imput of energy into the system. Long penetrat-
ing kink zones are characteristic of small scale kink bands, but
this feature is generally attributed to an initial elastic straining
through the rock packet as a whole followed by rapid kinking
redeploying this elastic strain into the permanent deformation Competent layer
of the kink band. Such a dynamic model does not fit well with
the stress and strain situation likely to exist during thrusting.
It is well known that the sideways deflection of incompetent Contact strain model -variable layer thickness controlled
material around a buckled layer progressively decreases by competent layer kinematics
away from the surface of the buckled layer (in the zone of
contact strain. Fig. 6), because such a geometry gives the least
work energy configuration of the system (Ramberg 1963).
Such decreasing fold amplitudes in layers surrounding the
folded competent rock in the hangingwall anticline is often
seen in natural thrust fault systems: the inner arcs of the
competent layer showing cuspate forms, the outer arcs with
more rounded forms and the amplitudes of the incompetent
rocks above the decreasing away from the bent competent
layer (Fig. 7B). Such a form is associated with a quite
different distribution of finite strains from those of the stand-
ard kink fold model. The kink model has no layer parallel
shortening and strain arises by simple shear displacements
predominantly in the kinked sectors of the structure. This Figure 7. Comparison of the classic fault bend fold model with kink folds
model predicts that initially unstrained or slightly strained in the hangingwall and a possible modification where the competent layer
strata should pass into a kink zone, and as they pass through shows buckle fold geometry with deamplifying folds in the zone of the
contact strain.
194
GEOMETRICAL PROBLEMS WITH RAMP-FLAT THRUST MODELS
semi-ductile and become involved in the structural edifice of colleagues into acceptance of a particular cross-section con-
the cover strata, and competence contrasts between rocks of struction, being of the opinion that, just because they possess
differing composition may be modified as a result of minera- seismic data, their views represent perfect truth.
logical changes during metamorphism (Ramsay 1983). The
models used to describe near surface structural features in the
crust are mostly constrained by plane strain with no material Thrust tectonics at Kimmeridge Bay, Dorset
passing in or out of a specific section plane. Such an outlook
is perfectly acceptable in this environment, but it would be In this section of the paper the geometric features of extremely
unwise to use such a model without very careful thought in well exposed sets of conjugate thrusts seen in dolostones and
internal orogenic zones. Metamorphic rock fabrics, backed shales exposed in the Upper Jurassic Kimmeridgian of the
up with strain measurements, often indicate that flattening or type locality at Kimmeridge Bay in South Dorset are described.
constrictional types of strain ellipsoids predominate, and it is The sediments consist of regularly bedded dolostones of
often surprising how such different types of strain can vary uniform thickness separated by dark brown or black organic
from place to place. The total displacement pattern in these shales (House 1969). On the west side of Kimmeridge Bay
regions is commonly built up by the superposition of several the foreshore is dominated by the wide exposure of a very
phases of deformation, superposed fold geometry and mul- gently inclined dolostone layer of about 50 cm thickness,
tiple cleavages clearly evidence this. In such situations, even known as the Flats Stone Band, situated near the core of a
where the individual deformations are of the plane strain type, regionally important anticline, the Purbeck Anticline
their combination can only be a plane strain under excep- (Donovan & Stride 1961; House 1969). This structure is
tional circumstances. Successive and differently oriented overturned to the north, having a steeply or vertically inclined
shear displacements on individual fault planes can bring northern limb and a gently southward dipping southern limb,
together rocks which were originally far separated and the and is generally interpreted as resulting from the reactivation
footwall and hangingwall of a fault in any one cross section of underlying pre-Albian extensional fault planes as com-
cannot be simply matched. Seismic sections are a great help pressional thrusts during the late Tertiary. The Flats Stone
in interpreting structural geometry in foreland regions but Band dolostone shows the development of complicated
they have severe limitations in basement rocks because the interferring thrust systems (Bellamy 1977; Leddra et al.
methods only allow us to identify reflectors with low incli- 1987).
nations, steeply inclined reflectors give rise to transparent The dolostone shows the development of four distinct
data windows. Even in foreland regions seismic records still types of deformation structures which, from their intersection
require interpretation and it should not be forgotten that, and interference relationships, appear to be formed at differ-
without borehole data, the reflection data can sometimes be ent times during the structural evolution of these rocks:
made to fit several kinematic models equally well. There is I. Low angle synsedimentary normal faults with displacements
a tendency today for some geologists to browbeat their of a few centimetres. The fault surfaces are injected by the
Figure 8. Thrust surface in the doloslone of the Flats Stone Band, Kimnieridge Bay showing overlapping white calcite fibres on the thrust surface.
195
J. G. RAMSAY
0 10 20 30 metres
Figure 9. Geological map of part of the surface of the Flats Stone Band, Kimmeridge Bay showing the thrust faults.
surrounding shaly material. grouped into two differently oriented conjugate sets with
2. Low angle thrusts forming two groups of conjugate sets average strikes of 110 and 45° respectively (Figs 9,10 & 11).
with displacements varying from zero to around 50 cm. The At some localities the triangular shaped block between two
fault surfaces are quite distinct from those of the parallel conjugate thrusts is lifted up as a type of horst mass
synsedimentary faults, being coated with overlapping fibres (Fig. 12andLeddraera/. 1987, Figs 3D &4D). The positions
of white calcite (Fig. 8). The overlap sense and lengths of where differently oriented hangingwall antiforms and their
these fibres is clearly in accord with the thrust displacements underlying thrusts meet show geometric relations suggesting
indicated by the displacements on the dolostone-shale con-
tacts (Dumey & Ramsay 1973).
3. Conjugate normal faults with displacements of up to a few
metres with general N-S or NNE-SS W strikes dipping east or
west at angles of about 50°. These clearly cut and displace the
thrust structures (2).
4. Regular sub-vertical planar or curved sets of joint surfaces
cut through all other structures. At least three major trends are
seen (striking 80, 0 and 150°), and all show plumose and
parabolic fracture surface markings indicative of generation
by rapid brittle fracture mechanisms.
The geometric features of the conjugate thrusts (3) are now
discussed further. On the exposed beach platform each thrust
shows a local uplift of the hangingwall above the regional
level of the upper dolostone surface very similar to that of the
classic fault bed fold model, although the outer arcs of the
hangingwall anticlines are more rounded than are the inner
arcs, conforming better to the model shown in Figure 7. A Poles to NE-SW sinking overthnjst faults
Individual antiforms can be traced along their hinges and d Poles to NW-SE striking overttinist faults
change amplitude and form in accordance with the extent of • Extension calcite fibre orientations, NE-SW faults
the displacements on the underlying thrust surface. They can o Extension calcite fibre orientations, NW-SE faults
196
GEOMETRICAL PROBLEMS WITH RAMP-FLAT THRUST MODELS
Shale
Dolostone
Figure 11. Schematic diagram of the geometric features of fold and faults developed in the Flats Stone Band doloslone of Kimmeridge Bay.
that the two sets of conjugate structures were formed at the the E-W direction.
same time. Although the resulting overall interference In the cliffs behind the foreshore good profile sections of
geometry has been decribed as polygonal (Leddra elai. 1987) these fold and thrust structures can be observed, and the two
it would be better described as lozenge shaped because the sides always show a mirror image downbending of the
polygons are not as regular and equidimensional as, for footwall and elevation of the hangingwall (Fig. 13) in accord
example, seen in mud crack- or basalt cooling-polygons. with the model shown in Figure 2C. The folding seen in the
Leddra et al. also suggested that there was a slight orientation shale layers around the dolostone is not kink like and the folds
effect {1987, Figs 2 & 5) and both data sets show a significant progressively deamplify around the folded dolostone layer in
anisotropy with a shorter axis in a N-S direction than that in the manner shown in Figure 7. ,. , .
Figure 12. Thrust elevated triangular wedge of dolostone between conjugate thrust planes. Note the arcuate fractures developed in the frontal zones of
both thrusts, and the concave form of these fractures toward the relative forward movement direction of the upper thrust plate. The fold and fault structures
are cut across by late joints.
197
J. G. RAMSAY
Figure 13. Ramp thrufit fault in the Flats Stone Band dolostone showing mirror symmetric folds in the hangingwall and footwall (cf. Fig. 2, Mode! 3).
The overlapping calcite fibres, when studied in thin sec- isolated angular fragments of dolostone occur in the form of
tion, show inclusion bands and small screens of wall rock dolomite-calcite cemented fault breccias. The structure of
material characteristic of the crack-seal mechanism of fibre this carbonate cement is very different from that of the fault
growth (Ramsay 1980). This structure is generally inter- surface fibres, the crystals forming radiating aggregates
preted as arising during tectonic fibre development where growing sub-perpendicularly to the walls of the dolostone
vein opening and fibre growth occur periodically over long fragments, and with a comb-structure typical of vug fillings
time periods. At some positions along the thrust faults (Bateman 1942).
Figure 14. Thrust of the Flats Stone Band dolostone showing calcite filled fault surface and calcite fibre lineations of the leading edge of the hangingwall
antiform (parallel to the pencil). Sub horizontal calcile veins branch off from the calcite filling of the thrust surface and appear to represent echelon extension
veins formed contemporaneously with the thrust development.
198
GEOMETRICAL PROBLEMS WITH RAMP-FLAT THRUST MODELS
The frontal parts of each hangingwall anticline and footwall Fault bend fold model
overall shear between hanging and tootwalls
syncline are striated and also may sometimes show fibrous Faults with continuity,
calcite fibres indicative of thrust sense movements (Fig. 14). amount ol slip varies from front to back
of fold complex
Such localized sliding could be the result of the bedding plane
slip on the steepened bedding surface according to the classic
fault bend fold model, however it might also arise from the
effect of wedging of the cut off tips of the hanging and
footwalls into the surrounding shales. Other special features
"Kimmeridge" model overall irrotational bulk strain
seen in hanging wall anticlines are curving sets of fractures Strain in incompetent layers
clearly related to the bending processes set up in these regions
(Fig. 12). These fractures are always concave towards the
shear sense direction. They appear to be extension structures,
and their curved form seems to suggest that various sectors of
the overthrust advance preferentially while adjacent parts of B
the thrust sheet are held back from thrust advance. Thrust wedges in competent layer
199
J. G. RAMSAY
survey of the sea floor south of the Dorset coast gives further models and to force too readily the geometry of naturally
information of the longitudinal nature of the Purbeck Anti- developed thrust systems to fit these models. It has been
cline (Donovan & Stride 1961). It appears that the Anticline pointed out that some geometric features of these models are
has an axial culmination about 15 km west of Kimmeridge difficult to explain in terms of mechanical behaviour of rock
Bay and a progressively increasing plunge to the east when materials and, probably more important for the practical
traced eastwards. The fold axis must therefore be curved and geologist, many well exposed examples of such structures do
the Kimmeridge Clay formation must be situated on the inner not fit too well the predicted geometry. Seismic data is not in
arc of a bowed hinge line. One might therefore expect similar every case a useful tool for resolving the dilemma because so
(but lesser) bedding plane shortenings in an E-W direction often a model forms the skeleton for interpreting the geological
just as there are N-S layer parallel shortenings in the fold flesh. It would be appropriate for geologists to show a little
profile section. If the axial culmination was developing at the more flexibility in the model choice and, in the future, to make
same time as fold amplification the local strains set up at much more accurate investigations of well exposed examples
Kimmeridge bay could account for the geometry of the of these types of structures to discover the exact geometric
observed thrust faults. forms of hanging wall antiforms and how these forms change
at distance from the thrust surface. Although realising that
Conclusions strain data is not always obtainable in natural systems it could
provide a critically valuable key for the kinematic and dy-
In this paper, it has been suggested that there has been a namic interpretation of thrusts and their associated fold
tendency to overplay the geometry of the classic fold and fault features.
References
Bateman, A. M. 1942. Economic mineral deposits. Wiley & Sons, New Knipe, R. J. & Needham, T. 1986. Deformation processes in accretionary
York. wedges - examples from the SW margin of the Southern Uplands,
Bellamy, J. 1977. Surface expansion megapolygons in Upper Jurassic Scotland. In: Coward, M. P. & Ries, A. C. (eds) Collision Tectonics.
dolostone (Kimmeridge, UK). Journal ofSedimentary Petrology, 47,973- Geological Society of London Special Publication, 19, 51-65.
978. Leddra, M. J., Yassir, N. A., Jones, C. & Jones M. E. 1987. Anomalous
Beutner, E. C , Fisher, D. M. & Kirkpatrick, J. L. 1988. Kinematics of compressional structures formed during diagenesis of a dolostone at
deformation at a thrust fault ramp (?) from syntectonic fibres in pressure Kimmeridge Bay, Dorset. Proceedings of the Geological Association, 8,
shadows. Geological Society of America Special Paper 222, 77-88. 145-155.
Boyer, S. E. & Elliott, D. 1982. Thrust systems. American Association of Mitra, S. 1986. Duplex structures and imbricate thrust systems: geometry,
Petroleum Geologists, 66, 1196-1230. structural position, and hydrocarbon potential. American Association of
Butler, R.W.H. 1983. Balanced cross-sections and their implications for the Petroleum Geologists Bulletin, 70, 1087-1112.
deep structure of the northwest Alps. Journal of Structural Geology, 5, Morrow, D.W.I 982. Diagenesis 1. Dolomite, The chemistry of dolomitization
125-137. and dolomite precipitation, Geoscience Canada, 9, 5-13
Collomb, P. & Donzeau, M. 1974. Relations entre kink-bands decametriques Price, R. A. 1981. The Cordilleran foreland thrust and fold belt in the
et fractures de socle dans I'Hercynian des Monts d'Ougarta (Sahara southern Canadian Rocky Mountains. In: McClay, K. R. & Price, N. J.
occidental, Algerie). Tectonophysics 24, 213-242. (eds) Thrust and Nappe Tectonics. Geological Society of London Special
Coward, M. P. 1983. Thrust tectonics, thin skinned or thick skinned, and the Publication, 9, 427-448.
continuation of thrusts to deep in the crust. Journal of Structural Geol- Ramberg, H. 1963. Strain distribution and geometry of folds. Geological
ogy. 5, \\3-\23. Institute of the University of Uppsala Bulletin, 42, 1-20.
Davis, D., Suppe, J. & Dahlen, F. A. 1983. Mechanics of fold-and-thrust belts Ramsay, J.G. 1966. Folding andfracturing ofrocks. McGraw Hill Book Co.
and accretionary wedges. Journal of Geophysical Research, 88, 1153- 1976. Displacement and strain. Philosophical Transactions ofthe Royal
1172. Society of London. A 283, 3-25.
Dieterich, J. H. 1970. Computer experiments on mechanics of finite ampli- 1980 Thecrack-sealmechanismofrock deformation. Wa/u/r,284,135-
tude folds. Canadian Journal of Earth Science, 7, 467-476. 139.
Donovan, D. T. & Stride. 1961. An acoustic survey of the sea floor south of 1983. Rock ductility and its influence on the development of tectonic
Dorset and its geological interpretation. Philosophical Transactions of structure in mountain belts. In: Hsii, K. (ed.) Mountain building proc-
the Royal Society of London. B 244, 299-330. esses. Academic Press, 111-127
Dumey, D. W. & Ramsay, J. G. 1973. Incremental strains measured by , Casey M. & Kligfield, R. 1983. Role of shear in the development of
syntectonic crystal growths. In: De Jong, K. A. & Scholten, R. (eds) the Helvetic thrust belt of Switzerland. Geology 11, 439-442.
Gravity and Tectonics, John Wiley & Sons, 67-96. & Huber, M. 1.1983. The techniques ofmodern structural geology, Vol
Elliott, D. 1976. The energy balance and deformation mechanisms of thrust I, Strain analysis. Academic Press, 1-307.
sheets. Philosophical Transactions ofthe Royal Society ofLondon, A2S3, & Huber, M. L,I987. The techniques of modern structural geology,
289-312. Vol.2, Folds and faults. Academic Press, 309-700.
Graham, R. H. 1978. Quantitative deformation studies in the Permian rocks Reks, I. J. & Gray, D. R. 1982. Pencil structure and strain in weakly deformed
of the Alpes-Maritimes. Goguel Symposium, B.R.G.M. 220-238. mudstone and siltstone. Journal of Structural Geology, 4, 161-176.
Heim, A. 1921. Geologic der Schweiz. Band II/I. Tauchnitz, Leipzig, Rich, J. L. 1934. Mechanics of low-angle overthrust faulting as illustrated by
House, M. R., 1969. The Dorset coast from Poole to the Chesil Beach. Cumberland thrust block, Virginia, Kentucky and Tennessee. American
Geologists Association Guide, 22. Association of Petroleum Geologists Bulletin, 18, 1584-1596.
, 1989. Geology of the Dorset Coast. Geologists Association Guide. Suppe, J. 1983. Geometry and Kinematics of fault-bend folding. American
Journal of Science, 283, 684-721.
200
The duplex model: Implications from a study offlexural-slipduplexes
P.W.G. Tanner
Abstract: Criteria for the recognition of duplexes developed on the limbs of flexural-slip folds
are summarized and the morphology of these flexural-slip duplexes is compared with that of well-
documented, fully-exposed examples of duplexes from thrust belts. The two groups of structures
have many features in common regardless of scale or of the rock type in which they occur. They
exhibit flat or gently curved floor and roof thrusts; low-angle thrusts in both the tip and rear of the
structure; and limited rotation of horses in the middle portion. In marked contrast to current
geometrical models for duplex formation, the development of ramp anticlines above newly-formed
thrusts at the tip of the duplex is relatively unimportant in natural examples. A revised model for
duplex formation is proposed in which new horses are accreted to the tip of the wedge-like structure
on low-angle thrusts, are then rotated between active floor and roof thrusts as the structure grows,
and become sigmoidal in the interior part of the duplex. The duplex maintains its streamlined
profile by internal adjustments, some oblique to the main movement direction, on both link thrusts
and backthrusts; all movements are accompanied by the growth of quartz or calcite fibre veins on
thrust surfaces.
The first imbricate structures were reported by Peach et al. The aim of this short contribution is to summarize the main
(1907) from the Moine Thrust zone, NW Scotland but the features of newly-discovered flexural-slip duplexes and, by
term 'duplex' was not applied to such structures until much using this information in combination with that from previ-
later (Dahlstrom 1970). Since then duplexes have been ously-described duplexes, to suggest a revised model for the
described from a wide range of geological environments formation of these structures in general.
including thrust belts (cf. Boyer & Elliott 1982), strike-slip
fault zones (Woodcock & Fischer 1986), extensional fault
systems (Gibbs 1984), and rocks affected by soft-sediment Flexural-slip duplexes
deformation (Shanmugam et al. 1988).
This paper is concerned with contractional duplexes, These are duplexes which form when, after the rocks have
which are rather brittle, easily eroded structures generally become lithified, packets of welded beds slip over one an-
found in gently dipping strata in thrust belts. As a result, the other during the development of flexural-slip folds (Tanner
complete duplex structure is seldom exposed and well-docu- 1989). They have been found recently in turbidite facies
mented examples of duplexes are rare. They include the sandstones and shales of Upper Carboniferous age from
Basse Normandie duplex (Cooper et al. 1983); the cleavage coastal cliff sections in North Devon and North Cornwall
duplexes described by Nickelsen (1986); internal duplexes between Hartland Quay and Bude (SW England) (Fig. 1). At
from the Lewis Thrust zone (McClay & Insley 1986); and a Hartland Quay the rocks have been folded into a single
small duplex from the Moine Thrust zone (Bowler 1987). generation of upright, slightly curvilinear chevron folds as a
Partially exposed, large-scale duplexes whose geometry is result of high-level deformation during the later stages of the
well known include the Haig Brook duplex (Fermor & Price Variscan (end-Carboniferous) orogeny (see Tanner 1989 and
1976); the Foinaven (Elliott & Johnson 1980) and Lighthouse references therein for details). Farther south these folds have
(Coward 1984) duplexes from the Moine Thrust zone; the been affected by S-directed simple shear (Sanderson 1979)
Mountain City window duplex (Boyer & Elliott 1982), and a and, a few kilometres south of Widemouth Bay, become tight,
duplex from the Makran accretionary prism (Piatt & Leggett recumbent structures (Fig. 1). Nearly all of the duplexes
1986). reported here occur on the limbs of folds which are either
Due to this scarcity of natural examples, our conception of unaffected (Hartland Quay), or little modified (Maer Cliff),
what a duplex looks like, i.e. the mental template used to by the later deformation. There is some evidence that an early
guide section construction in poorly exposed ground or on phase of N- directed thrust movement took place in these
seismic reflection profiles, is much influenced by theoretical rocks prior to the development of the chevron folds (Enfield
models, particularly that of Boyer & Elliott (1982). This et al. 1985; Whalley & Lloyd 1986) and the possibility that
model was based on the ramp anticline concept of Rich the duplexes described here are of pre-folding origin has been
(1934) whereas later modelling (Mitra 1986; Groshong & fully considered by the author, but rejected for the reasons
Usdansky 1988) used the Suppe kinematic model (Suppe outlined below. It should be stressed however that it is not the
1983) to generate the ramp-related folds and hence the duplex structural setting in which these duplexes formed but their
geometry. mode of formation and morphology which are of most
201
P.W.G. TANNER
Movement
importance here. horizons
The duplexes range in scale from a centimetre to a metre
thick; involve thin bands of slate, single sandstone beds, or displaced marker—p,
packages of beds; and are associated with other structures
such as simple ramps, blind thrusts, and imbricate structures
(Fig. 2). The detailed geometry and morphology of the Slate
duplexes, including slate duplexes (Nickelsen 1986), will be duplex
202
THE DUPLEX MODEL
FEATURES OF FLEXURAL-SLIP DUPLEXES cross-section can be area balanced. Small backthrusts are
POINTED (BACKTHRUST common in some duplexes; they have slickenfibre lineations
~',TOE'
SHEAR SENSE shown by fibre steps which lie approximately in the same plane as those on the link
thrusts and appear to be kinematically related to them. Stacked
SMALL duplexes are often found in which duplexes on different
DISPLACEMENTS scales formed from, for example, a single sandstone bed
on link thrusts
several centimetres thick, and from a laminated siltstone-
LENGTH: shale unit a few millimetres thick, are intimately associated
SOME a few cm to > 6 m
BACK ROTATION and form a single structural unit (Fig. 3).
SUB-DUPLEX
FORMATION
RELATIVELY FLAT" Comparison with other duplexes
FLOOR A N D ROOF
THRUSTS
The profiles of typical flexural-slip duplexes are compared
Q U A R T Z - C A R B O N A T E VEINS. C O M M O N L Y FIBROUS
with those of well-documented duplexes from various thrust
belts, starting with the simplest (? least evolved) of these
structures (Fig. 4). The Foinaven duplex (Fig. 4a), first
Figure 3. Summary in cartoon form of the main features of flexural-slip
described from the Moine Thrust zone by Peach et al. (1907),
duplexes.
has the Moine Thrust as a roof thrust. It is a foreland-
thrusts, and a rear part with gently dipping planar thrusts. propagating structure and the normal stratigraphic thickness
Curved bedding traces are seen within the horses in the of Pipe Rock and An t-Sron Formation -122 m, is represented
central parts of some duplexes, as shown diagrammatically in by a structural thickness of 510 m in the duplex, a total
Figure 3. shortening by area balance of -49% (Boyer & Elliott 1982).
Displacements on individual link thrusts are small and In overall morphology the Foinaven duplex closely re-
consistently show a thrust sense. The dip directions of the sembles the foreland portion of the Basse Normandie duplex
link thrusts and the trends of the slickenfibre lineations on from Boulonnais, France (Cooper et al. 1983), which was
their surfaces are commonly oblique to the movement direc- probably the first fully-exposed duplex to be described and
tions seen on the floor and roof thrusts. As this obliquity analysed. In this duplex (Fig. 4b) a 2-m thick sequence of
varies from place to place in the duplex the structure has thin-bedded limestones with two marker horizons has been
developed under conditions of variable 3-D strain and no affected by layer-parallel shortening (-22%), followed by a -
NW
An t-Sron Fm.
a Ikm
10m
/slate
Figure 4. Comparison of a simple flexural-slip duplex (c) with two duplexes of similar morphology from thrust belts, (a) the Foinaven duplex, Moine Thrust
zone, NW Scotland (after Elliott & Johnson 1980, Fig. 8); (b) the lower duplex, Basse Normandie, Boulonnais (after Cooper etal. 1983, Fig. 6); (c), a flexural-
slip duplex from near Berry Cliff, north of Hartland Quay, N Devon (SS 225258). The structure dips at 49°S but has been rotated to the horizontal in this
figure for comparison with a and b.
203
P.W.G. TANNER
^ : ^ ^ ^ X : : E ^ ^ ^ ^
m
erosive base
Figure S. Comparison of two flexural-slip duplexes (d & e) with duplexes from thrust belts, (a) duplex from the Makran accretionary prism, SW Pakistan
(after Piatt & Leggett 1986, Fig. 7); (b) duplex from the Eriboll area of the Moine Thrust zone, NW Scotland (after Bowler, 1987, Fig. 3); (c) an internal
duplex from the Lewis thrust sheet, Canadian Rocky Mountains (after McClay & Insley 1986, Fig. 11); (d) & (e) flexural-slip duplexes from Maer Cliff,
north of Bude, SW England (National Grid Reference SS 202082). Both profiles have been rotated to the horizontal for ease of comparison with a - c and
the N-S line represents the present-day horizontal.
27% shortening due to imbrication (Cooper et al. 1983). and contains 14 horses, consists of a single imbricated sand-
Calcite fibre growth was reported from the thrust surfaces and stone bed which is 5 cm thick near the base of the exposure,
the undeformed limestone sequence is seen at the north-east up to 9 cm thick within the duplex, and some 4.5 cm thick
end of the section (here shown in Fig. 4b as a mirror image of beyond the tip of the duplex. It is not known whether these
the original). thickness variations are a result of layer-parallel shortening
These two hinterland-dipping duplexes (Boyer & Elliott prior to, or during, duplex development or were sedimentary
1982) are compared with a flexural-slip duplex found near in origin. Traces of bedding within the horses show that little
Berry Cliff (SS 225258), north of Hartland Quay, North rotation of these has occurred, and displacements are on the
Devon, which dips at 49°S on the south limb of an anticline order of a few millimetres to 1 cm on each link thrust. Fibre
(Fig. 4c). The duplex, which is about 2m long (only the lower steps on 5 link thrusts give the sense of displacement shown
part is accessible and it has been drawn from photographs) in Figure 4c and confirm the direction of propagation of the
204
THE DUPLEX MODEL
duplex. It is of significance to the later discussion on the argillaceous limestone which is 14 m long and contains some
origin of duplexes that the roof thrust of this particular 27 horses. It has an overall geometry very similar to that
example continues to the limit of exposure at the foot of the described above and shows a progressive increase, then
cliff and that it is marked throughout the exposure by a thick decrease, in the maximum degree of clockwise rotation of
laminated quartz fibre vein up to 9 mm thick. From a study each link thrust from the floor thrust, as measured from the tip
of adjacent rocks in this section (Tanner 1989) it can be to the rear ofthe duplex (McClay & Insley 1986,Fig. 11). The
inferred that a bedding-parallel vein of this thickness is slip vectors marked by grooving, slickenside lineations and
indicative of a considerable amount of movement parallel to slickolites on the floor and roof thrusts are consistent in
the roof thrust. direction, whereas the slip directions on the link thrusts vary
Features of these simple duplexes which are of particular considerably.
significance are the (a) flat floor and roof thrusts, (b) constant For comparison with these duplexes described from thrust
ramp angles and lack of any ramp anticline development, and belts, two duplexes of flexural-slip origin from Maer Cliff,
(c) lack of evidence of significant rotation of the horses within near Bude in North Devon (Fig. 1) are shown in Figure 5, d
the duplex. & e and Figure 6. Both were drawn from detailed photographs
Well-exposed examples of duplexes showing a greater supplemented by field sketches and occur in interbedded
degree of internal rotation of horses are now described and are sandstones and slates dipping at 30-35° to the south, on the
compared with two examples of flexural-slip duplexes from southern limb of an anticline. Example d has 24 horses, is
the Bude area of North Cornwall (Fig. 1). They are all internal approximately 2 m long, and formed by the imbrication of a
duplexes (McClay & Insley 1986) in that they are contained 5-cm thick package of beds. Duplex e consists of over 18
within one stratigraphic formation and repeat a single bed- horses, is 54 cm long and formed from the imbrication of an
ding unit. homogeneous 4-cm thick sandstone bed. All of the thrust
The first example (Fig. 5a) is of an exceptionally well surfaces in d and e are marked by quartz fibre veins. Both
exposed duplex reported from the Makran Coast Ranges of duplexes are exposed on flat joint surfaces but another duplex
southwest Pakistan by Piatt & Leggett (1986), which occurs of similar size nearby is seen in 3-D and slickenfibres show
in well-bedded turbidite facies rocks of Tertiary age. It lies that movement on the link thrusts was oblique to that on the
beneath a major bedding-parallel thrust, the Garuk Kaur fault, floor and roof thrusts. In the latter duplex some ofthe internal
is 30m long, and was formed from a 30-cm thick 'turbidite thrusts which dip in the opposite direction to the main link
bed' shortened by -66% (Piatt & Leggett 1986). The frontal thrusts preserve fibre steps indicating a backthrust sense of
ramp dips at about 15°, the bed shows progressive back- movement.
rotation away from the tip of the duplex and the roof thrust has In the examples described so far, crucial information on
remained planar (Fig. 5a). The front portion of this duplex the form of the bedding surfaces within the horses in each
was figured on the cover of the Journal ofStructural Geology duplex is generally lacking; in the Bude-Hartland Quay area
for 1986 and it is the best example described so far of such a the best evidence of this seen so far is given by a partially
structure. Calcite fibre growth was reported from thrust eroded duplex found north of Widemouth Bay (Fig. 1)
surfaces in this area but was not described specifically from (National Grid Reference SS 198037). The duplex consists
the duplex. of a bedded sandstone unit repeated by imbrication and is
The next example (Fig. 5b) is of a small, completely over 6 m long and up to a metre thick, but only the hindward
exposed duplex from the north end of the Moine Thrust zone portion is preserved (Fig. 7). The shear sense and other
near Loch Eriboll described by Bowler (1987). It has formed features shown by this duplex are consistent with a flexural-
from the imbrication of a 30-cm thick quartzite unit, and is slip origin. A series of photographs, taken at right angles to
13m long and up to 1.2m thick (Fig. 5b), but it is not clear from the parallel fibre lineations seen on the floor and roof thrusts,
the description why a slice of quartzite apparently intervenes were used in the field to record details of the internal struc-
between the roof thrust (Bowler 1987, Figs 3 & 4b) and the ture. The resultant profile is only approximate in places
'roof defined by the aligned upper ends of the stacked horses. because of perspective problems but it shows that bedding in
This duplex shows the internal features which here are individual horses is generally only slightly sigmoidal, and
considered to be characteristic of duplexes in general: a there is only one clear example (at a, Fig. 7) of a possible
gently dipping frontal thrust; back rotation of horses in the ramp-related anticline. As with the other duplexes described
front-to-mid part of the duplex; and gently inclined thrust above, slickenfibre lineations on the link thrusts indicate that
surfaces in the rear portion. The link thrusts are mostly some out-of-section movement has taken place but a line-
curved and somewhat oblique to the movement direction and balanced restored section using two stratigraphic marker
the horses are lensoid in 3 -D. It is noteworthy that two or three horizons (contacts between beds A, B and B, C in Fig. 7)
sets of slickenfibre lineations occur on some link thrusts and within the duplex shows that the total shortening is -45% and
that fibre orientations in general are rather variable in ori- that the link thrusts were initially approximately parallel to
entation (Bowler 1987). one another and had a hindward dip of <20°. Because of
A series of small-scale imbricates and duplex structures uncertainty in the positions ofthe bottom of bed A and the top
were described from the Lewis thrust sheet in the Canadian of bed C in the unrestored section it is not known whether the
Rocky Mountains by McClay & Insley (1986). That repro- link thrusts were planar or slightly concave upwards.
duced in Figure 5c is a completely exposed duplex in All of these carefully documented duplexes have a similar
205
P.W.G. TANNER
•it-H
.,..fq. 'b-.-.:
,^urt, ' ; > • '
,,-! - -< j i . ji
Figure 6. The flexural-slip duplexes from Maer Cliff (National Grid Reference SS202082) shown in Figure 5d & e. The scale in (a) is 25 cm long and that
in (b) is 5 cm long.
external shape and internal geometry regardless of the fact 1). Planar or gently curved floor and roof thrusts bound the
that they have developed in rock types as diverse as lime- duplex and it has an asymmetric profile.
stone, sandstone, and quartzite, and that there is a 60-fold 2). Back rotation of horses to steeply dipping attitudes is
variation in scale for those shown in Figure 5. It should be restricted to the middle portion of the structure.
stressed that these examples have not been selected: to the 3). Link thrusts at the tip of a duplex have a low angle of
best of the author's knowledge, they represent the complete dip, as do those at the rear end of the structure.
set of well-documented examples of fully exposed duplexes 4). Displacements on individual link thrusts are oblique to
(apart from some flexural-slip duplexes similar to those the main transport direction and backthrusts are commonly
described here) and any omissions are unintentional. developed within the duplex.
Based on these examples, the features that have to be 5). All thrust surfaces are marked by calcite- or quartz-
explained by any model for the formation of a duplex are: fibre growth, depending upon host rock composition.
206
THE DUPLEX MODEL
ramp
anticline
roof thrust.
sub-duplex
207
P.W.G. TANNER
References
Bowler, S. 1987. Duplex geometry: an example from the Moine Thrust belt. McClay, K. R. & Insley, M. W. 1986. Duplex structures in the Lewis thrust
Tectonophysics, 135, 25-35. sheet, Crowsnest Pass, Rocky Mountains, Alberta, Canada. Journal of
Boyer, S. E. & Elliott, D. 1982. Thrust systems. American Association of Structural Geology, 8, 911-922.
Petroleum Geologists Bulletin, 66, 1196-1230. Mitra, S. 1986. Duplex structures and imbricate thrust systems: geometry,
Cooper, M. A., Garton, M. R. & Hossack, J. R. 1983. The origin of the Basse structural position, and hydrocarbon potential. American Association of
Normandie duplex, Boulonnais, France. Journal of Structural Geology, Petroleum Geologists Bulletin, 70, 1087-1112.
5, 139-152. Mitra, G. & Boyer, S. E. 1986. Energy balance and deformation mechanisms
Coward, M. P. 1984. A geometrical study of the Amaboll and Heilam thrust of duplexes. Journal of Structural Geology, 8, 291-304.
sheets, NW of Ben Amaboll, Sutherland. Scottish Journal of Geology, Nickelsen, R. P. 1986. Cleavage duplexes in the Marcellus Shale of the
20,87-106. Appalachian foreland. Journal of Structural Geology, 8, 361-371.
Cruikshank, K. E., Neavel, K. E. & Zhao, G. Z. 1989. Computer simulation Peach, B. N., Home, J., Gunn, W., Clough, C. T. & Hinxman, L. W. 1907.
of growth of duplex structures. Tectonophysics, 164, 1-12. The geological structure of the north-west Highlands of Scotland.
Dahlstrom, C.D.A. 1970. Structural geology in the eastern margin of the Memoir Geological Survey of Great Britain.
Canadian Rocky Mountains. Bulletin of Canadian Petroleum Geology, Piatt, J. P. & Leggett, J. K. 1986. Stratal extension in thrust footwalls, Makran
18, 332-406. accretionary prism: implications for thrust tectonics. American Asso-
Elliot, D. & Johnson, M.R.W. 1980. Structural evolution in the northern part ciation of Petroleum Geologists Bulletin, 70, 191-203.
of the Moine thrust belt, NW Scotland. Transactions of the Royal Society Ramsay, J. G. 1980. The crack-seal mechanism of rock deformation. Wamee,
of Edinburgh: Earth Sciences, 71, 69-96. 284, 135-139.
Enfield, M. A., Gillcrist, J. R., Palmer, S. N. & Whalley, J. S. 1985. Structural Rich, J. L. 1934. Mechanics of low-angle overthmst faulting as illustrated by
and sedimentary evidence for the early tectonic history of the Bude and Cumberland thrust block, Virginia, Kentucky and Tennessee. American
Crackington Formations, north Cornwall and Devon. Proceedings of the Association of Geologists Bulletin, 18, 1584-1596.
Ussher Society, 6, 165-172. Sanderson, D. J. 1979. The transition from upright to recumbent folding in
Fermor, P. R. & Price, R. A. 1976. Imbricate structures in the Lewis thrust the Variscan fold belt of southwest England: a model based on the
sheet around the Gate Creek and Haig Brook windows, southeastern kinematics of simple shear. Journal of Structural Geology, 1, 171 -180.
British Columbia. Geological Survey of Canada Paper, 76-lB, 7-10. Shanmugam, G., Moiola, R. J. & Sales, J. K. 1988. Duplex-like structures in
Geiser, P. A. 1988. The role of kinematics in the construction and analysis submarine fan channels, OuachitaMountains, Arkansas. Geology, 16,229-
ofgeological cross sections in deformed terranes. In: Mitra, G. & Wojtal, ' 232.
S. (eds) Geometries and Mechanisms of Thrusting. Geological Society Suppe, J. 1983. Geometry and kinematics of fault-bend folding. American
of America Special Paper, 222, 47-76. Journal of Science, 283, 684-721.
Gibbs, A. D. 1984. Structural evolution of extensional basin margins. Tanner, P.W.G. 1989. The flexural-slip mechanism. Journal of Structural
Journal of the Geological Society, London, 141, 609-620. Geology, 11, 635-655.
Groshong, R. H. & Usdansky, S. I. 1988. Kinematic models of plane-roofed 1990. The flexural-slip mechanism: Reply. Journal of Structural Ge-
duplex styles. In: Mitra, G. & Wojtal, S. (eds) Geometries and Me- ology, 12, 1085.
chanics ofThrusting. Geological Society of America Special Paper, 222, Whalley, J. S. & Lloyd, G. E. 1986. Tectonics of the Bude Formation, north
187-206. Comwall - the recognition of northerly directed decollment. Journal of
Knipe, R. J. 1985. Footwall geometry and the rheology of thrust sheets. the Geological Society, London, 143, 83-88.
Journal of Structural Geology, 7, 1-10. Woodcock, N. H. & Fischer, M. 1986. Strike-slip duplexes. Journal of
Structural Geology, 8, 725-735.
208
Palaeomagnetic techniques applied to thrust belts
A. M. McCaig' & E. McClelland^
Abstract: Palaeomagnetic data can be used to determine rotations of thrust sheets which cannot
be detected by conventional structural techniques. In basement rocks this may be the only way to
constrain rotations. A detailed knowledge of rotations is necessary for accurate section balancing
and can help in understanding the evolution of structures and the mechanisms of thrust sheet
deformation. Previously unsuspected rotations about steeply plunging axes have recently been
identified in many thrust belts around the world, and we present examples of the use of both pre-
and syntectonic remanences in both sedimentary and basement rocks. However, not all areas are
suitable for palaeomagnetic studies, and it is vital that appropriate structural and palaeomagnetic
tests are used to establish the ages of remanence components, and that the uncertainties inherent
in palaeomagnetic data are fully appreciated.
The study of the structural evolution of thrust belts is in large no meaningful remanence components; (6) statistical com-
measure the study and interpretation of finite rotations. This parisons within and between sites; (7) interpretation of data.
is normally accomplished by referring folded and tilted There is potential for error in all stages of this process, and
sedimentary rocks to an assumed initial horizontal orienta- palaeomagnetic programmes must be both planned and ex-
tion. This assumption is implicit in the construction of ecuted with great care. It is inevitable that some sites will not
balanced and restored cross-sections (Dahlstrom 1969; yield meaningful data, and it is particularly important that
Hossack 1979), where all rotations are normally assumed to sites be included in the dataset on purely rational criteria.
take place about axes normal to the plane of section. The Unfortunately, this has not always been the case, with the
occurrence of other rotations will automatically invalidate result that palaeomagnetism has acquired an unjustifiably
the assumption of plane strain within the section. bad reputation amongst many geologists.
Conventional structural studies in sedimentary rocks cannot The most suitable lithologies for palaeomagnetic work are
detect rotations which do not fold or tilt the bedding, and in volcanic rocks, shallow intrusives and iron-rich sediments,
polydeformed or intrusive basement rocks thrust-related ro- but with the advent of cryogenic magnetometers practically
tations are usually impossible to constrain. Palaeomagnetic any rock may yield significant results. For structural studies,
studies have the power to identify rotations not previously young (Tertiary) rocks are generally best since the range of
suspected in sedimentary rocks, and may be the only way of possible field directions is small. This means that any
detecting rotations in basement rocks. Palaeomagnetic studies significant deviations from the present day field direction are
are an important tool in the understanding of the structural almost certainly due to local tectonic rotations.
evolution of both internal and external parts of mountain Selection of sample sites should always be accompanied
belts. However, because of the time consuming nature of by field mapping and structural measurements with a view to
palaeomagnetic work and difficulties in establishing the age using palaeomagnetic tests to establish the age of remanence
of remanence, areas for such study need to be selected with as discussed below. Best results are obtained by using a rock
some care. drill to collect core samples in the field. Use of oriented
In this paper, the principles and techniques of blocks introduces an additional source of error and it is
palaeomagnetic study in deformed rocks willfirstbe reviewed. difficult to collect sufficiently fresh material. Cores can be
Examples will then be given of the application of very accurately oriented using a specially adapted sun com-
palaeomagnetism to thrust belts in both sedimentary and pass (Collinson 1983), eliminafing problems associated with
basement rocks. magnetic rocks. At least six and preferably eight samples
must be collected at each site if statistically significant results
are to be obtained.
Palaeomagnetic techniques and methods It is important that as many samples as possible be fully
demagnetized to determine the vector structure of the
Data collection
remanence. Whether Alternating Field or Thermal Treatment
The stages in the collection of a palaeomagnetic dataset can or a combination of both is used depends on the magnetic
be summarized as follows: (1) selection of field area; (2) mineralogy and the correct treatment is determined by pilot
selection of sample sites; (3) collection and orientation of experiments (Collinson 1983). Magnetic anisotropy may
samples; (4) demagnetisation of samples; (5) establishment deflect remanence components and should be tested for in
of stable remanence components, and rejection of sites with deformed samples (Lowrie et al. 1986).
209
A. M. MCCAIG & E. MCCLELLAND
Methods for establishing statistically significant remanence Palaeomagnetic studies of thrust belts in
components both within a sample and within a site were sedimentary rocks
established by Fisher (1953). Sites can be compared using
Previous work.
standard fold and contact tests as described below.
On the scale of an erogenic belt, much palaeomagnetic effort
has been directed towards the question of 'oroclinal bending'
Establishing the age of remanence
(e.g. Irving & Opdyke 1965; Van der Voo & Channell 1980;
Most rocks contain several components of remanence ac- Lowrie & Hirt 1986). Smaller scale studies of rotations in
quired at different times and by different mechanisms. thrust belts have been less common, but nevertheless rotations
Depositional remanence (DRM) arises from the settling of about previously unsuspected axes have been recognised
magnetic minerals in the Earth's field. Thermal remanence from several areas.
(TRM) results from cooling of igneous or metamorphic One of the earliest such studies was by Kotasek & Krs
minerals through the Curie point, while chemical remanence (1965), who were able to refute arguments for the
(CRM) components can arise whenever new minerals grow autochthonous nature of parts of the western Carpathians on
within a rock during diagenesis, metamorphism or defor- the basis of palaeomagnetic evidence. Local rotations of
mation. During thermal demagnetisation, different compo- thrust sheets around salients in the Wyoming-Idaho thrust
nents held in different minerals will be stripped off at different belt have been attributed to the buttressing effects of base-
temperatures. However, it cannot be assumed that the highest ment highs (Grubbs & Van der Voo 1976; Schwartz & Van
temperature components are the earliest, or even that they der Voo 1984; Eldredge & Van der Voo 1988). Around the
formed at higher temperatures (O'Reilly 1984). It is often Wyoming salient, the rotations vary more or less smoothly,
very hard to relate a particular component to a particular while large variations in rotation around the Helena salient
mineral growth episode in the rock, and indirect tests must be imply that here the thrust sheets broke and rotated as indi-
used to establish the age of magnetisation. vidual pieces. Kent (1988) has suggested that similar results
The most useful test in structural studies is the fold test around the Pennsylvania salient in the Appalachians are due
(McElhinny 1964). In a positive fold test, a remanence to a tightening of originally arcuate structures. In Nevada,
component on the two limbs of a fold is grouped significantly studies by Geissman etal. (1982), Hudson & Geissman (1984)
more closely after unfolding about the fold axis than in situ, and Geissman et al. (1984) indicate that rotation has occurred
showing that remanence acquisition predated folding. A due to the oblique incidence of thrusting onto a shelf margin,
negative/oW test indicates that remanence acquisition post- compounded by buttressing effects of irregularities in the
dated folding. In some cases, one component may predate shelf margin. Opdyke et al. (1982) and Klootwijk et al.
folding while another may be acquired during folding, and (1986) have presented evidence that the northwestern and
become most closely grouped after partial unfolding central Himalayan regions have been thrusted over the Indian
(McClelland-Brown 1983). shield with a coherent large-scale clockwise rotation. In
Mallorca, palaeomagnetic data has been used to rule out
The contact, or dyke test is useful wherever sedimentary or
differential rotation as an explanation for different orientations
basement rocks are cut by later intrusions (e.g. Allerton &
of structures in different thrust sheets (Freeman et al. 1989).
Vine 1987; McClelland & McCaig 1989). If a particular
Channell et al. (1990) describe palaeomagnetic data from a
remanence direction is present in both a dyke and its imme-
thrust pile in western Sicily which show up to 140° of
diate wall rocks, but disappears further away, then that
clockwise rotation in the uppermost unit, and decreasing
component almost certainly dates from the time of dyke
amounts of rotation in structurally lower units, providing
intrusion. Contact tests are usually more successful with
important new information for palinspastic reconstructions
small intrusions than with large ones.
of the western Mediterranean region. Finally, Dinares et al.
Other useful tests include the reversal test and the con- (1991, this volume) have used palaeomagnetic data to define
glomerate test (cf. McElhinny 1973). In the former, the lateral variability in rotations of carbonate thrust sheets from
presence of reversals of magnetization within a sedimentary the southern Pyrenees.
sequence indicates that the sequence has not been remagnetized
since the original remanence was acquired. In the latter, the
presence of randomly oriented remanence directions within
blocks of the same type indicates that this remanence, which Scope and limitations of the method
may be identifiable within the source formations of the Pre-deformational remanences. If pre-deformational
conglomerate, must pre-date deposition. If the remanence in remanences of known age are present, the evolution of
blocks and underlying formation is uniform in direction, then complex structures can be constrained, since reversal of the
both conglomerate and source have been remagnetized. deformation path must restore both bedding to a horizontal
In more complex fold structures, it is useful to test whether attitude and the remanence to its reference direction (cf.
the angle between a given remanence and the bedding re- Bonhommet ef a/. 1981). A useful concept here is that of the
mains constant across the structure. If it does, the remanence finite rotation pole (RiA&s 1989). This is the pole about which
must predate folding, and can be used to constrain the the bedding rotates to horizontal and simultaneously a
evolution of the structure even where two phases of folding remanence direction of known age rotates into coincidence
have occurred (cf. Bonhommet et al. 1981). with the reference palaeopole for this age. Figure 1 illustrates
210
PALAEOMAGNETIC TECHNIQUES APPLIED TO THRUST BELTS
211
A. M. MCCAIG & E. MCCLELLAND
N - S C R O S S S E C r r O N T H R O U G H ISABENA FOLD
Halt unfolded
Remanence ^ ^
directions ^^*---
parallel
Figure 3. Schematic cross-section
through fold in Triassic redbeds,
Unfolded Isabena valley, Pyrenean Axial Zone,
showing how intermediate and high
Remanence
i^^
directions dispersed
Remanence i^zzi^.
directions parallel
blocking temperature components
group on incremental unfolding.
of the fold and after the rotation of the thrust sheet, as the effects can be removed and total rotation poles can still be
partially unfolded mean is indistinguishable from the Alpine established (Cogne & Perroud 1985). Deflection due to
age reference direction. This data can be used to show that the magnetic anisolropy is a more difficult problem. Recent
rotation of this area finished before folding was complete. work in southwest Wales (Ogden etal. 1990; Stearns & Van
Internal strain of rocks can cause problems in interpreting der Voo 1987) suggests that this may have affected some of
syn-deformational remanences {Kligfieldf/a/. 1983; Cogne the Devonian redbeds studied by McClelland-Brown (1983)
et al. 1986). Kodama (1988) has suggested that some and McClelland (1987), where syn-deformational remanences
remanences interpreted to be 'syn-folding' may have resulted were identified. Further work is in progress to try to resolve
either through deflection of earlier remanence directions due this problem, but clearly anisotropy of magnetic susceptibil-
to internal strain or deflection of newly formed remanences ity must be tested for, and results from anisotropic rocks must
away from the field direction due to magnetic anisotropy. be viewed with caution. Syn-folding remanences certainly
Where two or more components are present, deflection due to do occur however, as shown by the very detailed study of
strain should affect all components which are affected by the Hudson e/<3/. (1989).
folding. If the strain can be measured independently, its
"<y:it-:
'^a Permian
pTv-'-N Stophnnlan
Palaeozoic baaamant
Figure 4. Cross-section 3 km west of line A-B in Fig. 5(a), constructed using the map of Mey (1969) and field data from this study. Note reactivation of
normal faults to explain absence of lower Trias, Permian and Stephanian in sheet 5, and out-of-sequence thrust which truncates fold structures in Triassic
footwall rocks.
212
PALAEOMAGNETIC TECHNIQUES APPLIED TO THRUST BELTS
213
A. M. MCCAIG & E. MCCLELLAND
(A)
-42° 34'
- 42° 30'
OMO'
(B)
0 1km
Figure 5. (a) Mean dyke remanence directions from sites or groups of sites in an antifomial stack cutting basement rocks in the southern Pyrenean Axial
Zone, updated after McClelland & McCaig (1989). The cone opens towards the actual declination of the remanence (point of cone is at site), width of cone
is twice alpha 95 (the cone of confidence), number by cone indicates inclination (positive numbers indicate polarity down plunge). Large arrow indicates
Late Carboniferous reference field direction. A-B and C-D are lines of sections in Fig 5(b), (b) Schematic cross-sections showing palaeomagnetic estimates
of tilt about a horizontal axis. Arrows are oriented at the estimated angle of lilt obtained from the group of dykes immediately below the arrow, the number
next to the arrow is the tilt estimate. Note that both normal and reversed polarities exist in the dykes.
214
PALAEOMAGNETIC TECHNIQUES APPLIED TO THRUST BELTS
SSW, with no sign of any arcuate pattern (McCaig & Evolution of structures
McClelland unpubl. data), and it seems that part of the
As discussed earlier, syntectonic remanences may provide
rotation may have occurred through domino faulting.
valuable information about intermediate points on the
strainpath leading to a particular structure. Lateral propaga-
Section balancing tion of folds may lead to tilting and then untilting of beds
along the fold axis, and similar effects may occur as thrust
Balanced cross-sections are only valid if all strain is confined
sheets move over ramps and flats. A number of studies have
to the plane of section. Rotations about axes not perpendicu-
attempted to address this type of problem (McClelland-
lar to the plane of section automatically invalidate the tech-
Brown 1983;Bates 1989), but although syntectonic remanence
nique, although the extent to which errors are introduced
components have been identified, little useful data for con-
depends on the amount of rotation and the orientation of
straining thrust sheet evolution has so far emerged, which was
bedding when rotation occurs. For example, rotation of a flat
not already available from study of pretectonic remanences.
lying thrust sheet about a vertical axis need not introduce any
The problem is that extremely detailed studies are required to
excess area into the plane of section, whereas rotation of a
constrain the relationship between remanence components
steeply dipping or irregular sheet almost certainly will.
which may have been introduced at slightly different times at
The area studied in the Pyrenees lies astride the section different locations in the structure.
line studied and balanced by Williams (1985) and Williams
& Fischer (1984). These authors did not suspect the rotations
documented here, and there is no doubt that small errors in Fluid movement during thrusting
their shortening estimate will have resulted. However, greater
New reinanence components are introduced into rocks as a
errors were probably associated with the unrealistic geometry
result of new mineral growth or the recrystallisation of
they proposed for the Nogueras Zone antiformal stack, in
existing minerals. Numerous studies in the fields of diagenesis
particular the failure to document out of sequence thrusts
and metamorphism have shown that such mineral growth
which have been identified by field mapping (Fig. 4).
often occurs as a result of fluid infiltration. Palaeomagnetism
Ultimately, the only way to incorporate palaeomagnetic
therefore offers a means of constraining fluid movement
data into such studies is by volume balancing (Butler 1990).
pathways through deforming rocks, as well as the age of fluid
This requires an exceedingly precise 3-D model of the present-
movement relative to particular tilting events (cf. Bates
day geometry of a portion of a thrust belt which is not
1989). Such studies may prove particularly valuable in the
generally possible where data is only available from surface
petroleum industry, where the age of a diagenetic event
mapping. It is recommended that future studies of volume
which affects reservoir permeability relative to the age of
balancing should concentrate on areas where extensive sub-
formation of structural traps can be of great importance.
surface data is available and where palaeomagnetic studies
can also be undertaken. It should always be borne in mind,
however, that the main purpose of such studies is not to This work was supported by NERC grant GR3/5933 to A. McCaig and E.
endlessly refine shortening estimates but to understand the McClelland. Reviews by P. Geiser and C. Jackson helped to improve the
manuscript.
mechanical processes involved in the evolution of thrust belts
and thrust geometries.
References
Allerton, S. & Vine, F. J. 1987. Spreading structure of the Troodos ophiolite, Collinson, D. W. \9ii3. Methods in Rock Magnetism and Palaeomagnetism.
Cyprus: some palaeomagnetic constraints. Geology, 15, 583-597. Chapman and Hall, London, 503p.
Bates, M. P. 1989. Palaeomagnetic evidence for rotations and deformation Courtillot, v., Chambon, P., Brun, J. P., Rocheite, P. & Matte, P. 1986. A
in the Nogueras Zone, Central Southern Pyrenees, Spain. Journal of the Magnetotectonic study of the Hercynian Montague Noire (France). Jn:
Geological .Society of London, 146, 4.59-476. McClelland, E.A., Courtillot, V. & Tapponier, P. E. (eds)
Bonhommet, N., Cobbold, P. R., Perroud, H. & Richardson, A. 1981. Magnetoteclonics. Tectonics, 5, 733-751.
Palaeomagnetism and cross-folding in a key area of the Asturian Arc Dahlstrom.C.D. 1969. Balanced cross-sections. Ca«ad/'an,/o«™fl/o/£arr/i
(Spmn). Journal of Geophysical Research, 86, 1873-1887. Sciences, 6, 743-757.
Butler, R.W.H. 1990. Evolution of fold-thrust complexes: a linked kinematic Dinares, J.. McClelland, E. & Santanach, P. 1991. Contrasting rotations
approach. In: Mitra S. (ed.) Structural geology of fold and thrust belts. within thrust sheets and kinematics of thrust tectonics as derived from
Johns Hopkins University Press, Baltimore (in press). palaeomagnetic data: an example from the Southern Pyrenees (this
Channell, J.E.T., Oldow, J. S., Catalano, R. & D'Argenio, B. 1990. volume).
Palaeomagnetically determined rotations in the western Sicilian fold and Elderedge, S. & Van der Voo, R. 1988. Palaeomagnetic study of thrust sheet
thrust belt. Tectonics, 9, 641-660. rotations in the Helena and Wyoming salients of the Northern Rocky
Cogne, J. P. & Perroud, H. 1985. Strain removal applied to palaeomagnetic Mountains. Geological Society of America Memoir, 171, 319-332.
directions in an orogenic belt; the Permian redbedsof the Alpes Maritimes, Elliot, D. & Johnson M.R.W. 1980. Structural evolution in the northern part
France. Earth and Planetary Science Letters, 72, 125-140. of the Moine thrust beltofNW Scotland. Tranracr/o/iio/V/jeftoya/Socief>'
Perroud,H.,Texier,M. P. & Bonhommet N. 1986. Strain reorientation of Edinburgh, Earth Sciences, 71B, 69-96.
of hematite and its bearing upon remanent magnetization, /H; McClelland, Fisher, R. A. 1953. Dispersion on a sphere. Proceedings of the Royal Society
E. A., Courtillot, V. & Tapponier, P. E. (eds). Magnetotectonics. Tec- of London. A, 217, 295-305.
tonics, 5, 753-767.
215
A. M. MCCAIG & E. MCCLELLAND
Freeman, R., Sabat, F., Lowrie, W. & Fontbote, J.-M. 1989. Palaeomagnetic Lowrie, W. & Hirt, A. M. 1986. Palaeomagnetism in arcuate mountain belts.
results from Mallorca (Balearic Islands, Spain). Tectonics, 8, 591-608. In: F.-C. Wezel (ed.) The Origin of Arcs. Elsevier, Amsterdam.
Geissman, J. W., Van der Voo, R. & Howard, K. L. 1982. A Palaeomagnetic , & Kligfield, R. 1986. Effects of tectonic deformation on the
study of the structural deformation in the Yerington district, Nevada, II. remanent magnetization of rocks. In: McClelland, E.A., Courtillot, V. &
Mesozoic basement units and their total and pre-OIigocene tectonism. Tapponier, P.E. (eds) Magnetoteclonics. Tectonics, 5, 713-722.
American Journal of Science, 282, 1080-1109. McClelland-Brown, E. 1983. Palaeomagnetic studies of fold development
Callian. J.T., Oldow, J. S. & Humphries, S. E. 1984. Palaeomagnetic and propagation in the Pembrokeshire OldRed Sandstone./n.McClelland-
assessment of oroflexural deformation in west-central Nevada and Brown, E. & Vandenberg, J. (eds) Palaeomagnetism of Orogenic Belts.
relations to emplacement of allochthonous assemblages. Tectonics, 3,179- Tectonophysics, 98, 131-149.
200. McClelland, E. 1987. Palaeomagnetic results from the Lower Devonian
Grubbs, K. L. & Van der Voo R., 1976. Structural deformation of the Idaho- Llandstadwell formation, Dyfed, Wales - Discussion. Tectonophysics,
Wyoming overthrust belt (USA), as determined by Triassic 143,335-336.
palaeomagnetism. Tectonophysics, 33, 321-336. & McCaig, A. M. 1989. Palaeomagnetic estimates of rotations in
Hossack, J. R. 1979. The use of balanced cross-sections in the calculation of compressional regimes and potential discrimination between thin-skinned
orogenic contraction: a review. Journal of the Geological Society of and deep cruslal deformation, //i; Kissel, C. & Laj, C. (eds) Palaeomagnelic
London, 136,105-711. Rotations and Continental Deformation. NATO ASI Series C. 254,365-
Hudson, M. R. & Geissman, J. W. 1984. Preliminary palaeomagnetic data 379.
from the Jurassic Humboldt lopolith, west-central Nevada: evidence for McElhinney, M. W. 1964. Statistical significance of the fold test in
thrust belt rotation in the Fencemaker allochthon. Geophysical Research palaeomagnetism, Geophvsical Journal of the Royal Astronomical Soci-
Letters, I, H2S-S3\. ety, 8, 338-340.
Reynolds, R. L. & Fishman, N. S. 1989. Synfolding magnetization in 1973. Palaeomagnetism and plate tectonics. Cambridge University
the Jurassic Preuss sandstone, Wyoming-Idaho-Utah thrust be\l.Journul Press.
of Geophysical Research, 94, 13,681-13,705. Mey, P.H.W. 1969. Geology of the upper Ribagorzana and Tor valleys,
Irving, E. & Opdyke, N. D. 1965. The palaeomagnetism of the Bloomsberg centra\Pyrenees,Spa.m.LeidseGeologischeMededelungen,41,229-292.
redbeds and its possible application to the tectonic history of the Molnar, P. & Lyon-Caen, H. 1989. Fault plane solutions of earthquakes and
Appalachians. Geophvsical Journal of the Royal Astronomical Society, active tectonics of the Tibetan Plateau and its margins. Geophysical
9, 153-167. Journal International. 99, 123-153.
Jackson, J. & McKenzie, D. P. 1989. Relations between seismicity and Ogden, K. L., Torsvik, T. H. & McClelland, E. 1990. The relationship
palaeomagnetic rotations in zones of distributed continental deforma- between syn-deformational remanences and magnetic fabric in the Old
tion, [n: Kissel, C. & Laj, C. (eds) Palaeomagnetic Rotations and Red Sandstone of S W Dyfed, Wales. Geophysical Journal International,
Continental Deformation. NATO ASI Series C, 254, 33-42. 101, 283 (abst.).
Kent, D. V. 1988. Further palaeomagnetic evidence for oroclinal rotation in Opdyke, N. D., Johnson, N. H., Lindley, E. H. & Tahirrkhehi, L. 1982.
the central folded Appalachians from the Bloomsburg and Mauch Chunk Palaeomagnetism of the Middle Siwalik formation of Northern Pakistan
formations. Tectonics, 7, 749-759. and rotation of the Salt Range decollement. Palaeogeography.
Kligfield, R., Lowrie, W., Hirt, A. & Siddans, A.B.W. 1983. Effect of Palaeoclimatology and Palaeoecology, 37, 1-16.
progressive deformation on remanent magnetization of Permian redbeds O'Reilly, W. 1984. Rock and Mineral Magnetism. Blackie, London, 220p.
from the Alpes Maritimes. In: McClelland Brown, E. & Vandenberg, J. Ramsay, J. G. 1967. Folding and fracture of rocks. McGraw Hill, 568p.
(eds) Palaeomagnetism of Orogenic Belts. Tectonophysics, 98, 43-57. Ron, H., Freund, R., Garfunkel, Z. & Nur, A. 1984. Block rotation by strike-
Klootwijk, C. T., Sharma, M. L., Gergan, J., Shah, S. K. & Gupta, B. K. 1986. slip faulting: structural and palaeomagnetic evidence. Journal of Geo-
Rotational overthrusting of the north western Himalaya: further physical Research, 89, 6256-6270.
palaeomagnetic evidence from the Riasi thrust sheet, Jammu foothills, Schwartz, S. Y. & Van der Voo, R. 1984. Palaeomagnetic study of thrust
India. Earth and Planetary Science Letters, 80, 375-393. sheet rotation during foreland impingement in the Wyoming-Idaho
Kodama, K. P. 1988. Remanence rotation due to rock strain during folding overthrust belt. Journal of Geophysical Research, 89, 10,077-10,086.
and the stepwise application of the fold test. Journal of Geophysical Van der Voo, R. & Channell, J.E.T. 1980. Palaeomagnetism in Orogenic
Research,93,3357-331l. Belts. Revues of Geophysics and Space Physics, 18, 455-481,
Kotasek, J. & Krs, M. 1965. Palaeomagnetic study of tectonic rotation in the Williams, G.D. 1985. Thrust tectonics in the south central Pyrenees.youraa/
Carpathian mountains of Czechoslovakia. Palaeogeography. of Structural Geology, 7, 11-18.
Palaeoclimatology and Palaeoecology, 1, 39-49. & Fischer, M.W. 1984. A balanced cross-section across the Pyrenean
orogenic belt. Tectonics, 3, 773-780.
216
Evolution of crystalline thrust sheets in the internal parts of mountain
chains
Robert D. Hatcher, Jr.' & Robert J. Hooper^
Abstract: Two kinds of crystalline thrust sheets form in the internides of mountain chains. Type
C megathrust sheets are internally brittle slabs of intact crust (composite basement) that detach
within the thermally softened ductile-brittle transition (DBT) and, once formed, behave as thin-
skinned thrust sheets. Type F thrust sheets are fold-related lobe-shaped thrust sheets that form
below or within the DBT by attenuation of the common limb between antiforms and synforms in
passive- or flexural-flow folding; transport is controlled by ductile flow. Type C megathrust sheets
form by continent-continent or arc-continent collision accompanying A-subduction; Type F
sheets form via A- or B-subduction below the DBT. Both result in crustal thickening. Individual
Type C megathrusts are very strong (compared to large foreland sheets), with maximum size and
displacement attained where crystalline thrusts ramp into weak zones in platform sedimentary
rocks. Here crustal thickness may be duplicated, but a (basal detachment) and |J (surface slope)
angles remain constant (and near zero) because of slab geometry. Coefficient of internal friction
along the base of nascent Type C megathrust sheets would be low (<0.3), while the sheet itself
would be very strong and coherent, with a high coefficient of internal friction (>0.85). Foreland
thrusts are driven ahead of Type C sheets as crystalline and foreland thrusts merge into the
Coulomb wedge of the foreland. Behaviour modes also merge here, because, once formed. Type
C sheets commonly ramp onto the platform and propagate along the basal detachment of the
deforming platform wedge. Thin platform assemblages on continental promontories restrict the
size and displacement of Type C and foreland sheets. Crustal duplexes form in ramp zones as
thrusts can no longer propagate along the DBT. Duplexes of platform sedimentary rocks (±
basement) may also form beneath the crystalline sheet and arch the .sheet above. Late macro- and
meso-scale structures (isolated domes, out of sequence thrusts, open folds, some crenellations, and
ductile shears) related to thrust emplacement may form within the sheet. Thrust-related meso- and
microfabrics are mostly concentrated in or near the fault zone. Character of fault rocks varies with
location, rate of motion on the fault, availability of fluid, and ambient T-P conditions in the fault
zone.
Fault rocks vary from cataclasite formed near the surface or at depth at high strain rate, to retrograde
mylonite formed in low-T or high strain rate zones, to annealed mylonite formed in high-T zones
or zones where recrystallization/recovery rates exceed strain rate, to absent along the base of some
Type F thrust sheets. Type F thrust sheets fomi in the middle to lower crust in upper greenschist
to granulite facies conditions. They may be thicker (or thinner) and have less regular shape than
Type C sheets because of association with folding, but are not as extensive because of lower overall
strength. Mesofabrics within the sheets are commonly more penetrative and representative of
transport and deformation of the entire sheet, with sheath folds a common meso- and macro- style,
but cylindrical folds with orogen-parallel lineations normal to transport or mixed-mode linear
fabrics may be present.
Crystalline and foreland thrust systems were once thought to geometry, appear quite similar. Thus the importance of both
be separate and independent structures, thus mechanical the similarities and differences should therefore be empha-
models were developed that characterized them separately sized in any mechanical model (Hatcher & Williams 1986).
(e.g. Armstrong & Dick 1974; Chappie 1978; Boyer& Elliott Early divisions centered around the involvement of crys-
1982; Davis era/. 1983). Certain characteristics - maximum talline rocks (or basement) in thrust systems. If crystalline
size, location in orogens, and initial detachment properties - rocks or basement were involved, the thrust system was thick-
are certainly different; other properties, however, including skinned; if crystalline rocks were not involved, the system
*Operated by Martin Marietta Energy systems, Inc., for the US Department of Energy under contract No. DE-AC05-840R21400.
217
R. D. HATCHER & R. J. HOOPER
was thin-skinned. As shown herein, however, particular gested that differences occur within foreland thrust systems;
kinds of crystalline thrust sheets exhibit thin-skinned behav- the mechanical characteristics of foreland fold-thrust belt
iour. The division of thrust systems into thick- or thin- thrust systems also differ significantly from those formed in
skinned based solely on the involvement of basement thus has accretionary wedge complexes because of fundamental dif-
only historical significance and should probably be aban- ferences in the mechanical properties of the materials being
doned for attempting to understand thrust mechanics. Inter- deformed (Woodward 1987).
estingly, several classic studies where many of the fundamen- The purpose here is to explore the processes related to the
tal geometric properties of thrusts were first observed were in formation and evolution of crystalline thrust sheets that form
fact conducted on crystalline thrusts - for example, the Moine in the internal parts of mountain chains (Fig. 1). Most kinds
thrust in the Northwest Highlands of Scotland and the thrust of crystalline thrust sheets may be grouped into two funda-
at Areskutan in Sweden (see references in Hatcher & Wil- mental types. Type C crystalline megathrust sheets form at
liams 1986). Whereas they may be geometrically similar, and the ductile-brittle transition (DBT), and possibly the base of
crystalline and foreland thrust sheets evolve behaviourally the lithosphere, and transport slabs of previously consoli-
toward each other during transport, there are important me- dated upper crust and cover sedimentary rocks. Type F
chanical differences in the way several kinds of crystalline crystalHne sheets form below the DBT in the realm of ductile
thrusts are generated and are transported. It has been sug- flow. These two kinds comprise most of the crystalline thrust
Trajectory of
future master
decollement
Late Proterozoic
and Future
Paleozoic Hayesville fault out-of-sequence
platform Cambrian rift
(Tactonic suture) thrust Colliding
basins
margin (Brevard fault) mass
North American
continental crust
previously thinned
by rifting—now/
restored to normal
thickness
b)
Cross section
Map view
Figure 1. (a) Ideal Type C sheet prior to formation. Dashed line indicates the projected location of detachment as it propagates along the ductile-brittle
transition (DBT) and onto the continental platform. Note the similarity of this diagram for a contractional regime to Figure 1 of Gibbs (1990), which shows
a linked fault model for an extensional regime, (b) Formation of a Type F sheet by differential flow in a ductile mass forming a fold, then shearing out of
the common limb between the antiform and synform forming a thrust. Cross section view is above; map view of lobate thrust sheets is below.
218
EVOLUTION OF CRYSTALLINE THRUST SHEETS
where e is strain rate, A is a material constant, a is normal Figure 2. (a) Depth-stress diagram showing the variation in properties of
stress, and n depends on the flow mechanism. A creepcreej law ""'^'^l ^""^ "^^"'''^ materials, assuming the crust is wet and quartz-bearing
that describes shear strain rate, Y can be expressed as in the upper part, dry and quartz-bearing in the lower part and the mantle is
composed of olivine and pyroxene. DBT - ductile-brittle transition. Dj -
thermal detachment. Dp - differential flow-related detachment. D^ - me-
chanical/compositional detachment, (b) Mechanical/compositional de-
tachment, (c) Thermal detachment, (d) Detachment related to differences
Y= Ax "e RT (2)
in flow rate.
219
R. D. HATCHER & R. J. HOOPER
result from contrasts in rock type in the depositional environ- Thermally induced anisotropy may be the principal
ment that form weak zones - bedding planes and mechanical mechanism worldwide for detachment in crystalline rocks in
(structural-lithic) unit contacts - within an otherwise strong orogenic belts and zones of crustal extension (Fig. 2c). The
sequence of sedimentary (or sedimentary-volcanic) rocks. zone of maximum anisotropic contrast occurs where thermal
These provide the sites for the classical detachment zones in softening of the crust reaches a threshold where ductile flow
foreland fold-thrust belts and accretionary wedge complexes is initiated - the DBT.
where weak rocks (shale, evaporite, overpressured zones) The third detachment mechanism - anisotropy induced by
localize detachment in sequences interlayered with strong differences in flow or strain rate - occurs below the DBT in
rocks (massive carbonates, sandstones, lava flows). Layer- rocks being deformed by uniform (passive) ductile flow (Fig.
ing is commonly subhorizontal and weak zones easily ac- 2d). Evidence for this mechanism is abundantly preserved in
commodate propagation of faults parallel to bedding. Excep- rocks from the internal parts of mountain chains formed
tions do occur: detachment has been noted in strong units in during at least the past 3 Ga of Earth history. Faults are
sedimentary sequences, as in the Keystone thrust in Nevada formed either compressionally by excision and simple shear
(Burchfiel ef a/. 1974), and attributed to high fluid pressure, producing a thrust fault along the common limb between an
but Brock & Engelder (1977) showed that thrust sheets are antiform and synform, or tensionally by ductile necking and
too permeable to maintain other than local overpressured simple shear producing ductile normal faults. Differences in
zones. flow or strain rate must be present to produce folds, or
Original compositional anisotropy may also arise in the structures other than foliation, in the realm of uniform flow.
lithosphere at the base of the continental crust (Fig. 2b), thus Detachment initiated from instability created by folding
providing an opportunity for detachment. This mechanism propagates as displacement increases and thrusts with tens of
has been proposed (for example Oxburgh 1972, 1974; Price kilometres of displacement may be produced. This mecha-
1986; Sacks & Secor 1990), but not documented, for crustal nism is discussed in greater detail below.
thrust detachment below the DBT, and forms an integral part
of both simple and pure shear models for crustal extension
(Wernicke 1985; Lister ef a/. 1986). Support for the idea of Fault rocks
crustal delamination in both extensional and contractional
crustal environments is provided by deep seismic reflection The character of faultrocks developed along thrusts vary with
data (e.g. Allmendingere?a/. 1987; Choukroune £•?«/. 1990). the thermal regime of the fault zone, the amount of fluid
Soper & Barber (1982), from BIRPS seismic reflection data, available in the fault zone, the shear strain rate, and the nature
suggested the Moine thrust formed as a detachment by of the rocks involved, including their composition, grain size,
delamination along the base of the continental crust. Butler and strength. While no unifying theory has been developed
& Coward (1984), using BIRPS seismic reflection data from to fully explain the processes related to the formation of fault
the MOIST survey, suggested an additional detachment oc- rocks, several recent papers address specific aspects of the
curred in the deep crust beneath the Moine thrust producing problem. The relationships between formation of mylonite
a crustal duplex. and cataclasite, and static and dynamic recrystallization of
600
- 500
Recovery Rate
• 425 ALMANDINE
Strain Rate
400
Biotite
300
T
200
0.01
(Ma)
Figure 3. Relationships between recrystallization-recovery rate, strain rate, temperature and total strain and the products of deformation along a fault. Point
A represents a rock that was deformed under conditions of increasing temperature. Mylonitic fabric would probably be annealed following cessation of
deformation, unless the rock mass cooled without increasing the temperature further so the existing textures would be preserved. Point B represents a rock
deformed following the thermal peak. Mylonitic (or metamorphic) texture would be preserved, and much of the strain in the rock may not be annealed.
220
EVOLUTION OF CRYSTALLINE THRUST SHEETS
mylonite and cataclasite have been discussed by Sibson below the DBT to more cool parts of the crust. For that
(1977, 1983) and Wise et al. (1984); Tullis & Yund (1985) reason, mylonite may be superposed by cataclasite along the
have discussed the relationships between feldspar deforma- same fault; alternatively, recurrent motion separated by long
tion, dynamic recrystallization, and the development of duc- time periods may erosionally unload a thrust sheet so that it
tile shear zones. Significant progress has been made in may deform brittlely when it is reactivated. Type F sheets
understanding the development of microfabrics in ductile form below the DBT so that fault rocks are commonly
fault rocks. Lister & Snoke (1984) addressed the develop- prograde (annealed) or retrograde mylonite. Occasionally,
ment of S-C fabrics (shear bands) and devised a classification both types of crystalline sheets develop with few or no fault
of mylonites that is rock-type dependent: quartzofeldspathic rocks along the contacts for significant distances. Either they
rocks produce type I S-C mylonite, whereas mica-rich rocks never developed or fault rocks formed, ceased to form, and
produce type II S-C mylonite. Hatcher (1978) and Hatcher & were removed by tectonic erosion along the fault as it moved.
Hooper (1981) first suggested that the ratio of recovery (or
recrystallization) rate to deformation (strain rate) (R/D) is
important in determining the resultant fault rock type (Fig. 3). Types of crystalline thrust sheets
Wise et al. (1984) considered this ratio a determining factor.
Passchier & Simpson (1986) and Hooper & Hatcher (1987), Two kinds of crystalline thrust sheets form in the intemides
in an attempt to explain the development of porphyroclast of mountain chains (Fig. 1): Type C - internally brittle slabs
systems, considered the R/D ratio a controlling factor in of intact crust (composite basement) that detach within the
determining the type of system that develops. The R/D ratio thermally softened ductile-brittle transition (DBT) and, once
is possibly the single most important parameter affecting the formed, behave as thin-skinned thrust sheets; and Type F -
character of fault rocks produced because the ratio incorpo- fold-related lobe-shaped thrust sheets that form below or
rates most of the variables affecting rock deformation. The within the DBT by attenuation of the common limb between
recrystallization rate, for example, is controlled by mineral- antiforms and synforms in passive- or flexural-flow folding
ogy, temperature, prior deformation history, fluid pressure, where transport is controlled by ductile flow. Type C thrust
confining pressure, and deviatoric stress. If the R/D ratio is sheets are commonly thought to form by continent-continent
very small, as might occur in the upper crust, brittle faults collision - or possibly arc-continent collision - accompany-
would develop. Ductile faults - and fault rocks - will form ing A-subduction, whereas Type F sheets form via A- or B-
under lower crustal conditions where the R/D ratio is larger subduction below the DBT. Ophiolite sheets, a subset of
(Fig. 3). Type C thrust sheets, are probably the products of B-
The timing of thrusting relative to the thermal peak also subduction. Crustal thickening accompanies formation of
affects the character of fault rocks that form. Thrusting may both types of thrust sheets.
occur prior to, during, or after the thermal peak in an orogen
affecting the degree (and rate) of recrystallization of fault
rocks that form along a crystalline thrust. The character of Type C sheets - crystalline megathrust sheets
fault rocks (texture, degree of recrystallization, incorporation
of ductile fragments in cataclasite) also provides clues about Individual Type C (composite) thrust sheets are very strong
the movement history of the fault. compared to thrust sheets in foreland fold-thrust belts, and, as
Most discussions of fault rocks deal with retrograde a consequence, are generally much larger than the latter, with
mylonite without considering their relationships to the proc- area to (restored) thickness ratios of from -3,000:1
esses that form prograde regional metamorphic rocks. A (Austroalpine, Moine) to -30,000:1 (Blue Ridge-Piedmont).
clear interrelationship exists between ambient temperature Maximum size and displacement are attained where detach-
and pressure (including fluid pressure) and strain rate along ment occurs over a wide area and the crystalline thrust ramps
a moving fault and the fault rocks produced (Fig. 3). into one or more weak zones in platform sedimentary rocks.
Mylonites formed prior to a thermal peak would initially Here crustal thickness increases significantly during thrust-
have the same retrograde character as those formed after the ing, but basal detachment (a) and surface slope (P) angles
peak, but available thermal energy would increase the rate of remain constant (near zero) - except in ramp zones - because
dynamic or static recrystallization so that the accumulated of the slab geometry (Figs la & 4) and non-Coulomb behav-
lattice strain and microtexture are annealed. Mylonites iour of the thrust mass. Foreland thrusts are driven ahead of
formed during the thermal peak may also anneal and produce a Type C sheet where crystalline and foreland thrust geom-
a microfabric that contains little evidence of shear zone etry merge and Coulomb wedge behaviour prevails above the
deformation. The megascopic fabric would, however, sur- DBT - as predicted by Davis et al. (1983).
vive providing most of the evidence of ductile shear. Retro- Thrust-related mesofabrics are mostly concentrated in or
grade mylonites preserve both a mega- and microfabric that near the fault zone at the base of the thrust sheet. One of the
is indicative of ductile deformation. best examples is the Jotun thrust sheet in southern Norway
Propagation of a thrust from the more ductile to the more where Middle Proterozoic structures and textures are pre-
brittle parts of the crust at the base of a Type C megathrust served except near the base of the sheet where all earlier
sheet may involve a long history of recurrent movement as fabrics become transposed into the strong L-S fault-related
well as the movement of rocks that were at one time within or fabrics formed during emplacement of the sheet (Milnes &
221
R. D. HATCHER & R. J. HOOPER
TypeC
crystalline sheet
Foreland fold—thrust Colliding
belt continent
or arc
Non-Coulomb
behavior
Coulomb
behavior
Detachment in
layered Detachment
rocks in
DBT
Figure 4. Diagrammatic representation of the relationships between a Type C thrust sheet that exhibits non-Coulomb behaviour and the transition to
Coulomb behaviour in the foreland fold-thrust belt being pushed ahead of the crystalline sheet, a - dip of basal decollement. p - surface slope angle.
Koestler 1985). More subtle late macro- and meso-scale It was emplaced as a product of Alleghanian (Late Carbonif-
structures (isolated domes, out-of-sequence thrusts, open erous-Permian) collision between Africa and North America.
folds, some crenellations, and ductile shears) related to thrust The BRP thrust sheet extends through the southern and
emplacement may occur within Type C sheets. central Appalachians from Alabama to Pennsylvania and
Thin platform assemblages restrict the size and displace- contains a number of earlier thrusts, as well as a series of
ment of Type C sheets and foreland fold-thrust belts in Alleghanian strike-slip faults. Once the crystalline sheet
general. The total width and displacement of the US ramps onto the platform, foreland deformation is driven in
Cordilleran thrust belt narrows as the aggregate thickness of front of and beneath the crystalline sheet (Fig. 5). The central
the stratigraphic section decreases southward. The width of Piedmont suture was once considered the eastern boundary of
the thrust belt is at the minimum in southeastern California the thrust sheet, but attempts to balance and palinspastically
where the Winters Pass thrust formed as one of the few restore the BRP sheet (Hatcher 1989; Hatcher et al. 1989)
basement thrusts in this part of the orogen. The width of the require that it root east of the central Piedmont suture, so the
thrust belt increases again southward paralleling the increase eastern edge of the BRP is thus probably the Alleghanian
in thickness of the platform assemblage (Burchfiel & Davis suture now buried beneath the Coastal Plain. Palinspastic
1971, 1975). A similar phenomenon occurs in the central reconstruction yielded a minimum displacement of 350 to
Appalachians. Displacement in the foreland fold-thrust belt 400 km for the part of the sheet with the greatest displace-
diminishes from the south to the Reading Prong and Hudson ment.
Highlands in the north where the basement and thin cover The western edge of the BRP thrust sheet consists of the
become involved in thrusting (Drake etal. 1988; Hatcher 1989) eastern part of the foreland fold-thrust belt. This segment
The overall geometry of the fault systems carrying Type C exhibits classic thin-skinned behaviour, as the master detach-
crystalline thrust sheets is similar to that of crustal-scale fault ment ramped from the basement and older (Upper Protero-
systems recognized in extensional terranes. Gibbs (1990) zoic-Lower Cambrian) rift-drift facies sedimentary rocks
provided a general model for large-scale linked fault systems into the Lower Cambrian master decollement of the foreland.
in extensional terranes where discrete brittle faults above the Most of the BRP consists of Palaeozoic basement generated
DBT, forming asymmetrical half-graben basins, are linked to during earlier orogenies (Grenville, Taconian, Acadian) that
detachments in the middle to lower crust at or below the DBT. mostly behaved as an intact homogeneous crust during the
Strain is taken up in the lower crust either along a shear zone Alleghanian event. Structures and plutons contained within
(or detachment) for throughgoing simple shear of the crust, or it were therefore cut through by the BRP master decollement.
in some cases by more continuous deformation in a broad The BRP detachment propagated westward along the ductile-
zone of ductile pure shear. The overall geometry of the brittle transition in a thickened softened crust of Grenville
extensional systems (see Gibbs 1990, Fig. 1) is remarkably and pre-Alleghanian Palaeozoic metamorphic rocks before it
similar to the fault systems in contractional orogens associ- ramped from the more western (cooler? & stronger?) base-
ated with Type C thrust sheets where ductile faults at or below ment into weak units in the sedimentary cover (Fig. 5b).
the DBT may merge with a foreland fold-thrust belt in the Most of the internal parts of the BRP thrust sheet contain
upper crust (Fig. 1). little evidence that it is a huge intact thin (2-12 km thick)
sheet of crystalline rock, largely because few obviously
Alleghanian structures are present in the internal parts of the
Examples
sheet. The Brevard fault, a complex fault zone having a
Blue Ridge-Piedmont sheet, Southern and Central Appala- possible earlier history of dip-slip motion and an Alleghanian
chians. The Blue Ridge-Piedmont (BRP) thrust sheet (Fig. history of both dip and strike-slip motion, is a major
5) is one of the largest intact Type C thrust sheets in the world. Alleghanian out-of-sequence structure that resides in the
222
a)
10 0 so MILES
10 0 100 KILOMETERS
Suture?
A'
A\T
Platform
Sedimentary rocks
Moho- Crustal-
duplex
50 100 km
Figure 5. (a) Outline map of the southern and central Appalachians showing the extent of the Blue Ridge-Piedinont thrust sheet. MCW - Mountain City window. GMW - Grandfather Mountain window. PM - Pine Mountain
window. Blue Ridge-Piedmont thrust sheet is shown by a diagonal line pattern. A-A' is the location of the section in (b). D-D' is the line of section in Figure 8(a). The labelled boxes indicate the locations of Figures 8(b)
& 9(a). (b) Section through the southern Appalachians at the widest point of the Blue Ridge-Piedmont thrust sheet (patterned) showing only Alleghanian (Late Carboniferous-Permian) structures and the dominance of the
Type C crystalline sheet in this part of the orogen.
R. D. HATCHER & R. J. HOOPER
BRP sheet. The carbonate slices from the Brevard fault zone anomalies may be recognized that are related to surface
provide strong evidence that the Palaeozoic carbonate plat- geological features. East of the central Piedmont suture
form extended at least as far east as the western Piedmont. numerous surface geological features are directly correlative
This evidence, coupled with the presence of Lower Cambrian with magnetic and gravity anomalies indicating the BRP
rocks within the leading edge of the BRP thrust sheet, thrust sheet is much thicker (>15 km) (Hatcher & Zietz 1980;
requires that the BRP sheet root east of the easternmost extent Iverson & Smithson 1982).
of the early Palaeozoic carbonate bank succession - probably A combination of surface geological and seismic reflec-
east of the present location of the central Piedmont suture tion data indicate antiformal-stack duplexes (ASD) formed in
(Hatcher er a/. 1989). the platform sequence in response to emplacement of the
Numerous windows in the BRP sheet provide direct evi- Type C thrust sheet. ASDs arch the Type C sheet into late
dence of the nature of the footwall or intermediate horse domes that deform the composite sheet by interaction of the
rocks, P-T conditions during emplacement of the thrust sheet overriding BRP thrust sheet with the footwall rocks. Duplex
(anchizone to lower greenschist), and also provide evidence formation in the footwall domes the overriding thrust sheet by
for minimum displacement. Rocks present in windows in the folding or arching. Isolated domes in Type C sheets previ-
BRP thrust sheet belong to the North American margin. This ously thought to be formed by fold interference may have
evidence confirms that the crystalline BRP sheet was em- formed by this mechanism. Extensive duplexing beneath
placed onto North American platform margin rocks. several unbreached domes imaged by seismic reflection data
Seismic reflection profiles indicate packages of seismic and erosionally breached antiforms - now windows such as
reflectors present beneath the BRP sheet that are identical to the Grandfather Mountain and Mountain City windows -
reflector packages in the foreland where reflectors in the within the BRP sheet and along the external fringes of the
sedimentary succession are traceable to the surface (Costain sheet are cored by duplexes. A similar arching mechanism
etal. 1989). The reflector packages correlate with the Lower may also have played a part in formation of the Engadine and
Cambrian through Lower Ordovician clastic-carbonate suc- Tauem windows in the Alps.
cession. These reflectorpackages are traceable with certainty Austroalpine sheet, Eastern Alps. The eastern Alps are
using the COCORP data (Cook et al. 1979,1983) at least to dominated by the composite Austroalpine thrust sheet (Fig.
the central Piedmont suture. Potential-field (gravity and 6) that was formed during the Eocene. Variscan basement and
magnetic) data also provide information about the relative cover of the Lombardy (or Camic) terrane were transported
thickness of the BRP thrust sheet (Hatcher & Zietz 1980). In westward transport accompanied emplacement of internal
general, the BRP west of the central Piedmont suture is sheets (Schliniq thrust), unroofing of the Austroalpine sheet
relatively thin (< 10 km), because only subdued magnetic around the windows. Schmid & Haas (1989) associated the
Figure 6. (a) Outline map of the Alps with the patterned area showing the distribution of the Austroalpine sheet and related structures (patterned) and external
basement massifs (black). B-B' is the section in Figure 6(b). C-C is the section in Figure 7. (After Laubscher 1988.) IL - Insubric fault zone. E - Engadine
fault. EN - Engadine window. TW - Tauem window. S - Schlinig thrust. Q/. - Otztal nappe. T - Turin. M - Milan. G - Genoa. Mu - Munich. V - Venice. B
- Belgrade. K - Kishinev.
224
EVOLUTION OF CRYSTALLINE THRUST SHEETS
225
R. D. HATCHER & R. J. HOOPER
Anligorio
STMPLON.TICINO NAPPES
Figure 7, Section through part of the western Alps showing the structural style of Type F Pennine thrust sheets and structures to the west. (From Homewood
etal. 1980.)
and more representative of transport and deformation of the higher grade central and southern portions of the zone (Spicher
entire sheet than mesofabrics in Type C sheets, with sheath 1972).
folds a common meso- and macro-style, although cylindrical Appalachian Inner Piedmont. The Appalachian Inner
folds with orogen-parallel lineations normal to transport or Piedmont and adjacent Chauga belt (Fig. 8a) consists of a
mixed-mode linear fabrics may be present (Hatcher & Tabor stack of amphibolite facies (garnet to sillimanite zone) re-
1989). The close relationships of internal fabrics to displace- cumbent Type F sheets, several of which record a history of
ment on large faults and formation of strongly noncylindrical post-metamorphic emplacement. Griffin (1971, 1978) rec-
macroscopic folds is restricted to faults that are emplaced at ognized that the Inner Piedmont is dominated by several large
or near the thermal peak. Fault fabrics may be totally Type F structures forming the core and flanks of the zone. He
annealed or nonexistent because the entire moving rock mass also showed that a strong component of penetrative
is in the same P-T state. inhomogeneous strain affected the rocks here and produced
Post-metamorphic faults in these internal zones will trun- a dominance of non-cylindrical mesofolds. The largest of
cate the more penetrative fabrics and form retrograde, albeit these structures, the Six Mile-Alto thrust sheet is late- orpost-
frequently ductile, assemblages and new fault fabrics along metamorphic and was emplaced in a semi-rigid condition
the later faults. These effectively behave as Type C sheets (Hopson & Hatcher 1988), truncating earlier more ductile
because they have cooled appreciably, although frequently structures. This sheet therefore evolved from a Type F to a
only to lower amphibolite or greenschist facies conditions. Type C sheet after it formed and cooled during transport.
Fault rocks may contain stable garnet-biotite that retrogrades Because quartz-mica mineral lineations are of mixed charac-
middle to upper amphibolite facies - or higher grade - assem- ter (Fig. 8b, Davis et al. 1989) - some fold-related, others
blages. transport-related - heterogeneous ductile flow was involved
in the emplacement of the early recumbent structures here.
Examples The underlying Walhalla nappe (Figs. 8a & 9a) remained as
a Type F thrust sheet during its entire history of movement,
Pennine Alps. The structure of the Pennine zone in the and was later overridden by the Six Mile-Alto sheet. Part of
Alps (Fig. 7) is dominated by several large recumbent Type the Walhalla nappe is faulted along the common limb, while
F fold-thrust structures that have the typical map (e.g. Spicher part of it was not faulted (Fig. 9b).
1972) and cross-section structure described above. They Central Gneiss Belt Domain, Grenville Province. The
consist mostly of bulbous fold nappes of isoclinally folded central gneiss belt of the late Middle Proterozoic Grenvilte
sediments, remobilized Variscan basement and Mesozoic province of eastern North America (Fig. 10a) contains a
cover sediments, and deep ocean sedimentary and ophiolitic series of lobate NW-directed Type F sheets (Culshaw et al.
assemblages (Milnes 1974; Trlimpy 1980). All units thin 1983; Davidson et al. 1982; Rivers et al. 1989). These
markedly into the root zone (Fig. 7). Transport directions are structures, like many in similar zones in other mountain
strongly imprinted in the mesofabrics, particularly in the chains, are probably large noncylindrical to sheath structures
226
EVOLUTION OF CRYSTALLINE THRUST SHEETS
a)
j"" ""•'"•*••~^ Root
"^''^--•i. "---, Zone
Carolina
y\\ T Terrane
Figure 8. (a) Section through the Inner Piedmont in South Carolina. Type F sheets have been modified by later open folding and movement on the
Alteghanian Brevard, Towaliga, and Blue Ridge-Piedmont faults. The Walhalla, Six Mile, and Anderson nappes were named by Griffin (1978).
b)
N
Figure 8. (b) Multiple lineation trends in the western Inner
Piedmont in North Carolina immediately north of the South
Carolina border (from unpublished data courtesy of T. L. Davis
and J. R. Tabor). (Locations shown in Fig. 5(a).)
0 1 Mile
1 0 1 Km
35° 15'
82° 22' 30
227
iililiWm Mile
mm •^_
^/ 34" 45"
83"
^'y\
W^'H-
Figure 9. (a) Geological map of the western Inner Piedmont and Chauga bell (including Brevard fault zone} in South Carolina and Georgia showing
the major structures. (Location shown in Fig. 5(a).)
S
^
EVOLUTION OF CRYSTALLINE THRUST SHEETS
Figure 9. (b) Cross sections E-E' and F-F' through the Walhalla nappe (locations shown in Fig. 9(a)). am - amphibolite, hornblende gneiss, and biotite
gneiss, tf - Tallulah Falls Formation, gg - Palaeozoic granitoids (cross-hatch pattern), hg - Henderson Gneiss (stippled), pm - Poor Mountain Formation, cr
- Chauga River Fonnaiton. mg - mylonite gneiss. Oc - Ordovician platform carbonate horse (in Brevard fault zone). Six Mile thrust sheet and Alto allochthon
are vertical lined. Toothed lines are thrusts (teeth on upthrown side).
229
R. D. HATCHER & R. J. HOOPER
Figure 10. (b) Structural trends reflecting strong body deformation in parts of the Algonquin Seguin,
Rousseau, and Moon River Type F sheets in the southern Grenville Province in Ontario. (From Culshaw
etal. 1983.)
sequence thrusts, but the number that exist falls far below the sheet. The latter make it impractical to ignore the cohesion
number required by the sand-box models and theory. Part of term, S,,, in the Coulomb equation.
the inability to apply Coulomb theory to real thrust systems
resides in the difficulty in estimating coefficients of friction |x| = S„ + |i(a„ - a) (3)
for the materials involved and the strengths of the different
elements that compose the thrust system. Dahlen et al. (1984) where x is shear stress, u is the coefficient of friction, o is
discussed this problem and noted that the coefficient of normal stress, and Oj is pore fluid stress. As the thrust moves
friction at the base of a moving thrust sheet, \i^ may be as low and crosses the first ramp into the rifted margin and platform
as 0.3 to 0.6, based on experiments with clays and clay-rich sedimentary rocks, the sheet enters the wedge assemblage
fault gouge, whereas the coefficient of friction of the sedi- and will assume the behaviour of foreland sheets, except for
mentary rocks of the sheet averages 0.85, as predicted by their dimensions. The footwall at this point will be localized
Byerlee's law. in either an ordinary weak unit in the platform succession, or
The applicability of Coulomb wedge theory to crystalline will climb over footwall ramps like foreland sheets.
thrust sheets has many of the same difficulties as for foreland The geometry of Type C sheets could therefore be mod-
fold-thrust belts. Type C megathrusts are generated along the elled like foreland sheets, as Elliott & Johnson (1980) have
DBT as non-Coulomb detachments with low basal coefficients assumed for the Moine thrust. Butler (1986) suggested that
of friction (probably <0.3), but with high strengths and for the Moine, there is thrust-related internal deformation
coefficients of internal friction (probably >0.85) within the (ductile shears and thrusts) in the basement rocks thereby
230
EVOLUTION OF CRYSTALLINE THRUST SHEETS
Hybrid
TypeC
TypeF
Crustal Duplexes
Ductile Brittle
Figure 11. Crustal thiclcening processes involving (a) Type C, (b) Type F, Figure 12. Rheological interrelationships between different kinds of thrust
and (c) crustal duplex structures. sheets. Note that some thrust sheets may evolve toward more brittle
behaviour during transport and emplacement. Hybrid behaviour includes
elasticoviscous and elastic-plastic behaviour; ductile includes both viscous
indicating greater displacement than was previously thought. and plastic.
Wojtal & Mitra (1986) related internal deformation to strain
softening and strain hardening of thrust sheets, and used the
occurrence of ductile shears and fracture sets coated with size and composition, Type F thrust sheets may form in a
low-temperature mineral assemblages as evidence of intemal more ductile realm as ductile internally deforming sheaths,
deformation in the western part of the Appalachian Blue then evolve during progressive deformation and transport
Ridge-Piedmont Type C megathrust sheet. This kind of into more coherent sheets that exhibit Type C hybrid behav-
deformation is also present in the basement of the Appalachian iour.
Reading Prong in Pennsylvania and New Jersey, but here it is
more difficult to find and to show that it is related to the master
Type C megathrust in the intemal parts of larger sheets, like Conclusions
the Blue Ridge-Piedmont sheet. Away from the thinner
leading edges of such sheets they appear to have remained I). Crystalline thrust sheets form as products of A- or B-
intact with less intemal deformation, although ductile de- subduction processes.
formation zones with young ""Ar/^'Ar ages (Dallmeyer 1989) 2). Crystalline sheets form mostly by non-Coulomb behav-
have been recognized in the intemal parts of the Blue Ridge- iour, but may evolve during transport toward Coulomb be-
Piedmont megathrust sheet. haviour.
Crystalline thrust sheets can be related to each other and to 3). Detachment occurs along zones of original mechanical
other kinds of thrusts in a rheological continuum that ranges weakness, thermally induced anisotropy, and by variation in
from brittle behaviour at one end to ductile (plastic or vis- rates of ductile flow.
cous) behaviour at the other (Fig. 12). Foreland thrust sheets 4). Type C sheets are generally slabs of consolidated base-
plot closer to the elastic end, while ideal Type F sheeets would ment that detach from the DBT or the base of the lithosphere
plot near the ductile end. Lava flows could be considered a as rigid masses that drive foreland deformation.
kind of superductile Type F thrust sheet, because they move 5). Type C sheets are always larger than the largest foreland
over a discontinuity of simple shear (the ground surface) and sheets in the same orogen, and thus are some of the largest
are internally deformed by inhomogeneous simple shear - structures on Earth.
hence their lobate shape in map view that resembles the 6). Type F sheets are lobate, sheath structures that form below
shapes of flattened sheath folds and Type F thrust sheets. the DBT by variation in rate of inhomogeneous ductile flow.
Flow patterns are similar, but the rheology is obviously 7). Type C and F sheets function in crustal thickening and
different. underplating processes. Crustal duplexes form as byproducts
Many thrusts that form as Type C or Type F thrust sheets of the formation of Type C sheets and also function in the
undergo a change in behaviour mode during progressive thickening process.
deformation and transport. This change is probably driven by 8). Crystalline thmst sheets can be related to each other and
decreasing temperature and pressure as the sheet moves to other kinds of thrusts in a rheological continuum that
upward in the crust. Type C megathmst sheets may become ranges from elastic behaviour at one end to ductile (plastic or
more elastic as they ramp onto a continental platform and viscous) behaviour at the other. They may form in one mode
become geometrically indistinguishable from thrust sheets in and evolve into another during progressive deformation and
the foreland formed by fault-bend folding. Except for their transport.
231
R. D. HATCHER & R. J. HOOPER
Research on crystalline thrust sheets has been supported by the US National Fund, administered by the American Chemical Society. Comments on an
Science Foundation grants EAR-7615564, EAR-7826316, EAR-7911802, early version of the manuscript by Rod Gayer, Ken McClay and an anony-
EAR-810852, EAR-8206949, EAR-8305832, EAR-8417894, and EAR- mous reviewer resulted in considerable improvement of the manuscript.
8816343 to RDH. Support by University of Tennessee/Oak Ridge National Secretarial assistance by Nancy L. Meadows and graphics assistance by
Laboratory Center of Excellence stipend is also gratefully acknowledged. Donald G. McClanahan in preparation of this paper are very much appreci-
Research by RJH has been supported by Donors of the Petroleum Research ated. The authors remain culpable for all errors of fact or interpretation.
References
Allmendinger, R. W., Hauge, T. A., Hauser, E. C , Potter, C. J., Klemperer, Dallmeyer, R.D. 1989. Polymetamorphic evolution of the western Blue
S. L., Nelson, K. D„ Kneupfer, P. & Oliver, J. 1987. Overview of the Ridge allochthon: Evidence from '"'Ar/^'^Ar mineral ages. Geological
COCORP 40° N transect, western United States: The fabric of an Society of America Abstracts with Programs, 21, 11.
erogenic belt. Geological Society of America Bulletin, 98, 308-319. Davidson, A., Culshaw, N. G. & Nadeau, L. 1982. A tectoncv-metamorphic
Ando, C. J., Cook, F. A., Oliver, J.E., Brown, L. D. & Kaufman, S. 1983. framework for part of the Grenville Province, Parry Sound region,
Crustal geometry of the Appalachian orogen from seismic reflection Ontario. Geological Survey of Canada Current Research Part B, Paper
studies. In: Hatcher, R. D., Jr., Williams, H. & Zietz, I. (eds) Contri- 82-lA, 175-190.
butions to theTectonicsandGeophysics ofMountain Chains, Geological Davis, D., Suppe, J. & Dahlen, F. A. 1983. Mechanics of fold-and-thrust
Society of America Memoir 158, 83-102. beltsandaccretionarywedges./()«rna/o/G^op/ivi/ca//?M«'arc/j,88,ll53-
Armstrong, R. L. & Dick, H. J. B. 1974. A model for the development of thin 1172.
overthrust sheets of crystalline rock. Geology, 1, 3 5 ^ 0 . Davis, T. L., Tabor, J. R. & Hatcher, R. D., Jr. 1989. Orogen-parallel to
Boyer, S. E. & Elliott, D. 1982. Thrust systems. American Association of orogen-oblique ductile deformation and possible late Palaeozoic (?)
Petroleum Geologists Bulletin, 66, 1196-1230. ductile deformation of the western Piedmont, southern Appalachians.
Brock, W. G. & Engelder, T. 1977. Deformation associated with the move- Geological Society of America Abstracts with Programs. 21, A-65.
ment of the Muddy Mountains overthrust in the Buffington window, Drake,A. A.,Jr.,Hall.L.M. &Nelson,A.E. 1988. Basement and basement-
southeastern Nevada. Geological Society ofAmerica Bulletin. 88,1667- cover relation map of the Appalachian orogen in the United States. US
1677. Geological Survey Map, I-1655, scale 1:1,000,000.
Brunei, M. & Kienast, J-R. 1986. Etude petro-structurale des chevauchements Dusel-Bacon, C. & Aleinikoff, J. N. 1985. Petrology and tectonic signifi-
ductiles himalayens sur la transversale de I'Everest - Makalu (Nepal) cance of augen gneiss from a belt of Mississippian granitoids in the
oriental. Canadian Journal of Earth Sciences, 23, 1117-1137. Yukon-Tanana terrane of east-central Alaska. Geological Society of
Burchfiel, B. C. & Davis, G. A. 1971. Clark Mountain thrust complex in the America Bulletin, 96, 411^25.
Cordillera of southeastern California: Geological summary and field trip Elliott, D. & Johnson, M.R.W. 1980. Structural evolution in the northern part
guide. In: Elders, W. A. (ed.) Geological excursions in southern Cali- ofthe Moine thrust belt, NW Scotland. Transactions ofthe Royal Society
fornia, University of California Riverside Campus Museum Publica- of Edinburgh: Earth Sciences. 71, 69-96.
tions No. 1, 1-38. Fleuty, M. J. 1964. Tectonic slides. Geological Magazine. 101, 452-456.
1975. Nature and controls of Cordilleran orogenesis, western United Gibbs, A. D. 1990. Linked fault families in basin formation. Journal of
States: Extensions of an earlier synthesis. American Journal of Science, Structural Geology, 12, 795-801.
275-A,363-396. Griffin, V. S., Jr. 1971. The Inner Piedmont of the southern crystalline
, Fleck, R. J., Secor, D. T., Vincelette, R. R. & Davis,G. A. 1974. Geology Appalachians. Geological Society of America Bulletin. 82, 1885-1898.
ofthe Spring Mountains, Nevada. Geological Society ofAmerica Bulle- 1978. Detailed analysis of tectonic levels in the Appalachian Piedmont.
lin,»5, 1013-1022. Geologische Rundshau, 67, 180-201.
Butler, R.W.H. 1983. Balanced cross-sections and their implications for the Hatcher, R. D., Jr. 1978. Reply. Eastern Piedmont fault system: Speculations
deep structure of the northwest Alps. Journal of Structural Geology, 5, on its extent. Geology, 6, 580-582.
125-137. 1989. Tectonic synthesis of the US Appalachians, Chapter 14, In:
1986. Structural evolution ofthe Moine thrust of northwest Scotland; Hatcher, R. D., Jr., Thomas, W. A. & Viele, G. W. (eds) The Appala-
a Caledonian linked thrust system? Geological Magazine, 123, 1-11. chian-Ouachita orogen in the United Slates. Geological Society of
& Coward, M. P. 1984. Geological constraints, structural evolution, America, Boulder, The Geology of North America. F-2, 511-535.
and deep geology ofthe northwest Scottish Caledonides. Tectonics, 3,347- & Hooper, R. J. 1981. Controls of mylonitization processes: Relation-
365. ships to orogenic thermal/metamorphic peaks. Geological Society of
Chappie, W. M. 1978. Mechanics of thin-skinned fold-and-thrust belts. America Abstracts with Programs, 13, 469.
Geological Society of America Bulletin, 89, 1189-1198. & Tabor, J. R. 1989. Orogen-parallel coaxial mineral lineations and
Choukroune, P., Roure, F., Pinet, B. & ECORS Pyrenees Team 1990. Main folds - Products of buckling and passive folding in coaxial and noncoaxial
results ofthe ECORS Pyrenees profile. Tectonophysics, 173,411^23. flow. Geological Society of America Abstracts with Programs. 21, A—
Cook, F. A., Albaugh, D. S., Brown, L. D., Kaufman, S., Oliver, J. E. & 177.
Hatcher, R. D., Jr. 1979. Thin-skinned tectonics in the crystalline & Williams, R. T. 1986. Mechanical model for single thrust sheets Part
southern Appalachians: COCORP seismic-reflection profiling of the I: Crystalline thrust sheets and their relationships to the mechanical/
Blue Ridge and Piedmont. Geology, 7, 563-567. thermal behaviour of orogenic belts. Geological Society of America
, Brown, L. D., Kaufman, S. & Oliver, J. E. 1983. The COCORP Bulletin. 97,975-985.
seismic-reflection traverse across the southern Appalachians. American & Zietz, I. 1980. Tectonic implications of regional aeromagnetic and
Association of Petroleum Geologists Studies in Geology. 14, 1-61. gravity data from the southern Appalachians. In: Wones, D. R. (ed.) The
Costain, J. K., Hatcher, R. D., Jr. & goruh, C. 1989. Appalachian ultradeep Caledonides in the USA. Virginia Polytechnic Institute Department of
core hole (ADCOH) Project site investigation regional seismic lines and Geological Sciences Memoir, 2, 235-244.
geological interpretation. In: Hatcher, R. D., Jr., Viele, G. W. & Thomas, Thomas, W. A., Geiser, P. A., Snoke, A. W., Mosher, S. & Wiltschko,
W. A. (eds) The Appalachian-Ouachita orogen in the United States. D.V. 1989. Alleghanianorogen. In: Hatcher,R.D.,Jr.,Thomas, W. A.
Geological Society of America, Boulder, The Geology of North America, & Viele, G. W. (eds) The Appalachian-Ouachita orogen in the United
F-2, Plate 8. States. Geological Society of America, Boulder, The Geology of North
Culshaw, N. G., Davidson, A. & Nadeau, L. 1983. Structural subdivisions of America, F-2,233-318.
the Grenville province in the Parry Sound - Algonquin region, Ontario. Homewood, P., Gosso, G., Escher, A. & Milnes, A. 1980. Cretaceous and
Geological Survey of Canada Current Research Part A, Paper 83-1B, Tertiary evolution along the Besan^on-Biella traverse (Western Alps).
243-252. Eclogae geologicae Helvetiae, 73, 635-649.
Dahlen, F. A., Suppe, J. & Davis, D. 1984. Mechanics of fold-and-thrust Hooper, R. J. & Hatcher, R. D., Jr. 1988. Mylonites from the Towaliga fault
belts and accretionary wedges: Cohesive Coulomb theory. Journal of zone, central Georgia; Products of heterogeneous non-coaxial deforma-
Geophysical Research. 89, 10,087-10,101. tion. Tectonophysics, 152, 1-17.
232
EVOLUTION OF CRYSTALLINE THRUST SHEETS
Hopson, J. L. & Hatcher, R. D., Jr. 1988. Structural and stratigraphic setting Sacks, P. E. & Secor, D. T., Jr. 1990. Delamination in collisional orogens.
of the Alto allochthon, northeast Georgia. Geological Society of America Geology. 18,999-1002
Bulletin, 100, 339-350. Schmid, S. M. & Haas, R. 1989. Transition from near-surface thrusting to
Iverson, W. P. & Smithson, S. B. 1982. Master decollement root zone intrabasement decollement, Schliniq thrust, eastern Alps. Tectonics, 8,
beneath southern Appalachians and crustal balance. Geologv, 10, 241- 697-718.
245. Selverstone, J. 1988. Evidence for east-west crustal extension in the eastern
Laubscher, H. 1988. Material balance in Alpine orogeny. Geological Society Alps: Implications for the unroofing history of the Tauem window.
of America Bulletin. 100, 1313- 1328. Tectonics. 7, 87-105.
Lister, G. S. & Snoke, A. W. 1984. S-C mylonites. Journal of Structural Sibson, R.H.I 977. Fault rocks and fault mechanisms. Geological Society of
Geology. 6, 617-638 London Journal. 133, 191-213.
, Etheridge, M. A. & Symonds, P. A. 1986. Detachment faulting and the 1983. Continental fault structure and the shallow earthquake source.
evolution of passive continental margins. Geology, 14, 246-250. Geological Society of London Journal. 140, 741-767.
Milnes, A. G. 1974. Structure of the Pennine zone (central Alps): A new Soper, N. J. & Barber, A. J. 1982. A model for the deep structure of the Moine
working hypothesis. Geological Society of America Bulletin, 85, 1727- thrust zone. Geological Society of London Journal. 139, 127-138.
1732. Spicher,A. 1912.TektonischeKartederSchweiz.'Wepfa.ndCompany,Basel.
&Koestler,A.G. 1985.Geological structureofJotunheimen, southern Schweizerische Geologische Kommission. scale 1/500,000.
Norway (Sognefjell-Valdres cross section). In: Gee, D. G. & Sturt, B. Suppe, J. 1985. Principles ofStructural Geology. Prentice-Hall, Englewood
A. (eds) The Caledonide orogen - Scandinavia and related areas. John Cliffs, New Jersey, 537p.
Wiley & Sons, London, 457^74. Templeman-Kluit, D. J. 1974. The Yukon crystalline terrane: Enigma in the
Milnes, A. G. & Pfiffner, O. A. 1980. Tectonic evolution of the Central Alps Canadian Cordillera. Geological Society of America Bulletin, 87, 1343-
in the cross section St. Gallen-Como. Eclogae geologicae Helvetiae, 73, 1357.
619-633. 1979. Transported cataclasite, ophiolite and granodiorite in Yukon:
Monger, J.W.H. 1984. Cordilleran tectonics: A Canadian perspective. Societe Evidence of arc-continent collision. Geological Survey of Canada
Geologique de France Bulletin. 26, 255-278. Paper, 79-14. 27.
Mortensen, J. K. & Jilson, J. A. 1985. Evolution of the Yukon-Tanana Triimpy, R. 1975. Penninic-Austroalpine boundary in the Swiss Alps: A
terrane: Evidence from southeastern Yukon Territory. Geology. 13, presumed former continental margin and its problems. American Jour-
806-810. nal of Science. 275-A, 209-238.
Mulugeta, G. 1988. Modeling the geometry of Coulomb thrust wedges. (ed.) 1980. Geology of Switzerland, a guidebook. Part A: An outline
Journal of Structural Geology. 10, 847-859. of the geology of Switzerland. Wepf & Company, Basel, Schweizerische
Nicolas, A. 1987. Principles of Rock Deformation. D. Reidel Publishing Geologische Kommission. I04p.
Company, 208p. Tullis, J. & Yund, R. A. 1985. Dynamic recrystallization of feldspar: A
Oxburgh, E. R. 1972. Flake tectonics and continental collision. Nature, 239, mechanism for ductile shear zone formation. Geology, 13, 238-241.
202-204. Wernicke, B. 1985. Uniform-sense normal simple shear of the continental
1974. The plain man's guide to plate tectonics. Proceedings of the lithosphere. Canadian Journal of Earth Sciences, 11, 108-125.
Geological Association. 85, 299-357. Willis, B. 1893. The mechanics of Appalachian structure. US Geological
Passchier, C. W. & Simpson, C. 1986. Porphyroclast systems as kinematic Survey ISlh ArmualReport 1891-1892. Part 2. 212-281.
indicators. Journal of Structural Geology. 8, 831-843. Wise, D. U., Dunn, D. E., Engelder, J. T., Geiser, P. A., Hatcher, R. D., Jr.,
Price, R. A. 1986. The southeastern Canadian Cordillera; Thrust faulting, Kish, S. A., Odom, A. L. & Schamel, S. 1984. Fault-related rocks:
tectonic wedging, and delamination of the lithosphere. Journal of Suggestions for tenninology. Geology, 12, 391-394.
Structural Geology. 8, 239-254. Wojtal S. & Mitra, G. 1986. Strain hardening and strain softening in faults
Rivers, T., Martignole, J., Gower, C. F. & Davidson, A. 1989. New tectonic zones from foreland thrusts. Get)/o^(ca/5o(;7>fyo//lfl(OTcaS«/to(«. 97,
divisionsoftheGrenville province, southeast Canadian Shield. rfc'WrtK'i-, 674-687.
8, 63-84. Woodward, N. B. 1987. Geological applicability of critical-wedge thrust-
Rutter, E. M. 1986. On the nomenclature of mode of failure in rocks. belt models. Geological Society of America Bulletin, 99, 827-832.
Tectonophysics. Ill, 381-387.
233
PART FOUR
Pyrenees
Evolution of a continental collision belt: ECORS-Pyrenees crustal
balanced cross-section
Josep Anton Muiioz
Abstract: Construction of a crustal balanced cross-section across the Pyrenean chain shows a
minimum shortening of 147 km, 112 km of which are related to stacking of basement thrust sheets
in the southern Pyrenees. Metamorphic conditions of the basement rocks, as well as thrust
geometry, indicate the maximum depth for the detachment level to be at 15 km. In the restored
cross-section, the upper crust is 110 km longer than the lower layered crust. The lower crust was
subducted together with the lithospheric mantle into the asthenospheric mantle and has not been
imaged by geophysical data probably because an increase of density through eclogitic meta-
morphism. The upper crust constitutes an orogenic lid mainly deformed by thrust structures. The
balanced cross-section has been constrained by the ECORS deep reflection seismic profile as well
as detailed surface data and available commercial seismic and oil well data. The restored cross-
section provides a better picture of the middle Cretaceous combined strike-slip and extensional
fault system as well as of the Hercynian crust. Hercynian geological features have been used as
an additional tool for the restoration of the basement thrust sheets.
The Pyrenees form an Alpine collision belt located in be- Crustal balanced cross-sections can be constructed simi-
tween the Iberian and European plates. Formation of the larly to the sections of the frontal parts of the orogenic belts
Pyrenean orogen is related to the kinematics of the Iberian (Dahlstrom 1969; Hossack 1979), but the assumptions nor-
plate, largely dependent on the motion of the larger neigh- mally adopted in the balancing methodology strongly limit
bouring plates (Eurasia and Africa). Knowledge of the past the restoration of complete orogenic belts. In most of the
positions of the plates on both sides of the Pyrenees would collision belts, the deep crustal structure is poorly constrained
constrain the amount of convergence and the amount of and the internal parts are thermally remobilized and highly
shortening involved in the collision. Convergence occurred deformed by ductile structures. In the case of the Pyrenees,
from Campanian to Early Miocene time, and resulted in a the ECORS profile provides data to constrain the crustal
partial closure of the Bay of Biscay along the North Spanish geometry at depth. Moreover, deformation and thermal
subduction zone (Boillot & Capdevila 1977), and into conti- processes in the internal parts were not strong enough to
nental collision in the Pyrenean chain. The exact separation destroy the pre-Alpine features completely. In fact, the
distance between the Iberian and European plates at the Pyrenees are one of the few mountain chains where no
beginning of the convergence is not very well constrained, metamorphic or plutonic processes have occurred during
although several reconstructions of Iberia deal with the collision. All these peculiar characteristics of the Pyrenean
amount of rotation and sinistral displacement of the Iberian chain allow us to construct a complete crustal balanced cross-
plate with respect to Europe during Cretaceous times. Differ- section which has provided further insight into the tectonic
ent solutions point to a separation between 100 km and 150 evolution of the chain.
km (Grimaud et al. 1982; Olivet et al. 1984; Boillot 1986). Complete crustal balanced cross-sections have been drawn
Thrust structures and related synorogenic materials are during the last decade for the Alps (Beach 1981; Butler 1986;
very well preserved and, as a consequence, the Pyrenean Menard & Thouvenot 1987), the Himalayas (Butler & Coward
chain has recently been the focus of much detailed structural 1989) and the Pyrenees (Deramond et al. 1985; Seguret &
work. During 1985-1986, a deep seismic survey across the Daignieres 1986). Most ofthem are based on the determination
Pyrenees was completed adding information about this orogen of the depth and geometry of the sole thrust, deduced from
(ECORS Pyrenees team 1988; Choukroune e/a/. 1989). The conventional balanced cross-sections of the external parts of
250 km long deep reflection seismic profile (ECORS-PYR- the orogenic belt together with the depth of the Moho below
ENEES) traverses the main Pyrenean structural units across the chain, determined by refraction seismic data. These
the central Pyrenees (Fig. 1). Knowledge of thrust structure sections are not very well constrained by deep seismic data
at the surface and at moderate depths from available com- and the geometry of the crustal thrust system is poorly
mercial seismic and oil-well data, together with the ECORS determined. As a result, some of the initial solutions expressed
profile permits construction of a reasonably well constrained in these sections have been reconsidered, once deep reflec-
crustal cross-section. tion seismic profiles have been obtained. Crustal balanced
235
J. A . MuNOz
236
EVOLUTION OF A CONTINENTAL COLLISION BELT
MORRERES
BACKTHRUST
', t UPF TH\
SOUTH PYRENEAN ' """' GAVARNIE TH. NORTH PYRENEAN NORTH PYRENEAN FRONTAL
MAIN TH SERRES MARGINALS MALADETA A COUFLENS
•\ TH SHEETS! ^ FAULT /
; SUB PYRENEAN TH
3 26 +
SOUTH PYRENEAN
CENTRAL UNIT AXIAL ZONE ANTIFORMAL STACK \ NORTH PYRENEAN
3 60-r UNIT
NPF CAMARADE
/ ,| » ARIZE ,, '!,MZ PAS
UPPER EOCENE-LOWER
I OLIGOCENE 1 SILURIAN-DEVONIAN
1 EOCENE-OLIGOCENE iJ LOWER CRETACEOUS J LOWER CARBONIFEROUS — GRANULITES
UNDIFERENTIATED
^^3 CAMBRO-ORDOVICIAN HERCYNIAN BASEMENT
Figure 2. Strip map and cross-section along the ECORS profile. Dashed broken line in the strip map corresponds to the trace of the seismic line.
constrained from the Comiols oil well data and from the Hercynian basement thrust sheets
cutoff points in its hangingwall and footwall (Fig. 2). The
Upper Cretaceous carbonate series is significantly thicker North of the South Pyrenean Central Unit, Hercynian base-
(1500 m) in the Montsec thrust sheet than in the northernmost ment rocks have been involved in Alpine thrusts. Knowledge
outcrops of the Serres Marginals thrust sheets, below the of the Alpine structure of the basement is the cornerstone in
Ager basin (Martinez & Pocovi 1988). deciphering the Pyrenean thrust system at a crustal scale.
The Boixols thrust sheet is located between the Tremp This structure is dominated by thrusting, Alpine cleavage and
basin and the northern thrust of the Upper Thrust sheets (Fig. folds being very subordinate. Basement rocks constitute
2). This thrust sheet consists of a thick (over 5000 m) thrust sheets which are piled on top of each other forming the
Mesozoic series, mainly Lower Cretaceous in age, the thick- Axial Zone antiformal stack (Fig. 2).
ness of the Lower Cretaceous marls and limestones being the The uppermost of these thrust sheets is the Nogueres Zone,
main stratigraphic difference with respect to the Montsec (Dalloni 1913, 1930) where thrusts affect basement and
thrust sheet. The southern boundary of the Boixols thrust Triassic cover rocks. In the Nogueres Zone thrusts and
sheet is along most of its cartographic expression, a buried Triassic beds are steepened and, as a result, hangingwall
thrust which is overlapped by the Maastrichtian Aren sand- anticlines display a downward facing fold geometry. Two
stone formation (Souquet 1967; Garrido-Megias & Rios units can be differentiated: the Upper Nogueres units and the
1972). The syntectonic character of the Aren sequence is Lower Nogueres units. The basement of the Upper Nogueres
clearly demonstrated by the fan attitude of the sandstone beds units consists of Silurian, Devonian and Lower Carbonifer-
and by their onlap disposition over the southern limb of the ous series (Fig. 2). This series is affected by Hercynian
San Cornell anticline (Simo & Puigdefabregas 1985). The thrusts and folds evidenced by the unconformity disposition
frontal part of the Boixols thrust sheet presents a complicated of the Triassic red beds. Thrusts are the main Hercynian
structure, resulting from the inversion of previous Lower structural feature in this unit. Triassic red beds and Devonian
Cretaceous extensional faults. The northern fault contact of limestones are affected by a north-dipping Alpine cleavage
the Boixols thrust sheet corresponds to a backthrust (Morreres associated with E-W trending open folds. The Lower Nogueres
backthrust). This thrust truncates previously developed folds units are composed of several small thrust sheets with a
and very often causes an omission of the Mesozoic stratigraphy. hectometric to kilometric displacement to the south. These
Nevertheless, cleavage and shear zones demonstrate its thrust units form a duplex at the southern limb of the Axial Zone
character. The Morreres backthrust represents a passive-roof antiformal stack. The Devonian series and its Hercynian
thrust (Banks & Warburton 1986) related to the stacking of structure differ significantly from those of the Upper Nogueres
basement thrust sheets below. units.
237
J. A . MuNOz
238
EVOLUTION OF A CONTINENTAL COLLISION BELT
0.m_
147 km
offset of the Moho below the North Pyrenean fault revealed The northernmost Pyrenean structure cortesponds to an
by refraction seismic data (Daignieres et al. 1982) would be anticline interpreted at depth as a duplex involving Upper
in agreement with this hypothesis. The Moho step would be Cretaceous flysch series (Deramond et al. 1990).
the result of differential thickening of the Iberian crust, with
a normal thickness, and the European crust, thinned before
the collision at both sides of the North Pyrenean fault. Pyrenean thrust system at crustal scale
Deformation of the Pyrenean belt is best explained by a
North Pyrenean thrust sheets
crustal thrust system which forms an orogenic lid displaying
North of the North Pyrenean fault zone, north-directed thrusts a characteristic geometry: a basement-involved antiformal
involve basement and cover rocks (Fig. 2). The Hercynian stack in the middle of the chain (Axial Zone), bounded by
basement forms culminations (Fischer 1984), the so called imbricated thrust systems, mainly south-directed thrusts to
North Pyrenean massifs (Trois Seigneurs and Arize massifs the south (Cover Upper thrust sheets in the Central Pyrenees)
along the ECORS section). The non metamorphic and and mainly north-directed thrusts to the north (Fig. 4). The
weakly deformed character of the Upper Cretaceous flysch basement antiformal stack forms a crustal wedge that moved
series which unconformably overlie basement rocks of the to the south as it developed. The southern cover succession
Trois Seigneurs massif contrasts with the strongly deformed has been wedged northwards and up the antiformal stack. The
Jurassic and Lower Cretaceous metamorphic rocks result has been a tectonic delamination (Price 1986) between
outcropping in the North Pyrenean Fault zone. the basement and the cover rocks along a bedding parallel
The Arize massif (Fig. 2) forms a pop-up structure detachment zone in the Middle Triassic evaporites and lutites
overthrusting, at both sides, Lower Cretaceous turbiditic (Keuper facies).
series. This massif is allochthonous over Mesozoic rocks at The Axial Zone antiformal stack only involves upper
depth as corroborated by commercial seismic data (Souquet crustal rocks. In the ECORS cross-section non-metamorphic
& Peybernes 1987) and by the ECORS seismic line to low grade metamorphic rocks are exposed at surface. The
(Choukroune e? a/. 1989). deepest outcropping basement rocks in the Axial Zone con-
Northwards the cover rocks of the Arize massif, the Albo- sist of high grade gneisses, but still correspond to middle
Cenomanian Camarade basin occurs (Fig. 2). This basin, crustal domains (formation depths about 15 km). Alpine
filled by breccias and turbiditic series several thousands of metamorphism is lacking and only some retrogradations at
metres thick, formed as a result of the sinistral displacement greenschist facies related to Alpine faults have been de-
of Iberia. The northern fault boundaries of the Camarade scribed (McCaig & Miller 1986). In the ECORS deep seismic
basin controlled the location of the North Pyrenean frontal profile, the antiformal geometry of the Axial Zone is evi-
thrust (Baby et al. 1988). This thrust follows on the surface denced by well defined reflectors at upper crustal levels
the initial geometry of the basin. Basement short-cuts are (Choukroune et al. 1989). Below these reflectors the Iberian
observed in the hangingwall of the North Pyrenean frontal lower layered crust remains undisturbed by thrusts, at least
thrust as a result of the inversion of the previous steeper underneath the southern half of the Axial Zone, and dips
strike-slip and extensional Early Cretaceous faults. increasingly below the European crust (Fig. 4). All these data
239
J. A. MuNoz
240
EVOLUTION OF A CONTINENTAL COLLISION BELT
transport direction N-S to NNE-SSW. This transport direc- northernmost outcropping Lower Cretaceous rocks reflect a
tion has recently been corroborated by rotations of the footwall progressive northwards shallowing of the Boixols basin
and hangingwall along oblique ramps, as deduced by (Berastegui ^/a/. 1990). This shallowing is corroborated by
palaeomagneticresults (Burbankef a/. 1991). Palaeomagnetic the northern provenance of fine-grained elastics which char-
results (Dinares et al. 1991, this volume) have also demon- acterized the last sequence of the basin, as exposed more to
strated the absence of relative rotations both between the the west of the ECORS section. There is no evidence to
Main Upper thrust sheets along the ECORS profile and assume the existence of Early Cretaceous basins located
between these thrust sheets and the foreland. Orientations of between the Boixols basin and the North Pyrenean fault (Fig.
thrust structures in the North Pyrenean units demonstrate a 4).
transport direction to the NNE, also close to the direction of The Axial Zone antiformal stack consists of three main
the northern ECORS profile (Fig. 2). thrust sheets, all of them outcropping along the ECORS
The cartographic arrangement of the Pyrenean basement profile: Nogueres, Orri and Rialp thrust sheets (Fig. 4). The
thrust sheets also suggests a transport direction to the S or to restored cross-section obtained will depend on the initial
the SSW, a direction which has been corroborated in other arrangement of these thrust sheets. Two extreme solutions
southern Pyrenean sections (in the eastern and western Pyr- are possible: a pre-collisional geometry with very little or no
enees) where no detachment level exists between the basement overlapping of these units and a pre-collisional geometry
and cover rocks (Muiioz 1985; Mmozetal. 1986). Rotations with maximum imbrication of these thrust sheets.
of the Nogueres Zone units (Bates 1989) may have been the A restored cross-section with minimum shortening has
consequence of a strong longitudinal structural variation been constructed (Fig. 4). Asa consequence, basement thrust
which resulted in the relay of small units and the existence of sheets have been initially superposed as much as possible.
several oblique and lateral structures. This superposition, and the maximum depth for the detach-
The ECORS cross-section has been restored using line- ment is constrained by the maximum exposed thickness of the
length balancing techniques for the cover thrust sheets and for upper thrust sheet (Nogueres) in the northern limb of the
the basement units with an attached lower Triassic and Axial Zone antifonnal stack and by the Hercynian metamor-
Permian series. The basement thrust sheets have been areally phic conditions shown by the basement rocks. After the
balanced, but always taking into account their internal Alpine Hercynian orogeny the basement rocks experienced a pro-
and Hercynian features. The restored section not only has to gressive shallowing, as demonstrated by the rocks observed
conserve the areas of the balanced one, but more importantly, below the post-Hercynian unconformity along the Axial
should present a coherent section concerning the precollisional Zone. A post-Hercynian deepening of basement rocks may
geological features. This methodology provides a powerful only have been restricted to the hangingwall of late to post-
tool for discriminating between the different possible re- Hercynian extensional faults (Permian, Triassic and Early
stored cross-section solutions. The ECORS restored cross- Cretaceous) and implies the existence of thick sedimentary
section will mainly depend on the proposed solution for the basins. Alternatively, there is no evidence of thick sedimentary
restoration of the Axial Zone antiformal stack. The final sequences over the Axial Zone thrust sheets.
restored section for the northern Pyrenees will be determined The restored cross-section (Fig. 4) has been constructed
by the interpretation of the reflectors located below the North assuming a maximum overlapping of the Orri thrust sheet
Pyrenean fault, and by the geometry of the previously thinned over the Rialp one - the two thrust sheets being separated by
European crust adopted in the restored section. a low-dipping Early Cretaceous extensional fault (Boixols
The south Pyrenean Upper thrust sheets have been re- fault). The Cambro-Ordovician rocks of the Orri thrust sheet,
stored independently of the Axial Zone antiformal stack as located over the Rialp window, show anchimetamorphic
imposed by the existence of the detachment level into the conditions, with a calculated maximum temperature of 250-
Keuperevaporites. The Serres Marginals and Montsec thrust 300°C (Bons 1989). The maximum reasonable depth by the
sheets have been restored taking the first synorogenic deposits end of the Hercynian orogeny, taking into account the mineral
as a horizontal stratigraphic reference level. As the thrusts parageneses shown by these rocks and the Hercynian crustal
mainly developed in a piggy back fashion, the horizontal configuration (Zwart 1979, 1986), must have been about 9
reference level gets progressively older and becomes asymptotic km, which could have increased by 1-2 km during the Early
upwards to the hinterland (Fig. 4). To the north of the Boixols Cretaceous basin formation. Cambro-Ordovician rocks ex-
thrust the horizontal reference corresponds to the base of the posed over the floor thrust of the Nogueres thrust sheet
Upper Cretaceous succession. The break-back reactivation (Pallaresa anticlinorium) show low grade greenschist facies
of the Serres Marginals and Montsec thrusts during Late metamorphic conditions (Bons 1989) and, with the proposed
Eocene-Early Oligocene times has to be taken into account in restoration, these rocks would have been initially located at a
the restoration (Verges & Murioz 1990), mainly in the cross- depth of 15 km. This depth is excessive because it is the
section thrust sheets, where structural units with intermediate expected depth for the location of high grade gneisses and
stratigraphies can be thrusted completely out-of-sequence, because the related Hercynian metamorphic gradient would
resulting in strong differences in the stratigraphy across the be too low. To avoid this problem, a downwards movement
thrusts (Martinez & Pocovf 1988). of the Nogueres thrust sheet to the north overLate-Hercynian
Restoration of the Boixols thrust sheet reveals the geom- and Early Cretaceous extensional faults could be invoked.
etry of the Early Cretaceous extensional basin (Fig. 4). The The maximum depth reached by these rocks determines the
241
J. A. MuNoz
NPF
50 km
I I
^ ^ ^ > ^ X
LOWER GNEISSES CAMBRO- SIUURIAN- SILURIAN- STEPHANO- TARDl-HERCYNIAN
CRUST ORDOVICIAN LOWER LOWER PERMIAN GRANITES
CARBONIFEROUS CARBONIFEROUS
(Lower structural (Upper structural
levels) levels)
Figure 6. (a) ECORS restored cross-section showing the pre-collisional geological features at Middle Cretaceous times, before the convergence. This
section corresponds to the minimum shortening solution, (b) Interpretative cross-section in Permian times once the Early Cretaceous extensional fault system
is restored. A Late-Hercynian extensional fault system is inferred from the relationships between thrust sheets adopted in the above restored section in order
to minimize the shortening.
maximum depth of the detachment, which in the restored (Choukroune et al. 1990) but also as Permian extensional
cross-section has been located at 15 km depth (Fig. 4). A faults, as suggested by well data. A Late-Hercynian extensional
deeper location is not possible because the sole thrust climbs system could have been responsible for a crustal thinning and
northwards, thus defining the wedge geometry of the North a first shallowing of mantle rocks (Iherzolites) in the northern
Pyrenean thrust sheets. In any case, a deeper detachment Pyrenees, as proposed by Duee et al. (1984). The total cal-
level would only be restricted to a narrow area southwards of culated shortening for the Axial Zone antiformal stack, taking
the North Pyrenean fault. the North Pyrenean fault as a northern reference, is 112 km.
Devonian rocks of the Nogueres Zone have been located Restoration of the North Pyrenean thrust sheets depends
in the restored section above the Devonian and Cambro on the geometry predicted for the thinned crust in middle
Ordovician rocks of the Orri thrust sheet in order to minimize Cretaceous times. Thinning is deduced by the emplacement
the displacement of the Upper Nogueres units (supposed to be of lower crustal and mantle rocks to upper crustal levels, and
only 11 km). As a result, the Nogueres thrust sheet has been by the thermal metamorphism along the North Pyrenean fault
restored by overlapping the Orri one (Fig. 4). This proposed zone. Restoration also depends on the geometry deduced for
restoration imposes a Late-Hercynian extensional geometry. the lower crustal rocks that have been assumed to exist at
The unmetamorphosed Devonian rocks deformed by middle crustal depths below the North Pyrenean fault
Hercynian thrusts (the shallowest exposed structural level) (Choukroune ef a/. 1989). The North Pyrenean fault has to be
are directly located over slightly metamorphosed Devonian located in the restored section south of the European crust
rocks deformed by very tight syn-cleavage folds (Fig. 6). imaged by the ECORS profile. This implies a minimum
Both Devonian units are unconformably overlapped by displacement backwards of 22 km. Restoring the section,
Triassic red beds. This Late-Hercynian extensional fault taking this minimum displacement into account, the crust
could be synchronous with the Stephano-Permian basins and below the North Pyrenean Zone becomes 25 km thick, which
their related volcanic activity. Permian basins outcrop close is probably too thick to explain the geological features cited
to the proposed Permian extensional fault between the Upper above. If the reflectors below the North Pyrenean fault are
and Lower Nogueres units (Fig. 6). The mid-crustal reflec- considered as lower crustal rocks and removed to the south,
tors observed in the northern ECORS profile (Choukroune et the crust below the North Pyrenean sole thrust becomes
al. 1989) cannot only be interpreted as Hercynian thrusts thinner and the total displacement along this thrust increases.
242
EVOLUTION OF A CONTINENTAL COLLISION BELT
A limitation of the extent of this horse of lower crustal rocks display a listric geometry over the lower layered crust. This
is imposed by the cut-off position of the autochthonous geometry has been observed in the undeformed parts of the
contact between the basement and cover (Fig. 4). In the ECORS profile, deduced after the restoration or by
balanced section presented here, only a small unit of lower comparision with other areas (Fig. 6). Most of the BIRPS
layered crust which presents the same seismic fabrics as the deep reflection profiles across the Mesozoic extensional
autochthonous lower layered crust has been considered in basins of northwestern Europe show a highly reflective lower
contrast with previous interpretations (Roureef a/. 1989). The crust which is not penetrated by the extensional faults (Cheadle
crust below the North Pyrenean massifs has been restored to et al. 1987). Although the lower layered crust has been
a lower initial thickness than in the deformed state, because interpreted as the result of extensional processes at depth and
some ductile thickening is expected to have occurred. The the faults do not show listric geometry, they must join a
calculated shortening for the North Pyrenean thrust sheets is detachment level over the lower crust. The same geometry is
36 km and the thickness of the crust below the North Pyrenean observed in the Bay of Biscay and Galicia Banks continental
fault zone in the restored section is 15 km (Fig. 4). margins where the Early Cretaceous extensional faults merge
The total calculated shortening for the Pyrenean belt along into a detachment level located over the lower crust (Le
the ECORS profile from the presented crustal balanced cross- Pichon & Barbier 1987; Boillot & Malod 1988). The location
section is 147 km. This value is coherent with the proposed of these discontinuities favoured the delamination of the
northward displacement of Iberia with respect to the Euro- crust, the upper part forming an orogenic lid shortened by an
pean plate during the Pyrenean collision (Grimaud et al. upper crustal thrust system. The crust below this middle
1982; Boillot 1986). The most remarkable result of the crustal detachment was subducted beneath the European
restoration of the Pyrenean belt is the apparent discrepancy crust (Fig. 4).
between the length of the upper crust and the length of the
lower layered crust.
In the restored cross-section, the Iberian crust below the
foreland and the more external south Pyrenean thrust sheets,
presents a constant thickness of about 34 km (Fig. 4), which
is in agreement with the crustal thickness of the non-de-
formed Hercynian areas of the Iberian Massif (Banda et al.
1981). Can this portion of the apparently undeformed Iberian
crust be thinned in the restored section in order to fill the 110
km hole of the missing middle and lower crust of northern CENOMANIAN 92 Ma
Iberia (below the Orri and Nogueres thrust sheets)? (Fig. 4). \N\Al / //
This assumption is not reasonable because the resulting
Iberian crust is too thin, both in the foreland (less than 30 km)
and in the inner parts (over 20 km). The amount of crustal
thinning experienced by the northern Iberian crust before the PALEOCENE 68 Ma
Pyrenean collision is difficult to determine. This thinning
could have taken place during Permian, Late Triassic and
Early Cretaceous times. Nevertheless, the cover sediments
which overlay the Hercynian basement of the Axial Zone
thrust sheets do not demonstrate a significant attenuation of LUTETIAN 47 Ma
the underlying crust. Moreover, deep seismic reflection
profiles (ECORS, BIRPS) of genetically related close areas
(during Early Mesozoic stretching events), which have not
been reworked by the Alpine collision, show thinning of the
UPPER
crystalline crust below the deepest basins of up to 10 km, and OLIGOCENE 30 Ma
a shallowing of the Moho of only a few kilometres (Cheadle (present section)
etal. 1987). The crust in the restored section (Fig. 4) has been
reduced to a thickness of 15 km below the Cretaceous basins
north of the North-Pyrenean fault and to 28 km below the
Boixols basin in the southern Pyrenees. To the north of this
basin, the crust has been drawn with a constant thickness of
28 km although this is too thin if we consider that Lower
Cretaceous extensional basins did not occur over the Nogueres
thrust sheet.
In the restored cross-section, the main discontinuities
which affected the Pyrenean crust before the collision
(Hercynian cleavage, Hercynian thrusts, Late-Hercynian Figure 7. Evolution of the Pyrenean crust along the ECORS profile from
extensional faults and Early Cretaceous extensional system) the Early Cretaceous to the end of the collision event.
243
J. A. MuNoz
References
Baby, P., Crouzet. G., Specht, M., Deramond, J., Bilotte, M. 0., Debroas, E. Bayer, R., Cazes, M., DalPiaz, G. V., Damotte, B., Elter, J., Gosso, G., Hirn,
J. 1988. Role des paleostructures albo-cenomaniennes dans la geometric A., Lauza, R., Lombardo, B., Mugnier, J. L., Nicolas, A.. Nicolich, R.,
des chavauchements frontaux nord-pyreneens. Compres Rendus de Polino, R., Roure, P., Sacachi, R., Scarascia, S., Tabacco, I., Tapponnier,
I'Acudemie des Sciences, Paris. 306, 307-313. P., Tardy, M., Taylor, M., Thouvenot, F., Torreilles, G. & Villien, A.
Banda, E.,Suriiiach, E., Aparicio, A., Sierra, J. & Ruiz de la Parte. E, 1981. 1987. Premiers resultats de la traversee des Alpes occidentales par
Crust and upper mantle structure of the central Iberian Meseta (Spain). sismique reflexion verticale (Programme ECORS-CROP). Comptes
Geophysical Journal of the Royal Astronomical Society, 67, 779-789. Rendus de I'Academic des Sciences, Paris, 305, 1461-1470.
& Berastegui, X. 1990. Seismic fabrics in deep seismic profiles. In: Beach, A. 1981. Thrust tectonics and crustal shortening in the external
Pinet, B. et al.. (eds), The potential of deep seismic profiling for hydro- French Alps based on a seismic cross-section. Tectonophysics, 79,T I-T6.
carbon exploration. Editions Technip, Paris, 73-76. Berastegui, X., Garcia, J. M. & Losantos, M. 1990. Structure and sedimen-
Banks, C. J. & Warburton. J, 1986. 'Passive-roof duplex geometry in the tary evolution of the Organya basin (Central South Pyrenean Unit, Spain)
frontal structures of the Kirthar and Sulaiman mountain belts, Pakistan. during the Lower Cretaceous. Bulletin de la Societe Geologique de
Journal Structural Geology, 8, 229-237. France, S,25\-2M.
Bates, M. 1989. Palaeomagnetic evidence for rotations and deformation in Boillot, G. 1986. Comparison between the Galicia and Aquitaine margins.
the Nogueras Zone, Central Southern Pyrenees. Journal of the Geologi- Tectonophysics 129, 243-2.'55.
cal Society, London, 146, 459-476. & Capdevila, R. 1977. The Pyrenees, subduction and collision ? Earth
and Planetary Science Letters, 35. 251-260.
244
EVOLUTION OF A CONTINENTAL COLLISION BELT
& Malod, J. 1988. The north and north-west Spanish continental Duee, G., Lagabrielle, Y., Coutelle, A. & Fortane, A. 1984. Les Iherzolites
margin: a review. Revista Sociedad Geologica. Espana, 1, 295-316. associees aux chainons bearnais (Pyrenees occidentales): mise a
Bons, A.-J. 1989. Very low-grade metamorphism of the Seo Formation in the I'affleurement ante-dogger et resedimentation albo-cenomanienne.
Orri Dome, South-Central Pyrenees. Geologie en Mijnhouw. 68,303-312. Comptes Rendus de TAcademic des Sciences, Paris, 299, 1205-1210.
Burbank, D. W., Verges, J., Munoz, J. A. & Bentham, P. 1991. Coeval ECORS Pyrenees team. 1988. The ECORS deep reflection seismic survey
hindward- and forward-imbricating thrusting in the Central Southern across the Pyrenees. Nature, 331, 508-511.
Pyrenees, Spain, Timing and rates of shortening and deposition. Geo- Farrell, S. G., Williams, G. D. & Atkinson, C. D. 1987. Constraints on the age
logical Society of America Bulletin (in press) of movement of the Montsech and Cotiella Thrusts, south central
Butler, R.W.H. 1986. Thrust tectonics, deep structure and crustal subduction Pyrenees, Spain. Journal of the Geological Society of London, 144,907-
in the Alps and Himalayas. Journal of the Geological Society, London. 914.
143, 857-873. Fischer, M. W. 1984. Thrust tectonics in the North Pyrenees. Journal of
1990. Balancing sections on a crustal scale: a view from the western Structural Geology, 6, 721-726.
Alps. Proceedings of the European Geotraverse Sixth workshop, 157- Garrido-Megi'as, A. & Rios, L.M . 1972. Si'ntesis geologica del Secundario
164. y Terciario entre los ri'os Cinca y Segre (Pirineo central de la vertiente
& Coward, M. P. 1989. Crustal scale thrusting and continental during surpirenaica, provincias de Huesca y Lerida). Boletin GeologicoyMinero
Himalayan collision tectonics on the NW Indian plate. In: Sengor, A.M.C. de Espana, 83, 1-47.
(ed.). Tectonic Evolution of the Tethyan Region. NATO ASI, series C: Goldberg. J. M. & Maluski, H. 1988. Donnees nouvelles et mise au point sur
Mathematical and Physical Sciences, 259, 387-413. Fage du metamorphisme pyreneen. Comptes Rendus de TAcademie des
Casas, J. M. & Poblet, J. 1989. Essai de restitution de la deformation dans une Sciences, Paris, 306, 429-435.
zone avec piis et chevauchaments, le 'synclinal de Llavorsi' dans le Grimaud. S., Boiilot. G., Collette, B. J., Mauffret, A., Miles, P. R. & Roberts,
Pyrenees centrales (Espagne). Comptes Rendus de l'Academic des Sci- D. B. 1982. Western extension of the Iberian-European plate boundary
ences, Paris, 308, 427-433. during the Early Cenozoic (Pyrenean) convergence, a new model.
Camara. P. & Klimowitz, J. 1985. Interpretacion geodinamica de la vertiente Marine Geology. 45, 63-17.
centro-occidental surpirenaica. Esiudios Geologicos, 41, 391-404. Hossack, J. R. 1979. The use of balanced cross-sections in the calculation of
Cheadle, M. J., McGeary, S., Warner, M. R. & Matthews, D. H. 1987. orogenic contraction: A review. Journal of the Geological Society of
Extensional structures on the western UK continental shelf, a review of London,136,705-711.
evidence from deep seismic profiling. In: Coward, M. P., Dewey, J. & Laubscher, H. 1988. Material balance in Alpine orogeny. Geological Society
Hancock, P. L. (eds) Continental Extensional Tectonics. Geological of America Bulletin, 100, 1313-1328.
Society of London Special Publication, 28. 445-465. Laubscher. H. P. 1990. Seismic data and the structure of the central Alps.
Chopin, C. 1984. Coesite and pure pyrope in high-grade blueschists of the Proceedings of the European Geotraverse Sixth workshop, 149-156.
Western Alps, a first record and some consequences. Contributions to Le Pichon, X. & Barbier, F. 1987. Passive margin formation by low-angle
Mineralogy and Petrology, 86, 197-218. faulting within the upper crust. The Northern Bay of Biscay margin.
1987. Very-high-pressure metamorphism in the western Alps, impli- Tectonics, 6, 133-150.
cations for subduction of continental crust. Philosophical Transactions Martinez, M. B. & Pocovi, J. 1988. El amortiguamiento frontal de la
of the Royal Society of London, series A, 321, 183-197. estructura de la cobertera surpirenaica y su relacion con el anticlinal de
Choukroune, P. 1976. Structure et evolution tectonique de la zone nord- Barbastro - Balaguer. Acta Geologica Hispanica, 23, 81-94.
pyreneenne (analise de (a deformation dans une portion de chaine a Mattauer, M. 1985. Presentation d'un modele lithospherique de la chaine des
schistosite subverticale). Memoires de la Societe Geologique de France, Pyrenees. Comptes Rendus de V Academic des Sciences. Paris, 300, 71-
127, 116p. 74.
, Seguret, M. & Galdeano, A. 1973. Caracteristiques et evolution 1986. Intracontinental subduction, crust-mantledecoUement and crustal-
structurale des Pyrenees. Bulletin de la Societe Geologique de France, 1, stacking wedge in the Himalayas and other collision belts. In: Coward,
601-611. M. P. & Ries, A. C.(eds) Collision Tectonics. Geological Society of London
& ECORS Team. 1989. The ECORS Pyrenean deep seismic profile Special Publication, 19, 37-50.
reflection data and the overall structure of an orogenic belt. Tectonics, 1990. Vue autre interpretation du profil ECORS Pyrenees. Bulletin de
8, 23-39. la Societe Geologique de France, VI, 307-311.
, Pinet, B., Roure. F. & Cazes. M. 1990. Major Hercynian structures McCaig, A. M. & Miller, J. A. 1986. 40Ar-39Ar age of mylonites along the
along the ECORS Pyrenees and Biscaye Lines. Bulletin de la Societe Merens fault, Central Pyrenees. Tectonophysics, 129, 149-172.
Geologique de France, 8, VI, 313-320. McClelland, E. A. & McCaig, A. M. 1989. Palaeomagnetic estimates of
Dahlstrom, C.D.A. 1969. Balanced cross sections. Canadian .lournal Earth rotation in basement thrust sheets. Axial Zone, Southern Pyrenees.
Sciences, 6, lii-151. Cuadcrnos de Geologia Iherica, 12, 181-193.
Daignieres, M., Gallart, J., Banda, E. & Him, A. 1982. Implications of the Menard, G. & Thouvenot, F. 1987. Coupes equilibrees cmstales, methodologie
seismic structure for the orogenic evolution of the Pyrenean Range. et application aux Alpes occidentales. Geodinamica Acta, 1, 35-45.
Earth and Planetary Science Letters, 57, 88-100. Montigny, R., Azambre, B., Rossy, M. & Thuizat, R. 1986. K-Ar study of
Dalloni, M. 1913. Stratigraphie et tectonique de la region des Nogueras Cretaceous magmatism and metamorphism from the Pyrenees: age and
(Pyrenees centrales). B»//er/n de la Societe Geologique de France, 4,243- length of rotation of the Iberian peninsula. Tectonophysics, 129,257-274.
263. Mugnier, J. L., Gueilec, S., Menard, G. Roure, F., Tardy, M. & Vialon, P.
1930. Etude geologique des Pyrenees Catalanes. Annates Facultee 1990. Crustal balanced cross-section through the external Alps deduced
Sciences Marseille, XXVI, 373pp. from the ECORS profile. Bulletin de la Societe Geotogique de France (in
Debroas, E-J. 1987. Modele de bassin triangulaire a I'intersection de press)
decrochements divergents pour le fosse aibo-cenomanien de la Ballongue Mufioz, J. A. 1985. Estructura alpina i herciniana a la vora sudde la zonaaxial
(zone nord-pyreneenne, France). Bulletin de la Societe Geologique de del Pirineu oriental. Tesi doctoral. University of Barcelona, 305p.
France. 8, 887-898. , Puigdefabregas, C. & Fontbote, J. M. 1984. El cicio alpino y la
1990. Le Flysch noir albo-cenomanien temoin de la structuration estructura tectonica del Pirineo. In: Comba, J. A., Lihro Juhilar J. M.
albienne a senonienne de la zone nord-pyreneenne en Bigorre (Hautes Rios, Geologia de Espana, 2. IGME, Madrid, 185-205.
Pyrenees. France). Bulletin de la Societe Geologique de France. 8,273- . Martinez, A. & Verges. J. 1986. Thrust sequences in the eastern
285. Spanish Pyrenees. Journal of Structural Geology, 8, 399-405.
Deramond, J., Graham, R. H. Hossack. J. R. Baby. P. & Crouzet, G. 1985. Mutti.E..Rosell. J., Allen,G.P.,Fonnesu,F.&Sgavet'ti,M. 1985. r/if Eocene
Nouveau modele de la chaTne des Pyrenees. Comptes Rendus de haronia tide dominated delta-shelf system in the Ager Basin. Exc. Guide-
I'Academic des Sciences, Paris, 301, 1213-1216. book 6th European Regional Meeting. Lerida, Spain, 579-600.
, Baby, P., Specht, M. & Crouzet, G. 1990. Geometric des Olivet, J. L., Bonnin, J., Beuzart, P. & Auzende, J. M. 1984. Cinematique de
chevauchements dans la zone nord-pyreneenne ariegoise precisee par le I'Atlantiqiie nord et central. Centre National pour Texploration des
profil ECORS. Bulletin de la Societe Geologique de France, 8,287-294. oceans. Rapports scientifiques et techniques, 54, 108p.
Dinares, J., McClelland, E. & Santanach, P. 1991. Contrasting rotations Pocovi', J. 1978. Estudio geologico de las Sierras Marginales Catalanas
within thrust sheets and kinematics of thrust-tectonics as derived from (Prepirineo de Lerida). Acta Geologica Hispanica, 13, 73-79.
palaeoniagnetic data: an example from the southern Pyrenees (this Price, R. A. 1986. The southeastern Canadian Cordillera, thrust faulting,
volume). tectonic wedging, and delamination of the lithosphere. Journal of Struc-
tural Geology, 8, 239-245.
245
J. A. MUNOZ
Roure, F., Choukroune, P., Berastegui, X., Munoz, J.A ., Villien, A., & Peybemes, B. 1987. Allochtonie des massifs primaires nord-
Matheron, P., Bareyt, M., Seguret, M., Camara, P. & Deramond, J. 1989. pyreneens des Pyrenees Centrales. Comptes Rendus de I'Academie des
ECORS Deep Seismic data and balanced cross-sections, geometric Sciences. Paris, 305, 733-739.
constraints to trace the evolution of the Pyrenees. Tectonics, 8, 41-50. Tome, M., De Cabissole, B., Bayer, R., Casas, A., Daignieres, M. & Rivero,
Schreyer, W. 1988. Subduction of Continental Crust to Mantle Depths, A. 1989. Gravity constraints on the deep structure of the Pyrenean bell
Petrological Evidence. Episodes, 11, 97-104. along the ECORS profile. Tectonophysics, 165, 105-116.
Seguret, M. 1972. Etude tectonique des nappes et series decollees de la Verges, S. & Munoz, J. A. 1990. Thrust sequences in the Southern Central
partie centrale du versant sud des Pyrenees.PubVications deVUniversite Pyrenees. Bulletin de la Societe Geologique de France, 8, 265-271.
de Sciences et Techniques de Languedoc, serie Geologic Structural, n. Vielzeuf, D. & Komprobst, J. 1984. Crustal splitting and the emplacement
2, Montpellier. of Pyrenean Iherzolites and granulites. Earth and Planetary Science
& Daignieres, M. 1986. Crustal scale balanced cross-sections of the Letters, 67, 383-386.
Pyrenees, discussion. Tfcfonop/i^'s/ci, 129, 303-318. Williams, G. D. & Fischer, M. W. 1984. A balanced section across the
Simo, A. & Puigdefabregas, C. 1985. Transition from shelf to basin on an Pyrenean erogenic belt. Tectonics, 3, 773-780.
active slope, upper Cretaceous, Tramp area, southern Pyrenees. Exc. Zandvliet, J. 1960. The geology of the Upper Salat and Pallaresa valleys.
Guide-book 6th European Regional Meeting, Lerida, Spain, 63-108. Central Pyrenees, France/Spain. Leidse GeologischeMededelingen, 25,
Souquet, P. 1967. Le Cretace superieur sud-pyreneen en Catalogue, Aragon 1 -27.
et Navarre. These Doctoral Sciences Naturales. Faculte Sciences Tou- Zwart, H. J. 1979. The geology of the Central Pyrenees. Leidse Geologische
louse, 529p. Mededelingen, SO. 1-44.
1986. The Variscan Geology of the Pyrenees. Tectonophysics, 129,9-
27.
246
Thrusting and foreland basin evolution in the Southern Pyrenees
C. Puigdefabregas', J.A. Muiioz- & J. Verges'
Abstract: The geometry and the infill of the south Pyrenean foreland basin mainly depend on the
tectonic subsidence history due to the flexure of the crust, which in turn can be related to the
structural evolution of the mountain chain at a crustal scale. Characteristics of the infill of the basin
as well as the relationships between structures and synorogenic deposits allow distinction of four
stages in the evolution of the south Pyrenean foreland basin. These stages can be related to the
structural evolution of the crust as deduced from partial restored cross-section construction through
the central Pyrenees. Stage I (Upper Cretaceous) is characterized by strong subsident turbiditic
troughs deposited over a thinned crust. Uniform distribution of continental facies and a crust
restored to its initial thickness corresponds to the stage II (Palaeocene). During Stage III, turbiditic
troughs developed synchronously with the initial subduction of the lower crust (Lower and Middle
Eocene). Stage IV (Uppermost Eocene-Oligocene) is characterized by continental deposition
coeval with the increase of the crustal cross-sectional area, produced by both piggy-back and
break-back thrust sequences.
The development of a thrust system in a mountain chain takes model for the progressive unconformities in the northern
place .synchronously with the accumulation of sediments in margin of the Ebro basin and correlated these with the
the related foreland basins. This implies that basin geometry synchronous emplacement of the Pyrenean thrust system.
and sedimentation patterns are controlled by the development Some authors have attempted to distinguish sedimentary
of the thrust system. sequences in both the central and eastern southern Pyrenean
The Pyrenees, mainly in their southern part, are one of the foreland basins, and relate them to the tectonic evolution
best examples to illustrate the interaction between tectonics (Soler & Puigdefabregas 1970; Garrido-Megias 1973;
and sedimentation because the shallow erosion level is such Puigdefabregas 1975;Nijman&Nio 1975; Puigdefabregas ef
that the relationships between thrusts and their related deposits al. 1986; Labaume et al. 1987, and Mutti et al. 1988).
are well exposed. The Pyrenean orogen was developed as a The aim of this paper is to present a general outline of the
consequence of an Alpine age continental collision that geometry and infill of the south Pyrenean foreland basin, its
ranged from the Upper Cretaceous to Miocene times. The relationship to the thrust system and finally to correlate this
associated thrust system includes (Fig. 1): a basement with the crustal evolution of the whole crust along the
antiformal-stack bounded by cover imbricated thrust systems ECORS profile.
(Munoz, this volume). Thrusts of the central antiformal stack
are south-directed and the southern imbricate thrust system
involves more shortening than the northern one. Moreover, Stages in the evolution of the foreland basin
the floor thrust of the southern thrust system is the sole thrust
of the Pyrenean chain, which implies that in the central and The characteristics of the infill of the Pyrenean foreland basin
eastern Pyrenees, Iberia has been subducted to the north allow the determination of several stages in its evolution
below Europe. regardless of longitudinal variations of facies, and mainly
The south Pyrenean foreland basin is a triangular shaped dependent on the pattern of thrusting. These stages, well
feature between the Pyrenees to the north and the Catalan and established in the central and eastern Pyrenees, have been
Iberian Ranges to the SE and SW respectively. Most of the used in the present paper in order to describe the main features
area represents the last stage of the basin fill known as Ebro of the foreland basin deposits and their relationships with the
foreland basin, whereas the earlier stages involved in the evolution of the Pyrenean chain.
south Pyrenean thrust system occurred as piggy-back basins
(Ori & Friend, 1984). An outline of the Ebro basin with the
Stage I (Upper Santonian-Maastrichtian)
first observations on the relationships between tectonics and
sedimentation was given by Riba (1967), who described Prior to the Alpine compression of the Pyrenees, a Lower
features such as basin asymmetry and the outward migration Cretaceous extensional rift basin developed (Puigdefabregas
of facies depocentres. Riba (1973) also proposed a kinematic & Souquet 1986). The Upper Cenomanian transgressive
247
C . PUIGDEFABREGAS ET AL.
s N
t APTIAN 114 Ma
\v\v •'/
"^ ^ '^
CENOMANIAN 92 Ma
V-vAl / »>
SM Montsec Boixols
PALEOCENE 68 Ma Stage 11
Figure 1. Structural sketch of the Pyrenees and the ECORS crustal cross-
section.
UPPER
OLIGOCENE 30 Ma (present section) Stage IV
platform deposits unconformably overlie the syn-rift depos-
7/
its and related normal faults (Berasteguie? a/. 1990). During
this period, thermal subsidence was still active as indicated by
the superposition of Lower and Upper Cretaceous depocentres,
suggesting a pure shear extensional model for the Lower
Cretaceous rifting.
The first Pyrenean thrusts developed as a result of the
reactivation of the Lower Cretaceous extensional faults during
Figure 2. Partially restored cross-sections showing the Lower Cretaceous
Upper Santonian times. In front of these thrusts, deep E-W extensional configuration and thrust geometry at crustal scale during the
elongated basins were formed and infilled with Upper stages of the foreland basin evolution.
Santonian to Campanian age turbiditic deposits (Vallcarga
Formation of Mey et al. 1968) which are up to 6000 m in the unconformities are related to the Boixols thrust.
north (Dubois & Seguin 1978)and2000min the south. These The first thrusts developed over an already stretched crust
turbidites unconformably overlie the earlier sequences and (Fig. 2). As a consequence, both the limited tectonic subsid-
their deposition seems to be controlled by the development of ence induced by the inversion of the lower Cretaceous
the first Pyrenean thrust sheets (Boixols; Fig. 1). Similar extensional basins and the remnant thermal subsidence con-
relationships were documented westwards for the Turbon tributed to the formation ofturbiditic troughs as suggested by
thrust sheet which moved from the Uppermost Turonian to Brunei (1984) in the NW Pyrenees. Moreover, a thin and
Maastrichtian (Souquet & Deramond 1989) and also for warm crust would favour the formation of narrow and deep
thrusts in the northern Pyrenees (Baby et al. 1988; Desegaulx foreland basins in front of the thrust sheets (Kamere/a/. 1983).
etal. 1990). As thrusting progressed during the Maastrichtian,
progressive unconformities (Garrido-Megias 1973) affected
Stage II {Uppermost Maastrichtian-Palaeocene)
the shallow marine sequences of the Aren Group. The
geometry and distinct depositional sequences within this During this stage the south Pyrenean foreland basin was
formation have been described by Nagtegaal et al. (1983), characterized by the deposition of fluvial and lacustrine red
Simo & Puigdefabregas (1985), Fondecave etal. (1989) and beds, which overlay and in part interfingered with the Aren
Mutti & Sgavetti (1987) who also discuss eustatic or tectonic Group. These deposits are known as Garumnian (Tremp
models for their formation. A number of unconformity- Formation by Mey et al. 1968) or Tremp Group (Cuevas
bounded depositional sequences can be distinguished, in 1989). They are up to 1000 m thick in places and extend over
which relative disposition of facies, depocentres and angular the whole central and eastern Pyrenees, in the north as well as
248
THRUSTING AND FORELAND BASIN EVOLUTION IN THE SOUTHERN PYRENEES
in the south (Plaziat 1975). Conglomerate formations to generate strongly subsident troughs by thrust sheet loading
unconformably overlie the inverted extensional faults (Boixols (Beaumont 1981; Jordan 1981).
thrust) and are coeval with the southwards piggy-back propa- In the western Pyrenees, the equivalent sediments to the
gation of the southern thrust system in the central Pyrenees. Garumnian shallow-water deposits are represented by 350 m
Thickness and facies distributions demonstrate the contem- of deep-water carbonate and siliciclastic sediments deposited
poraneity of the Garumnian sedimentation with the new- in continuity with the Upper Cretaceous turbidites (Pujalte et
formed thrusts (Montsec thrust; Figs 1 & 2). al. 1989). Thus, the evolution proposed for Stage I, charac-
The Lower Cretaceous extensional faults were completely terized by thrusts superimposed on a thinned crust, continued
inverted during Palaeocene times; thus the stretched upper in the western Pyrenees through the Uppermost Cretaceous-
crust recovered its initial pre-Cretaceous length, and prob- Palaeocene and probably until Early Eocene times. Thick-
ably the crustal thickness also attained its pre-Cretaceous ness variations of the series and the occurrence and thickness
crustal thickness. This conclusion has also been deduced of volcanic rocks (Montigny et al. 1986) demon.strate a
from the partial restoration of the ECORS balanced cross- greater extension of the western Pyrenean crust during Early
section for Palaeocene times (Mufioz 1991, this volume; Fig. Cretaceous times. As a result, during convergence the crust
2). in the western Pyrenees would have recovered its initial
Although thrust structures continuously developed from thickness later than that in the central and eastern Pyrenees.
Stage I, the change in the infill of the foreland basin could be This concept is in agreement with the proposed crustal and
related to the crustal thickness compensation, inferred to foreland basin evolution in the central Pyrenees.
occur at the end of the thermal subsidence. As a result, during
this stage the foreland basin is characterized by uniform
Stage III (Lower and Middle Eocene)
shallow-water deposits occupying a wide area. Involvement
of basement rocks in thrust structures was still too restricted After the widespread Ilerdian transgression, the south Pyrenean
S-N
stage III
stage IV
Axial zone
Antiformal stack
Serres marginals CAMPODARBE GROUP
Cardona evaporites
25 km
Figure 3. Partially restored cross-sections at the time of stages III & IV. These sections correspond to the southern part of the ECORS profile. The syntectonic
sediments related with the stage have been shaded as well as the previous foreland basin deposits stippled. A: Aren Group (Upper Cretaceous); T: Tremp
Group; AM: Ager and Montanyana Groups; LN: Lower Nogueres Units; UN: Upper Nogueres Units.
249
C. PUIGDEFABREGAS ET AL.
—p?»«™^
--- ---''c-^^'5
'^-^—T^ -T""
N-S Palaeocurrents
Figure 4 Longitudinal E-W schematic cross-sections during stages III & IV. See location in Figure 3.
Mediterranean
Sea
Figure 5. Schematic map showing the facies distribution during stage III.
250
THRUSTING AND FORELAND BASIN EVOLUTION IN THE SOUTHERN PYRENEES
251
C. PUIGDEFABREGAS ET AL.
ment, with lacustrine deposits in the centre and clastic wedges obvious out-of-sequence disposition (Fig. 7).
on the basin margins continued until the end of the Oligocene. The contribution of these out of sequence thrusts in the
The structural evolution was controlled by the growth and total displacement of the thrust system was very small. To the
development of the crustal scale antiformal stack in the inner contrary, the vertical component of movement was significant
part of the chain (Axial Zone Antiformal Stack). The piggy- and therefore the structural relief increased.
back southwards migration of the Pyrenean thrust system The basement antiformal stack and synchronous with its
involved both the SPCU and the Palaeogene sequences of the formation, the cover thrust sheets (SPCU), have been wedged
previous stages, in the Gavamie and Cadf thrust sheets (lower backwards (to the north) over the Morreres passive roof
thrust sheets; Munoz etal. 1986). This thrusting motion was backthrust (Munoz, this volume). This backthrust together
synchronous with the incorporation of basement units into with a set of out of sequence thrusts are related with three
the Axial Zone Antiformal Stack and controlled the charac- successive conglomerate sequences (Mellere, pers. comm.)
teristic progressive, forward migration of depocentres. of the Collegats Fm. (Mey et al. 1968). The conglomerates
Break-back thrust sequences developed synchronously which documented the out-of-sequence thrusts of the Collegats
with the overall piggy-back displacement of the Pyrenean and Oliana area are approximately of the same age.
thrust sheets, as characterized by the external structures. This
break-back thrusting in the external areas was equivalent to
the basement antiformal stacking in the inner part of the chain Discussion and conclusions
in that both tend to increase the taper in order to allow the
progression of the orogenic wedge (Davis etal. 1983; Dahlen The geometry and major infilling of the south Pyrenean
& Suppe 1988). This particular structural situation signifi- foreland basin was mainly dependent upon both the surficial
cantly increased the cross-sectional area, so that erosion and and subcrustal forces which produced tectonic subsidence.
supply of coarse elastics was therefore enhanced. The surficial forces were mainly due to thrust sheet loading
Several examples of break-back thrusting sequence can be and the subcrustal forces result from thermal evolution of the
unambiguously deduced from the relationships between the lithosphere and from its flexure caused by the traction of the
structures and the related synorogenic conglomerates subducting slab (Brunei 1986; Molnar & Lyon-Caen 1988).
(Martinez ef a/. 1988; Verges &Muiioz 1990). In the eastern The geometry of the thrust sheets might have been a factor
oblique boundary of the SPCU (Oliana area) these relation- modifying the tectonic subsidence but this mainly generated
ships between structures and clastic deposits are well illus- longitudinal variations in the infilling of the foreland basin.
trated (Verges & Mutioz 1990). The break-back thrust Finally, as far as the tectonic control on the foreland basin
system is constituted by a set of thrusts, each of them covered depositional sequences is concerned, the influence of local
by a conglomerate formation, which in tum is thrust by the structures must be taken into account, in addition to other
younger thrust formed hindwards of the previous one, in an factors such as global sea level changes. Although it is not the
Conglomeratic Units
Upper Eocene
UPPER
CRETACEOUS
JURASSIC
TRIASSIC
^CARDONA EVAPORITES
+ -t
+ + IGUALADA MARINE MARLS
.f + + 4-+ + + + + + LOWER
+ + + f + t-+- + + > + + +-4- + +-
+ + + + + + + + + + t +- + EOCENE
0 2 k m +^-^• + t + ^- + ^ - ^ t ^ - + ^ - ^ - t ^ • + ofi?„,^iVr + + + + + *^ GARUMN AN
Y f j " ' " * <• -f + + . » * - * • * • + + + + » - + + + + BASEMENT *• t ^ * •>• *
^^^^^^^^ ^ + + + 4 + + + + + + + + + f + + + -rT + + + + f + +4. + +
Figure 7. Geological cross-section of the eastern termination of the SCU showing the relationships between out-of-sequence thrusts and synorogenic
conglomeratic deposits.
252
THRUSTING AND FORELAND BASIN EVOLUTION IN THE SOUTHERN PYRENEES
here determine the depositional geometries of the syntectonic characteristic of Stage III (Lower and Middle Eocene).
conglomerates (Verges & Muiioz 1990). The sudden increase of the tectonic subsidence during this
Recent papers have modelled the flexure of the lithosphere period (Burbank, pers. comm.) is due to both a greater
of the Pyrenees and the geometry of the foreland basin influence of the surficial loading and by the existence of
(Brunet 1986; Desegaulx & Moretti 1988; Desegaulx et al. subcrustal forces. During this period the basement rocks are
1990) using the physical parameters calculated or deduced involved in thrust sheets and the load increases. In addition
from the present state of the lithosphere. The problem of the Iberian crust began to subduct below the European crust
modelling old mountain chains is to know how these physical giving rise to a pull by the subducted slab.
parameters have evolved during the formation of the orogen The lithospheric flexure value was probably the same all
and also during the post-tectonic evolution. Partial crustal along the chain but turbiditic troughs only developed in the
restored cross-sections can be used as an additional tool. This footwall of the SPCU because the thrust sheet of the SPCU
kind of work through the Central Pyrenees allows investigation partially fill the trough in the central Pyrenees. During this
of the relationships between the foreland basin geometry and stage, there is an increase of the thrusting rate with respect to
the crustal profile as deduced from partial restored cross- the earlier one (Labaume et al. 1985; Verges & Martinez
sections (Mufioz 1991, this volume). Crust below the Pyrenean 1988).
belt has evolved during the formation of the mountain chain Stage IV (Upper Eocene-Oligocene) is characterized by
from an initial thinned crust to a thickened crust with a coarse grained alluvial fan deposition as an erosional re-
subducted slab of about 100 km long (Munoz 1991, this sponse to the increase of the cross-sectional area which
volume). As a consequence, not only the thickness of the resulted from the growth of the basement antiformal stack in
crust below the belt has changed through convergence, but the hinterland and to the development of the break-back
also its elastic parameters, which control the geometry of the thrusting sequences in the frontal areas. In spite of the taper
foreland basin. increase and consequently the surficial weight, the tectonic
The major subdivisions that have been made in the south subsidence during this stage decreases with respect to previous
central Pyrenean foreland basin are related to stages with stages (Burbank e? a/. 1991). As a consequence, a subcrustal
deposition of turbidites in strongly subsident narrow troughs explanation such as thermal re-equilibration or collapse of the
and to stages with deposition of shallow-water to continental subducting slab has to be considered.
sediments occupying a wider area. The former indicate a It must be noted that this evolution in four stages, which
small wavelength of the lithospheric flexure, the latter a large lasts 50 Ma, is proposed for this particular transect and is not
wavelength. necessarily synchronous with other transects, first because of
The turbiditic troughs characteristic of Stage I, (Upper differences in structural style and strength of the crust, and
Santonian-Maastrichtian) have been interpreted as a result of secondly because of the diachronous development of the belt
a restricted surface weight but acting over a warm and thin (Choukroune 1976).
crust. In addition, subcrustal forces due to the thermal These four stages may be difficult to distinguish in other
subsidence inherited from the previous extensional history mountain chains because of differences in preservation of the
have contributed to the strong tectonic subsidence. During foreland deposits or because of a different foreland basin
this stage, the crustal shortening has not compensated the history. As most of the collisional belts evolved from
initial crustal thickness. previous stretched crust to a final thickened crust, one might
Stage II, (Uppermost Maastrichtian-Palaeocene), charac- expect that both surficial and subcrustal processes for the
terized by a rather uniform facies distribution, corresponds to evolution of the foreland basin would be similar to those
the period where the thickening of the crust is compensated described in this paper.
by crustal shortening, so that there are insignificant subcrustal
forces. The only contribution in the tectonic subsidence is We are grateful to Ken McClay for comments and revision of the original
due to the surficial loading by the thrust sheets themselves. manuscript. This work has been partially supported by the CICYT project
GEO89-0254.
The creation of elongated turbiditic troughs is the main
References
Anadon, P. & Marzo, M. 1986. Sistemas deposicionales eocenos del margen Brunet, M. F. 1984. Subsidence history of the Aquitaine basin determined
oriental de la cuenca del Ebro: sector Igualada - Montserral. Libra guia from subsidence curves. Geological Magazine, 121, 421 -428,
Exc. XI Congreso Espahol de Sedimentologia, 4.1 -4.59. 1986. The influence of the evolution of the Pyrenees on adjacent basins.
Baby, P., Crouzet, G., Specht, M., Deramond, J., Bilotte, M. & Debroas, E. Tectonophysics. 129, 343-354.
J. 1988. Role des paleostructures albo-cenomaniennes dans lageometrie Burbank, D. W., Verges, J., Mutioz, J. A. & Bentham, P. 1991. Coeval
des chevauchements frontaux nord-pyreneens. Compies Rendus de hindward- and forward-iinbricating thrusting in the Central Southern
I'Academie des Sciences, Paris, 306, 307-313. Pyrenees, Spain: Timing and rates of shortening and deposition. Geo-
Beaumont, C. 1981. Foreland basins. Geopftysical Journal of tlie Royal logical Society of America Bulletin (in press).
Astronomical Society, 65, 291-329. Choukroune, P. 1976. Structure et evolution tectonique de la zone nord-
Berastegui, X., Garcia, J. M. & Losantos, M. 1990. Structure and sedimen- pyreneenne (analise de la deformation dans une portion de chaine a
tary evolution of the Organyabasin (Central South Pyrenean Unit, Spain) schistosite subverticale). Memoires de la Societe Geologique de France,
during the Lower Cretaceous, Bulletin de la Societe Geologique de ni, 116p.
f/wjcf, 8, VI;251-264.
253
C. PUIGDEFABREGAS ET AL.
Cuevas, J. L. 1989. La Formacion Talarn: estudio estratigrafico y Nagtegaal, P.J.C., Van Vliet, A. & Brouwer, J. 1983. Syntectonic coastal
sedimentoldgico de las facies de un sistema aluvial en el transito offlap and concurrent turbidite deposition: the Upper Cretaceous Aren
Mesozoico-Cenozoico de la Conca de Tremp. Tesis de Licencitura, Sandstone in the South-Central Pyrenees, Spain. Sedimentary Geology,
Universitat de Barcelona, 107p. 34, 185-218.
Dahlen, F. A. & Suppe, J. 1988. Mechanics, growth, and erosion of mountain Nijman, W. & Nio, S. D. 1975. The Eocene Moniafiana delta (Tremp-Graus
belts. Geological Society of America Special Paper, 218, 161-178. Basin, provinces of Lerida and Huesca, Southern Pyrenees, N. Spain). In:
Davis, D., Suppe, J. & Dahlen, F. A. 1983. Mechanics of fold-and-thrust belts IX Congres de Sedimentologie. Nice, 18p.
and accretionary wedges. Journal of Geophysical Research, 88, 1153- Ori, G. G. & Friend, P. F. 1984. Sedimentary basins formed and carried
1172. piggyback on active thrust sheets. Geology, 12, 475-478.
De.segaulx, P. & Moretti, I. 1988. Subsidence history of the Ebro basin. Plaziat, J. C. 1975. LTIerdien a I'interieur du Paleogene languedocien; ses
Journal of Geodynamics, 10, 9-24. relations avec le Spamacien, I'llerdien sud-pyreneen. I'Ypresien et le
, Roure, F. & Villien, A. 1990. Structural evolution of the Pyrenees: Paleocene. Bulletin de la Societe geologique de France, 7, 168-182.
Tectonic heritage and flexural behavior of the continental crust. In: Puigdefabregas, C. 1975. La sedimentacion molasica en la cuenca de Jaca.
Letouzey. J. (ed.) Petroleum and Tectonics in Mobile Belts, 31-48. Pirineos, 104, 188p.
Dubois, P. & Seguin, J-C. 1978. Les flyschs cretace et eocene de la zone , Munoz, J.A. & Marzo, M. 1986. Thrust belt development in the
commingeoise et leur environnement. Bulletin de la Societe Geologique Eastern Pyrenees and related depositional sequences in the southern
de la France, 1, 6il-(fl 1. foreland basin. In: Allen, P. A. & Homewood, P. (eds) Foreland basins.
Fondecave-Wallez, M. J., Souquet, P. et Gourinard, Y. 1989. Enregistrement Special Publication of the International Association of Sedimentologists,
sedimentaire de I'eustatisme et de la tectonique dans la s^rie turbiditique 8, 229-246.
du Cretace des Pyrenees centro-meridionales (Groupe de Vallcarga, & Souquet, P. 1986. Tectosedimentary cycles and depositional sequences
n.gr.,Es,pagne).ComptesRendus del'Academic des Sciences, Paris,309, oftheMesozoic and Tertiary from the Pyrenees. Tectonophysics, 129,173-
137-144. 203.
Garrido-Megi'as, A. 1973. Estudio geologico y relacion entre tectonica y , Collinson, J., Cuevas, J. L., Dreyer, T., Marzo, M., Mellere. D.,
sedimentacion del Secundario y Terciario de la vertiente meridional Mercade, L., Mufioz, J. A., Nijman, W. & Verges, J. 1989. Alluvial
pirenaica en su zona central (prov. Huesca y Lerida). Tesis Doctoral, deposits of the successive foreland basin stages and their relation to the
Facultad de Ciencias. Granada, 395p. Pyrenean thrust sequence. 4th International Conference on Fluvial
Jordan, T. E. 1981. Thrust loads and foreland basin evolution. Cretaceous, Sedimentology. Excursion Guidebook. (Marzo & Puigdefabregas, eds),
Western United States. American Association of Petroleum Geoloi-ists 175p.
Bulletin, 65, 2506-2520. Pujadas, J., Casas, J. M., Munoz, J. A. & Sabat, F. 1989. Thrust tectonics and
Karner, G. D., Sleekier, M. S. & Thome, J. A. 1983. Long-term Paleogene syntectonic sedimentation in the Empordaarea, southeastern
thermomechanical properties of the continental lithosphere. Nature, 304, Pyrenees. Geodinamica Acta, 3, 195-206.
250-253. Pujalte, v., Robles, S., Zapata, M., Orue-Etxebarria, X. & Garcia-Portero, J.
Labaume, P., Seguret, M. & Seyve, C. 1985. Evolution of a turbiditic 1989. Sistemas sedimentarios, secuencias deposicionales y fenomenos
foreland basin and analogy with an accretionary prism: Example of the tectoestratigraficos del Maastrichtiense superior-Eoceno inferior de la
Eocene South-Pyrenean basin. Tectonics, 4, 661-685. cuenca vasca (Guipuzcoa y Vizcaya). XII Congreso Espahol de
, Mutti, E. & Seguret, M. 1987. Megaturbidites: A depositional model Sedimentologia, Bilbao. Guia de E.xcursiones, 47-88.
from the Eocene of the SW-Pyrenean foreland basin, Spain. Geo-Marine Riba, O. 1967. Resultados de un estudio sobre el Terciario continental de la
Letters,7,9\-l0l. parte este de la depresion central catalana. Acta Geologica Hispanica, 11,
Martinez, A., Verges, J. & Mufioz, J. A. (1988) Secuencias de propagacion 1-6.
del sistema de cabalgamientos de la terminacion oriental del manto del 1973. Las discordancias sintectonicas del Alto Gardener (Prepirineo
Pedraforca y relacion con los conglomerados sinorogenicos. Acta Catalan). ensayo de interpretacion evolutiva. Arra Geologica Hispanica,
Geologica Hispanica, 23, 119-128. 8, 90-99.
Mey, P.H.W., Nagtegaal, P.J.C. Roberti, K. J. & Hartevelt, J.J.A. 1968. Saez. A. & Riba, O. 1986. Depositos aluviales y lacustres paleogenos del
Lithostratigraphic subdivision of post-hercynian deposits in the south- margen pirenaico Catalan de la cuenca del Ebro. Libro guia Excursiones.
central Pyrenees, Spain. Leidse Geologische Mededelingen,41,221 -228. XI Congreso Espahol de Sedimentologia. Barcelona, 6.1 -6.29.
Molnar, P. & Lyon-Caen, H. 1988. Some simple physical aspects of the Seguret, M. 1972. Elude teclonique des nappes et series decollees de la
support, structure, and evolution of mountain belts. Geological Society partiecentraleduver.santsuddes Pyrenees. Ptiblic-dlionadeVUniversite
of America Special Paper, 218, 179-207. de Sciences et Techniques de Languedoc, serie Geologie Structurale n.
Montigny, R., Azambre, B., Rossy, M. & Thuizat, R. 1986. K-Ar study of 2, Montpellier, I55p.
Cretaceous magmatism and metamorphism in the Pyrenees: age and Simo. A. & Puigdefabregas, C. 1985. Transition from shelf to basin on an
lengthof rotation of the Iberian Peninsula. Tectonophysics, 129,257-273. active slope, upper Cretaceous, Tremp area, southern Pyrenees. E.xc.
Muiioz.J. A., Martinez, A. & Verges, J. 1986. Thrust sequences in the ea,stem Guide-book 6th European Regional Meeting. Lerida, Spain, 63-108.
Spanish Pyrenees. Journal of Structural Geology, 8, 399-405. Soler, M. & Puigdefabregas, C. 1970. Lineas generales de la geoiogi'a del
1991. Evolution of a Continental Collision Belt: ECORS-Pyrenees Alto Aragon Occidental. Pirineos, 96, 5-20.
Crustal Balanced Cross-section (this volume). Souquet, P. et Deramond, J. 1989. Sequence de chevauchemems et sequences
Mutti, E. & Sgavetti, M. 1987. Sequence stratigraphy of the Upper Creta- de depot dans un bassin d'avant-fosse. Exemple du sillon cretace du
ceous Aren strata in the Orcau-Aren region, south-central Pyrenees, versant sud des Pyrenees (Espagne). Comptes Rendus de l'Academic des
Spain: Distinction between eustatically and tectonically controlled depo Sciences, Paris, 309, 137-144.
sitional sequences. Annali delTVniversita di Ferrara. Sezionc Scienze Verges, J. & Martinez, A. 1988. Corte compensado del Pirineo oriental:
della Terra, 1, 22p. geometria de las cuencas de antepai's y edades de emplazamiento de los
, Seguret, M. & Sgavetti, M. 1988. Sedimentation and deformation in mantos de corrimiento. Ada Geologica Hispanica, 23, 95-106.
the Tertiary sequences of the southern Pyrenees. American Association & Munoz, J. A. 1990. Thrust sequences in the Southern Central
of Petroleum Geologists Mediterranean Basins Conference. Field Trip Pyrenees. Bulletin de la Societc geologiquc de France, 8. 265-271.
7, 153p.
254
South Pyrenean fold and thrust belt: The role of foreland evaporitic
levels in thrust geometry
J. Verges", J.A. Muiioz^ & A. Martinez'
Abstract: Combined surface and subsurface data were used to determine the geometry of the
foreland fold and thrust belt in the southeastern Pyrenees. In the south Pyrenean foreland basin,
three evaporitic horizons were deposited: the Beuda, the Cardona and the Barbastro Formations.
These evaporites range from Lutetian to Lower Oligocene in age, and their depocentres shifted
successively to the SW in front of the advancing Pyrenean thrust sheets. From Late Eocene to Late
Oligocene, a salt-basal foreland fold and thrust belt developed over a detachment level with a
staircase geometry controlled by the arrangement of the evaporitic basins. Synchronously, the
central Pyrenean thrust sheets were displaced southwards over the same detachment level. The
arrangement of the evaporitic basins, mainly the Cardona basin, the N-S transport direction of the
south Pyrenean thrust sheets, and the change in trend of the South Pyrenean main thrust (SPMT),
were all responsible for the different trend of structures in the foreland fold and thrust belt. The
limits of these folded regions coincide with the boundaries of the Cardona evaporitic basin. The
southernmost outcropping frontal structure of the Pyrenees is a backthrust related to the edge of
the Cardona massive salt level and is located in some places 40 km southwards of the SPMT. Well-
constrained timing of defonnation permits the interpretation that the shortening occurred from
Upper Eocene times and has minimum values of between 21 km and 25 km. These values represent
more than 15% of the 147 km of shortening deduced for the overall Pyrenean chain.
The southern Pyrenees are made up of cover and basement has been displaced to the south over a detachment level. The
thrust sheets which were displaced southwards from the Late southernmost structures affecting the eastern Ebro basin
Cretaceous to the Oligocene (Early Miocene in the western define an approximate E-W front with the SPMT of the
Pyrenees). The south Pyrenean structural units can be grouped central Pyrenees. The displacement and shortening involved
into Upper Thrust Sheets, which consist only of cover rocks, in these foreland structures have been probably transferred in
mainly a Mesozoic series, and Lower Thrust Sheets, formed the eastern extremity of the Pyrenean chain to dextral NW-SE
by basement rocks and a Palaeogene series unconformably strike-slip faults, later reactivated by Neogene extensional
overlying a reduced and thin cover sequence (Muiioz et al. faults (Fig. 2).
1986; Murioz 1991, this volume). In the eastern Pyrenees, The thrust and fold structures of the eastern Ebro basin are
Palaeogene sediments involved in the Lower Thrust Sheets strongly controlled by the geometry of the decollement level,
(Cadi thrust sheet) occupy a narrow area (Figs 1 & 2). These which is largely dependent on the location of weak evaporitic
sediments represent the northern part of the South Pyrenean horizons in the foreland series. The structural style of this
foreland basin, which has been incorporated into the region resembles the style observed in other detached fold
allochthonous units (Puigdefabregas et al. 1986; Verges & and thrust belts, such as the Appalachian Plateau (Rodgers
Martinez 1988). The Cadi thrust sheet is overthrusted by the 1963), and the Salt Range, in Pakistan (Jaume & Lillie 1988;
Pedraforca upper thrust sheet (Fig. 2). Westwards, in the Lillie <>/«/. 1987; Baker e/a/. 1988), amongst other examples.
central Pyrenees, upper thrust sheets (Boixols, Montsec and The main goal of this contribution is the study of the
Serres Marginals) cover a wide area over an autochthonous geometry of structures of the eastern foreland Ebro basin,
non-deformed Palaeogene series, which unconformably between the Pyrenean thrust sheets and the Catalan Coastal
overlies basement rocks (Mufioz 1991, this volume). The Ranges and their relationships with the arrangement of the
eastern termination of the Upper Thrust Sheets of the central different Palaeogene evaporitic basins. Regional work and
Pyrenees is the oblique Segre thrust zone, which limits with mapping showing general features of the area has been done
the Ebro foreland basin and, northwards, with the Cadi thrust (Riba 1967; Wagner e/a/. 1971). The available subsurface
sheet (Fig. 2). In the central and eastern Pyrenees, the data (drill holes and seismic lines) together with detailed field
emergent floor thrust of the south Pyrenean thrust sheets, data, enable the construction of accurate cross-sections. In
herein called South Pyrenean main thrust (SPMT), describes addition, the synchroneity between sedimentation and tec-
a step geometry at surface (Figs 1 & 2). The eastern foreland tonics permits us to date thrust movements and to calculate
Ebro basin, located southwards and eastwards of the SPMT the Pyrenean shortening since Late Eocene times.
255
J. V E R G E S ET AL.
Figure 1. Structural sketch of the Pyrenees and location of the study area (Fig. 2).
Figure 2. Structural map showing main features of the study area. The South Pyrenean Main thrust (SPMT) represents the southern boundary of the different
Mesozoic and Cenozoic cover rocks thrust sheets, which are depicted in different types of continuous lines. The foreland basin is represented in white.
Foreland map derived from a compilation of the map by Ramirez & Riba (1975) and from own observations. The crossed circles represent oil wells (Igme
1987).
256
SOUTH PYRENEAN FOLD AND THRUST BELT
water sequences deposited in narrow troughs in front of the Iguatada tpradalta marls)
Clastic systems, related to emergent thrusts, prograded south- B«uda Bauda Imarina avaporitas)
^
r
Arminolaa (akipa marls)
Per^ya (cartortala platform)
Coronaa (outar platform)
Cadi (cartxmata platform)
PonMs
J
age. In the southern areas, now the Ebro basin, continental ILERDIAN
CadI Sagnari (opan ntarlna nuria)
Orpf (eartMMiata platform)
sedimentation was general in Priabonian times. The boundary PALEOCENE
Tramp (rad bads)
between marine and continental deposits is marked by marine Madiona (rad bads)
257
J. VERGES ET AL.
2
LUTETIAN PRIABONIAN
Montsec
-r^??^:•x•:•^x•:•:•:•:|v^:^i;::^::
/--i- -<!<5:>ii::::>i-i»I^:^:::-x:-Sx-:-:::
aSBlMM
A
i*2Km ^fi^x-XvyBergaiX::::-;.:".::;::::;
Lleida
&
4
LOWER OLIGOCENE
PRESENT DAY
B
•^T^ Cadi„<-
,,-rr.r--.-.v;
D o a a
• LLeltla D D O O D a D
D c^Q n a a a a a o D Q D
O D O D D D D D D O O O D
" " o _a o • D D n
- " " A" ( " - " 20KI1)
a D d
BEUDA CARDONA 1 n o D BARBASTRO
Figure 4. PaUnspastic restoration during the deposition of the Beuda (Lutetian), Cardona (Priabonian), and Barbastro evaporites (Later Priabonian-Eariy
Oligocene), with the approximate amount of displacement of the SPMT. Last sketch corresponds to the present position of the evaporitic depocentres. Dashed
line represents the present position of the South Pyrenean main thrust (SPMT).
N
(Riba 1967; Wagner et al. 1971; Ramirez & Riba 1975;
Malmsheimer & Mensink 1979), which can be divided into
three main regions: the NE region with continuous structures
varying from E-W to ESE-WNW, the Central region with
NE-SW trending structures, and the SW region, which dis-
plays roughly E-W trending folds, such as the Barbastro-
Balaguer anticline, which continues westwards of the study
area (Fig. 2).
The northern structures of the NE region (Alpens, La Quar
and Berga) show an E-W trend slightly oblique to the SPMT. _0m
Southwards, the Prats and Solsona-Navas synclines and the Figure 5. Interpretative distribution of Cardona evaporitic facies and
Puigreig anticline stretch over a distance of 50 km with an thickness from the basin centre to the southern margin (simplified from
ESE-WNW trend (Fig. 2). Busquets et al. (1985). No horizontal scale.
258
SPMT PEDRAFORCA
P U I G R E I C A.
CARDONA A
CATALAN COASTAL
RANGES
gijiiiiiiiiii^
B-B
SANAUJA A.
SEGRE
OLIANA A.
THRUST
VILANOVA A
S
Sena i P l
/C«RD0«A • ^ BABBASTBO ,
-J — I..- I-., i. I 1 . L
H=V
A-A
SOLSONA 0
Lowtr OUg
BARBASTRO
IMIMinTTIIIUI
Bartorian
-
BEUDA
"^
1
Lutetian
Cuisian
Paleortne
CADI
VJ'-
S.MARG
ISINALS
HONTSEC BOIXOLS
.,,„„,
Triassic
Basemtrt
'^1 "'r>
Figure 6. N-S balanced geological cross-sections showing the general features of the entire folded foreland. See Figure 2 for location.
J. VERGES ET AL.
Figure 8. Geological map of the Sanaiija antiform, which separates the folded Central region and the non-deformed S W region. See Figure 2 for location.
Crossed circles are oil wells. Black circles are towns. Dotted unit is the Barbastro Formation.
260
SOUTH PYRENEAN FOLD AND THRUST BELT
FONTS SERO T over the Cardona evaporites (Verges & Murioz 1990).
In the Central region, over the flat detachment located in
the Cardona salt, most of the area is occupied by gently
dipping beds corresponding to broad synclines that separate
narrow anticlines. These anticlines are cored by salt which
flows from the synclines. The load ofthe synorogenic Upper
Eocene-Lower Oligocene clastic rocks (Solsona Formation),
deposited in the synclines, contributed to the salt flowage.
The synchronism of clastic deposition with salt growth is
documented by progressive unconformities in the limbs of
the anticlines. Once the salt attained a sufficient amplitude,
it continued to flow diapirically, thus breaching the previ-
Figure 9. N-S geological cross-section across the culmination zone between ously-formed anticlines. This can be observed in the Cardona
the Sanaiija and Vilanova anticlines. See Figure 8 for location. anticline, where salt is still flowing at present (Ramirez &
Riba 1975).
Geological cross-sections The southern part of the deformed Ebro foreland displays
north directed structures (backthrusts). These backthrusts
Two geological cross-sections across the Bbro foreland fold (Old and Sanaiija) coincide with the southern edge of the
and thrust belt have been included in this paper: the eastern Cardona evaporitic basin (Figs 2 & 6). They branch with the
one (Figs 2 & 6, B-B') across the Lower and Upper Thrust detachment close to the southern end ofthe thick Lower Salt
Sheets, the deformed Ebro foreland and the Catalan Coastal Member (Fig. 10). The 016 branch line forms the southern
Ranges: and the western section (Fig. 6, A-A') across the front ofthe eastern Pyrenees, while westwards ofthe Sanaiija
Upper Thrust Sheets and the deformed Ebro foreland. In the backthrust, the detachment climbs up to the Barbastro
latter, basin fold and thrust structures occur above a detach- evaporites and deformation continues southwards (Fig. 6).
ment level, which is located in the different evaporitic hori- South of the 016 backthrust, a buried tip line can be traced
zons ofthe basin. The geometry ofthe structures is controlled with an approximate E-W trend, merging with the Barbastro-
by the geometry of the detachment level. Palaeogene rocks Balaguer anticline tip line to the west (Munoz et al. 1984;
underlying the decoUement remain undeformed (Fig. 6). In Williams 1985).
the cross-sections, a double vergence can be observed. In the
northern part, all structures are south directed (Oliana. Puigreig,
and Cardona anticlines), while in the southern part they are Timing
north directed (Siiria, Old, and Sanaiija structures). The folds
of the middle area, such as the Vilanova anticline, show no Well-constrained geometric relationships between structures
clear vergence. and syntectonic sediments, combined with biostratigraphic
The geometry ofthe detachment level, the southern con- and magnetostratigraphic data, enable us to calculate the age
tinuation of the south Pyrenean sole thrust, displays a flat and of the structures.
ramp geometry controlled by the arrangement of the Beuda, The Oliana anticline (duplex) developed in Priabonian
Cardona and Barbastro evaporites. In the eastern section
(Fig. 6, B-B"), the detachment climbs up from the Beuda to N
the Cardona evaporites. In the western section (Fig. 6, A-A'), PASSIVE BACK-THRUST BURIED FRONTAL
TIP LINE
the decollement climbs up from the Cardona to the Barbastro
evaporites in the southernmost extremity. The NE study
region studied is detached over the Beuda Formation, the
Central region over the Cardona Formation, and the SW
region over the Barbastro Formation (Figs 2 & 4). The
Cardona evaporites are the most suitable decollement level,
due to the existence of a thick salt package (Fig. 5). This
staircase geometry contrasts with the flat detatchments found
in many thin-skinned mountain belts with basal salt
decollements.
A duplex structure formed above the ramp that connects
the Beuda and Cardona horizons. The Puigreig anticline is
the surface expression of this duplex, as can be inferred from
seismic lines across the anticline (Fig. 6, B-B'). The Oliana
anticline has been described as a duplex at depth formed Figure 10. Interpretative tectonic model for the southern boundary of the
above a ramp which connected the floor thrust of the Upper Cardona evaporitic basin drawn in Figure 5 (with the same legend). The
Thrust Sheets with the foreland detachment (Fig. 6, A-A'), disappearance of the Lower Salt member increases the basal friction of the
similar to the Puigreig anticline, later displaced southwards decollement and, as a consequence, a passive backthrust was developed.
261
J. VERGES ETAL.
times, between 40Ma and 36Ma, at the same time as the (Oliana anticline) of the NE-SW trending Segre thrust dem-
break-back imbricate stack in the hangingwall of the Segre onstrate that these structures are oblique with respect to the
thrust (Burbank et al. 1991). The Oliana anticline has rotated transport direction (Burbank et al. 1991; Dinares et al.
20° anticlockwise (Burbank etal. 1991; Dinares etal. 1991, 1991, this volume). In addition, palaeomagnetic data suggest
this volume). As a consequence, it formed with a trend of at that no rotations occurred in the central Pyrenean Upper
least N70°E. Eastwards, movement along the E-W trending Thrust Sheets (Dinares etal. 1991, this volume), which does
SPMT south of the Pedraforca thrust sheet and growth of the not agree with the idea of aprogressive change in the transport
E-W trending Berga-Bellmunt anticline took place at the direction of the south Pyrenean units (Nijman 1989). All the
same time (Mato & Saula 1991). The Oliana anticline rotated above point to an approximate N-S transport direction as the
and was displaced southwards over the Cardona evaporites most suitable for the development of the structures of the
during Earliest Oligocene, synchronous with the growth of deformed Ebro foreland basin. This is in agreement with the
the Puigreig anticline. Deformation progressed southwards, arrangement of the different trending structures of the study
as demonstrated by the sedimentological characteristics of area (Fig. 2).
the Lower Oligocene Solsona sequence. Both limbs of the In the study area, where the majority of the trends of the
Cardona anticline show lower Solsonadistal alluvial sediments structures are oblique with respect to the thrusting transport
(coeval with coarser elastics related to the growth of the direction, an accurate calculation of the shortening would
Oliana anticline) without facies, thickness or palaeocurrent involve a 3-D restoration. Although only two cross-sections
directions (Saez 1987). Consequently, southern structures of have been included, they have been restored, despite the
the Central region and Sanaiija antiform are Early Oligocene possible errors of such restorations, if only to give an idea of
and younger than the northern ones. In the study area, there the shortening involved. Bed-length techniques were used
are no available criteria to determine the age of the Barbastro for balancing competent layers (Dahlstrom 1969), while
anticline, but westwards, it has been dated as Late Oligocene evaporitic rocks have been areally restored (taking into
(Pardo & ViUena 1979; Riba etal. 1983). account only the N-S migration of salt). The pin lines are
located at the end of the deformed zone (black points in cross-
sections, Fig. 6). The eastern section (Fig. 6, B-B') is drawn
Shortening normal to the trend of the structures in each segment of the
section, whereas the western one (Fig. 6, A-A') is oblique to
The calculation of shortening of the Pyrenean chain through the structures, but parallel to the thrust transport direction.
its eastern realm should take into account the shortening In this paper, only the shortening that occurred after the
involved in the structures of the deformed Ebro foreland deposition of Early Priabonian Cardona marine evaporites
basin. The construction of balanced and restored cross- has been analysed. Both cross-sections show similar amounts
sections across this area poses some difficulties due to the of shortening: 25 km in the A-A' and 21 km in the B-B' cross-
different strikes of the structures. As far as the determination sections (Fig. 6). This is a minimum value, as neither erosion
of transport direction is concerned, which is crucial for (the hangingwall ramp of the SPMT is eroded in both the A-
accurate construction of balanced cross-sections, questions A' and B-B' cross-sections) nor internal deformation such as
arise. Can all these structures be formed by a unique transport the strong cleavage observed in the overturned Lower and
direction? If so, which of these structures are oblique and Middle Eocene marls outcropping in front of the SPMT (B-
which are frontal? B' cross-section) have been taken into account. The total
As previously described, structures of different trends amount of shortening can be partitioned in two, that is, the
developed at the same time in the deformed foreland, as for shortening that occurred in the foreland structures, and the
example, NE-SW structures of the Central region and the shortening due to the movement of the SPMT since Priabonian
NNW-SSE structures of the NE region. Moreover, neither times. In the A-A' cross-section, the shortening south of the
cross-cutting relationships between structures with different SPMT is 21 km while the shortening due to the motion of the
trends nor interference patterns, with the exception of the SPMT is a minimum of 4 km. In the B-B' cross-section, the
Sanaiija antiform, are observed, which is in agreement with deformed foreland southwards of the SPMT shows 11.5 km
a synchronous development of different trending structures. of shortening, whereas the movement of the SPMT is a
In the eastern Pyrenees, a N-S transport direction can be minimum of 9.5 km. In the former section, there is a big
deduced for the emplacement of the Cadi thrust sheet and difference between shortening due to the motion of the SPMT
lower units (Mufioz et al. 1986). In the Pedraforca thrust and deformation in foreland deposits, whereas in the latter
sheet, a N180°E to N195°E transport direction has been section, both values of shortening are similar. These differ-
calculated from the arrangement of frontal and oblique struc- ences in the relative amount of shortening between that in the
tures, as well as from the strike of tear faults (Verges & foreland basin and along the SPMT, observed in the two
Martinez 1988; Martinezefa/. 1988). In the Sierras Marginales cross-sections, can be explained by the arrangement of the
thrust sheets, equivalent to the lower Pedraforca units, similar boundaries of the NW-SE Cardona evaporitic basin (Fig. 4).
criteria allow us to deduce a N-S transport direction from The south Pyrenean sole thrust reached the western part of the
structures developed synchronously with the folds and thrusts northern margin of the Cardona salt (Fig. 6, A-A' cross-
of the deformed Ebro foreland. Anticlockwise rotations in section) before reaching the eastern margin (B-B' cross-
the structures of both the hangingwall (imbricates) and footwall section). When the sole thrust climbed up to the Cardona
262
SOUTH PYRENEAN FOLD AND THRUST BELT
decollement level, the movement along the SPMT, in the The oblique arrangement of the evaporitic basins with
Segre zone, stopped and a hindward thrusting sequence with respect to the thrust transport direction, which was controlled
no important displacements developed in the hangingwall of by the thrusting evolution of the south Pyrenean thrust sheets,
the SPMT, while the Oliana anticline developed in its footwall determined the existence of numerous oblique structures in
(Verges & Munoz 1990). In contrast, in the eastern part (B- the foreland, mainly those of the Central region. The trend of
B' cross-section), the SPMT overthrusted the overall sedi- the structures of each region coincides with the direction of
mentary pile, which was involved in the growth of the Oliana the SPMT located hindwards of them, as can be seen in the
anticline (Fig. 6). structural map of Figure 2. In the NE region, structures strike
parallel to the E-W segment of the SPMT. In the Central
region, NE-SW structures are parallel to the oblique Segre
Conclusions thrust (oblique segment of the SPMT) and, in the SW region
fold structures strike parallel to the E-W segment of the
The geometry of the structures of the study area is character- SPMT of the central Pyrenees.
istic of foreland fold and thrust belts with a weak evaporitic The greater extent to the south of the thrust sheets of the
decollement level at depth (Davis & Engelder 1985). Anti- central Pyrenees with respect to the eastern Pyrenean units
clines are narrow and salt-cored and synclines are broad. (step geometry of the SPMT) is partially due to a major
Thrusts are mainly blind and backthrusts coexist with for- displacement of the central thrust sheets once they reached in
ward directed thrusts. Evaporites that accumulated in the Late Eocene the Cardona salt. The salt was never reached by
anticline cores move diapirically, thus overprinting previously the eastern part of the SPMT, instead the SPMT climbed up
developed structures. This has led all these structures to be section and developed the structures of the foreland above the
interpreted as having developed by diapirism alone and not evaporitic detachment levels (Fig. 6). The NE-SW structures
being related to the Pyrenean contractional structures. above the Cardona detachment of the Central region formed
The Central region, located above the Cardona decollement, by an oblique component of shortening coeval with further
has undergone shallow erosion as can be deduced from areal movement to the south of the SPMT along the oblique Segre
distribution of the youngest sediments outcropping in the zone.
synclines. The present topographic slope of the Central Well-constrained relationships between tectonics and
region is less than 1°, the basal slope is 3°, but more con- sedimentation enable one to deduce that the deformation
clusively, syntectonic sediments of the same age are located started in the study region later than deposition of the Cardona
approximately at the same altitude in the centre of the evaporites (Burbank ef a/. 1991). Shortening produced after
synclines. This demonstrates a constant narrow taper of the the Cardona deposition by both the motion of the SPMT and
Central region through its evolution, as shown by the salt- the deformation in the foreland basin is between 21 km and
basal fold and thrust belts. 25 km, both minimum values. This represents approximately
A peculiarity of the salt-basal Ebro foreland fold and thrust the 15% of the total 147 km of shortening for the whole
belt is the existence of structures with different strike, which Pyrenean belt (Munoz 1991, this volume).
developed from the Late Eocene to the Early Oligocene along The overall geometry of the southern central and eastern
a N-S thrust transport direction. Superposition of the struc- Pyrenees is mainly the result of the disposition of the Cardona
tural map over the evaporitic basin distribution (Figs 2 & 4) salt. The central Pyrenean thrust sheets moved further to the
clearly shows that there is a close relationship between the south, synchronously with the development of the foreland
defined structural regions and the evaporitic basin distribu- fold and thrust belt to the east. The displacement of the central
tion. The NE region coincides areally with the Lutetian part of the SPMT, over the Cardona sah, is at least of the same
Beuda evaporites, the Central region with the Priabonian magnitude of the shortening involved in the deformed Ebro
Cardona evaporites and the S W region with the Upper Eocene- foreland. Frontal structures, emplaced forward of the Cardona
Lower Oligocene Barbastro evaporites. The three horizons salt, in continuation with the Barbastro-Balaguer tip anticline,
act as a decollement level, the Cardona being the most are located 40 km southwards of the eastern part of the SPMT
suitable due to the existence of a thick salt member. The trace. Eastwards of the Cardona salt, the shortening is
detachment steps up from the northern and lower (oldest) accomplished by folds, thrusts and cleavage normal to the
evaporitic horizon to the southern and upper (youngest) one. beds, and the southern Pyrenean front is only located 10 km
Ramps are located at the edge of the Cardona salt basin and south of the SPMT.
correspond at surface to the boundaries of the different
structural regions defined by their different strikes. In the We would like to thank Oriol Riba, Juan Jose Pueyo and Alberto Saez for
northern edge of the Cardona basin, a duplex structure de- their helpful comments. We have found the observations of our two
anonymous referees very useful. The English was revised by Michele
veloped above the ramp (Oliana and Puigreig anticlines) Pereira. This work was supported by the CAICYT project PB85-0098-C04-
while in the southern edge, a set of backthrusts formed, 03 and by the Servei Geologic de Catalunya.
defining a frontal wedge.
263
J. VERGES ET AL.
References
Anadon. P., Colombo, F., Esteban. M., Marzo, M., Robles, S., Santanach, P. Martinez. A. & Verges, J. 1986. Thrust sequence in the eastern
& Sole Sugraiies, L . 1979. Evolucion tectonoestratigrafica de los Pyrenees. Journal of Structural Geology, 8, 399-405.
Catalanides./lctoG^o/d^/caWj-pdn/ca.HomenatgeaLlufsSoleiSabaris. PuigdeRibregas, C. & Fontbote, J. M. 1984. El ciclo alpino y la
14, 242-270. estructura tectonica del Pirineo. In: Comba, J. A., Libra Juhilar J. M.
Baker, D. M., Lillie, R. J., Yeats, R. S., Johnson, G. D., Yousuf, M. & Hamid, Rios, Geologica de Espafia, 2. IGME, Madrid, 185-205.
A. S. 1988. Development of the Himalayan frontal thrust zone: Salt Nijman, W. 1989. Thrust sheet rotation? The south Pyrenean Tertiary basin
Range, Pakistan. Geology, 16, 3-7. configuration reconsidered. Geodinamica Acta, 3, 17-42.
Burbank, D. W., Verges, J.. Munoz, J. A. & Bentham, P. 1991. Coeval Pardo, G. & Villena, J. 1979. Aportacion a la geologia de la region de
hindward- and forward-imbricating thrusting in the Central Southern Barbastro. /4r'/a Geoldgica Hispanica. Homenatge a LIuis Sole i Sabaris.
Pyrenees, Spain: Timing and rates of shortening and deposition. Geo- 14, 289-292.
logical Society America Bulletin (in press). Pueyo, J. J. 1975. Estudio petrologico y geoqui'mico de los yacimientos
Busquets, P., Orti, F., Pueyo, J. J., Riba, O.. Rosell, J, Saez, A., Salas. R. & potasicos de Cardona, Siiria, Salient y Balsareny (Barcelona, Espafia).
Tabemer, C. 1985. Evaporite deposition and diagenesis in the saline Tesis Doctoral. Universidad de Barcelona, 3l5p.
(potash) Catalan basin, Upper Eocene. Excursion Guide-hook 6th Euro- Puigdefabregas, C , Muiioz, J. A. & Marzo, M. 1986. Thrust belt develop-
pean Meeting, Lleida, Spain. 13-59. ment in the Eastern Pyrenees and related depositional sequences in the
Dahlstrom,C.D.A. 1969. Balanced cross sections. Ca«ai^(on7oH/-«a/o/£a;Y/i southern foreland basin. In: Allen, P. A. & Homewood, P. (eds) Foreland
Sciences, 6, 743-757. basins. Special Publication of the International Association of
Davis, D. M. & Engelder, T. 1985. The role of salt in fold and thrust belts. Sedimentologists, 8, 229-246.
Tectonophysics, 119, 67-88. Ramirez, A. & Riba, 0.1975. Bassin polassique Catalan et mines de Cardona.
Dinares, J., McClelland, E. & Santanach, P. 1991. Contrasting rotations IX Congres Inter, de Sedimentologie, Nice 1975. Livret-guide Ex. 20,
within thrust sheets and kinematics of thrust-tectonics as derived from 49-58.
palaeomagnetic data: an example from the southern Pyrenees (this Reguant, S. 1967. El Eoceno marino de Vic (Barcelona). Memorias del
volume). Instituto Geologico y Minero de Espana, 68, 1 -350.
IGME. 1987. Contribucion de la exploracion petrolifera al conocimiento de Riba, O. 1967. Resultados de un estudio sobre el Terciario continental de la
la geologia de Espafia. Instituto Geologico y Minero de Espana, 1 -465. parte este de la depresion central catalana. Acta Geoldgica Hispdnica.
Jaume, S. C. & Lillie, R. J. 1988. Mechanics of the Salt Range-Potwar II, 1-6.
Plateau, Pakistan: A fold and thrust belt underlain by evaporites. Tectonics, Reguant, S. & Villena, J. 1983. Ensayo de sintesis estratigrafica y
7,57-71. evolutiva de la cuenca terciaria del Ebro. In: Comba, J. h.,LihroJuhilar
Jurado, M. J. 1988. El Triasico del .subsuelo de la cuenca del Ebro. Tesis J. M. Rios, Geologia de Esparia, 2, IGME, Madrid, 131-159.
doctoral. Univ. de Barcelona, 259p. Rodgers, J. 1963. Mechanics of Appalachian foreland folding in Pennsylvania
Lillie, R. J.. Johnson, G. D., Yousuf, M., Hamid, A. S. & Yeats, R. 1987. and West Virginia. American Association of Petroleum Geologists Bul-
Structural development within the Himalayan foreland fold and thrust letin, 47, l527-\5i6.
belt of Pakistan. Canadian Society of Petroleum Geologists Memoir, 12, Rosell. L. & Pueyo, J. J. 1984. Caracteri'sticas geoqui'micas de la formacion
379-392. de sales potasicas de Navarra (Eoceno superior). Comparacion con la
Malmsheimer, K. & Mensink, H. 1979. Der geologische Aufbau des cuenca potasica catalana. Acta Geoldgica Hispanica, 19, 81-95.
Zentralkatalanischen Molassebeckens. Geologische Rundschau, 68,121 - Saez, A. & Riba, O. 1986. Depositos aluviales y lacustres paleogenos del
162. margen pirenaico Catalan de la cuenca del Ebro. Libro guia Excursion. XI
Martfnez-Pefia, M. B. & Pocovi, J. 1988. El amortiguamiento frontal de la Congreso Espaiiol de Sedimentologia. Barcelona, 6.1-6.29.
estructura de la cobertera surpirenaica y su relacion con el anticlinal de 1987. Estratigrafi'a y sedimentologia de las formaciones lacustres del
Barbastro - Balaguer. Acta Geologica Hispanica, 23, 81-94. transito Eoceno - Oligoceno del NE de la cuenca del Ebro. Tesis doctoral.
Marlmez, A., Verges, J. & Munoz, J. A. 1988. Secuencias de propagacion del Universitat de Barcelona, 352p.
sistema de cabalgamientos de la terminacion oriental del manto del Verges, J. & Martinez, A. 1988. Corte compensado del Pirineo oriental:
Pedraforca y relacion con los conglomerados sinorogenicos. Acta geometria de las cuencas de antepals y edades de emplazamiento de los
Geoldgica Hispanica, 23, 119-127. mantos de corrimiento. Ada Geoldgica Hispdnica, 23, 95-106.
, Clavell, E. & Kennedy, J. 1989. Stratigraphic framework of the & Muiioz, J. A. 1990. Thrust sequences in the Southern Central
thrust geometry and structural inversion in the southeastern Pyrenees; La Pyrenees. Bulletin de la Societe geologique de France, 8, 265-271.
Garrotxa area. Geodinamica Acta, 3, 185-194. Wagner, G., Mauthe, F. & Mensik, H. 1971. Der Salzstock von Cardona in
Mato, E. & Saula, E. 1991. Los ciclos sedimentarios del Eoceno medio y Nordospanien. Geologische Rundschau, 60, 970-996.
superior en el sector NE de la depresion del Ebro (zona de Berga-Vic). Williams, G. D. 1985. Thrust tectonics in the south central Pyrenees./owrna/
Caracterizacion de sus h'mites. Geogaceta (in press). of Structural Geology. 7, 11-17.
Munoz, J. A. 1991. Evolution of a Continental Collision Belt: ECORS-
Pyrenees Crustal Balanced Cross-section (this volume).
264
Contrasting rotations within thrust sheets and kinematics of thrust
tectonics as derived from palaeomagnetic data: an example from the
Southern Pyrenees
J. Dinares' 2, E. McClellandS P. Santanach^
'Department of Earth Sciences, University of Oxford, Parks Rd., Oxford 0X1 3 PR,
UK
^Departament de Geologia Dindmica, Geofisica i Paleontologia, Universitat de
Barcelona, Zona Universitaria de Pedralbes, 08028 Barcelona, Spain
Palaeomagnetism has been used as a powerful tool for iden- island of Mallorca (Spain) which show very different struc-
tifying structural rotations about any axis in fold and thrust tural patterns (one area associated with frontal ramf>s and
belts. This technique becomes specially valuable for quanti- another area associated with oblique-lateral ramps). They
fying structural rotations about the vertical axis and can found that palaeomagnetic directions after tectonic correction
therefore be used to document thrust sheet rotation in mobile were very similar in both areas, thus indicating no differential
belts. Magnetotectonic studies from several orogenic belts rotation between them.
have shown that such rotations are common (e.g. Kotasek & Structural rotation about the vertical axis causes material
Krs 1965; Schwartz & Van der Voo 1984; McClelland & to move into or out of section, and 2-D sections cannot be
McCaig 1989) due to complexities such as oblique and lateral expected to balance. Palaeomagnetism is the only way of
ramps, and e.g. the existence of an arcuate foreland margin, quantifying such rotations and is thus an essential test before
Rotations in the Idaho-Wyoming overthrust belt (USA) 2-D section balancing is allowable. The South-Pyrenean
have been studied by a number of authors (Grubbs & Van der Central Unit (SCU) as defined by Seguret (1972) is a major
Voo 1976; Schwartz & Van der Voo 1984; Eldredge & Van trapezoid-shaped thrust sheet consisting of several separate
der Voo 1988) to attempt to discriminate between various units (Fig. 1). Its western and eastemmost boundaries have
models for the origin of the observed arcuate thrust belt, been interpreted as oblique ramps and therefore a question
These authors conclude that buttressing edge effects due to a arises as to whether rotations might have occurred within the
Precambrian massif in the foreland were the cause of the unit. A deep seismic profile (French-Spanish ECORS group)
documented rotations. McClelland & McCaig (1989) have has recently been shot in the Pyrenees and this cross-cuts the
recently postulated that the growth of an antiformal stack in SCU through its central-eastern part in an approximately N-
the Axial Zone of the Pyrenees caused differential rotation to S section (ECORS Pyrenees Team, 1988; Choukroune et al.
occur in this thrust pile. Freeman et al. (1989) compared 1989). The section has also been balanced by Roure et al.
palaeomagnetic directions from two thrust sheets on the (1989). Moreover, a putative counterclockwise rotation of
265
J. DiNARES ET AL.
Figure 1. Structural sketch of the central and eastern Pyrenees. Boxes indicate location of Figure 4 & Figure 8. Sections A, B, and C are the location of
cross-sections from Figure 2.
the SCU has been postulated by Nijman (1990). Boixols unit. It extends southwards from the southern edge
This paper presents an extensive palaeomagnetic study in of the Nogueres zone. Further to the south the Montsec unit
the south-central cover thrust sheets of the Pyrenees in order constitutes the second thrust sheet. The Sierras Marginales
to identify and quantify structural rotations within those unit is the lowermost and its floor thrust has followed the
units. Additional attention has been given to the Pedraforca Tertiary sediments of the Ebro foreland basin. The southern
area piggy-back thrust pile which is one of the constituent limit of the Boixols unit corresponds mostly to a blind thrust
parts of the so called Upper Thrust Sheets together with the overlapped by detritic sediments of the Maestrichtian Aren
SCU and the Ampurda units (Muiioz et al. 1986). The fore- depositional sequence (Souquet 1967; Garrido-Megias &
land Oliana anticlinal stack located in the eastern margin of Rios 1972). The syntectonic character of this sequence is
the SCU has also been studied. demonstrated by the onlap geometry over the frontal part of
the Boixols unit (Simo & Puigdefabregas 1985). The age of
the Montsec emplacement in the ECORS transverse is
Geological setting and sampling Ypresian (Williams & Fischer 1984; Mutti et al. 1985; Farrell
etal. 1987). The Montsec thrust can be connected westwards
The South-Pyrenean Central Unit (SCU) includes a succes- below the Eocene conglomerates with the Cotiella Thrust
sion of south-directed thrust sheets involving Mesozoic- along the N-S striking Mediano anticline (Fig. 1). Towards
lower Tertiary cover rocks. These are thrust over the E the Montsec thrust branches onto the Sierras Marginales
autochthonous Palaeogene rocks that lie directly over either thrust and both trend NE-S W (oblique zones) in the so called
the Palaeozoic basement of the Ebro foreland basin or over a Segre thrust (Verges & Muiioz 1990). The Sierras Marginales
thin Mesozoic cover (Camara & Klimovitz 1985; ECORS unit is formed by an imbricate fan thrust system and contains
Pyrenees Team 1988;Roureefa/. 1989). The major Pyrenean a limited Mesozoic series. The stratigraphic relationships
thrust belt developed mainly by foreland propagating thrust- between syntectonic deposits and structures led Verges &
ing between Upper-Cretaceous and Miocene times (Munoz et Munoz (1990) to establish a two-stage evolution of the unit.
al. 1986; Martinez et al. 1986; Verges & Martinez 1988). They postulate an initial emplacement between the Lower
Coeval break-back and piggy-back sequences have also been Eocene and the Upper Eocene and later development of the
reported recently (Verges & Munoz 1990). structures in the Upper Eocene-Lower Oligocene.
In the ECORS profile three main thrust sheets can be The lithostratigraphy of the SCU sheets is characterized
distinguished (Fig. 2). The uppermost sheet is called the by the almost complete absence of Palaeozoic basement
266
CONTRASTING ROTATIONS WITHIN THRUST SHEETS
rocks and also by the thinning of the Mesozoic cover from the include the maximum numberof fold tests, where the age of
northern unit towards the south (Puigdefabregas & Souquet remanence can be identified as being pre-, syn- or post-
1986). The Boixols unit is characterized by a thick Mesozoic folding (see McCaig & McClelland 1991, this volume, for
series of about 5000 m. The Lower Cretaceous series is further discussion). Samples for the palaeomagnetic study
thinner in the Montsec unit and is completely absent in the were collected with a portable drill and oriented with both
Sierras Marginales unit where Upper Cretaceous rocks di- magnetic and solar compasses. Typically, five to twelve
rectly overlie Jurassic. samples from different horizons were taken per site (105
In the Pedraforca area three separate thrust sheets can be sites. 748 samples). Sites were chosen to yield spatial and
distinguished (Fig. 2). The lowermost one belongs to the temporal distribution of possible structural rotations in both
Lower Thrust Sheets of the Pyrenees and is called the Cadf N-S and E-W transverses. Several sedimentary rock-types
thrust. It is constituted by basement rocks (Devonian and ranging from Jurassic to Eocene in age were sampled from
Lower Carboniferous), a thick Stefano-Permian series and a different tectonic units, but it was not always possible to find
reduced Mesozoic and a thick Palaeogene series (Garumnlan similar and contemporaneous lithologies in each structural
facies and Lower and Middle Eocene). The Pedraforca thrust unit due to the pre-thrusting basinal facies distribution, spe-
is subdivided into two units, the Lower Pedraforca thrust cially in a N-S section. 35% of the samples were found to be
sheet, and the Upper Pedraforca thrust sheet, structurally unsuitable for our purpose due to either a weak intensity of the
equivalent to the Boixols sheet (Verges & Martinez 1988). natural remanent magnetization (NRM) or an erratic behaviour
The purpose of this palaeomagnetic study is to define any of the magnetic moment upon demagnetization. These un-
structural rotations which have occurred during the Alpine suitable samples were mostly the continental red-sandstones
thrusting. The aim was to find rocks which carry remanence and lacustrine limestones of the Garumnian facies (uppermost
components pre-dating the deformation. These magnetiza- Cretaceous-lowermost Palaeocene) which outcrop widely in
tion vectors would act as passive markers and record any such the Tremp piggy-back basin in the Montsec thrust sheet and
rotation. The sampling strategy was, therefore, designed to also in the Sierras Marginales unit. Similarly, most of the
11 / / Upper Cretaceous
•A:U\
MES020IC UNITS
UD»l CrtlKlOui
Figure 2, (a) Balanced cross-section of the cover thrust sheets across the south central Pyrenees (southern part oftheECORS profile), (Verges & Munoz
1990); (b) Structural cross-section of the eastern termination of the Montsec and Serres Marginals thrust sheets through the Oliana anticline, (Verges &
Muiioz 1990); (c) Balanced cross-section through the eastern Pyrenees (Pedraforca thrust pile), (Verges & Martinez 1988). See Figure 1 for location.
267
J. DiNARES £7 AL.
W,UP W,UP
/ 1'
/
0 1 (1)250
/ \ N,N
// \
H
s,s ^ N,N NHM 0 831 mA/M
1 ( 1 i 1
45000 \ z*-^^^
/ ^ NRM 2.011mA/M
- \
\ a 89J21-4a Si
1mA/M
- ^
89J31 -4b \
E.DOWN
b
E.DOWN
• -e- - J^RMJ0.769mA/M
0,3mA/M
^ d
JDT39-2a
E.DOWN
I 89J38-4b
I
I
E.DOWN
Figure 3. Typical orthogonal plots of stepwise thennal demagnetization for four representative specimens: (a) from site 89J31 (Lower Jurassic): (b) from
site JDT39 (Aptian); (c) from site 89J21 (Eocene); (d) from site 89J38 (Eocene). Solid (open) symbols represent projections on the horizontal (vertical)
plane. Vector end-points are plotted in in-situ coordinates. H stands for high-temperature characteristic component, and L stands for low-temperature recent
overprint component.
shallow-water limestones mainly of Upper Cretaceous age demagnetization was followed by thermal cleaning. The bulk
were found to be unsuitable for a palaeomagnetic analysis. magnetic susceptibility of pilot specimens for each site was
The bulk of suitable sites consisted of grey marls and marly- monitored by using a low-field bridge during the thermal
limestones of Lower Cretaceous and Lower Eocene in age treatment in order to detect any mineralogical change. The
(Figs 3 & 4). The intensity of the NRM for those suitable rock- results from the demagnetization sequence of each specimen
types ranges between 1 x IQ-^A m ' and 2 x 10' A m '. The were plotted on orthogonal vector plots (Dunlop 1979). Most
directions and intensities of the samples were measured using of the samples showed a multicomponent structure of the
a two-axis Cryogenic Consultants Ltd. magnetometer which NRM upon demagnetization (Fig. 3). Characteristic
can measure a minimum magnetization of 1 x 10^ A m'' with remanence components were calculated by numerical inter-
adequate accuracy. active computer line-fitting in the orthogonal vector-projec-
tions (Torsvik 1986) and/or using the Linefind algorithm
(Kent et al. 1983). Fisher statistics (Fisher 1953) were used
Palaeomagnetic methods and results to compute the mean directions.
Since most of the Mesozoic lithologies sampled fall into
All characteristic directions were determined after specimen the normal-polarity Cretaceous quiet zone, no positive reversal
demagnetization (2 or 3 specimens per sample). Specimens test was found at any site. The primary nature of the
were subjected to either progressive alternating field (AF) characteristic component has been demonstrated by a positive
demagnetization using an apparatus capable of peak fields of fold test on Lower Cretaceous sites from the Boixols thrust
55 mT (550 Oe) or stepwise thermal demagnetization using sheet. The Eocene sites from the Oliana anticline pass both
a non-magnetic furnace. Both instruments were shielded reversal and fold tests. Similar behaviour of the NRM upon
from the Earth's magnetic field. In some instances AF demagnetization of the same rock-types in areas where no
268
CONTRASTING ROTATIONS WITHIN THRUST SHEETS
•oauERES
UPPER CKETACBOUS
FLYSCH
BASIN tUJamiCHTlAN
LOHER CRETACEOUS
MEN Fm.
Figure 4. Simplified structural map of the eastern termination of the Boixols thrust sheet and location of the eight sites considered for the fold test (See
text) and location of cross sections from Figure 5.
fold test can be performed allows confidence in the assumption magnetization (IRM) in increasing magnetic fields and the
of the primary nature of the characteristic component. In-situ AF and/or thermal demagnetization of a simple or composite
primary declinations and inclinations have been corrected for IRM has proved to be a great help in understanding the
dip of the strata by rotation of the bedding plane to the character of some NRM demagnetization patterns. This will
horizontal about the strike. Where plunging structures were be the subject of a publication elsewhere.
observed, the tectonic correction (unfolding) incorporated
unplunging followed by unbuckling. The reversed order did
not produce substantial differences. This is probably due to Boixols thrust sheet
the particular geometric relationship of the remanence and The internal structure of the Boixols thrust sheet consists of
structures. In any case the fold axes plunged no more than 15 kilometric-scale E-W trending folds (Boixols and St. Cornell
degrees. anticlines, Santa Fe syncline), (Fig. 4). These affect a thick
Rock-magnetic experiments were also carried out in order Mesozoic series and are the result of the modification by the
to determine the mineralogy involved in each rock-type. The Alpine compression of extensional structures associated with
study of the progressive acquisition of isothermal remanent the formation and the evolution of the Organya basin
J30
Figure 5. N-S cross-sections with the location of eight Lower-Cretaceous sites from the Boixols thrust sheet that pass the fold test (see Fig. 4 for location).
269
J. DIN ARES £7 AA.
270
CONTRASTING ROTATIONS WITHIN THRUST SHEETS
Montsec thrust sheet possible to conclude that there is no evidence for a post-
Triassic rotation of the unitfrom Van Dongen's data. One site
Ten reliable sites are located in the Montsec thrust sheet (Fig.
was sampled in the uppermost part of the thrust sheet (Lower
4). Five are located in the central-eastern part (one in the
Eocene) in order to test the validity of that conclusion and to
Lower Jurassic, one at the Jurassic-Cretaceous boundary, one
obtain an Eocene palaeodirection to compare with other
in the Lower Cretaceous, and two in the Upper Cretaceous).
Eocene directions from different structural units in the south-
In the western part of the Montsec thrust sheet four sites were
em Pyrenees. Two component remanences were found, and
sampled (one in Upper Cretaceous rocks, and three in Eocene
the site-mean directions of the high-temperature component
sediments). Only one site (Upper Cretaceous) is located in
(H) before and after tectonic correction for the sites and the
the easternmost part of the Montsec thrust sheet.
tectonic-corrected palaeodeclination of the primary compo-
All sites mentioned above showed a two-component be-
nents are shown in Figure 8 (see discussion below).
haviour of the NRM upon demagnetization plus a soft viscous
component removed at the first treatment step. The unblock-
ing temperature spectrum for the two characteristic compo- Oliana foreland anticline
nents differs slightly for each site but we can differentiate a
A break-back thrust sequence coeval with forward-imbricat-
low-temperature (150-300°C), low-coercivity (10-25 mT)
ing thrusting has been postulated in the Oliana area in the
component (L), and a high-temperature (250-580°C), high-
easternmost part of the Montsec thrust sheet (Verges &
coercivity (20-55mT) component (H). The L component
Muiioz 1990). The break-back sequence is deduced from the
always falls close to the present Earth's magnetic field before
relationship between thrusts and four related synorogenic
any tectonic correction (in-situ coordinates) and therefore is
conglomerate units, each unconformably overlying the
interpreted as a recent overprint. Sometimes the soft viscous
previous one and truncated by a more hinterland thrust.
component can mask the L component. Neither a positive
These unconformities suggest a synchronous growth of the
fold test nor a within-site positive reversal test has been found
Oliana anticline with the development of the break-back
in the Montsec thrust sheet. The lack of a positive fold test is
sequence. The anticline is formed by the piling-up of horses
due to the similar tectonic attitude of the strata in most areas
over the sole thrust and thus accommodating the foreland
of the thrust sheet (Fig. 2). An open syncline (Tremp-Graus
deformation (Verges & Murioz op. cit.). Timing of the events
piggy-back basin) constitutes the thrust sheet, but no similar
and rates of deformation has been attempted by means of a
suitable lithologies can be sampled from both limbs. Some of
magnetostratigraphy study (Burbank et al. 1991).
the Eocene sites carry a reversely magnetized component H
and a between-site positive reversal test thus exists. Rock- Samples were obtained from the uppermost part of the
magnetic properties, similarity of demagnetization patterns foreland sequence, the Igualada marls (Bartonian-Priabonian)
and consistency of data from other areas with positive which outcrop in the core of the anticline (one site from each
palaeomagnetic field tests allows us to be confident of the limb) and from the Uppermost Eocene-Lower Oligocene
primary character of component H. continental conglomerates (one site in palustrine sandstones
from the western limb). The remanence consists of two
All 15 sites sampled in the southernmost unit of the SCU
components, L and H. The component H from the two
(Sierras Marginales Unit) proved unsuitable for
Igualada marls sites passes the reversal test between sites and
palaeomagnetic purposes.
thus confirms the primary character of that component oth-
erwise already argued by a positive fold test (Fig. 7).
Pedraforca area thrust pile
Three main separate thrust sheets can be distinguished in the
Pedraforca area (Cadi, Lower Pedraforca and Upper OLIANA ANTICLINE
Pedraforca thrust sheets). As a pilot study one site in each of
these thrust sheets was sampled. The age of the lithologies
sampled were different in each thrust sheet due to their
different lithostratigraphy. Lower Eocene marls were sam-
pled in the Cadi unit (site 89J21), Upper Cretaceous sandy-
limestones in the Lower Pedraforca thrust sheet (site 89J20)
and Lower Cretaceous marly-limestones in the Upper
Pedraforca thrust sheet (site 89J11), proved unsuitable.
Some of the Permian and Triassic rocks from the Cadi unit
were the objective of an early study (Van Dongen 1967). At
that time the Cadi area was considered autochthonous and the AFTER
palaeomagnetic results were interpreted in the light of previ- COMPONENT H
ous results from rocks of the same age in Europe and Iberia.
Those results were roughly consistent with other Permian and
Figure 7. Equal-area projections showing site mean directions and their
Triassic palaeodirections from the rest of the Iberian plate. respective cones of 95% confidence of H component for three sites from the
Since the Cadf unit is accepted to be allochthonous it is now Oliana anticline. Site mean directions are shown (a) in-situ, (b) after tectonic
271
J. DiNARES ET AL.
Discussion NW (D/I = 291/57, a,j = 3.2 and D/I = 306/67, a,, = 6.8 re-
spectively) which are rotated about 25° and 10° in a
The results of the palaeomagnetic investigation of the SCU counterclockwise sense when compared with the expected
and adjacent areas are plotted in Figure 8. These local declination calculated from the Hauterivian-Barremian
palaeodeclinations have been compared with expected refer- palaeopole from Galdeano era/. 1989. The fact that site 89J5,
ence directions for stable Iberia, recalculated for the location immediately east of sites 89 J8 and 89J9, and site 89J7 located
of the SCU (centred at Tremp) from palaeopoles given by south of those two sites and both of Aptian age show no
several authors (Jurassic: Shott et al. 1981; Hauterivian- significant rotation when compared with the expected local
Barremian and Aptian: Galdeano et al. 1989; Upper Creta- value, leads one to suspect the apparent rotation of sites 89J8
ceous: Van der Voo & Zijderveld 1971; Eocene: Westphal et and 89J9. This would imply that either the single Iberian
al. 1986). A significant structural rotation may be considered palaeopole determined for the Lowermost Cretaceous is not
to have occurred when there is no overlap of the confidence adequately constrained or, alternatively, two more possibilities
cones from the reference direction and the local direction could explain the declination offset between the Hauterivian-
from rocks of that age (rotation usually >15°). Alternatively, Barremian site-mean directions and the Aptian-Albian site-
one can look for relative rotations between sites of the same mean directions: (1) effect of the anticlockwise rotation of the
or similar age from the same or different tectonic units. In this Iberian plate as a whole during the lowermost Cretaceous; (2)
way any uncertainty in the determination of the reference local rotation prior to both thrusting and sedimentation of the
directions can be avoided, as the Jurassic and Lower Creta- Aptian-Albian sedimentary sequences. Further
ceous reference directions are determined from single stud- palaeomagnetic work is needed in both the Pyrenean belt and
ies. the rest of the Iberian plate to assess this question properly.
It is obvious from Figure 8 that substantial rotation occurs Departures from the general pattern can be related to local
within and between tectonic units. The salient features of the structure. The Lower Cretaceous site 89J14 located in a
data indicate clockwise and counterclockwise rotation at the syncline structure (Peracal§ syncline) between the Flamisell
western and eastern edges of the units respectively, this and Noguera Pallaresa rivers (Fig. 8) shows a 45° clockwise
rotation decreasing to zero in the central part. This general rotation, in contrast to no rotation observed from other sites
pattern is seen in both the Montsec and Boixols units, al- in the vicinity. The Peracal? syncline is bounded by the
though the more complex internal structure of the Boixols Morreres backthrust in the north and by a syn-sedimentary
unit is reflected in more variable rotations inferred from the extensional fault to the south (Berastegui et al. 1990). The
palaeomagnetic data. The palaeomagnetic declinations from clockwise rotation of site 89J14 is interpreted as a local
the central/eastern part of the Montsec thrust sheet (sites rotation of the Peracal9 syncline during the Alpine compres-
JDT48, Jurassic; JDT49, Lower Cretaceous; sites JDT58 and sion. Likewise, to the west of the Flamisell river, the Lower
JDT60, Upper Cretaceous) show no significant rotation when Cretaceous site 89J40 shows an anticlockwise rotation of 37°,
compared with the expected directions. This is supported by whereas further to the west, site 89J31 located in Lower
unpublished data from a magnetostratigraphic study on the Jurassic rocks of the Alins subunit and site 89J32 located in
Lower Eocene further East in the Montsec thrust sheet Lower Cretaceous rocks of the Las Aras subunit show 45° and
(Noguera Pallaresa river) which shows that no rotation oc- 38° clockwise rotation, respectively. It is interesting to note
curred there (Pascual, pers.com.). In contrast, the Lower that the discordant anticlockwise rotation of site 89 J40 occurs
Eocene sites 89J36,89J38,89J12 from the westem part of the in an area where macrostructures tend to be oriented N-S (Fig.
Montsec thrust sheet show a significant clockwise rotation of 8), in contrast to the general E-W trend. This feature and the
31°, 26° and 15° respectively. Note that the rotation de- detailed structure north of this area (Serra Fallada, Perves) are
creases towards the E. The 24° counterclockwise rotation not yet well understood and make detailed interpretation of
from a single site in Upper Cretaceous rocks (89J24) located the palaeomagnetic results difficult. Sites 89J29 and 89J30
at the eastern part of the Montsec thrust sheet N of the Oliana located in a thrust slice containing Lower Cretaceous rocks
anticline is not statistically significant due to the high within- (Fig. 8) show large amounts of rotation (123° ami- and 128°
site dispersion (0,^=32°). clockwise, respectively). The westemmost site sampled in
The distribution of the determined palaeodeclinations the SCU (89J35) is located in Upper Cretaceous rocks from
from 27 reliable sites throughout the Boixols unit and related a structural subunit related to the Boixols thrust sheet N of
subunits (Serra Fallada, Perves, Las Aras, Alius) is more Campo. This site shows a 61° clockwise rotation when
complex. There is palaeomagnetic evidence that no substan- compared with the expected palaeodeclination.
tial rotation occurred in the central-eastern part of the thrust
sheet (between the Segre and Flamisell rivers) when site-
mean directions from 12 Lower Cretaceous sites and 3 Upper Structural interpretation
Cretaceous sites are compared with the expected directions.
A single Lower Cretaceous site (89J4) located east of the The construction of balanced sections as discussed by several
Segre river in the eastern edge of the Boixols thrust sheet authors (e.g. Dahlstrom 1969; Hossack 1979) is a widespread
shows 23° counterclockwise rotation. It should be noted that practice in orogenic belts especially in their external parts.
site 89J9 (Hauterivian) and site 89J8 (Upper Barremian) both Balanced and restored sections are particularly valuable in
located in the Segre valley show declinations towards the W- that they can be used to calculate estimates of bulk shortening
272
Permian Jurassic Low. Cret. Upp. Cret. Eocene
4 t ! ! t
10 15 Km
Figure 8. Palaeomagnetic data on structural map of the central eastern Pyrenees. Arrows represent the site-mean or group of sites-mean magnetic declinations ( site location at the base of the arrow); numbered directions
are sites from this study (underlined sites for prefix JDT otherwise 89J), lettered arrows correspond to data from other studies (V, Van Dongen, 1967; B, Burbank & Puigdefabregas, 1985; M, McClelland & McCaig, 1989)
J. DiNARES ET AL.
in orogenic belts. Usually, balanced sections are choosen to estimates of displacement can be made.
trend perpendicular to orogenic strike since the maximum The palaeomagnetic rotations from the Boixols unit can be
displacement and transport direction are generally perpen- explained in a similar way but additional complications may
dicular to orogenic strike. The basic approaches to section arise west of the Flamisell river. In fact, an interference
balancing assume plane strain, or conservation of cross- pattern between two systems of folds and thrusts has been
sectional area. In other words, the area of section has not described recently not much further W by Souquet &
changed during deformation. A section which crosses ob- Deramond (1989) and Fondecavez-Wallez et al. (1989).
lique thrust ramps cannot be restored to the undeformed state The 20° anticlockwise rotation (Fig. 8) of the foreland
and balancing oblique sections is inherently difficult as area Oliana duplex-anticline has to be regarded as a consequence
may not be maintained in the section plane. of its growth. The translation of the horses across the footwall
There are several features that can produce area reduction ramp may account for such rotation if the transport direction
in across-section: e.g. strike elongation ortectonic compaction. is not perpendicular to the thrust trace. This kind of rotation
Therefore, it is possible to take account of estimated volume observed in the foreland basin immediately adjacent to a
loss when constructing balanced cross-sections, although the major oblique ramp (the Segre thrust) has been also reported
basic assumption of plane strain leads to a minimum estimate in the vicinity of the oblique ramp of the Pedraforca thrust
of shortening (Hossack 1979). In addition to the volume loss sheet in a preliminary magnetostratigraphic study (Burbank
effects, the existence of structural rotations about the vertical & Puigdefabregas 1985). There, an anticline similar to the
or inclined axis can also induce errors in balanced section Oliana anticline (Vilada anticline) developed in the footwall
calculations. The estimate of shortening would be an under- of an imbricate stack (Martinez et al. 1988).
or over-estimate depending on whether material has moved The palaeomagnetic results of the pilot study in the
into or out of section. Pedraforca area thrust pile are twofold: the Eocene site
In the palaeomagnetic study of the SCU one can compare located in the Cadi thrust sheet shows no significant rotation
areas within the same structural unit which relate in different when compared with the expected reference value) and the
ways to the thrust plane. There are areas (central/eastern part palaeodeclination is different when compared with other
of the Montsec and Boixols thrust sheets) that relate to thrust Eocene paleodeclinations determined from other structural
planes trending E-W. Since the transport direction for those units in this study (Fig. 8). The Eocene directions from the
thrusts has been established to be N-S, those areas are thus Cadi thrust sheet agree with previous data from Permian and
associated with frontal ramps. The Montsec thrust sheet can Triassic rocks from the Cadi unit (Van Dongen 1967) which
be connected westwards below the Eocene conglomerates also show negligible rotation when compared with the expected
with the Cotiella thrust along the N-S striking Mediano values. On the other hand the results from a single site in the
anticline. The trace of the thrust plane varies from an E-W Lower Pedraforca thrust sheet shows 50° anticlockwise rota-
direction in the frontal part to a NW-SE swinging to a near N- tion when compared with the expected Upper Cretaceous
S direction in the Mediano anticline (dextral oblique-ramp). direction. These two facts suggest that the rotation might
Towards the E the Montsec thrust sheet merges with the have occurred due the piling up of the units. A piggy-back
Sierras Marginales thrust and constitutes the NE-S W striking thrusting sequence has been demonstrated for the Pedraforca
Segre thrust (sinistral oblique-ramp). The Boixols thrust culmination stack (Verges & Martinez 1988). This mecha-
plane has a similar pattern to that of the Montsec thrust sheet. nism could be responsible for the rotation of the higher and
It can be extended westwards in a NW-SE trending direction older structural units whereas rotation of the younger units
below the N-S striking Turbon anticline. The sinistral ramp could be minimal. Such a hypothesis has not been fully tested
in the eastern edge of the thrust sheet is less evident and is in the Pedraforca pile since the pilot site located in the Upper
shorter than the Montsec one. Pedraforca thrust sheet did not yield reliable results. Further
The palaeomagnetic rotations from the Montsec thrust palaeomagnetic work will be necessary before any conclu-
sheet can be explained in terms of structural rotations about sion can be made.
a vertical axis related to the oblique ramps. Clockwise Palaeomagnetism has proven to be a valuable tool for
rotations occur in areas associated with dextral oblique- identifying structural rotations about the vertical axis in the
ramps whereas anticlockwise rotations concentrate in areas southern Pyrenean thrusts. The presence of structural rota-
related to sinistral oblique-ramps. Note that rotation in- tions at the edges of the thrust sheets would hamper any
creases westwards in the former case and reflects the curva- attempt of cross-section balancing. However, the fact that
ture of the thrust plane (Fig. 8). Rotation can arise due to there is evidence for no rotation in the central part of the
existence of drag produced by translation of the thrust block South-Pyrenean Central Unit allows the balancing techniques
over the ramp. This will require differential displacement of to be applied to sections across that area (e.g. the ECORS
the block at the different points but rotation is not the only seismic profile) which in any case would be the most rational
response to such difference of displacement and the role of choice for constructing a section, avoiding areas containing
local structures will have to be taken into account before any oblique ramps.
274
CONTRASTING ROTATIONS WITHIN THRUST SHEETS
References
Berastegui X., Losantos, M. & Garcia Senz, J. M. 1990. Structure and Martinez, A., Verges, J. & Mufioz, J. A. 1986. Secuencias de propagacion del
sedimentary evolution of the Organya basin (Central South Pyrenean sistema de cabalgamientos de la terminacion oriental del manto del
Unit, Spain). Bulletin de la Sociele geologique de Fiance, 8, 251-263. Pedraforca y relacion con los conglomerados sinorogenicos. Acta
Burbank, D. W. & Puigdet^bregas, C. 1985. Chronologic investigations of Geologica Hispdnica, 23, 119-128.
the South Pyrenean basins: preliminary magnetostraligraphic results McElhinny, M. W. 1964. Statistical significance of the fold test in
/)om(fte/?i/)o//Bairn. International Association of Sedimentologists, 6th paleomagnetism. Geophysical Journal of the Royal Astronomical Soci-
European Regional Meeting, Abstracts, Lleida, Spain, 66-69. ety of London, 8, 338-340.
Burbank, D. W., Verges, J., Mufioz, J. A. & Bentham, P. 1991. Coeval McClelland Brown, E. 1983. Palaeomagnetic studies of fold developement
hindward- and forward-imbricating thrusting in the Central Southern and propagation in the Pembrokeshire Old Red Sandstone.
Pyrenees, Spain: Timing and rates of shortening and deposition. Geo- Tectonophysics, 98, 131-149.
logical Society America Bulletin (in press). McClelland, E. & McCaig, A. 1989. Palaeomagnetic estimates of rotations
Camara, P. & Klimowitz, J. 1985. Interpretacidn geodinamica de la vertiente in compressional regimes and potential discrimination betwen thin-
centro-occidental surpirenaica (Cuencas de Jaca-Tremp). Estudios skinned and deep crustal deformation, /n: Palaeomagnetic rotations and
Geo/o,i;/coi, 41, 391-404. continental deformation. NATO ASI series C, 254, 365-379.
Choukroune, P. & ECORS Team. 1989. The ECORS Pyrenean deep seismic Mufioz, J. A.. Martinez, A. & Verges, J. 1986. Thrust sequences in the eastern
profile reflection data and the overall structure of an orogenic belt. Spanish Pyrenees. Journal of Structural Geology, 8, 399-405.
Tectonics, 8, 23-39. Mutti, E., Rosell, J., Allen, G. P., Fonnesu, F. & Sgavetti, M. 1985. The
Dahlstrom, C. D. 1969. Balanced cross sections. Canadian Journal of Earth Eocene Baronia tide dominated delta-shelf system in the Ager basin.
Sciences, 6, 743-747. Exc. Guide-book, 6th European Regional Meeting, Lleida. Spain, 579-
Dinares Turell, J. 1989. Palaeomagnetic evidence for non-rotational thrust- 600.
ing in the South Pyrenean Central Unit. International Association of Nijman, 1990. Thrust sheet rotation ? - The south Pyrenean Tertiary basin
Geomagnetism and Aeronomy, (I.A.G.A), 6th Scientific Assembly. Ab- configuration reconsidered. Geodinamica Acta, 4, 17-42.
stracts, Exeter, UK., 236. Puigdefabregas, C. & Souquet, P. 1986. Tecto-sedimentary cycles and
Dunlop, D. J. 1979. On the use of Zijderveld vector diagrams in depositional sequences of the Mesozoic and Tertiary from the Pyrenees.
multicomponent palaeomagnetic studies. Physics of the Earth and Plan- Tectonophysics, 129, 173-203.
etary Interiors, 20, 12-24. Roure, F., Choukroune, P., Berastegui, X., Mufioz, J. A., Villien, A.,
ECORS Pyrenees team 1988. The ECORS deep reflection seismic survey Mathemon, P. Baryet, M., Seguret, M., Camara, P. & Deramond, J. 1989.
across the Pyrenees. Nature, 331, 508-511. ECORS deep seismic data and balanced cross-sections: Geometric
Elderedge, S. & Van der Voo, R. 1988. Paleomagnetic study of thrust sheet constraints on the evolution of the Pyrenees. Tectonics, 8, 41 -50.
rotations in the Helena and Wyoming salients of the northern Rocky Schott, J. J , Montigny, R.. Thuizat, R. 1981. Paleomagnetism and potas-
Mountains. Geological Society of America Memoir, 171, 319-332. sium-argon age of the Messejana Dike (Portugal and Spain): angular
Farrell, S. G., Williams, G. D. & Atkinson, C. D. 1987. Constraints on the age limitation to the rotation of the Iberian Peninsula since the Middle
of movement of the Montsech and Cotiella thrusts, south central Pyr- Jurassic. Eai-th and Planetary Science Letters, 53, 457-470.
enees,Spain. Journal of the Geological Society of London,114,901-914. Schwartz, S. Y. & Van der Voo, R. 1984. Paleomagnetic study of thrust sheet
Fisher, R. A. 1953. Dispersion on a sphere. Proceedings of the Royal Society rotation during foreland impingement in the Wyoming-Idaho overthrust
of London. A- 217, 295-305. belt. Journal of Geophysical Research, 89, 10077-10086.
Fondecava-Wallez, J. M., Souquet, P. & Gourinard, Y. 1989. Enregistrement Seguret, M. 1972. Etude tectonique des nappes et series decodes de la partie
sedimentaire de I'eustatisme et de la tectonique dans la serie turbiditique centrale du versant sud des Pyrenees. Publications de I'Universite de
du Cretacedes Pyrenees centro-meridionales (Groupe de Vallcarga, n. Sciences et Techniques de Languedoc, serie Geologie Structurale, 2,
gr., Espagne). Comptes Rendus de l'Academic de Sciences. Paris, 308, Montpellier. 155p.
Serie II, 1011-1016. Simd, A. & Puigdefabregas, C. 1985. Transition from shelf to basin on an
Freeman, R., Sabat, F., Lowrie, W. & Fontbote, J. M. 1989. Palaeomagnetic active slope, upper Cretaceous, Tremp area, southern Pyrenees. Exc.
results from Mallorca (Balearic Islands, Spain). Tectonics, 8, 591-608. Guide-book. 6th European Regional Meeting, Lleida, Spain, 63-108.
Galdeano, A., Moreau, M., Pozzi, J. P.. Berthou, P. Y., & Malod, J. A. 1989. Souquet, P. 1967. Le Cretace superieur sud-pyreneen en Catalogne, Aragon
New palaeomagnetic results from Cretaceous sediments near Lisboa et Navarre. These Doctoral Sc. Nat., Fac. Sc. Toulouse, 529p.
(Portugal) and implications for the rotation of Iberia. Earth and Plan- Souquet,P. & Deramond, J. 1989. Sequencedechevauchementsetsequences
etary Science Letters, 92,95-106. de depot dans un bassin d'avant-fosse. Exemple du sillon cretace du
Garrido-Megias, A. & Rios, L. M. 1972. Si'ntesis geologica del Secundario versant sud des Pyrenees (Espagne). Comptes Rendus de TAcademic de
y Terciario entre los rios Cinca y Segre (Pirineo central de la vertiente Sciences, Paris, 309, 137-144.
surpirenaica, provincias de Huesca y Leridn). Boletin Geologico v Minero, Torsvik, T. 1986. lAPD - Interactive Analysis of Palaeomagnetic Data
83, 1-47. (User-guide and program description). Internal Publication, University
Grubbs, K. L. & Van der Voo, R. 1976. Structural deformation of the Idaho- of Bergen, Institute of Geophysics.
Wyoming overthrust beU (USA). Tectonophysics, 33, 321-336. Van der Voo, R. & Zijderveld, J.D.A. 1971. Renewed paleomagnetic study
Hossack, J. R. 1979. The use of balanced cross sections in the calculation of of the Lisbon volcanics and implications for the rotation of the Iberian
orogenic contraction: A review. Journal of the Geological Society of Peninsula. Journal of Geophysical Research, 76, 3913-3921.
London, 136,105-711. Van Dongen, P. G. 1967. The rotation of Spain: palaeomagnetic evidence
Hudson, M. R., Reynolds, R. L. & Fishman, N. S. 1989. Synfolding from the eastern Pyrenees. Palaeogeography, Palaeoclimatology and
Magnetization in the Jurassic Preuss Sandstone, Wyoming-Idaho- Utha Palaeocology, 3,417-432.
Thrust Belt. Journal of Geophysical Research. 94, 13681 -13705. Verges, J. & Martinez, A. 1988. Corte compensado del Pirineo oriental:
Kent, J. T., Briden, J. C , & Mardia, K.V . 1983. Linear and planar structure geometria de las cuencas de antepais y edades de emplazamiento de los
in ordered multivariate data as applied to progressive demagnetization of mantos de corrimiento. Acta Geologica Hispdnica, 23, 95-106.
palaeomagnetic remanence. Geophysical Journal of the Royal Astro- & Mufioz, J. A. 1990. Thrust sequences in the southern central
nomical Society, 75, 593-621. Pyrenees. Bulletin de la Societe geologique de France, 8, 265-272.
Kotasek, J. & Krs, M. 1965. Palaeomagnetic study of tectonic rotations in the Westphal, M., Bazhenou, M. L., Lauer, J. P., Pechersky, D. M. & Sibuet, J.
Carpathian Mountains of Czechoslovakia. Palacogeography. C. 1986. Paleomagnetic implications on the evolution of the Tethys belt
Palaeoclimatology and Palaeocology. 1, 39-49. from the Atlantic Ocean to the Pamirs since the Triassic. Tectonophysics,
Li, Z. X., Powell, McA. & Schmidt, P. W. 1989. Syn-deformational remanent 123, 37-83.
magnetization of the Mount Eclipse Sandstone, central Australia. Geo- Williams, G. D. & Fischer, M. W. 1984. A balanced section across the
physical Journal Internadonal, 99,205-222. Pyrenean orogenic belt. Tectonics, 3, 773-780.
Losantos, M., Berastegui, X. Mufioz. J. A. & Puigdefabregas, C. 1988. Corte
geologico cortical del Pirineo Central (perfil ECORS): Evolucion
geodinamica de la cordillera Pirenaica. In: Simposio sobre cinturones
orogenicos. SGE, 1988, II Congreso Geologico de Espaiia, Granada,
1988,7-16.
275
PART FIVE
Alps
The Alps - a transpressive pile of peels
H. Laubscher
Abstract: One of the striking features of recent reflection seismograph traverses in the Alps is the
strong reflection band of the 'Penninic front'. It cuts discordantly through the complex geometry
of the Pennine nappes and ties laterally into the Miocene Simplon line. The early Miocene ductile
component of this line continues into the base of the roof zone of the Lepontine dome with its
retrograde EW shearing and stretching. A kinematic model compatible with the data may be
constructed on the concept of an erogenic lid subjected to dextral transpression by the early
Miocene Adriatic indenter (Insubric-Helvetic phase). In this model, the Penninic front reflections
mark the shear zone at the base of the orogenic lid. The frontal part of this lid is the fold-and-thrust
belt of the Helvetic nappes and equivalent contemporaneous thrusts such as those of the Prealps.
The lid was deformed and partly eroded during the subsequent Windows phase which pushed up
the External Massifs. The Insubric-Helvetic lid was subdivided into a mosaic of sublids which
accommodated the divergence of translation at the NW-comer of the early Miocene Adriatic
indenter. The movement of this indenter had a component of about 150 km translation to the west
(dextral strike-slip along the Insubric line), and 100 km to the north (estimated translation and
shortening of the Helvetic nappes). As a consequence, that part of the brittle orogenic lid north of
the Insubric line (Silvretta sublid) was pushed in a northerly direction, while that at the western
front of the Adriatic indenter (Gran Paradise sublid) moved mere to the west. In between, the
Lepontine sublid was stretched axially by normal faulting and uplifted by tectonic underplating
due to particularly intensive N-S compression in the narrowest part of the central Alps. It
participated as a pull-apart domain in the dextral motion along the Insubric line and was subject
to rapid tectonic denudation comparable to that in metamerphic core complexes. Adjacent to the
west (external side of the lid), the Prealps sublid accommodated axial stretching largely through
complementary strike-slip along its bordering transverse zones (Giffre and Kander lines). Below
the orogenic lid, the middle and lower crust were deformed disharmonically. In the westem Alps,
parts of the lower crust and even the upper mantle were peeled off from the subducted slab and
wedged into the middle crust, whereas in the central Alps, during this particular phase, subductien
of lower and some middle crust seems to have been dominant, except for the most intensely
squeezed Lepontine domain, where middle crust was piled up und produced an uplift of up to 20
km. The pile of Alpine nappes participating in this phase therefore had three components: those
inherited from previous phases and forming passively the erogenic lid; those developing at the base
and the front of the lid; and those forming dishannonically in the middle and lower crust below the
lid.
One of the new and unexpected tectonic features brought to 1903): at its base it produces peel nappes through predomi-
light by the recent reflection seismic traverses of the Alps is nantly simple shear, and at its leading edge frontal fold and
that giving rise to the reflection band of the 'Penninic front' thrust belts.
(Bayer et al. 1987). This band of strong reflections cuts This paper focuses upon the quantitative kinematics of this
discordantly the complex structures of the Penninic nappes early Miocene orogenic lid in the central Alps and their
and therefore marks a shear zone that is younger. To the north surroundings, with the aim of arriving at a new perspective of
and east itjoins the Simplon-Rhone valley line (Bearth 1956, some long-standing problems of Alpine tectonics.
Steck 1984, Mancktelow 1985) which is equally discordant
to the Penninic nappes and is essentially of Miocene age
(Mancktelow 1985). The cataclastic Simplon fault was T h e p r o b l e m of quantification in Alpine tectonics
preceded by a ductile shear zone of some thickness which
extends to the roof of the Lepontine dome (Merle e?fl/. 1989) Quantification in tectonics, when properly done, proceeds
and is there dated as (latest Oligocene to) early Miocene from a set of discrete observations, by interpolation, to
(Hurforde?a/. 1989). Thisassociationofthe'Fenninicfront' continuous geometric bodies, then to a history of these
reflection band with the brittle-ductile transition character- bodies, from an initial position in an undeformed state to their
izes the base of an'otogenic lid'(Laubscher 1983), the cold present position and state (kinematics), and finally to an
andstrongupper 10-20kmofacollisionzonethatisthemost explanation of this historical process in terms of causal
effective stress guide between the colliding segments of relations, mainly applying theoretical mechanics (dynam-
continental crust. It acts as 'traineau ecraseur' (Termier ics). This procedure is necessarily an iterative, trial-and-error
277
H. LAUBSCHER
\ -^^ Zurich
- v ,- A
/
50 km
^
Bern e ^ -
^sty^ =J a e 0 U
\i e
—/Co 3
Milan
^ ^PProx/mate subsurface border o'^
Southern Alps
/o B a s i n
Fore-, Austroalpine — j Lepontine-Simplon-
Alpine elements
Hinterland J
Rhione Valley
( LSR) domain
Tertiary amphi- External Tertiary cz> Strain ellipses (open;
bolite facies Massifs Intrusives ^ Steck (1987); black:
Merle etal.,(1989))
Boundary of Boundary of Base of Insubric-
. -y Front of Insubric-
SouthernAlps ~ ' L S R domain " Helvetic lid and
Helvetic lid
sublid boundaries
one. A particular difficulty in this process is the diversity of of thin thrust sheets or nappes (Lugeon 1902), involving a
scale and of the quality and the nature of data; geophysical series of original basins and intervening highs ('geosynclines'
data must be used, but they involve larger scales and other and 'geanticlines'). Even rough quantifications were at-
properties of the rocks than those used by the surface geolo- tempted. As a result, it was inferred that shortening of several
gist. Microscopic features such as shear indicators must be hundred kilometres had reduced the original width of the
integrated into domains of regional dimensions (Laubscher Alpine geosyncline to the present narrow belt. Later, plate
1990a). tectonics helped by defining the Alpine geosyncline as the
The search for plausible initial states is easiest for sedi- Tethy s ocean and its margins, and by tying Alpine kinematics
mentary bodies, particularly those involved in foreland thrust to that of the Atlantic (Dewey era/. 1973, 1989). Still, there
belts. Here, the construction of balanced sections as a first is a marked lack of precision.
step to a quantitative kinematics is feasible with some preci- In the history of Alpine tectonics, those elements attribut-
sion. It relies on conservation principles, general ones such able to strike-slip have proved particularly elusive and con-
as material conservation or more restrictive ones such as area troversial. A considerable part of this research effort has been
balancing or even bed-length and bed-thickness balancing aimed at resolving the role of strike-slip in the kinematic
(Suppe 1983;Kligfieldera/. 1986). These by no means trivial puzzle of the Alps.
but conceptually rather simple procedures fail for complex
bodies such as the Alps; for these even a cursory inspection
reveals the importance of lateral migration of material and of Identification and quantification of strike-slip
transformations, e. g. by high pressure, which may disguise components
material beyond recognition by indirect, geophysical means
(Laubscher 1970, 1988, 1990b; Schreyer 1988). The com- The concept that strike-slip must have played a considerable
parative ease in handling thrust belts in sediments led to the role in the kinematics of the narrow arcs of the Alps is readily
realization, about 100 years ago, that the Alps consist of a pile apparent, and the search for the main elements is not difficult
278
THE ALPS - A TRANSPRESSIVE PILE OF PEELS
(Laubscher 1988; Coward & Dietrich 1989). Faults such as (1990a, b) argues for the scheme represented in Figure 2: it is
the Insubric line or the Giudicarie line that obliquely dissect compatible with the basic data and provides a viable kinemat-
Alpine structures are obvious candidates. Qualitatively, ic scheme. These phases are composed of numerous sub-
strike-slip may be verified by any of a number of shear sense phases which, however, are hard to correlate regionally.
indicators. These have been observed all along the Insubric Different opinions have been voiced. Roure et al. (1989)
line (Schmid etal. 1989) and also along the southern Giudicarie propose an important link between Late Oligocene-Early
(Laubscher, Schonbom & Schumacher, unpubl. data). Shear- Miocene Insubric motions and the Lombardic thrust belt,
sense indicators, however, have inherent limitations in that which, from the data alone, not considering their quantitative
they provide sense and direction but not the quantity of kinematic relations, seems neither impossible nor necessary.
translation. In addition, they may be late features that They base their argument on the stratigraphic record of the
camouflage earlier, possibly more important events. Sali Vercellese well in the westem part of the Po plain
To obtain quantitative indicators for strike-slip one must (compare Fieri & Groppi 1981), where ditch samples docu-
either find marker boundaries originally transverse (prefer- ment a gap between Aquitanian and Langhian (D. Bernoulli,
ably perpendicular but usually oblique) to the strike-slip written pers. comm., 1990). There is no doubt that the
(transfer) boundary of a displaced body of rock, or alterna- Insubric phase had its repercussions along the arc of the
tively establish the amount of shortening at the leading edge westem Alps, but the particular link across the Po basin with
or extension at the trailing edge of the body. Transverse the Lombardic thrust belt as proposed by Roure etal. (1989)
boundaries are almost invariably fuzzy, it is often uncertain seems arbitrary. As it cannot achieve the apparent
whether they existed before or developed during the strike- allochthoneity of the lorio-Tonale segment of the Insubric
slip deformation phase, and correlation across the fault is line (Laubscher 1990b), this link would appear to be of
therefore ambiguous. Under the circumstances trial and error doubtful quantitative kinematic significance.
interpretations must be used. Suspected boundaries may be
accepted provisionally, and when several of them result in
compatible kinematics, they may be integrated into a model Quantification and timing of strike-slip on the
pro tempore. Insubric line
279
H. LAUBSCHER
<U O (0
. a
to E
nappe correlation suggests that, in the Miocene, the Lepontine-
O 0) ra E
60 . Simplon-Rhone branch was more important.
5
Insubric L e p o n t i n e Helvetic
Line nappes
Insubric Helvetic
Line S t. M o r ( 2 nappes N
-r L- I
____l_^ ^^ra^^"*"^"
A D R I A I
I E U R O P A
I
^".'^.4:^^' ^j-v \_,
I
100 km
o -^^
Figure 5. The uplift of the Lepontine domain by tectonic underplating. Light shading - upper (brittle) crust; dark shading - lower (granulitic) crust; dots
- Molasse conglomerates of the fore-and hinterland. From the convergence of the Insubric line with the front of the Alps (from (b) to (a), compare Figs 1,
6,7) it is inferred that the gap between the subducted European plate and the Insubric line becomes narrower. This is actually borne out by the recent reflection
seismic surveys (Frey et al. 1989). Uplift of Lepontine may be due to piling-up of middle crust north of the Insubric line (tectonic underplating), see text.
Cooling assumed somewhat slower than uplift and tectonic denudation by extension of the lid, with steep thermal gradients as in metamorphic core
complexes.
280
THE ALPS - A TRANSPRESSIVE PILE OF PEELS
Another possibility for the quantification of the westward External massifs were pushed up. Consequently, the base of
translation of the Adriatic indenter is that of estimating the the deformed and partly destroyed Insubric-Helvetic orogenic
shortening of its leading edge in the western Alps. lid should at present be exposed immediately behind the
According to the seismic sections of ECORS-CROP (Bayer 'roots' of the Helvetic nappes, about parallel to the internal
et al. 1987) a crustal shortening of about 120 km minimum border of the External Massifs. This is the site where the
and possibly as much as 200 km seems to be required for the 'Penninic front' of the recent reflection seismograph surveys
imbrications of lower crust and upper mantle (compare Fig. emerges at the surface (Fig. 1). This reflection band may
7). The average would again be 160 km, and the same amount therefore be considered to mark the base of the early Miocene
would have had to be incurred by the otogenic lid. Timing Insubric-Helvetic lid. For this lid, crude estimates suggest
here is difficult, however. about 100 km of N- to NW-translation (Fig. 4).
A third group of estimates for the westward translation of In map view, the lid appears to be subdivided into amosaic
the early Miocene Adriatic indenter is based on post-Oligocene of sublids (Fig. 6) which seem to have accommodated the
sinistral displacement along its southern edge (Villalvemia- divergence of transport direction adjacent to the NW-comer
Varzi-Levanto line =VVL). This fault zone, originally de- of the Adriatic indenter. This divergence implies lateral
fined by Elter & Pertusati (1973), though the only fault stretching. The brittle behaviour of the lid requires a style of
separating the post-Oligocene Apennines from the pre- accommodation different from that in the ductile zone below
Oligocene Ligurian Alps, is often severely underestimated where most of the fabric data such as those in Figure 1 were
(compare Laubscher 1988). Apparent sinistral displacements obtained. Lateral stretching in the lid would be expected to
vary from about 80 km for the middle Penninic units of the be accomplished by systems of strike-slip or normal faults
western Alps with respect to those of the Ligurian Alps, and according to the Mohr-Coulomb criteria.
somewhat higher figures for the displacement of the ophiolite A central role in the mosaic of sublids is played by the
and Austroalpine units (Dent Blanche/Sestri-Voltaggio), to Lepontine-Simplon-Rhone domain of Figure 1, which in a
between 125 km (western tip of the Monferrato Apennines at first approximation may be represented by one sublid, called
TurinA'arzi area) and 150 km (Turin/meridian of Levanto). 'Lepontine' in Figure 5. In the Simplon area, the base of the
Although the estimates are not unambiguous, they appear to lid seems to rise to the top of the Lepontine 'dome', where
be systematically lower by several tens of kilometres than the amphibolite grade rocks were uplifted to the surface (Fig. 1).
estimates along the Insubric line. It is possible that the Qualitatively, this domain has many of the characteristics
discrepancy is due to the presence of early (pre-Burdigalian) of a 'metamorphic core complex' (Davis 1988; Dokka et al.
motions along IL but not VVL. Figures for the Insubric line 1986). In particular, there are indications of E-W extension
are probably more appropriate for the kinematics of the of the roof (e.g. Simplon line) as well as the ductile zone
Insubric-Helvetic lid. underneath (Steck 1984; 1987, Mancktelow 1985) coupled
In spite of these discrepancies, which do not affect the with rapid cooling (Hurford etal. 1989) and retrograde west-
order of the estimates, a remarkably uniform picture emerges, verging folds and thrusts (Merle & Le Gal 1988; Merle etal.
and an assumption of about 150kmpost-Rupelian dextral 1989). Quantitatively, a maximum vertical uplift of the
displacement along the Insubric line would appear justified Lepontine domain of about 20 km may be assumed (upper
on a pro tempore basis. amphibolite facies adjacent to surface rocks in the south).
The amount of E-W stretching is harder to come by but it may
have been more than 40 km (40 km original length of the
The Helvetic thrust belt and its erogenic lid dome stretched by a factor of 2). In the dextral Insubric
system this E-W stretching appears to play the role of a pull-
A large part of the motions of the Helvetic nappes thrust belt apart segment, and the 40 km or so of stretching would be the
seems to be coeval with dextral translation along the Insubric pull-apart contribution of the 150 km of total (late Oligocene
line (Insubric-Helvetic phase, Fig. 2; Schmid et al. 1989; to) early Miocene dextral displacement.
Laubscher 1990b). Such thrust belts develop at the frontal There must be a kinematic link between stretching and
part of an orogenic lid like that conjectured to overlie the rapid uplift in the sublid. Isostatic uplift is unrealistic because
Penninic front reflection band, according to Laubscher (1983). the domain is too narrow and the gradient of uplift is too steep.
The orogenic lid consists of the upper 10 to 20 km of strong, The alternative is some kind of underplating. As the uplift
brittle upper crust in their hinterland (Fig. 5). There is a sort took place at a particular tectonic position within the Insubric-
of simple shear at the base of this lid (for a pure shear Helvetic system, tectonic underplating seems more plausible
component see Dietrich & Casey 1989) which presumably than incidental magmatic underplating. The special tectonic
becomes somewhat diffuse in the high-T ductile domain position is that in front of the NW comer of the Adriatic plate,
(Laubscher 1983; Schmid & Haas 1989). The lid is in need because there the E-W trending Insubric fault converges
of continuous repair as it gets uplifted and eroded: the westward with the SW-NE trending northem border of the
cooling, formerly ductile parts are added to its base. The Alps, thus making the NW comer of the Adriatic plate marks
orogenic lid that was active during the Insubric-Helvetic the narrowest section of the Alps (Fig. 6). Here, the problem
phase was partly destroyed by subsequent uplift, beginning of crustal escape was the most severe (Figs 5 & 6). Arguably,
early on in the Lepontine domain, and culminating subse- it was easiest westward and upward. In particular, masses of
quently during the 'Windows phase' (Fig. 2), when the the middle European crust which were piled up against the
281
H. LAUBSCHER
282
THE ALPS - A TRANSPRESSIVE PILE OF PEELS
Helvetic kinematics rather late, (2) that south of the Rhone Lid kinematics versus middle and lower crust
river no trace of a higher and older Helvetic nappe is found kinematics
ouside the root zone, and (3) that north of Bonneville the
Prealps begin with a thrust on the Molasse, it seems to me that Figures 5 & 7 schematically illustrate the disharmony between
a strong case can be made for an important transverse zone. the tectonics of the upper crust lid and that of the middle and
Obviously the elements of the sublid mosaic moved some- lower crust. The deeper portions of the western Alps seem to
what independently: the position of ramps and flats changes be dominated by imbricated peels of lower crust and upper
across their boundaries, and these functioned as transfer mantle within an interval ordinarily considered as middle
zones. The most peculiar sub-lids are those in front of the crust. In the central Alps, on the other hand, a wedge of high
NW-comer of the Adriatic indenter: the Lepontine and Prealps velocity, layered material conjectured to be a peel of Adriatic
sublids. Of the border zones of the Prealps sublid, the Kander lower crust, fits better into the kinematics of the middle
line has a sinistral, the Giffre line a dextral component. Miocene Lombardic phase (Laubscher 1990 b) and is con-
Together, they delimit a domain of lateral stretching at the sequently not shown in Figures 5 & 7. Therefore, it would
'northern hinge of the arc of the Western Alps' (Laubscher seem thai, in the Central Alps, deep tectonics in the early
1982). This hinge is more sharply defined on the internal Miocene was dominated by eclogitization and subduction of
(Lepontine sublid) than on the external (Prealps sublid) side crustal masses, i.e. downward rather than upward escape,
(Fig. 1). The Lepontine sublid was raised differentially and except for the Lepontine sublid where upward and westward
stretched laterally; it was partly incorporated into the dextral escape was important. As mentioned before, this may be due
Insubric system. The spatial association of the Lepontine and to its special position at the NW-comer of the Adriatic
Prealps sublids suggests a causal connection, the details of indenter.
which, however, remain to be worked out. The position of the
The transpressive pile of peels of the Insubric-Helvetic
Gotthard sublid suggests that it too is associated with the NW
Alps thus may be subdivided into three parts: the already
comer of the Adriatic indenter. The three together compose
existing ones riding passively with the orogenic lid; those
the early Miocene Central Alps. The Simplon and Silvretta
formed at the base and front of the moving lid, and those of
sublids are the most substantial ones and define the main
the middle and lower crust or even upper mantle that escaped
trunks of the lower Miocene Insubric-Helvetic Alps: the
subduction and were added to the middle crust.
western and eastern Alps, respectively.
Figure 6. The sublids of the central Alps and their surroundings, Insubric-Helvetic phase.
283
H. LAUBSCHER
Figure 7. Simplified block diagram illustrating the 3rd dimension of Figures I & 6. In particular, the Lepontine-Simplon-Rhone valley (LSR) pull-apart
has been greatly simplified. The section through the Gran Paradise is according to Bayer etal. (1988), simplified. The External Massifs are omitted because
they did not exist in the lower Miocene. Compare Figures 1 & 5.
References
Bayer, R., Cazes, M., Dal Piaz, G.V., Damotte, B., Elter, G.,Gosso, G., Him, Bernoulli. D., Bertotti, G. & Zingg, A. 1989. Northward thrusting of the
A., Lanza, R., Lombardo, B., Mugnier, J.-L., Nicolas, A., Nicoiich, R., Gonfolite Lombarda {South-Alpine Molasse) onto the Mesozoic se-
Polino, R., Roure, P., Sacchi, R., Scarascia, S., Tabacco, I., Tapponnier, quence of the Lombardian Alps: Implications for the deformation history
P., Tardy, M., Taylor, M., Thouvenot, F., Torreilles, G. & VilHen, A. of the Southern Alps. Eclogae Geologicae Helveiiae, 82. 841 -856.
1987. Premiers resultats de la traverse des Alpes occidentales par Coward. M. P. & Dietrich, D. 1989. Alpine tectonics: an overview. In:
sismique reflexion verticale (Programme ECORS-CROP). Comptes Coward, M. P., Dietrich. D. & Park, R. G. (eds) Geological Society of
Rendus de I'Academie des Sciences, Paris, 305, Serie I!, 1461 -1470. London Special Publication, 45, 1-29.
Bearth, P. 1956, Geologische Beobachtungen im Grenzgebiet der Davis, G. A. 1988. Rapid upward transport of mid-crustal mylonitic gneisses
lepontinischen und penninischen Alpen. Eclogae Geologicae Helveiiae, in the footwall of a Miocene detachment fault. Whipple Mountains,
49,279-290. - ,.. , southeastern California. Geologische Rundschau, 11, 191-209.
284
THE ALPS - A TRANSPRESSIVE PILE OF PEELS
Dewey, J. F., Helman, M. L., Turco, E., Button, D.H.W. & Knott, S. D. 1989. Merle, O., Cobbold, P. R.& Schmid, S. 1989. Tertiary kinematics in the
Kinematics of the western Mediterranean. In: Coward, M. P., Dietrich, Lepontine dome. In: Coward, M. P., Dietrich, D. & Park, R. G. (eds)
D. & Park, R. G. (eds) Geological Society ofLondon Special Publication, Geological Society of London Special Publication, 45, 113-134.
45, 265-283. & Le Gal, Ph. 1988. Post-amphibolitic westward thrusting and fold
, Pitman, W. C , Ryan, W. B. F. & Bonin, J. 1973. Plate tectonics and vergence in the Ticino domain. Eclogae geologicae Helvetiae, 81,215-
the evolution of the Alpine system. Geological Society of America Bul- 226.
letin. S4,3137-31S4. Fieri, M. & Groppi, G. 1981. Subsurface geological structure of the Po plain,
Dietrich, D., & Casey, M. 1989. A new tectonic model for the Helvetic Italy. Centra Nazionale Ricerche, Progetto Finalizzato Geodinamica,
nappes. In: Coward, M. P., Dietrich, D. & Park, R. G. (eds) Geological 414.
Society of London Special Publication, 45, 47-63. Ramsay, J. G. 1989. Fold and fault geometry in the western Helvetic nappes
Dokka, R. K., Mahaffie, M. J. & Snoke, A. W. 1986. Thermochronologic of Switzerland and France. In: Coward, M. P., Dietrich, D. & Park, R. G.
evidence of major tectonic denudation. Tectonics, 5, 995-1006. (eds) Geological Society of London Special Publication, 45, 33-45.
Elter, P. & Penusati, P. 1973. Considerazioni sul limite Alpi-Appennino e Roure, F., Polino, R. & Nicolich, R. 1989. Poin^onnement, retrocharriages
sulle sue relazionicon I 'arco delle AlpioccidenidU. Memorie della Societa et chevauchements post-basculement dans les Alpes occidentales:
Geologica Italiana, 12, 359-375. evolution intracontinentale d'une chaine de collision. Comptes Rendus
Ferrazzini, B. & Schuler, P. 1979. Fine Abwicklungskarte des Helvetikums de I'Academie des Sciences de Paris, 309, Serie II, 283-290.
zwischen Rhone und Reuss. £f/o,i;ae Gfo/f>,?ii;af//f/vfr/of, 72,439-453. Schmid, S. M., Aebli, H. R., Heller, F.& Zingg, A. 1989. The role of the
Frei, W., Heitzmann, P., Lehner, P., Muller, St., Olivier, R., Pfiffner, A., Periadriatic line in the tectonic evolution of the Alps. In: Coward, M. P.,
Steck, A. &Vaiasek, P. 1989. Geolraverses across the Swiss Alps. Dietrich, D. & Park. R. G. (eds) Geological Society of London Special
Nature, 340, 544-548. Publication. 4S, 153-171.
Gelati, R., Napolitano, A. & Valdisturlo, A. 1988. La gonfolite Lombarda, & Haas, R.. 1989. Transition from near-surface thrusting to intrabasement
straligrafia e significato nella evoluzione oligo-miocenica del margine decollement, Schlinig thrust. Eastern Alps. Tectonics, 8, 697-718.
sudalpino. Rivisia italiana di Paleontologia (Stratigrafia), 94,285-332. Schreyer, W. 1988. Subduction of continental crust to mantle depths:
Gidon, M. 1976. Carte geologique simplifiee des Alpes occidentales du petrological evidence. Episodes, 11, 97 - 104.
Leman a Digne. 1 : 250 000. Bureau de Recherches Geologiques et Schweizerische Geologische Kommission (ed.). 1934. St-Maurice. Geo-
Minieres, editions Didier-Richard. logical Atlas of Switzerland 1:25 000, Sheet 8. Kiimmerly and Frey, Bern.
Giger, M. & Hurford, A.J. 1989. The Tertiary intrusives north of the Insubric 1937. Saxon-Morcles, Geological Atlas of Switzerland 1:25 000, Sheet
Line (Central Alps): Their Tertiary uplift, erosion, redeposition and 10. Kummerly and Frey, Bern.
burial in the South-Alpine foreland (Como, Northern Italy). Eclogae 1951. Finhaut. Geological Atlas of Switzerland 1:25 000, Sheet 24.
Geologicae Helvetiae, 82, 857-866. Kummerly and Frey, Bern.
Hurford, A. J., Flisch, M. & Jager, E. 1989. Unravelling the thermo-tectonic 1980. Geologische Karte der Schweiz I : 500 000. Bundesamt fur
evolution of the Alps; a contribution from fission track analysis and mica Landestopographie, Wabem.
dating. In: Coward, M. P., Dietrich, D. & Park, R. G. (eds) Geological Steck, A. 1984. Structures de deformations tertiaires dans les Alpes centrales
Society of London Special Publication, 45, 369-398. (transversale Aar-Simplon-Ossola). Eclogae Geologicae Helvetiae, 77,
Kligfield, R., Geiser, P. & Geiser, J. 1986. Construction of geologic cross 55-100.
sections using microcomputer systems. Geobyte, 2, 60-67. 1987. Le massif du Simplon-Reflexions sur la cinematique des nappes
Laubscher, H. P. 1970. Bewegung und Warme in der alpinen Orogenese. de gneiss. Schweizerische mineralogische und petrographische
Schweizerische mineralogische und petrographische Mitteilungen, 53, Mitteilungen, 67, 27-45.
565-596. Suppe, J. 1983. Geometry and kinematics of fault-bend folding. American
1982. A northern hinge zone of the arc of the western Alps. Eclogae Journal of Science. 283, 684-721.
Geologicae Helvetiae, 75, 233-246. Thomas, W. A. 1990. Transverse zones in thrust belts. Eclogae Geologicae
1983. Detachment, shear and compression in the Central Alps. Geo- Helvetiae, 83 (in press).
logical Society of America Memoir, 158, 191-211. Triimpy, R. 1969. Die helvetischen Decken der Ostschweiz: Versuch einer
1988. Material balance in Alpine orogeny. Geological Society of palinspastischen Korrelation und Ansatze zu einer kinematischen Ana-
America Bulletin, 100, 1313-1328. lyse. Eclogae Geologicae Helvetiae, 62,105-142.
1989. The tectonics of the southern Alps and the Austroalpine nappes: 1980. An outline of the geology of Switzerland. Geology of Switzer-
a comparison. In: Coward, M. P., Dietrich, D. & Park, R. G. (eds) land- a guide-book. 5 - 104. Schweiz. Geol. Kommission, Wepf & Co.
Geological Society of London Special Publication, 45, 229-242. Publ., Basel/New York. 334p.
1990a. The problem of the deep structure of the Southern Alps: 3-D Wheeler. R. L. 1980. Cross-strike structural discontinuities: possible explo-
material balance considerations and regional consequences. ration tool for natural gas in Appalachian overthrust belt. The American
Tectonophysics, 176, 103-121 Association of Petroleum Geologists Bulletin. 64, 2166-2178.
1990b. Deep seismic data from the central Alps: mass distributions and Withjack, M.O.& Scheiner, C. 1982. Fault patterns associated with domes-
their kinematics. Memoires de la Societe Geologique de France (in press). an experimental and analytical study. The American Association of Pe-
Lugeon, M. 1902. Les grandes nappes de recouvrement des Alpes du troleum Geologists Bulletin, 66, 302-316.
Chablais et de la Suisse. Bulletin de la Societe Geologique de France. 4, Zwahlen, P. 1986. Die Kandertal-Storung, eine transversale Diskontinuitat
1, 723-825. im Bau derHelvetischen Decken. Ph. D. Diss. Bern, Universitiitsdruckerei
Mancktelow.N. 1985. The Simplon Line: A major displacement zone in the Beni,103p.
western Lepontine Alps. Eclogae Geologicae Helvetiae, 78, 73-96.
285
Structural evolution of the western Chartreuse fold and thrust system,
NW French Subalpine chains
Robert W.H. Butler*
Abstract: Structural styles in the outer parts of the Alps of SE France, in common with much of
the Tethyan province, involve displacements on thrust faults together with important fold
structures. These deform the pre-existing sedimentary pile which contained normal faults and
displays significant lateral stratigraphic variations. The underlying crystalline basement is not
involved suggesting important detachment at the base of the Mesozoic cover. Abundant field
outcrops are used to construct a regional cross-section through the Chartreuse district of the
Subalpine fold and thrust belt, across the outermost structures of the western Alps. Thrusting
directions are estimated as being ESE-WNW, determined from the spherical mean azimuth (106°)
of the wide range of orientations of slip axis indicators (striae, shear fibres etc.). However, the
range in orientations of slip axis indicators probably does not reflect regional non-plane strain but
combinations of rotations of blocks about vertical axes within broader fault zones (but not thrust
sheets) and of earlier thrusts about horizontal axes during the development of thrust anticlines. The
fold-thrust belt (Mesozoic shelf sediments and Tertiary molasse) shows a gross imbricate
geometry but folding plays an important role, particularly preceding thrust ramp development,
which cannot be explained by simple fault-bend fold models. Locally ramps coincide with abrupt
changes in Mesozoic stratigraphy suggesting local normal fault controls. Thrust anticlines show
complex geometry developed in part by fault bend folding but substantially modified by large
numbers of minor thrusts concentrated in the forelimbs.
The Subalpine chains of France are a classic foreland thrust shallow marine environment (Demarcq 1970; Nicolet 1979).
belt, formed on the outer, European margin of the western Following the cessation of regional sedimentation in the
Alpine mountain belt. These structures are particularly well middle Cretaceous (Lemoine et al. 1986) the Subalpine shelf
exposed in the segment from Haul Giffre on the Franco-Swiss probably experienced two main episodes of burial; during the
border to Diois (Fig. 1). Formed in late Miocene times, the deposition of foredeep sediments (Oligo-Miocene) and dur-
thrust belt represents the last 20-30 km of Alpine shortening ing the subsequent thrusting in late Miocene times. Folding
(Butler 1989a). Since then the region has experienced several and thrusting effects both Mesozoic shelf sediments and
kilometres uplift caused by isostatic rebound during erosion Tertiary molasse deposits throughout most of the Subalpine
and unloading in the tectonic hinterland. Deep gorge sections chains and all of the Jura. Those thrusts which climb through
and over 2 km vertical relief combine to provide excellent the stratigraphy emerge into the molasse basin although
surface control on sub-surface structure. A compilation of many structures are blind, being imaged on seismic reflection
drill-hole, gravity and seismic refraction data by Menard profiles (e.g. Charollais et al. 1977) or predicted by section
(1980) provides a depth to basement map for the region. A construction (e.g. Butler 1989a). Only in the north of the
recently acquired deep seismic reflection line (Bayer et al. region, towards theircontinuation into the Helvetic nappes of
1987) and associated experiments (e.g. Mugnier et al. 1987) Switzerland, were the Subalpine Mesozoic rocks enveloped
show that basement is not involved directly beneath the thrust by the far-travelled Prealpine thrust sheets.
belt although it has been carried up on thrusts further east. The combination of significant vertical relief and continu-
The Subalpine chains themselves, and the more outlying ous gorge sections makes the Subalpine chains a particularly
Jura hills, are formed by folded and thrust Mesozoic shelf good natural laboratory to study the evolution of fold-thrust
sediments together with vestiges of the overlying foredeep complexes. This contribution discusses the structural evolu-
basin sequences. The shelf units are up to 4 km thick, tion of a small part of this thrust belt, in the western Char-
although more commonly lie between 2-3 km, and are made treuse massif which lies between the cities of Grenoble and
up of alternations of limestones and shales with several Chambery (Fig. 1). This region contains the southern ter-
regionally continuous carbonate platforms. The Tertiary mination of the vestigial Swiss foredeep basin, contained
foredeep sediments in contrast are classic molasse sandstones, between converging Jura hills and the main Subalpine chains.
grits and conglomerates mostly deposited in a tide-dominated It is possible therefore to examine a range of thrust structures
^Present address: Department of Earth Sciences, University of Leeds, Leeds, LS2 9JT, UK.
287
R.W.H. BUTLER
50 km
Figure 1. Location map of the Subalpine chains in the external part of the western Alps of France. The boxed area (a) is the Chartreuse massif (see Fig.
2).
from the most outlying (western) into the main thrust belt sections have been constructed through the Chartreuse using
along a single transect. The problems to be addressed include angular fault-bend folding models (Suppe 1983) by Mugnier
thrust localization, the variations in 'structural style' and et al. (1987). Cross-sections through surrounding regions are
regional deformation patterns on a large scale. provided by Doudoux et a/. (1982) and Butler (1989a,b). The
area is covered by 1:50,000 geological maps published by
BRGM (sheets: Domene, 1969; Montmelian, 1969; Voiron,
Stratigraphy and regional structure 1970; Grenoble, 1978; see also 1-.250,000 Lyon, 1980). The
explanatory booklets provide additional stratigraphic details.
The basic geology of the Chartreuse (Fig. 2), southern Jura The Chartreuse area has recently been studied using ex-
and surrounding regions has been well-known for many years plosion seismology, as discussed by Thouvenot & Menard
(e.g. Gignoux & Moret 1944) and has formed part of several (1990). This experiment used a single shot point to the north
excellent review articles (e.g. Ramsay 1963; Debelmas & of the Chartreuse, recorded along two longitudinal lines
Lemoine 1970; Debelmas & Kerckhove 1980). Structural running sub-parallel to strike and two fans which arc across
information is provided by Gidon (1964, 1981, 1988), strike. They obtained wide angle reflections from the as-
Doudoux etal. (1982) and Siddans (1983). Balanced cross- sumed top of basement and from Mesozoic limestones. From
288
WESTERN CHARTREUSE FOLD AND THRUST SYSTEM
Upper Cretaceous
and Urgonion
Vo(anginior)
cross-section ,
Tithonian
evapor/fes
these data Thouvenot & Menard suggest major basement
involvement in the thrusting, both passively as a step devel-
oped during the Mesozoic rifting history (discussed below) b) "Subalpine"
and by active WNW-directed thrusting. However, as the
Figure 3. Simplified stratigraphic columns for the Jura (a) and Subalpine
authors point out, the experimental design makes the correla-
chains (b) in the Chartreuse district of the external Alps. The brick ornament
tion, and therefore determination of position and orientation, represents carbonates, the pecked symbol is shale and stippled beds are
of reflectors rather difficult. The greatest limitations arise siliciclastics. The numbers adjacent to each column represent the cumulative
from the use of a single shot-point which prohibits back- thickness in kilometres.
shooting along the longitudinal sections. Wide angle experi-
ments are best suited to regions with laterally persistant 2 km (Fig. 3). Commonly the upper part of this sequence is
structure, not an attribute of the Subalpine chains (e.g. Doudoux missing, having been eroded through the Palaeogene. This
et al. 1982). The authors have not published ray-tracing unconformity is spatially related to a suite of extensional
computations to support their seismic models. Therefore faults which define the eastern margin of the Rhone-Bresse
their interpretations which correlate particular limestone basin and probably represents footwall uplift in response to
formations and basement steps with reflections are only this normal faulting. Oligocene basins are also preserved
weakly supported by the data. Nevertheless, the general within the thrust belt and, as we will see, the controlling
position for top basement is used in construction of the normal faults influenced the subsequent development of
regional section (Fig. 4). thrust structures. Different levels of top basement detected by
Thouvenot & Menard's (1990) seismic experiment may
reflect normal fault blocks in the subsurface. These faults are
Nature of the pre-thrusting template
overlain stratigraphically by the true foredeep fill; there is an
In the west the Jura platform is dominated by laterally additional unconformity at the base of the Miocene molasse
persistant carbonates with a cumulative thickness of about succession.
289
R.W.H. BUTLER
WNW ESE
Ratz anticline St Laurent basin Echaillon anticlines main Chartreuse hills
drill hole
^•;-*^f-srfs?^^*c-^:^P^
^•^^^^rm
_I2P basement
km
Figure 4. Outcrop and deep structural data available for the construction of regional balanced cross-sections (see Fig. 5). The section line is indicated on
Figure 2. The key is as Figure 5.
The eastward passage from the Jura into the Subalpine A cross section through the western Chartreuse
chains contains important facies and thickness variations in
the Mesozoic succession (Fig. 3). Lemoine et al. (1986) Using published maps, additional detailed and reconnais-
consider this to reflect an eastward increase in syn-sedimen- sance mapping by the author allied to studies of fault zone
tary rifting associated with the old European continental kinematics together with available regional seismic data
margin of Tethys and is illustrated schematically by Siddans (Menard 1980; Bayer et al. 1987), a cross-section was pro-
(1983). In the east of the study area (sillon subalpin, see Fig. duced through the frontal part of the thrust belt (Fig. 4). This
1) the Jurassic succession achieves a thickness of 2500-2800 covers the transition from the frontal thrust anticline (Ratz
m (BRGM, 1969b) in contrast with just 720 m in the Faramas anticline) into the main Chartreuse. The section is directly
borehole (BRGM, 1980a) located in Bas Dauphine (Fig. 1). observable in its upper part thanks to the substantial topogra-
Pelagic sedimentation increases in importance to the east; phy cut by deep gorges. The deeper levels are constructed to
there is however, no evidence that comparable subsidence produce a restorable structural geometry (Fig. 5) and hence
caused by major rift faulting occurred beneath the western the section runs WNW-ESE, parallel to the fault movement
Chartreuse. A more tangible effect is shown by the Berriasian axes defined by shear fibres (Ramsay 1980), slickensides and
thickness and facies from the Jura into the Subalps. In the Jura grooves on fault surfaces (Hancock & Barka 1987). These
the lowermost Cretaceous rocks are well bedded carbonates kinematic data are discussed later. The background strain
with only about 50 m of succession between the regionally state of thrust sheets is very low with no appreciable cleavage
extensive Tithonian and Valanginian limestones. Within the generation or penetrative fracturing within the massive car-
central Chartreuse the same stratigraphic position is taken by bonates. As discussed later, more intense deformation occurs
up to 500 m of interbedded limestones and shales. This on the (western) fore-limbs of folds and around thrust zones
change occurs abruptly across the front of the Chartreuse but these regions are of restricted extent. For the most part
hills. Although Lemoine et al. (1986) consider the lower simple line-length restoration methods (e.g. Goguel 1963;
Cretaceous to be a 'post-rift' sequence presumably deposited Dahlstrom 1969) can be employed. Parts of the Chartreuse
during thermal subsidence, the radical thickness changes massif contain NE-SW dextral oblique-slip fault zones
across a short horizontal distance, together with local evi- (Goguel 1948; Gidon 1964) which locally generate patches
dence for slumping (in the Guiers Mort valley. Fig. 2), sug- of non-plane strain within the thrust belt. The cross-section
gests active extensional faulting continuing into Berriasian (Fig. 5) was constructed away from these areas, concentrating
times. There is little evidence for radically differential on the western Chartreuse although the continuation of the
subsidence across the Chartreuse region following the topographic profile east of the Entremont thrust crosses an
Berriasian although ongoing sequence stratigraphic studies oblique-slip fault, in structures not considered here. Thrust-
on Cretacerous units may elucidate this. Much of the evi- ing in the more outlying structures illustrated on Figure 4
dence has been removed by the Palaeogene uplift which (west of the Entremont thrust) is considered to have been
effected the Jura but which was less marked to the east. plane strain.
The net result of the various Mesozoic subsidence histo- In contrast to sections through some other foreland thrust
ries and differential uplift was to generate a laterally variable belts, the western Chartreuse shows broad, open fold forms
multi-layer sequence formed by 'competent' limestones and (Fig. 4) rather than the geometric kink-bands predicted for
'incompetent' shales with the local offsets caused by normal some fault bend structures (see Gidon 1988). This observa-
faults. This type of template, rather than the idealized layer- tion is crucial in chosing the appropriate method of section
cake stratigraphies adopted for many foreland thrust belts, is construction. The kink-band model popularized by Suppe
characteristic of much of the Tethyan province (e.g. Lemoine (1983) has been applied to the Chartreuse (Mugnier et al.
etal. 1986; Butler 1989b). 1987). This approach has not been adopted here because it
290
WNW
Bas Dauphin e Ratz Qnticlinf Sf Laurent basm Epine anhclme
thrust sysfei^s
Figure 5. Balanced and restored section through the western Chartreuse, section line marked on Figure 2. Note that the two parts of the restored section join up (at x).
R.W.H. BUTLER
55W
km
Figure 6. Partially restored longitudinal section showing the relationship between footwall and hangingwall geometry of the Voreppe thrust (VT) together
with the predicted trajectory of the thrust beneath the Echaillon anticlines. Viewed looking back down the thrusting direction. The section runs approximately
along the present outcrop trace of the Voreppe thrust. Note that the upper parts of the section would be overlain by the Corbel thrust sheet (not illustrated).
Key as Figure 5. OHWR—oblique hangingwall ramp; FHWR—frontal hangingwall ramp; HWF—hangingwall flat; FWF—footwall flat; OFWR—oblique
footwall ramp; FFWR—frontal footwall ramp. These elements are defined by bedding orientations in the hangingwall and by the stratigraphic relationships
in the footwall. No vertical scale intended. The unconformity (wavy line) at the base of the Miocene molasse (stippled) does not reflect thrust-related erosion
prior to molasse deposition.
obscures the importance of additional faults, stratigraphic stones (BRGM, 1970). However the form of the anticline
complications and other structural styles which will be shown suggests several hundred metres of older stratigraphic suc-
to be appropriate for Subalpine tectonics. Nevertheless, the cession remain buried although still in the hangingwall to the
section (Fig. 5) balances in that it has been constructed to frontal Alpine thrust.
preserve formational cross-sectional area during deforma- Behind the Ratz there are several other anticlines that
tion with the added constraint of maintaining horizontal represent the Jura fold belt, ahead of the Voreppe thrust which
shortening at all depths (above the local detachment at the carries the Chartreuse. These structures are termed here the
base of the cover) in the section. Echaillon anticlines, after the gorge on the Guiers Vif river
which passes through them. These folds become well-
developed towards the north continuing into the Jura (see Fig.
Structural geometry in the Chartreuse
I) but just south of the cross-section line they vanish into the
The Chartreuse hills are defined by a series of large-scale St Laurent molasse furrow (Fig. 2). This lateral variation in
broad anticlines and synclines which have a general asymme- structure can be interpreted using longitudinal sections (Fig.
try with steep, west-facing fore-limbs and more gently- 6). In the north the Echaillon anticlines contain a large
dipping back-limbs. Fold hinges are generally rather smooth. proportion of the available Mesozoic succession but south-
The forelimbs contain several minor thrust breaks which wards, where they die out, there is insufficient space at depth
disrupt the structure and the anticlines themselves are well- to contain this volume of material. The disappearance of
known to be carried on regionally extensive thrusts which cut these folds into the molasse basin cannot be due to a simple
up into the molasse (Gignoux & Moret 1944). Several southward plunge of fold hinges. A more likely explanation
detachment horizons are known to exist within the Mesozoic is that the thrusts which carry these anticlines climb up
succession and the overlying molasse. The best known of stratigraphic section in their hangingwalls so that they carry
these is the detachment at the base of the cover sequence, less stratigraphy (Fig. 6). On the cross-section line the thrusts
within Triassic evaporites, which has transferred Alpine would detach in the lower Cretaceous. Further south the
shortening far out into the foreland to form the Jura hills. It thrusts would climb up into the molasse and presumably
is likely that this detachment has been activated to carry up the branch onto the Voreppe thrust. Perhaps this lateral transfer
Ratz anticline (Figs 2 & 4). The oldest rocks exposed in the of thrust displacement provides a model for the southward
core of the structure are Upper Jurassic (Tithonian) lime- termination of the Jura fold belt as a separate belt of structures
292
WESTERN CHARTREUSE FOLD AND THRUST SYSTEM
straintid "bead"
Fold-thrust relationships
tacal/sed fault
The hangingwall anticlines within the Chartreuse fold-thrust
belt have a characteristic form whereby the forelimbs are
steeply dipping, rather than showing the conventional mod-
erate (30°) dips predicted by simple fault-bend models (e.g.
localised fault strained "bead" Suppe's( 1983) mode 1 folds). The back-limb dips of, e.g. the
localised fault
Ratz anticline (Fig. 4) are consistent with these rocks overly-
JZJ^':::>:^^ 1 ing a footwall ramp (i.e. <45°). The forelimb dips vertically,
(, ; r £ i ^ ^ ^ );K:o-i::
•^ims:,viV:5;V/0x::fflv;. 1 with an interlimb angle of 90°. This suggests an additional
1 1 i 1 1 1 1 1 r 1 "p1 1 1 1 i 1
I 1 1 1 1 n^sci 1 1 strain component, rather than fault slip alone.
1
The simple fault-bend fold model which dates from Rich
(1934), formalized as the mode 1 structures of Suppe (1983)
Figure 7. Variations in structural style which may exist in the Alpine thrust
belt, (a) The classic fault bend fold (Suppe 1983) where displacements and since followed by numerous geologists (e.g. Crane 1987;
localize onto a discrete thrust, (b) A broad bead of strain developed above a De Paor 1988) is but one extreme of tectonic behaviour in a
detachment. 'Displacement' is distributed through a large volume of rock, foreland thrust belt. In it shortening is highly localized onto
(c) Combinations of fault-bend folding and distributed strain at different a discrete fault surface throughout the evolution of the struc-
levels in the thrust belt. This geometry is the simple tip-line strain pattern ture. The range of geometries which can be produced is
(e.g. Williams & Chapman 1983). (d) A derivation of (c) where structural
style alternates between strain and discrete thrusting. This is the most
extremely small, being a simple function of fault shape and
common deformation pattern in the thrust belt, in conjunction with folding.
consistent with the southern segment of the Voreppe thrust mean azimutfi-
accommodating extra slip which in the north is taken up on Figure 8. Stereonets of striae and shear fibres together with the poles to fault
planes measured on two splays (VTZ. 1 & VTZ.2) from the Voreppe thrust.
the Echaillon structures. This geometry supports the idea of
The data come from a road section c, 1 km east of St. Christophe en Guiers
thrust branching proposed above. Regionally it appears that (on the D520c). These and the following stereoplots are upper hemishere,
displacements across the southern Jura are transferred onto equal angle were calculated together with the statistics using the programme
the Subalpine chains from north to south. STEREO, marketed by Rockware Inc.
293
R.W.H. BUTLER
WNW ESE
~ 'te 20 m
©^
soft
Figure 10. Sketch section of a road cut (c. 1 km west of Corbel village on the D45) through the forelimb of the anticline at Corbel (see Figs 2 & 4 for location
of the structure). The large arrows represent stratigraphic way-up. The slip sense on thrusts is determined from small scale structures. Discussion of the
deformation mechanisms on these and other thrusts in the Chartreuse is reserved for a companion paper.
294
WESTERN CHARTREUSE FOLD AND THRUST SYSTEM
295
R.W.H. BUTLER
section towards the west. These geometries are now down- Hauterivian shales is preserved, offset by small fault zones
ward facing. One package of bedding has been thickened by with about 5 m of WNW-directed offset. The distance
closely-spaced imbricate thrusts (Fig. 10) which are also between the Valanginian and Urgonian limestones is just 15
downward facing. Presumably these structures pre-date the m along the road section, in contrast to the usual 100-150 m
major thrust anticline at Corbel. found in the Chartreuse area. The Valanginian limestones at
Currently low angle faults are also present which locally outcrop probably lie in the hangingwall to another E-dipping
dip both east and west (Fig. 10). These offset the steep fault which has carried them almost onto the Urgonian. The
Urgonian bedding and apparently show both WNE and ESE implied offset on this fault would be about 100 m. As with the
directed movements. These structures apparently post-date smaller thrusts, this fault would post-date the main Corbel
the major thrust anticline. Similar suites of low-angle thrusts anticline.
are found in the east of the section (Fig. 10) where well- The complexity of structural evolution at Corbel is re-
bedded Valanginian limestones are present. The stratigraphic flected in the distribution of the orientation of linear markers
top of the Valanginian limestones and their passage into the (striations, shear fibres) found on fault planes (Fig. 8b).
These lie crudely along a vertical plane parallel to the regional
thrusting direction but plunge by varying degrees towards the
WNW and ESE. This population includes early thrust
minor intraformational thrusting
lineations which now lie on downward facing structures on
the forelimb of the major Corbel anticline together with
gently-dipping, post-fold thrusts which dip ESE and break up
the forelimb.
296
WESTERN CHARTREUSE FOLD AND THRUST SYSTEM
297
R.W.H. BUTLER
References
Bayer, R., Cazes, M., Dal Piaz, G.V., Damotte, B., Elter, G., Hirn, A., Lanza, Gidon, M. 1964. Nouvelle contribution a I'etude du massif de la Grande
R., Lomardo, B., Mugnier, J-L., Nicolas, A., Nicolich, R., Polino, R., Chartreuse et de ses relations avec les regions avoisinantes. Geologic
Roure, F., Sacchi, R., Scarascia, S., Tobacco, I., Tapponnier, P., Tardy, Alpine, 40, 187-205.
M., Taylor, M., Thouvenet, F., Torrielles, G. & ViUien, A. 1987. 1981. Les deformations de la couverture des Alpes occidentales
Premiers resultats de la traversee des Alpes occidentales par sismique extemes dans la region de Grenoble: leurs rapports avec celles du socle.
reflexion verticale (Programme ECORS-CROP). Comptes Rendus de Comptes Rendus de TAcademic des Sciences, Paris, 292, 1057-60.
I'Academic des Sciences, Paris, 305, 1461-70. 1988. L'anatomie des zones de chevauchement du massif de la
Butler R.W.H. 1989a. The geometry of crustal shortening in the western Chartreuse (Chaines subalpines septentrionales, Isere, France). Geologic
Alps. In: Sengor, A.M.C. (ed.) Tectonic evolution of the Tethyan Region. Alpine, 64, 27-47.
Proceedings of NATO Advanced Study Institute, C239, 43-76. Gignoux, M. & Moret. L. 1944. Geologic Dauphinoise {ou V initiation a la
1989b. The influence of pre-existing basin structure on thrust system geologic par l'etude des environs de Grenoble). B. Arthaud, Grenoble,
evolution in the western Alps. In: Williams, G.D. & Cooper, M.A. (eds) 425p.
Inversion Tectonics. Geological Society of London Special Publication, Goguel, J. 1948. Le role des failles de decrochement dans le massif de la
44,105-22. Grande Chartreuse. Bulletin de la Societe Geologique de la France, 18,
in press. Evolution of Alpine fold-thrust complexes: a linked kinematic 277-235.
approach. //;.' Mitra, S. (ed.) Structural geology of fold and thrust belts 1963. Tectonics (English ed.). Freeman, San Fransisco, 384p.
(Elliott volume). Johns Hopkins University Press. Hancock, P. L. & Barka, A. A. 1987. Kinematic indicators on active normal
BRGM 1969a. Carte geologique de la France a 1:50000, feuille Domene. faults in western Turkey. Journal of Structural Geology. 9, 573-84.
Bureau des Recherches Geologiques et Minieres, Orleans Kligfield. R., Geiser, P. & Geiser, J. 1987. Construction of geologic cross-
1969b. Carte geologique de la France a 1:50000, feuille Montmelian, sections using microcomputer systems. Geohyte, 1. 60-6.
Bureau des Recherches Geologiques et Minieres. Orleans Knipe, R.J. 1985. Footwall geometry and the rheology of thrust sheets.
1970. Carte geologique de la France a 1:5(X)(X), feuille Voiron, Bureau Journal of Structural Geology, 7, 1-10.
des Recherches Geologiques et Minieres. Orleans Lemoine, M., Bas, T., Arnaud-VanneauU, A., Amaud, H., Dumont, T.,
1978. Carte geologique de la France a 1:50000. feuille Grenoble, Gidon, M., Bourbob, M., De Graciansky, P. C , Rudkiewsky, J. L.,
Bureau des Recherches Geologiques et Minieres, Orleans Megard-Galli J. & Tricart, P. 1986. The continental margin of the
1980a. Carte geologiques de la France ii 1:50000, feuille La-Tour-du- Mesozoic Tethys in the Western Alps. Marine and Petroleum Geology,
Pin, Bureau des Recherches Geologiques et Minieres, Orleans 3, 179-99.
1980b. Carte geologique de la France a 1:250000, feuille Lyon, Bureau Menard, G. 1980. Profondeur du socle antetriassique dans le sud-est de la
des Recherches Geologiques et Minieres, Orleans France. Comptes Rendus de I'Academic de Sciences, Paris, 290,299-302.
Charollais, J., Pairis, J-L. & Rosset, J. 1977. Compte rendu de I'excursion de Mugnier, J. L., Arpin, R. & Thouvenot, F. 1987. Coupes equilibrees a travers
la Societe Geologique Suisse en Haute Savoie (France) dulO au 12 le massif subalpin de la Chartreuse. Geodinamica Acta, 1, 125-37.
Octobre 1976. Eclogae Geologicae Helveliae, 70, 253-85. Nicolett, C. 1979. Le Bas Dauphine septentrionale, etude stratigraphique et
Cooper, M. A. & Trayner, P. M. 1986. Thrust-surface geometry: implica- sedimentologique. These 3eme cycle. University of Grenoble.
tions for thrust belt evolution and section balancing techniques. Journal Pfiffner, O. A. & Burkhard, M. 1987. Determination of paleo-stress axes
of Structural Geology, 8, 305-12. orientations from fault, twin and earthquake data. Annales Tectonicac. 1,
Crane, R. C. 1987. Use of fault cut-offs and bed travel distance in balanced 48-57.
cross-sections. Journal of Structural Geology, 9, 243-6. Ramsay, J. G. 1963. Stratigraphy, structure and metamorphism in the
Dahlstrom.C.D.A. 1969. Balanced cross-sections. C(jHorf/a«,/()H/na/o/£(ir/t/i western Alps. Proceeding of the Geological Association, 74, 357-91.
Science, 6, 743-57. 1980. The crack seal mechanism ofrock deformation. Wart(/T. 284,135-
Debelmas, J. & Kerchove, C. 1980. Les Alpes franco-italiennes. Geologic 9.
Alpine, 56,21-58. Rich, J. 1934. Mechanics of low angle overthrust faulting as illustrated by the
Debelmas, J. & Lemoine, M. 1970. The western Alp,s: Palaeogeography and Cumberland block, Virginia, Kentucky, Tennessee. American Associa-
structure. Earth Science Reviews, 6, 221-56. tion of Petroleum Geologists Bulletin. 18. 1584-96.
Demarcq, G. 1970. Elude stratigraphique du Miocene rhodanien. Memoires Rowan, M. G. & Ratliff, R.A. 1988. Use of fault cut-offs and bed travel
de la Bureau des Recherches Geologiques el Minieres, Orleans, 61. distance in balanced cross-sections: Discussion 1. Journal of Structural
De Paor, D. G. 1988. Balanced section in thrust belts. Part I: construction. Geology. 10,311-6.
American Association of Petroleum Geologists Bulletin, 72, 73-90. Siddans, A.W.B. 1983. Finite strain patterns in .some Alpine nappes.,/ourna/
Doudoux,B.,Mercier, B. & Tardy, M. 1982. Une interpretation nouvelle de of Structural Geology, 5, 441-8.
la structure des massifs subalpins Savoyards (Alpes occidentales): Nappes Suppe, J., 1983. Geometry and kinematics of fault bend folding. American
de charriage oligocenes et deformations superposees. Comptes Rendus Journal of Science, 283, 684-721.
de t'Academie des Sciences, Paris. 295, 63-8. Thouvenot, F. & Menard, G. 1990. Allochthony of the Chartreuse Subalpine
Elliott, D. 1976. The energy balance and deformation mechanisms of thrust massif: explosion-seismology constraints. Journal of Structural Geol-
sheets. Philosophical transactions of the Royal Society of London. A2H3, ogy, 12, 113-121.
289-312. Williams, G. D. & Chapman. T. 1983. Strain developed in the hangingwall
Fischer, M. W. & Coward, M. P. 1982. Strains and folds within thrust sheets: of thrusts due to their slip-propagation rate: a dislocation model. Journal
an analysis of the Heilam sheet, northwest Scotland. Tectonnphysics. 88. of Structural Geology, 5, 563-72.
291-312.
298
Kinematics of a transverse zone in the Southern Alps, Italy
Gregor Schonborn
Abstract: Transverse zones cut the south-verging thrust belt of the Southern Alps into different
segments. Detailed mapping together with computer-aided construction of balanced cross-
sections has enabled the kinematic models of the structures east and west of the Ballabio-Barzio
transverse zone to be expanded into the complex transverse zone itself Kinematic analysis reveals
three distinct stages of basement and sediment thrusting of 17 km, 25 km and >20 km shortening.
The shortening for each stage is the same in all cross-sections. In the Ballabio-Barzio zone, an
inherited normal fault caused the thrust of stage 1 to ramp onto the Upper Triassic and the thrust
of stage 2 to split into different branches. East of a reactivated sinistral north-south trending tear
fault, wedging of the Middle Triassic units led to backthrusting of the Upper Triassic sheet above.
The triangular tip of a basement fault-bend fold produced lateral escape on both sides contempo-
raneous with southward thrusting. Stage 3 deformation was an out-of-sequence thrust, the frontal
fault-bend fold of which was hindered by an oblique lineament. The subsequent stages of a thrust
breaking through the hinge of this fold, overturning and flattening the cut off frontal limb, are
shown by cross-sections through the frontal range at different distances to the oblique lineament.
The Insubric line is a steeply north dipping east-west trending transverse zone. Therefore they are presented briefly here.
fault zone, separating the Southern Alps from the rest of the The kinematic model of the transverse zone introduced
Alps (Fig. 1). The Lombardian part of the Southern Alps is here is quantitatively consistent with the development of the
a south-verging thrust belt consisting of sediments and base- Grigna (in the west) and the Valtorta segment (in the east) and
ment slivers, segmented by north-south striking transverse accounts for many features that have so far resisted explanation.
zones. Alpine metamorphism is absent to very low grade in
the deeper parts and brittle deformation dominates every-
where. Methodology
Fundamental work in the Lombardian Alps was done by
Dutch geologists (e.g. De Sitter & De Sitter-Koomans 1949). Unravelling the Lombardian Alps means dealing with at least
They advocated the concept of vertical basement uplift and six stages of brittle deformation: Permian normal faulting
gravitational sediment sliding and crustal shortening was not expressed as ENE oriented graben and pull-aparts (isopach
assumed at all until the late 1970s (Gaetani & Jadoul 1979; maps of the Permian graben fill indicate this orientation, De
Castellarin 1978). Publication of seismic lines (Fieri & Sitter & De Sitter-Koomans 1949); Liassic normal faulting
Groppi 1981) revealed a foothills belt deeply buried under due to E-W oriented extension; compressional stages 1 to 3
post-kinematic basin fill. Laubscher (1985) constructed (Cretaceous to Miocene?); a late Miocene compressional
balanced cross-sections to estimate the amount of shortening stage documented in the foothills below the Po plain. Addi-
of the exposed part of the Southern Alps, whereas Roeder tionally, the thickness of some Triassic strata changes up to
(1989; in press) linked them to the foothills belt. hundreds of metres within short distances. Therefore quan-
This work is part of a Swiss National Science Foundation titative kinematic models will hardly ever be accurate, but
project aimed at clarifying the kinematic role of the trans- they are still very useful for understanding complex tectonics,
verse zones in the Southern Alps. As this required knowledge since local observations are often incoherent, ambiguous,
of detailed tectonics not available in current publications, apparently contradictory or simply absent.
mapping at 1:5000 and 1:10000 of the whole area was carried A lot of local data have been collected, classified into
out. In order to maximize ease and accuracy of cross-section categories of significance and most of them incorporated into
construction, computer-aided balancing with the preserva- a 2-D kinematic model, which then was extended into the
tion of bedlength and thickness (GEOSEC - 20™©, Geo- third dimension.
Logic Systems, Inc., Kligfield et al. 1986) was applied.
Balanced cross-sections through the structures east and
west of the Ballabio-Barzio transverse zone (Fig. 2) were Stratigraphy
constructed involving both detached basement and sediments.
These have been published previously (Schonborn 1990) but As summarized in Figure 3, the area investigated involves a
serve as a starting point for unravelling the kinematics of the Hercynian and older metamorphic basement, Permian elastics
299
KINEMATICS OF A TRANSVERSE ZONE. SOUTHERN ALPS, ITALY
-; Rhaetian to
. Cretaceous
I Upper Tr.
Middle Tr. A' n
I Mezzoldo bas. unit i C
+ S. Marco basement unit 10 km
301
G . SCHONBORN
Wm^ Camian
The Valtorta segment is bordered to the west by the Faggio
line, a strike-slip fault with transpressional character in the
south and transtensional in the north. Sinistral sense of shear
is evidenced by some striations, slickolites and steps, clearly
Ladinian sinistral parallel faults (markers are more than 100 m dis-
placed), the geometry of pull-aparts and compressional fea-
tures where the line splits into different branches (Fig. 2b), the
Anisian
SEythian backthrust of the Upper Triassic unit in the east and the much
+++++++++ Upper Permian larger south-verging shortening of the Upper Triassic west of
++++++++++• rIOOOrr
++++++++++• Middle the line. Zanchi etal. (1988) observed dextral sense of shear,
Permian
+ + + + +++ + + ++++^ which is in disagreement with all of the author's data.
++++++++++++++++++• The trace of the section C (Fig. 6) is bent to account for
some eastward motion of the unit 2A (Barzio). The most
striking features are the two Upper Triassic rampfolds, the
Figure 3. Summary stratigraphic column. The competent layers of Middle
and Upper Triassic age are both outlined by detachment horizons. units IIA and IIB (Resegone and Camozzera), thrust south-
ward along the sinistral Faggio line in the east and dextral
Stage, about 25 km in the second stage. The model may be transpressional faults in the west. The unit 1 (Muschiada), the
tested calculating shortening separately for sediments and block above the sheet II A, is composed of a very condensed
basement. Since the unit 3 is cut obliquely by the Lecco line Middle Triassic sequence and topped at Mt. Muschiada itself
in Figure 4c, there is little information on stage 3. by a small remnant of the Upper Triassic sheet above (unit I).
The cross-section (Fig. 6) is simplified for better clarity: the
Middle Triassic is too thick.
The Valtorta section The S W dipping Middle Triassic unit 2 A is separated from
the lower basement unit by the Valtorta fault, which has a
East of the Ballabio-Barzio zone structural geometry changes marked vertical component of displacement there. SE trending
(Fig. 5). No large-scale mass transport along the transverse dextral strike-slip faults in this unit, partly in the form of
zone beyond the studied area is observable, therefore overall flower structures (e.g. east of the village of Barzio) may be
shortening in the east and the west has to be identical. interpreted as former continuations of the Valsassina line, but
During stage 1 the Orobic thrust ramped through both the cut off and transported eastward by the contemporaneously
Middle and Upper Triassic layers, without a flat in the active Valtorta fault (Fig. 2b).
Camian horizon between. The Middle Triassic unit 1 (Bruco)
is an equivalent of 1A and IB, and also remnants of the Upper
Triassic unit above (I, Zuccone unit, not exposed in the Kinematics
Grigna segment) may be observed. Figure 6a shows the restored version of the section C. The
The kinematics of the second stage were complicated by Upper Triassic units II A and II B are interpreted as being
backthrusting of the Upper Triassic sheet II (Aralalta, Figs 5b equivalent to the unit II (Aralalta-Albenza) from further east.
& 5c). After ramping through basement and Middle Triassic, The link is enabled by correlating the Albenza - Upper
the latter by two in-sequence thrusts, shortening was split into Triassic (southern part of II) with the Camozzera - Upper
a south-verging thrust (12 km), some internal deformation of Triassic (IIB): the overlying Rhaetian to Cretaceous strata of
the Upper Triassic and a backthrust (about 5.5 km). There- both units are continous (compare Figs 10 a & b). Unit IIA
fore the units 2A and 2B beneath the Upper Triassic II formed is the northern continuation of IIB, and restoring the different
a wedge. In order to accommodate backthrusting of the unit Upper Triassic units levels up exactly the northern termina-
II, the Middle Triassic unit 1 (Bruco) above is dissected in the tions of the units II A and II B in the west and n in the east
model by an out-of-sequence thrust, creating the unit 1' (compare Fig. 9c and the Upper Triassic ramp on Fig. 9b).
('Superbruco'). The Middle Triassic Muschiada unit (1 on section C) is
The backthrust was concealed by the Valtorta fault. Along explained as lateral continuation of the Bruco unit (1 on
this sinistral strike-slip fault eastward motion of the Valtorta section D) because both are thrust onto the Upper Triassic
segment took place. level II (the correlation is covered up by subsequent rotation
The out-of-sequence thrust active during stage 3 cut the and backthrusting of the eastern segment). Lying on top of
302
KINEMATICS OF A TRANSVERSE ZONE. SOUTHERN ALPS, ITALY
N CROSS•StCTlO'H^ X
^ 4&. e^SIJ&S
+ + + + + -t- + t + + + + + + + t-t- + +, . « . • >,-r + 4--t- + + + + -|-+ + + + + + + + + + + + + + + + +-(. + i--)- + + + +
- . >.^i-'+ + + ^- + + ^-^- + + ^. + -^- + + ^- + + + + + + + + ^. + + + + + ^.^.^. + ^.
+ + + + +-+ + + + + + + + + " '_'•'•'+ + + + •>- + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
+ + + + -,'+ + + + + + +++ + + +++ + ++ + + + + + ++4-4-+ + ++ + + + + + ++ +
20 km
b)
UPPER TR.
H =V
20 km
THRUSTS. STAGE 1 _ . _ 7 : f ' . _ , (OROBIC THRUST) STAGE 2 (COLTIGNONE THRUST) STAGE 3 (LEGCO THRUST)
Figure 4. Cross-section A. A kinematic model of the Grigna mountains, modified after SchSnbom (1990). In a first stage the upper basement unit, the Middle
Triassic unit 1A and then IB were transported along the Orobic thrust. In stage 2 the lower basement unit (Orobic anticline) and the Middle Triassic unit
2 were thrust on lop ofthe unit 3. Shortening of the first stage is 17 km, of the second stage 25.5 km. Basement unit I = San Marco. 2 = Mezzoldo; Middle
Triassic unit 1A = Grigna-N, i B = Grigna-S, 2 = Coltignone, 3 = Lecco.
H = V 20 km
THRUSTS: STAGE 1 _ . _ - ^ . _ . (OROBIC THRUST) STAGE 2 (COLTIGNONE THRUST) STAGE 3 (LECCO THRUST]
Figure 5. Cross-section D. A kinematic model of the Valtorta segment, modified after Schonbom (1990). Dashed lines are thrusts active in the subsequent
stages, (a) Palinspastic configuration before thrusting, (b) Situation after stage 1. Arrows indicate the relative movements ofthe next stage. The UpperTriassic
sheet II is thrust southward as well as northward. (c)Current cross-section after stage3.The units 2A and 2Bfomi a wedge below the backthrust UpperTriassic
unit II. Shortening as in Figure 4. Basement unit 1 - San Marco, 2 = Mezzoldo; Middle Triassic unit 1 = Bruco. 1' - 'Superbruco', 2A - Barzio. 2B =
'Infrabarzio', 3 - Lecco; UpperTriassic unit I - Zuccone. II = Aralalta-Albenza, II C = 'Gerola'.
303
G. SCHONBORN
the frontal limb of the lower basement unit the Middle sections further east (Figs 5 & 6): The ramp of the stage 1
Triassic unit 2A corresponds to the northern part of the unit thmst onto the Upper Triassic is originated by the inherited
2 (Coltignone), occupying the same position in the Grigna fault, as well as the splitting of the stage 2 thrust into two
segment. The unit 2B is not exposed but is required to ensure branches. The Ema block (2B of Fig. 7) shows the situation
the same amount of shortening as in the east and in the west. of the unit 2B further east during early stage 2 thmsting.
This unit is probably strongly imbricated since it vanishes Consequently the thmst in the westem part of the transverse
laterally. zone between the Ema unit and the Upper Triassic IIA above
Unlike in the section D (Fig. 5), the stage 1 thrust ramps had a much smaller displacement than the same thmst in the
through the Upper Triassic after a low angle detachment in eastem part. The southward motion of the II A thrust,
the Camian in the section C (Fig. 6b). The stage 2 thrust splits accounting for some 3.3 km in the section C, is reduced to
in the Camian into a high angle detachment through the perhaps a few hundred metres in the section B and vanishing
Upper Triassic (the Resegone thrust) and into a flat following towards W. The (projected) remaining folds around the tip of
the Camian beds cut by the subsequent thmst of stage 3. the unit 2 in the Grigna segment (Figs 4c & 9) are reminiscent
of the folds of the Mt. Moregallo which occupies a similar
tectonic position.
The western part of the Ballabio-Barzio zone The overturned sequence in the south is explained as the
overtumed limb of a fault-bend fold (see below). The Upper
Geometry
Triassic unit IIC, drilled into by the Gerola well, is depicted
The Middle Triassic unit 2A is strongly imbricated north of in Figure 7 as the hangingwall of the stage 2 thmst, cut by the
an Upper Triassic block called Due Mani (IIA west on Fig. out-of-sequence stage 3 thmst. Its length in N-S direction
2b). A steep fault (which is not exposed but a steep inclination must be at least 12 km (displacement of the location of the
is inferred by the limiting outcrops) marks the contact, future stage 3 thmst in the Upper Triassic during stage 2, Figs.
turning into a more gently south-dipping position (35°) as the 6b & 7b). Thrusting in the Jurassic strata of the 'Gerola unit'
result of minor backthrusting. This fault and constmcted should represent parts of stage 2 thmsting.
cross-sections (the base of the Upper Triassic is exposed The cross-sections A to D are summarized in a block
further south) point out that this block does not lie on top of diagram (Fig. 8) to clarify the 3-D geometries.
Middle Triassic 2A, but more or less on the same level. It was
thmst southward onto two Middle Triassic blocks that are
connected to the sheet 2 further west. The eastem one, called Chronological development
Mt. Ema, dips steeply WSW - SSW, discordant with its basal
thmst. Further to the south, an overturned sequence of Upper The chronological development is shown on schematic maps.
Triassic Dolomia Principale to Cretaceous Pontida flysch Figure 9a is a horizontal cut through the Triassic before
with many south to west-verging thmsts is one of the major thmsting: a ridge of Middle Triassic strata was bordered by
problems of local tectonics. A well near Gerola (Fig. 2b) inherited normal faults to the SW and to the E. 'Lecco fault'
revealed a nearly flat-lying sequence of right way up Upper means the eastem border of the Early Liassic Generoso basin
Triassic to Cretaceous strata (e.g. Jadoul & Gaetani 1986). in the Comizzolo area. Vertical displacement was more than
1 km along the Lecco and Fontana faults and unknown along
the Faggio fault.
Kinematics
Figure 9b illustrates the situation after stage 1: The Orobic
The restored version of the model (Fig. 7a) interprets the thmst ramped onto the Upper Triassic east of the Fontana
Upper Triassic unit U A as the hangingwall of an inherited fault, which made this step easier by lowering the Upper
normal fault (Fontana fault). This is the most simple and Triassic. Because the thmst split into two branches west of
kinematically viable explanation for the geometry described the Fontana fault (1A & 1B), the Middle Triassic got displaced
above, and it is supported by the comparison of Upper dextrally above this fault. Although the Upper Triassic west
Triassic cataclasites after the method of Froitzheim (1988). of the Faggio fault occupied a higher position than east of it,
The imbrication of the unit 2A may be produced by thmsts this healed fault was not reactivated during stage 1. The
active during stage 1 as well as during stage 2. The Ema block theoretically demanded strike-parallel shortening on the for-
with its bedding cut discordantly by the basal thmst is ward termination of the somewhat oblique Upper Triassic
depicted as sheared off and rotated frontal limb of the unit 2. ramp (Apotria etal. 1990) possibly influenced the forming of
The stage 1 thrust crosses the Upper Triassic at the the 1A thmst. The obliqueness of this ramp could be related
inherited fault, thmsting the (projected) Middle Triassic unit to ENE oriented prolongations of the Fontana fault. Also
1 (Muschiada) on top of the Upper Triassic unit 11 A. other important features in the Lombardian Alps are oriented
The stage 2 thmst splits into two thmsts near the inherited like this (e.g. the Orobic, Trabuchello and Cedegolo anti-
fault: one going on horizontally and then ramping through the clines, the Clusone fault and the unnamed one northeast of
Middle Triassic below the unit 2B, the other leading up San Pellegrino). Liassic normal faulting following an inher-
immediately and thmsting Upper Triassic n A onto 11B (Fig. ited Permian fault system and influencing Alpine thmsting
7b). could be an explanation for this phenomenon.
This section reveals the reason for the evolution of the Figure 9c elucidates stage 2: The triangular shape of the
304
KINEMATICS OF A TRANSVERSE ZONE, SOUTHERN ALPS, ITALY
N CROSS•S%CTI09i C
+ + + + + + + + + + + + + + + + + + +,-•*' + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
+ + + + + + + + ,] + + + + + + + _>^' • • • " " ^v'^ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
+ + + + + + + + + + + + + + ;.-F'+ + + + + + + + + + + + + + + + + + + + + + t + + + + + + + + + + + + +
-t- + + + + +^' ''+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + •!- + + + + + +
20 km
b)
CRETACEOUS
JURASSIC
PERMIAN
+ +++ + and
+ + + + -h
+ +++ +
BASEMENT 20 km
Figure 6. Cross-section C. A kinematic modeiof the eastern part of the Ballabio-Barzio transverse zone, (a) Restored section, (b) Situation after stage 1.
Arrows indicate movements of stage 2. (c) Current cross-section. Minor backthrusiing of unit 2A is located between the Permian and the basement. Some
eastward motion and extension along strike of unit 2A has to be added. Unit 1 is too thick and the Upper Triassic unit I above not depicted, neither is unit
IIC in Figure 6c, Shortening as in Figure 4. Basement unit 1 = San Marco, 2 = Mezzoldo; Middle Triassic unit 1 = Muschiada, 2A = Barzio, 2B = ' Infrabarzio',
3 - Lecco; Upper Triassic unit II A = Resegone, II B = Camozzera, H C = 'Gerola'.
c)
CRETACEOUS
JURASSIC
^^^B
•
UPPER THIASSIC
MIDDLE TRIASSIC
PERMIAN
+ + ++ + and
+ + ++ + BASEMENT
+ + ++ + 20 km
THRUSTS: STAGE1 _ . _ r ± " . _ . ^ " (OnOBICTHRUST) STAGE 2 . ™ ^ ™ J W , - ' (COLTIGNONE THRUST) STAGE 3 (LECCO THRUST)
Figure 7. Cross-section B. A kinematic model of Ihe western part of the Ballabio-Barzio transverse zone. The structures further east (Figs 5 & 6) were
originated at the inherited Fontana normal fault: the ramp of the stage 1 -thrust onto the Upper Triassic, the splitting of the stage 2-ihrust and internal thrusting
of the Upper Triassic during stage 2. The unit IIC represents the hangingwall of the stage2thrust, cutby theout-of-sequence stage 3-thrust. Shortening as
in Figure 4. Basement unit 1 - San Marco unit, 2 = Mezzoldo; Middle Triassic unit 1 = Muschiada, 2A = Barzio, 2B = 'Infrabarzio', 3 = Lecco; Upper Triassic
unit II A = Due Mani, II B = Camozzera, II C = 'Gerola'. P = Upper Triassic of 11 Pizzo, E = Jurassic of Erve.
305
G. SCHONBORN
Figure 8. Simplified block diagram of the Grigna segment to the left, the Vailorta segment to the right and the uplifted Ballabio-Barzio transverse zone
in between. Cross-sections A to D refer to Figures 4,7,6 & 5. The Upper Triassic unit I in the transverse zone and the frontal fault-bend fold of cross-section
A are not depicted. Basement unit 1 - San Marco ("Orobic crystalline'), 2 = Mezzoldo ("Orobic anticline'); Middle Triassic unit I - Muschiada in the
transverse zone, Bruco in the Vaitorta segment, lA = Grigna-N, IB = Grigna-S, 2 = Coltignone, 2A - Barzio, 2B -Ema/"Infrabarzio'; Upper Triassic unit
I = Zuccone, U (A&B) - Aralalta-Albenza, IIA = Due Mani and Resegone, IT B - Camozzera, IIC = 'Gerola', P = It Pizzo, I & II = not exposed Upper Triassic
above the Grigna.
tip of the lower basement unit deflected the local transport level 2. A larger rotation of the Vaitorta segment than
directions. The Grigna segment was compressed towards assumed here would make 2B unnecessary, but then prob-
SW (Grigna syncline), the Vaitorta segment was rotated lems would arise further west; in this case the Upper Triassic
clockwise producing extension in the northern part of the unit 11A must have been pushed south from below the Middle
transverse zone and the adjoining Vaitorta segment. Parts of Triassic unit lA. No indications supporting this interpreta-
the unit 1, sunk in a graben of the Upper Triassic II, verify a tion are at hand.
post-stage 1 age of the extension. During stage 2 the Faggio Between the units II A west and east transtensional fea-
fault was reactivated as a tear fault: the Vaitorta segment was tures in the northern part and transpressional in the southern
backthrust. The unit 2 A was transported somewhat eastward part provide evidence of minor variations in transport direc-
and northward with respect to the basement (i.e. l8to 19km tions, possibly as an effect of the rotation of the Vaitorta
of southward transport have to be added). Backthrusting took segment. Shortening along the thrust II A (Resegone thrust)
place to a very limited extent along thrust planes between increased eastward, intensifying the along-strike shortening
Permian volcanics and the basement. An opening gap be- near the Faggio line (N-S trending anticline in the unit II A
tween the units 2A and 2 because of the SE movement of the east).
unit 2 A and the SW movement of the unit 2 may have caused Stage 3 thrusting is localized in the southern part of the
the eastward dip of the unit 1A (above) near the transverse map. Here the influence of a transverse zone on thrusting can
zone. The swinging of the front of the unit 2A (shaded barbs) be observed directly. Figure 10 exhibits the frontal parts of
below the Upper Triassic indicates the area of existence of the the discussed sections. Further away from the oblique Lecco
unit 2B below. This probably strongly imbricated unit is lineament a fault-bend fold (or broken through fault-propaga-
necessary to ensure the shortening of the Middle Triassic tion fold, which can hardly be distinguished after some
306
KINEMATICS OF A TRANSVERSE ZONE. SOUTHERN ALPS, ITALY
T 10km B T 10 km B D
Idle J r . rami
Q 1A
A A A 4 i
1B
A • A
^ \
2
A A A..
A' C D'
Middle Triassic upper Triassic
B
transport) with a steep southern limb is the sign of a ramp In the section C through the Mt. Camozzera (Fig. 10b)
through the Upper Triassic to lowermost Cretaceous strata transport along the thrust below therampfold, 3a, was stopped
(Fig. 10a). The unexposed stage 3 thrust (Lecco thrust) by the Lecco lineament and another thrust, 3b, broke through
turned into a flat in the lower to middle Cretaceous, as it is the hinge of the fold, slightly overturning the frontal limb as
indicated by the fold style and the thrusts further south. West shown by the steeply north dipping Jurassic beds. Minor
of the Lecco lineament (e.g. near the Mt. Comizzolo) struc- backthrusts are demanded by the geometry and can actually
tures look similar to Figure 10a. be observed in the field. The changes between 10a and 10b
307
G. SCHONBORN
occur across the Faggio line and are responsible for the
intriguing picture of sinistral displacement high up in the
mountains (hangingwall of thrust 3b) and dextrai down in the
valley (foolwall of 3b) along the same fault.
Figure lOc is a cross-section through the eastern flank of
Mt. Barro, that is surrounded by various elements of the
Lecco lineament. The Upper Triassic of the overturned limb
was thrust onto the also steeply north dipping Jurassic along
thrust 3c.
The section B through II Pizzo (Fig. lOd) displays the final b) Section C (Camozzera)
308
KINEMATICS OF A TRANSVERSE ZONE. SOUTHERN ALPS, ITALY
NE Fa^gio Transpression SW
Line zone Lecco
0 Line
Lower Liassic
Upper Rhaetian
Upper Triassic
Middle Triassic
•..•,].••...il> '•">";
Figure 11. Cross-section E. SW trending cross-section in the southern part of the transverse zone. The hangingwall of the thrust 3b in Figure 10b (IIA &
IIB) was thrust onto the overturned footwall of this thrust in Figure lOd (P & E). The corresponding hangingwall is eroded on the southwestern side. Parallel
faults of the dextral Lccco line intersect the overturned strata. Upper Triassic unit II = Aralalta-Albenza, II A = Resegone, II B = Camozzera, P = Pizzo; E
= Jurassic of Erve.
fault and the area further east backthrust. and extension, which in turn has the same effect as inherited
The fault-bend fold of the stage 3 thrust was hindered by unevenness. In this way an inherited fault or other uneven-
an oblique lineament {Lecco line). A thrust broke through the ness is able to propagate its consequences.
hinge of the fold and overturned the frontal limb. The out-of- In the case of the Ballabio-Barzio transverse zone the
sequence stage 3 thrust cut off the hangingwall of the stage 2 oblique orientation of Permian graben with respect to subse-
thrust now lying below. quent thrusting led to en-echelon aligned rampfolds, which in
Shortening is some 17 km during the first, 25 km during turn produced lateral escape, which had the effect of {1) S ^ -
the second and at least 20 km during the third stage. deflection and in this way enabled reactivation of the Faggio
This transverse zones illustrates the severe influence of fauh as a tear fault and therefore backthrusting of the Valtorta
oblique inherited faults on subsequent thrusting: they may segment and (2) lateral extension in the northern part of the
provoke additional ramps (Fontana fault) or more complex transverse zone, along strike compression in the Brembo
thrust systems, trying to overcome the obstacle {Lecco fault). valley and possibly in the southern part of the transverse zone.
Additionally all sorts of oblique and lateral ramps may
originate near such faults, depending on fault geometry and This work benefited from discussions and field trips with H. Laubscher, M.
stratigraphy. Healed inherited faults oriented parallel to the Schumacher, Th. Noack and many others. Th. Widmer is thanked especially
for offering his field data. The paper has been improved by ihe advice of R.
transport direction may be not reactivated until deflection of Zoetemeijer, W. Sassi and an anonymous reviewer as well as by the editorial
the local S|-pattem produces favourable conditions {Mohr help of K. McClay. Support by the Swiss National Foundation, grant
criterion) for reactivation as strike-slip or transpressional numbers 20 - 4798.85 and 20 - 25560.88 is gratefully acknowledged.
faults. Every lateral movement, although small compared to
the main transport direction causes along strike compression
309
G . SCHONBORN
References
Apotria, T. G., Spang, J. H., Wiltschko, D. V. & Snedden, W. T. 1990. Three- & Doniselli, T. 1986. La successione del Lias inferiore di Morterone
dimensionalmodelling of transverse fault-bend deformation./w.McClay, (Lecchese). Memorie dellu Societu Geologicu Ituliunu, 32, 49-66.
K. R. (ed.) Thrust Tectonics 1990, Programme with Abstracts, Royal Kligfield, R., Geiser, P. & Geiser, J. 1986. Construction of Geologic Cross
Holloway & Bedford New College, University of London, 18. Sections using Microcomputer Systems. Geohyte, 1, 60-66.
Bernoulli, D. 1964. Zur Geologie des Monte Generoso (Lombardische Laubscher, H. 1985. Large-scale, thin-skinned thrusting in the Southern
Alpen). Beitrdge zur Geologischen Karte der Schweiz, NP118, 1-134. Alps: Kinematic models. Geological Society of America Bulletin, 96,710-
, Bertotti, G. & Zingg, A. 1989. Northward thrusting of the Gonfolite 718.
Lombarda ('South-Alpine Molasse') onto the Mesozoic sequence of the 1990. The problem of deep structure of the Southern Alps: 3-D material
Lombardian Alps: Implications for the deformation history of the balance considerations and regional consequences. Tectonoplnsics, 176,
Southern Alps. Eclogue Ceologicae Helvetiae, 82, 841-856. 103-121.
& Winkler, W. 1990. Heavy mineral assemblages from Upper Lichtensteiger, T. 1986. Stratigraphie und Tektonik der sudostlichen Alta
Cretacaceous South- and Austroalpine flysch sequences (Northern Italy Brianza. Ph.D. thesis, Basel, (unpubl.).
and Southern Switzerland): source terranes and paleotectonic implica- Pieri, M. & Groppi, G. 1981. Subsurface geological structure of the Po plain,
tions. Eclogue Geologicue Helvetiue, 83, 287-310. Italy. Progetto Finalizzato Geodinamica, Pubhlicazione del Consiglio
Bertotti, G. 1990. The deep structure of the Monte Generoso basin. In: Roure, Nazionale delle Ricerche, 414, 1-13.
P., Heitzmann, P. & Polino, R. (eds) Deep structure ofthe Alps. Memoire Roeder, D. 1989. South-Alpine thrusting and trans-Alpine convergence. In:
de la Societe geologique de France, 156; Memoire de la Societe Coward, M. P., Dietrich, D. & Park, R. G. (eds) Alpine Tectonics.
geologique de la Suisse, 1; Societa geologica d'ltalia, volume speciale, Geological Society Special Publication, 45, 211-227.
1, 290 - 303. in press. Thrusting and wedge growth. Southern Alps of Lombardia
Castellarin, A. 1978. II problema del raccorciamenti crostali nel Sudalpino. (Italy). Teclonophysics.
Rendiconti dellu Societu Geologicu Ituliunu, 1, ll-Ti. Schonbom, G. 1990. A kinematic model of the western Bergamasc Alps.
De Sitter, L. U. & De Sitter-Koomans, C. M. 1949. The Geology of the Eclogue Geologicae Helvetiae, 83, in press.
Begamasc Alps. Leidse Geologische Mededelingen, 14B, 1-257. & Laubscher, H. 1986. The suborobic imbrications near Taceno.
Proitzheim, N. 1988. Synsedimentary and synorogenic normal faults within Memorie della Societu Geologica Italiana, 32, 113-121.
a thrust sheet of the Eastern Alps (Ortler zone, Garubiinden, Switzer- Schumacher, M. 1990. Alpine basement thrusts in the Eastern Seengebirge,
land). Eclogue Geologicue Helvetiae, 81, 593-610. Southern Alps, Italy/Switzerland. Eclogue Geologicae Helvetiae, 83, in
Gaetani, M. 1982. Elementi stratigrafici e strutturali della galleria Bellano- press.
Varenna (Nuova SS 36) (Como). Rivista Italiana di Paleontologia e Snedden, W. T. & Spang, J. H. 1990. Geometric analysis of displacement
Stratigrafia,»S, 1-10. transfer between thrust faults using fault-bend and fault-propagation fold
& Poliani, G. 1978. II Toarciano e il Giurassico medio in Albenza. models. In: McClay, K. R. (ed.) Thrust Tectonics 1990, Programme with
Rivistu Itulianu di Paleontologiu e Strutigrafiu, 84, 349-382. Abstracts, Royal Holloway & Bedford New College, University
& Jadoul, F. 1979. The structure of the Bergamasc Alps. Rendiconti of London, 19.
dellu Accademiu Nazionule dei Lincei, Serie VIII, 66,411-416. Suppe, J. 1983. Kinematics of fault-bend folding. American Journal of Sci-
Gianotti, R., Jadoul, P., Ciarapica, G., Cirilli, S., Lualdi, A., Passeri, ence, 2»3,fyM-72\.
L., Pellegrini, M. & Tannoia, G. 1986. Carbonifero superiore, Permiano Winterer, E. L. & Bosellini, A. 1981. Subsidence and sedimentation on
e Triassico nell'area lariana. Memorie della Societu Geologica Italiunu, Jurassic platform margin. American Association of Petroleum Geolo-
32, 5-48. gists Bulletin, 65, 394-421.
Jadoul, P. 1985. Stratigrafia e paleografia del Norico nelle Prealpi Bergamasce Zanchi, A., Forcella, P., Jadoul, P., Bernini, M., Bersezio, R., Fomaciari, M.,
occidentali./JiVii(a/M/(a«fl diPuleontologia e Siratigrafia,9l,479-512. Rossetti, R. & Torazzi, S. 1988. The Faggio-Morterone transverse line:
& Gaetani, M. 1986. L'assetto strutturale del settore Lariano centro- Mesoscopic analysis and kinematic implications. Rendicontidella Societd
meridionale. Memorie della Societa Geologica Italiana, 32, 123-131. Geologica Italiana, 11, 279-286.
310
Hangingwall geometry of overthrusts emanating from ductile
decoUements
Peter Jordan & Thomas Noack
Abstract: The hangingwall geometry of thrusts emanating from decollement-type sole thrusts
strongly depends on the position, the thickness and the deformation style of the sole thrust and the
way the ductile basal layers are included in the hangingwall geometry. Analytical models for some
characteristic overthrust geometries are formulated and compared to field examples from the Jura
mountains. All these models are area balanced and within the portions which deform in a brittle
fashion are also line length balanced. Due to a wedge-like inclusion of ductile material in the thrust,
overthrusts nucleating from thick sole thrusts have back limbs that are longer and flatter than the
ramp. Furthermore, these backlimbs and the respective hinges rotate during deformation and the
flat-ramp hinge migrates in the opposite sense than bulk shear movement. Due to hinge rotation
and migration, internal destruction of those back limb areas that migrate through the hinges is kept
to a minimum. In these models, important portions of the backlimb do not even migrate through
the flat-ramp hinge.
A decollement is characterized by a stratal controlled sole layers in the vicinity of thrusts have an important influence on
thrust localized in particularly soft or incompetent horizons the geometry and the internal deformation of the hangingwall.
(e.g. Ramsay & Huber 1987, p. 518). In many cases, these For instance, they cause a hangingwall geometry that is not
horizons, e.g. evaporites, behave in a ductile manner while (directly) related to the geometry of the footwall ramp as it is
the overlaying strata, e.g. carbonates or clastic sediments, assumed by Suppe (1983), Jamison (1987) and Mitra &
behave in a brittle fashion (Laubscher 1961, 1973, 1986; Namson (1989), who use a sharp, infinitesimally thin sole
Davis & Engelder 1985; Malavieille & Ritz 1989). thrust for their constructions.
Decollement in ductile layers has been recognized by Buxtorf First attempts to quantify the geometry and deformation of
as early as 1907. Consequently, with the inclusion of ductile overthrusts emanating from thick ductile sole thrusts were
basal layers in the hangingwall of thrust systems, the amount made by Suter (1981) and Taboada et al. (1990). In the
of thickening or thinning has often been overestimated. following, it will be shown that the nearly infinite range of
Ductile materials have commonly been used to fill gaps natural decoUement-thrust geometries (e.g. Fig. 2e) can be
between the basement and deformed cover without regard to regarded as combinations of four major types (Figs 2a - d)
mass balance or thrust kinematics (Fig. la). In order to keep that differ in the localization and the deformation style of the
mass balance, other authors try to fill these gaps with duplexes sole thrust and the hangingwall geometry.
and basement structures (e.g. Boyer & Elliot 1982; Mitra The Type 1 Thrust is associated with a 'Top Decollement'
1986,1988; Gurler et al. 1987; Bitterli 1988). Their con- (Fig. 2a), characterized by a sharp (infinitesimally thin) sole
structions are based on a sharp sole fault at the base or within thrust at the top of the weak layer. No weak material is
the ductile layers, and a mass and line length conserving dip included in the thrust. The hangingwall is assumed to deform
domain structure (Suppe 1983, Groshong & Usdansky 1988) in a geometry according to Suppe (1983). This type of
in the hangingwall, viz. the ductile layers are assumed to deformation is quite common in the frontal duplex zone of
behave the same way as all other overlying brittle strata eastern Jura (e.g. Bitterli 1988).
(Fig. lb). The Type 2 Thrust is related to a' Simple Shear Decollement'
Natural examples show that sole thrusts are often not sharp (Fig. 2b), introduced by Malavieille & Ritz (1989). It is
faults but shear zones of a certain thickness (e.g. Jordan & characterized by homogeneous layer-parallel simple shear in
Niiesch 1989). Based on recent field work in the Jura and on a thick ductile sole thrust. The ductile material is squeezed
geometrical models, this paper will demonstrate that there are between the competent backlimb and the undeformed footwall,
accumulations and rolling-out of weak basal materials in producing a backlimb that is considerably flatter than ex-
decoUement-related overthrusts. However, they are in con- pected from the ramp angle. The horizontal crest and the
cordance with overall mass balance, and they are controlled forelimb still exhibit the angular relationship to the ramp
by the thickness of the sole thrust, its localization within the angle indicated by the model of Suppe (1983). Malavieille &
ductile layers, its deformation style and the way the ductile Ritz (1989) suggest that the Arc de Digne nappe of southern
material within and above the sole thrust is included in France is a good example for this geometry.
overthrusting. These local changes in thickness of ductile The Type3 Thrustis associated with a 'Basal Decollement'
311
p. JORDAN & T. NOACK
5_q(e Ihrust
Weak and Ductile Material
Basal Decollement
312
H A N G I N G W A L L GEOMETRY OF OVERTHRUSTS EMANATING FROM DUCTILE DECOLLEMENTS
Figure 3. Model and geometry of a thrust related to simple shear decollement (Type 2) (for discussion see text, explanation of symbols in Table 1). The
orientation of the passively rotated hinge C^' gives the amount of internal shear in the stiff backlimb as the hinge was originally oriented perpendicularly
to layering.
base of the stiff hangingwall segment has reached the upper Ramp models
footwall flat. Nevertheless, possible models for this latter
situation are discussed. All models are formulated in two dimensions, i.e. in the
transport direction assuming plane strain (cf. Suppe 1983).
The ramp models for Type 2 and 3 thrusts and thrusts
Table 1 Symbols used in text and equations. nucleating from combined decollements describe the
Ac, Be, Cc hinges of hangingwall above ductile basal layer situation from thrust nucleation until the arrival of the base
Ac', Be', Ce' inactive (passively transported) hinges of of the britde layer at the upper footwall flat.
hangingwall
Be* hinge between backlimb and foreland dipping
crest in 'shear model' Thrust related to a simple shear decollement (Type 2)
b distance JE = h/ tan (9)
C ramp-flat hinge of footwall = imaginary flat-ramp This model assumes that (Fig. 3): (1) all volumes are con-
bisector in basal ductile layer served; (2) within the competent layers, bed lengths and
h distance HJ = hop . sin (Be)
=/(As); distance EP; distance along initial
thicknesses are conserved; and (3) there is no displacement
'bas
position of top of basal ductile layer between along the interfaces between ductile and competent layers.
footwall ramp and basal point of hangingwall The thrust fault that cuts the competent layers at an angle9
Cc-hinge is suggested to have propagated rapidly at the very first step
'front length of 'sharp frontal shear zone' of deformation. Subsequently the hangingwall starts to thrust
'top distance HP; length of the top of basal ductile
layer in backlimb
over the footwall in such a way that the volume of the ductile
NP nucleation point of initial hangingwall Cc-hinge strata remains constant, i. e. the 'volume loss', Avol , is
source '
(= P for As=0)
thickness of (brittle) hangingwall layers above
compensated by a 'volume gain', Avol^.^ (Fig. 3), therefore:
Te
ductile basal layer
Td thickness of ductile basal layer 0.5 A s T d = 0 . 5 - Ibas-h
thickness of partial layer i
(1)
Ti
source sink
a dip of C-hinge of footwall
ae = 0.5.(jt-e); dip of Cc-hinge of hangingwall
y = As/Td; shear parallel to decollement
7> shear of partial layer i
where As is the shortening in the competent strata, Td the
7* additional shear in ramp triangle parallel to Cc
Aa = a - o c =0.5.(e-ec)
thickness of the ductile layer, 1^^^ the distance between P and
As shortening of all competent hangingwall layers E, (the lower end of the footwall ramp of the competent
Avol volume of source = volume of sink layers), and h the vertical component of the displacement
e ramp angle of footwall vector EH. During thrusting, the stiff hangingwall is rotated
dip of backlimb = apparent ramp angle of
by the angle 0^ around point P, resulting in a hinge C . In order
hangingwall
to keep layer thickness constant, C^ has to be the bisector of
the angle (ii-G). In contrast to the model of Taboada et
313
p. JORDAN & T. NOACK
314
HANGINGWALL GEON4ETRY OF OVERTHRUSTS EMANATING FROM DUCTILE DECOLLEMENTS
Ibas(NP) = Td/tan(9)
"c "c B
Be B{
Figure 7. Plausible models forthrust development afterarrival of the baseof the stiff backlimb at the top of the footwall: (a)'sharp frontal shear zone model';
(b) geometry of 'sharp frontal shear zone model' and kinematic sequence from step 0 (C^," etc.) to step 1 (C^,' etc.). The rotation of B^ is indicated by an
arrow as the rotation is to small to be drawn. C^andB^rotateparaliel to each other. Notice backward migration of C^ hinge. ' Shear models': (c) hangingwall
passes through the hinge A^,that is fixed, (d) hangingwall passes through new hinge B^.* that originates from B^, (after Taboada fl <j/. 1990). (e) Face of
hangingwall is equal to ramp situation (Suppe {1983)-type,cf. Fig. 3), rest of hangingwall passes through new hinge Bj,* that nucleates from Bj.(aflerMalavieille
& Ritz 1989).
315
^'^' p. JORDAN & T. NOACK iji.
i^...
4-
(8)
where y. is the shear in x direction (Fig. 3) in the partial layer '^' ^
i that has a thickness T , and T^= 0. Discrete thrusts (bottom,
intra and top decollements) have to be treated as partial layers
(with T y. = amount of displacement along thrust; and, for
the summing-up of thicknesses, T. = 0). The specific 'volume
loss' is normalized by
Avol
•-• • " ' source ^-i ^-1
fk = ^,^^ , where Td = ^ T j and As = ^ T J Y '
i=l i=]
(9a)
Figure 8. Riepel quarry (for locaiion see Fig. l):(a)cross section showing
fkA S' T d = 0.5- Ibas" h (9b) a sharp frontal shear zone-type thrust and an accumulation of ductile
i'j'i >: source sink evaporites (members I and 2) in front of the footwall thrust; (b) restoration
of latest deformation step (cross section constructed and restored using
Geosec"^'^. © Geo-Logic Systems Inc. 1989).
Tc
Ramp-flat m o d e l s »"*••• )c = arctan (11)
l^lbas + Tc/tan(e)
After the base of the back limb has reached the top of the The length of the frontal shear zone, 1 (Fig. 7b) can be
footwall (h=Tc), Type 2 and 3 models and all combined calculated noting that L +1 = As + l
c from lop bas
models are no longer constrained. There is an infinite variety
of possible geometries. In the following, we will discuss
front = As + 1bas
- '\/TC2 + ( Ibas + Tc/tan (6))
possibilities for on-going thrusting (Fig. 7) using a 'sharp
frontal shear zone model' and various 'shear models'.
(12)
The 'sharp frontal shear zone model' (Fig. 7a, b) assumes
a decoupling of ductile and rigid parts of the hangingwall And as a consequence of decreasing 0^ and increasing 1^^,
along their interface when the backlimb has reached the upper substituting equation (9) in (10) yields that the displacement
footwall flat. This results in adirect overriding of the footwall across the (infinitesimally thin) frontal shear zone is slightly
by the brittle part of the hangingwall along a frontal shear faster than the displacement across the sole thrust.
zone which is commonly formed as a very thin slice of ductile The transition of a basal decollement related thrust (Type
material, that might be developed as amylonite (Fig. 7d). The 3) into a sharp frontal shear zone-type thrust is shown in
triangle EHP (Fig. 3) (=Avol .^^ ) becomes bigger with in- Figure 5d. A good field example for a sharp frontal shear
creasing As (Fig. 7b) as almost no ductile material is trans- zone-type thrust is exposed at Riepel Quarry (north of Aarau,
ferred to the upper flat. Noting that h= EH sin(e)=Tc is Fig. 8). The deformation within the triangle (prism) below
constant, equation (9b) yields the backlimb is obviously compressive and accretionary
while the frontal shear zone is.developed as a shear zone.
Introduced by Malavieille & Ritz (1989), t h e ' shear mod-
Ibas = E P = 2 - f k A s-Td/Tc (10)
els' discuss possibilities of transferring ductile material onto
316
HANGINGWALL GEOMETRY O F OVERTHRUSTS EMANATING FROM DUCTILE DECOLLEMENTS
Figure 8(coni'd.). (c)Linedrawingofpanoramapiclure(ca. 30°, cf. Fig. 8a) of lowest level of Ricpcl quarry showing toolwall ramp (left), and accumulation
of hangingwall evaporiies (right). Top right shows nonh limb of the small fold in the southern part of Riepel Quany (cf. Fig. 8a). (d) Detail of mylonite
shear zone ('sharp frontal shear zone', size indicated by coin); (e) detail of fold in the accumulation domain in front of footwall thrust (size indicated by
hammer).
the upper footwall flat. Geometrically, they are all based on contrast, the model proposed by Taboada et al. (1990) (Fig.
a folding down of the crest and the upper backlimb parallel to 7d) is based on a new hinge, B^*, between the backlimb and
different hinges (A^ or B^*, Fig. 7c-d) that are all bisectors of the crest segment. The new hinge develops from the hinge B
the 'initial', unfolded geometry. All'shear models' conserve (initially a bisector) when the latter reaches the flat. The
the volume, however, line lengths and bed thicknesses of one hinge B migrates further while B * is fixed (however, it ro-
or the other segment of the brittle layers have urgently to tates in time). This model results in a flat and far ranging
undergo changes in order to keep geomed-ical constraints. forelimb and a foreland sloping crest segment. Line lengths
Kinematic ally, the three models differ in the function of the and bed thicknesses are conserved within the crest segment
hinges. but not in the forelimb. The last model (Fig. 7e), proposed by
In the model introduced here (Fig. 7c), the thrust front Malavieille & Ritz (1989), is a modification of the model of
passes through the fixed hinge A^. The hinge B^, initially a Taboada et al. (1990). Shear is restricted to the part between
bisector inherited from the ramp geometry, is deflected while B * and the tip of the ductile thrust triangle. The thrust face,
passing through A^. This model results in a steep forelimb and i.e. the frontal forelimb and the frontal crest (flat) are Suppe
a high crest. Line lengths and bed thicknesses are conserved (1983)-type and are conserved from the preceding ramp
in the forelimb but not in the domain between A and B . In thrusting.
317
p. JORDAN & T. NOACK
Conclusions footwall flat, the backlimb becomes again flatter in the 'sharp
frontal shear zone model', whereas it keeps on steepening in
The geometry of thrusts related to thick ductile sole thrusts the 'shearmodel'. The rotation ofbacklimb and hinges result
clearly differ from classical Suppe (1983)-type thrusts nucle- in a less severe deformation of the layers passing through the
ating from sharp sole faults and having a footwall ramp- hinges. A major part of the backlimb is deformed only by
controlled dip domain geometry. The differences include (1) layer-parallel slip, as it does not migrate through the flat-ramp
a backlimb that is much longer than the actual ramp; (2) a hinge. This solves some questions concerning field evidence,
lower angle of the back limb in respect to the ramp angle; (3) where it seems, that the backlimb of a fault-bend-fold has
a rotation of the backlimb; (4) a rotation of hinges B^ and C^; often remained more or less intact.
and (5) a migration of hinge C^ in the direction opposite to
hangingwall displacement. In the ramp situation, i.e. until the
base of the brittle part of the hangingwall has reached the We would like to thank Rick Groshong and an anonymous reviewer for their
upper footwall flat, the backlimb becomes steeper while the constructive suggestions and the Swiss National Science Foundation for
hinges B and C become flatter. After having reached the supporting this work (grants 4.904-0.85.20 and 20-25560.88).
References
Bitterli, Th. 1988. Die dreidimensionale Massenbilanz - ein wichtiges 1973. Jura Mountains. In: De Jong, K. A. & Scholten, R. (eds) Gravity
Hilfsmittel zum Verstandnis der regionalen Kinematik (Schuppenzone and Tectonics, 217-227, London (Wiley).
von Reigoldswil, Faltenjura). Eclogae geologicae Helvetiae, 81,415-431. 1986. The eastern Jura; Relations between thin-skinned and basement
Boyer, S. E. & Elliott, D. 1982. Thrust systems. American Association of tectonics, local and regional. Geologische Rundschau, 73, 535-553.
Petroleum Geologists Bulletin, 66, 1196-1230. Malavieille, J. & Ritz, J. F. 1989. Mylonitic deformation of evaporites in
Buxtorf, A. 1907. Zur Tektonik des Kettenjura. Berichte der Versammlung decollements: examples from the Southern Alps, France. Journal of
der Oberrheinischen geologischen Vereinigung, 30/40, 29-38. Structural Geology, 11, 583-590.
Davis, M.D. & Engelder, T. 1985. The role of salt in fold-and-thrust belts. Mitra, S. 1986. Duplex structures and imbricate thrust systems: Geometry,
Tectonophysics, 119, 67-88 structural position, and hydrocarbon potential. American Association of
Geiser, P. A. 1988. The role of kinematics in the construction and analysis Petroleum Geologists Bulletin, 70, 1087-1112
of geological cross-sections in deformed terranes. In: Mitra, G. & Wojtal, 1988. Three-dimensional geometry and kinematic evolution of the Pine
S. (eds) Geometries and mechanisms of thrusting with special reference Mountain thrust system, southern Appalachians. Geological Society of
to the Appalachians. Geological Society of America Special Paper, 222, America Bulletin, 100 , 72-95.
47-76. & Namson, J. 1989. Equal-area balancing. American Journal of Sci-
Groshong, R. H. & Usdansky, S. 1.1988. Kinematic models of plane-roofed ence. 289, 563-599.
duplex styles. In: Mitra, G. & Wojtal, S. (eds) Geometries and mecha- Noack, Th. 1989. Computergestiitzte Modellierung geologischer Strukturen
nisms of thrusting with special reference to the Appalachians. Geo- im ostlichen Jura: Konstruktion balancierter Profile, Gravimetric,
logical Society of America Special Paper, 222, 197-206. Refraktionsseismik. Ph.D. thesis Universitdt Basel (Switzerland)
Gsell, F. J. 1968. Geologic des Falten- und Tafeljuras zwischen Aare und (unpubl.).
Witnau und Betrachtungen zur Tektonik des Ostjuras zwischen unterem Ramsay, J. G. & Huber, M. I. 1983. The techniques of modern structural
Hauenstein im Westen und der Aare im Osten. Ph.D. thesis, Universitdt geology, vol I: Strain analysis. Academic Press. London
Zurich (Switzerland) (unpubl.). & Huber, M. 1.1987. The techniques ofmodern structural geology, vol.
Giirler, B., Hauber, L. & Schwander, M. 1987. Die Geologie derUmgebung 2: folds and fractures. Academic Press. London
von Basel mit Hinweisen iiber die Nutzungsmoglichkeiten derErdwarme. Suppe, J. 1983. Geometry and kinamatics of fault-bend-folding, American
Beitrdge zur geologischen Karte der Schweiz, N.F. 160 Journal of Science, 283, 684-721.
Jamison, W. R. 1987. Geometric analysis of fold development in overthrust Suter, M. 1981. Strukturelles Querprofil durch den nordwestlichen Faltenjura,
teiranes. Journal of Structural Geology, 9 , 207-219. Mt.-Terry-Randuberschiebung-Freiberge.£c/o^af;?eo/og!cae//?/ve;/ae,
Jordan, P. & Nuesch, R. 1989. Defonmation structures in the Muschelkalk 74, 255-275.
Anhydrites of the Schafisheim Well (Jura Overthrust, Northern Switzer- Taboada, A., Ritz, J. F. & Malavieille, J. 1990. Effect of ramp geometry on
land). Eclogae geologicae Helvetiae, 80, 429-454. deformation in a ductile decollement level. Journal of Structural Geol-
Laubscher, H. P. 1961. Die Femschubhypothese der Jurafaltung. Eclogae ogy. 12, 297-302.
geologicae Helvetiae, 54, 221-282 .
Appendix
State of deformation in the ductile layers of a Type 2 decollement:
^1 -t- Y sin (ac)cos(ac) -Y cos^(ac) yi Y
In a reference system normal and parallel to the original layering
(Fig. 3), the deformation of the ductile basal layer in the ramp Y* sin 2 (ttc) 1 Y* sin (ac)cos (Oc) }S> 1
triangle HPB (Fig. 3) can be formulated by the superposition of two
simple shear movements. The first is parallel to the decollement and
Noting that a = 7t/2-9 11 (angle of bisector C )
is characterized by
y=As/Td.
The second is parallel to the hinges C^ and B^ of the respective (l-cos (Be))^ A
deformation step and is characterized by 2 - cos (9c) Y(2-COS (GC))-
sin (Sc)
7*=2-sin(ec)/(l+cos(ec)) = 2-im(%J2).
V sin (9c) Y ^in (8c) + cos (8^)
where 0^=/(g) (equation (2, 5 & 6)). The finite strain as shown in
Fig. 2b resulting from a given g = As/Td is (cf Ramsay & Huber
1983, p. 290):
318
The Venetian Alps thrust belt
C. Doglioni
Abstract: The Venetian part of the Southern Alps (N Italy) is a Neogene south-vergent thrust belt.
The minimum shortening of the chain is 30 km. The thrusts trend N60°-80°E and show an inherited
N10°W-N10°E normal fault pattern of the Mesozoic continental margin. These earlier features
strongly controlled the evolution of the following oblique thrust belt. Structural undulations along
strike of folds and thrusts occur in correspondence to Mesozoic faults, thickness and facies
variations. The thrusts are arranged in an imbricate fan geometry. A frontal triangle zone laterally
ends at transfer faults. Earlier stages of the thrust belt were characterized by frontal triangle zones
which have later been involved and cut by the progression of the internal thrusts.
This paper describes the main structural grain of the Venetian to east: the eastem margin of the Asiago Plateau, the Seren
Alps (Figs 1 & 2), a part of the Southern Alps, south of the (Graben) Valley, the Cismon Valley alignment (clearly seen
Insubric Lineament. It is a SSE-vergent fold and thrust belt on satellite images), the Passo Rolle Line, the S Gregorio
of mainly Neogene age (Dal Piaz 1912; Leonardi 1965; alignment, and to the east the Col delle Tosatte - Fadalto
Castellarin 1979; Laubscher 1985, Massari ef a/. 1986; Roeder alignment (Figs 4 & 5). The Mesozoic alignments probably
1989; Doglioni 1987) probably produced by the dextral used inherited Variscan discontinuities as well. Platform and
transpression in the central-eastern Alps (Laubscher 1983). basinal Mesozoic facies do not coincide everywhere with the
The chain (Fig. 3) deformed a pre-existing Mesozoic passive old horst and graben structure, i.e. the drowned Trento
continental margin (Aubouin 1963; Bosellini 1965; 1973; Platform, which acts as a horst with reduced basinal sequences
Bernoulli et al. 1979; Winterer & Bosellini 1981). after the Middle Jurassic until the Late Cretaceous.
The study area has been shortened mainly during Neogene
times (Venzo 1939; Massari etal. 1986; Doglioni 1987) and
not deformed by the Dinaric chain which constituted the Structure of the thrust belt
unfolded foreland during Palaeogene times (Doglioni &
Bosellini 1987). The geometry of the thrust belt is that of an imbricate fan (Fig.
3) with a main envelope angle produced by the thrust slices
close to 7° (critical taper of wedge) according to the model of
Inherited structures Piatt (1988). The main thrusts are in order from the internal
parts to the foreland, the Valsugana Line, the Belluno Line,
The Southern Alps were part of a Mesozoic continental the Moline Line, the Tezze Line and the Bassano Line. The
margin, according to stratigraphic analysis, i.e. facies and thrust belt is not cylindrical in shape and the strain continu-
thicknesses changes (Bernoulli et al. 1979; Winterer & ously changes along strike. In a map view of the area (Figs
Bosellini 1981). The area can be divided into three main 2 & 5), the thrusts show an anastomosing pattern along strike,
structural sectors during the Mesozoic. These are, from west maintaining constant shortening which can conservatively be
to east: the Trento Platform, the Belluno Basin and the Friuli calculated as 30 km (Fig. 3).
Platform. True Mesozoic normal faults (i.e. Liassic) have The structural evolution of the thrust belt shows a general
been documented at the western border of the Trento Platform rejuvenation from the internal thrust to the external ones.
(Castellarin 1972; Doglioni & Bosellini 1987) and at its However the intemal thrust sheets seem to have been reac-
eastem margin (Winterer & Bosellini 1981; Bosellini et al. tivated also in recent times (Sleyko et al. 1987). The crys-
1981; Bosellini & Doglioni 1986; Masetti & Bianchin 1987, talline basement outcrops in the hangingwall to the Valsugana
Doglioni & Neri 1988). thrust and is composed of Variscan metamorphosed
The Mesozoic normal faults trend mainly N10° W-N10°E. greenschist facies rocks intruded by Late Carboniferous
It can be argued that the Trento Platform, the Belluno Basin granitic bodies. Basement depth in the Venetian Plain is
and the Friuli Platform were bounded by crustal normal inferred by magnetic data (Cassano et al. 1986) and by the
faults, mainly N-S trending, acting at different times and with assumption of the general hinterland dipping monocline
different displacements during Jurassic time and during at typical of thrust belts. This is consistent with the southward
least the Early Cretaceous (Fig. 4). The main Mesozoic rising of the basement discovered in the Assunta Well at 4747 m
tectonic features bordering the Belluno Basin are from west where Late Triassic dolomites onlap a Late Ordovician
319
C. DOGLIONI
Alpine thrust belt and was located in the foreland of the Palaeogene WSW- Tertiary sedimenis,
i.e. flysch and molasse
vergent Dinaric thrust belt. Note in the eastern side the overlapping between
Southern Alps and Dinarides. Figure 2. Simplified tectonic map of the Venetian Alps.
granite (Fieri & Groppi 1981). The basement is clearly The frontal part of the thrust belt is characterized by a
involved in the Valsugana Line, but balanced cross-sections triangle zone (Figs 2 & 3) which generates a general south-
would indicate a wider involvement by southern thrusts as ward dipping monocline characteristic of the Venetian foot-
well. Significant thickness variations of the sedimentary hills between Bassano and Vittorio Veneto. The frontal
cover and minor dips of the thrusts cannot be excluded and triangle zone is the most peculiar structure of the foreland and
these could considerably increase the amount of shortening its presence is indicated by: (1) the general absence of an
along the thrust belt (Roeder 1989). The main decollement of important thrust at the base of the mountains (Monte Grappa
the thrust belt appears to be located in the basement (15-20 km - Visentin Anticline); (2) the necessity of a thrust at the base
in depth) beneath the Dolomites as suggested by the construc- of the anticline to resolve the volume problem of the struc-
tion of balanced cross-sections (Doglioni 1987) and by focal tural high; (3) the south-dipping monocline in the frontal part
mechanisms of earthquakes indicating low angle thrust planes of the chain which is typical for the triangle zones (i.e. Bally
(i.e. the Siusi event, Slejko et al. 1987). Triangle zones are et al. 1966; Jones 1982; Boyer & Elliott 1982); (4) the
present along the Valsugana thrust where the basement is presence of north-vergent backthrusts (i.e. in the Possagno
sometimes wedged within the sedimentary cover, or it pro- and Follina areas, Braga 1970; Zanferrari et al. 1982). The
duces a triangle zone in the Valsugana Valley where the interpretation presented in Figure 3 of the triangle zone is
Valsugana thrust faces a north-vergent basement involved only one possibility: i.e. there are not clear indications of the
backthrust, to the north of the Asiago Plateau. Major undu- southward continuity of the decollement necessary to adsorb
lations along the Valsugana thrust occur as a result of inher- the amount of displacement. This could be in turn expressed
ited features, i.e. the sinistral N-10°E striking transpressive through pressure solution cleavage or more probably in
undulation of Borgo Valsugana which occurs at an inherited antiformal stack duplexes within the antlicline core of the
structural high as indicated by the reduced thickness of the triangle zone.
sedimentary cover. The thrust has a ramp trajectory and has
It is interesting to note that a similar triangle zone has been
not assumed a staircase geometry in the sedimentary cover,
reported for the northern part of the Alps in the Bavarian
probably due to the difficulty in generating flexural slip folds.
foreland (Miiller et al. 1988). The triangle zone between
Within the sedimentary cover the thrusts are characterized
Bassano (Schio?) and Vittorio Veneto seems to be connected
by cut-off angles ranging between 5° and 45°. Preferential
with a ramp-flat geometry of the deep-seated blind thrust
decollement layers are the Tertiary Possagno Marls, the Late
which generated a thrust-propagation fold (the Monte
Cretaceous Scaglia Rossa, and other, buried levels within the
Grappa-Visentin Anticline). This was active at least during
Late Permian and Triassic sequences. The thrust planes
Late Miocene times because Tortonian and Messinian
assume steeper angles when a footwall syncline is present.
sediments onlap with a gradually smaller inclination the
Footwall synclines are well developed in Cretaceous pelagic
southern limb ofthe anticline (Massarief a/. 1986). Sequence
thin-bedded rocks (B iancone and Scaglia Rossa) whose folding
boundaries in the southern fold limb are marked by angular
is accomodated by intense flexural slip. Chevron folds are
unconformities with decreasing angles toward the foredeep
particularly common in these two formations and their am-
suggesting the coeval activity of the frontal fold (Monte
plitude and wavelength decrease away from the thrust planes.
Grappa - Visentin Anticline). It is clear that the unconformities
320
THE VENETIAN ALPS THRUST BELT
are angular only along dip where the frontal fold is perpen- Timing of the thrust belt
dicular to the assumed regional maximum Neogene stress (a,:
N20°-30°W) and the fold axis has a 'cylindrical' trend. In the Venetian segment of the Alps the thrusts became active
Where there are structural undulations in the fold axis (i.e. the during Late Oligocene to Quaternary times. Tortonian
sinistral transpressive zones of Valdobbiadene-Comuda and sandstones are thrust by the Valsugana Line (Venzo 1939)
the greater Fadalto alignment) the unconformities are marked and Pliocene shales are folded along the frontal triangle zone.
by angular relationships along both dip and strike. In sum- Moreover Messinian-Pliocene onlap geometries in the south-
mary, structures control the nature of the unconformities. A em border of the chain support a mainly Neogene age of
growth fold, with constant horizontal axis, generated by pure deformation. The extension of the unconformities within the
compression produces angular unconformities only along molasse is a function of the thrust belt structure and reflects
dip, whereas a growth fold generated by transpression pro- areas of stronger uplift. A problem is represented by the style
duces angular unconformities both along dip and strike. and timing of the orogenic evolution: Has the thrust belt
To the north, the Belluno Line may have been a blind thrust grown southward in a regular fashion by continuous creep
generating a triangle zone during earlier stages of the defor- since Late Oligocene time and do the unconformities record
mation, later rising at the surface in the northern limb of the moments of sea level fall (low stand)? Or was the chain
Belluno Syncline. This is supported by the steep attitude of generated step by step so that the unconformities simply mark
the northem limb of the Belluno Syncline which is difficult to moments of tectonic activity? In general, plates move with a
explain geometrically as a simple footwall syncline. It is also regular velocity suggesting that in areas of deformation the
notable that triangle zones mainly occur in the Mesozoic tectonic evolution should follow an almost constant activity.
Belluno Basin (Fig. 5), rather than in the neighbouring If the tectonic evolution generated with a constant regularity
platforms. In fact the Belluno Syncline is developed in the and tectonic activity had a wavelength too short or too long
deepest structural zone with thick basinal lithofacies. This with respect to the eustatic sea level changes, then an interest-
earlier structural situation had an important influence in the ing problem appears in dating the thrust belt: the timing of the
morphology and source areas of the hydrographic pattern tectonic activity has been considered as the time missing at
during the Late Miocene. regional unconformities and the age of coarse-grained sedi-
ment supply (i.e. the Messinian conglomerate) onlapping the
NNW
SSE
Volsugana Line
o Belluno L.
-> <? .,
>--- Tezze L. Bassano L
^MolineL FONZASO
Figure 3. Balanced cross section across the Venetian Alps. See Figure 2 for location. Horizontal scale = vertical scale. Legend: C, crystalline basement;
H, Late Hercynian granite; T, Late Permian-Lower and Middle Triassic formations; P, Late Triassic (Dolomia Principale); G, Liassic platform facies (Calcari
Grigi) gradually southward passing to Liassic-Dogger basinal facies in the Venetian Plain (Soverzene Formation, Igne Formation, Vajont Limestone); R,
Dogger-Malm basinal facies (Lower and Upper Rosso Ammonitico, Fonzaso Formation); B, Early Cretaceous (Biancone); S, Late Cretaceous (Scaglia
Rossa); E, Palaeogene (Possagno Marls, etc.); N, Late Oligocene—Neogene Molasse; Q, Quaternary.
321
C. DOGLIONI
322
THE VENETIAN ALPS THRUST BELT
Asiago pop-up (Barbieri 1987) constitutes the western part of tion of the Friuli Platform. The study area was located in the
the study area, and is a wide plateau formed on the inherited foredeep of the Dinaric thrust belt during Palaeogene times
Trento Platform (horst). The deformation at the western end (Fig. 1) and suffered subsidence due to the load of the WS W-
of the Valsugana Line is transferred in the Trento area vergent Dinaric thrust sheets. A regional ENE-dipping
through the dextral transpressive Calisio Line (Fig. 2) to the monocline developed at that time and was inherited and
sinistral transpressive Giudicarie Belt to the west. This involved in the younger SSE-vergent Neogene Southalpine
undulation similarly runs around a minor inherited Late deformation. The variations along strike of the deformation
Palaeozoic and Mesozoic horst, within the wider Trento are reflected also in the Neogene and Quaternary foreland
Platform. On the basis of thickness and facies changes the basin.
amplitude of the Trento Platform was probably wider in the
east during Jurassic times (Seren Valley) and probably re-
treated (as horst, with basinal facies) by about 10 km during Conclusions
Cretaceous times (eastern margin of the Asiago Plateau,
Valsugana Valley). The Tezze Line develops at this final The good outcrops and the clear interference between inher-
eastern margin of the Trento horst and undulates in an oblique ited features and Alpine tectonics make the Venetian Alps a
and lateral ramp (sinistral transpression) at the intersection classic example of a thrust belt. Earlier Mesozoic features
with the inherited Seren Valley alignment (Fig. 5). The strongly influenced the evolution of the chain. Any kind of
Alpine deformation within the Belluno Basin is more diffuse, structural undulation along strike of the thrust belt is associated
characterized by a major number of thrust planes and reduced with pre-existing synsedimentary faults, thickness and/or
wavelength folds with respect to the lateral platform areas facies variations in the sedimentary cover. The thrusts are
(Fig. 5). The Belluno Line mainly develops to the east of the arranged in an imbricate fan geometry and show a frontal
Trento Platform. It branches the Valsugana thrust and shows triangle zone which was probably present at earlier stages of
an eastward increase in displacement and amplitude of the the thrust belt in more internal zones. The variations along
fault-propagation folding and fault-bend folding in the strike of the deformation are reflected also in the Neogene and
hangingwall. Commonly, the inherited tensional Mesozoic Quaternary molasse. The frontal triangle zone appears to be
areas have been reactivated in transpressive zones (Fig. 4) a growth fold from Late Oligocene to Quaternary time
and are transfer zones between two different styles of defor- because clastic sedimentation on the southern limb of the
mation. For instance the N-vergent backthrust in the anticline shows onlap geometries and reduced thicknesses.
hangingwall of the Belluno Line ends at the western margin According to this progressive evolution of the thrust belt, the
of the Vette Feltrine at the intersection with the inherited unconformities within the molasse could record lowstands of
tensional zone of the Cismon Valley (Fig. 5). The Caneva eustatic cycles.
Line and the Fadalto Line are two respectively dextral and
Thanks to R. Crane and three anonymous referees who revised the paper. I
sinistral transpressive zones at the eastern margin of the
also acknowledge K. R. McClay, A. Bally, G. Barbieri, D. Bernoulli, A.
Belluno Basin (Fig. 5). The Caneva Line represents the Bosellini, J. Channell, G.V. Dal Piaz, H. Laubscher, D. Masetti, F. Massari,
eastern transfer fault of the frontal triangle zone (Fig. 5). The I. Moretti, J. Piatt, E. Semenza and E. Zappaterra for useful comments and
Fadalto transpression was emplaced at the western termina- discussions. The research was supported by an MPI grant.
References
Auboin, J. 1963. Essai sur la paleogeographie post-triasique et revolution Boyer, S. E, & Elliott, D. 1982. Thrust systems. American Association of
secondaire et tertiaire du versant sud des Alpes orientales (Alpes meridi- Petroleum Geologists Bulletin, 66, 1196-1230.
onales; Lombardie et Venetie, Italie; Slovenie occidentale, Yougoslavie). Braga, G. P. 1970. L'assetto tettonico dei dintomi di Possagno (Trevigiano
Bulletin Societe Geologique de France, 7, 5, 5, 730-766. occidentale). Accademia Nazionale Lincei, 48, 451-455.
Bally, A. W., Gordy, P. L. & Stewart, G. A. 1966. Structure, seismic data and Buxtorf, A. 1916. Prognosen und Befunde beim Hauensteinbasis und
orogenic evolution of Southern Canadian Rocky Mountains. Bulletin of Grenchenbergtunnel und die Bedeutung der letzteren fiir die Geologic
Canadian Petroleum Geology, 14, 337-381. des Juragebirges. Verhandlungen der Natwforschenden Gesellschaft in
Barbieri, G. 1987. Lineamenti tettonici degli Altipiani Trentini e Vicentini Basel,21, 184-254.
tra Folgaria e Asiago (Prealpi Venete). Memorie Scienze Geologiche, 39, Cassano, E., Anelli, L., Fichera, R. & Cappelli, V. 1986. Pianura Padana,
257-264. Interpretazione integrata di dati geofisici e geologici. A^ip, int. report, 1-
Bernoulli, D., Caron, C , Homewood, P., Kalin, O. & Von Stuijvenberg, J, 27.
1979. Evolution of continental margins in the Alps. Schweizerische Castellarin, A. 1972. Evoluzione paleotettonica sinsedimentaria del limite
Mineralogische und Petrographische Mitteilungen, 59, 165-170. tra 'piattaforma veneta' e 'bacino lombardo', a nord di Riva del Garda.
Bosellini, A. 1965. Lineamenti strutturali delle Alpi Meridionali durante il Giornale Geologia, 37 (1970), s. 2, 1, 11-212.
Vervao-lnas,.Memorie Museo StoriaNaturale Venezia Tridentina, 15,3, 1979. II problema dei raccorciamenti crostali nel sudalpino. Rendiconti
1-72. Societd Geologica Italiana, 1 (1978), 21-23.
1973. Modello geodinamico e paleotettonico delle Alpi Meridionali Dal Piaz, G. 1912. Studi geotettonici sulle Alpi Orientali (Regione tra il
durante il Giurassico-Cretacico. Sue possibili applicazioni agli Appennini. Brenta ed il Lago di Santa Croce). Memorie Istituto GeologiaR. Universitd
In: Accordi, B. (ed.) Moderne vedute sulla geologia dell'Appennino, diPadova.l, 1-196.
Accademia Nazionale Lincei, 183, 163-205. Doglioni, C. 1987. Tectonics of the Dolomites (Southern Alps, Northern
& Doglioni, C. 1986. Inherited structures in the hangingwall of the Italy). Journal of Structural Geology, 9, 181-193.
Valsugana Overthrust (Southern Alps, Northern Italy). Journal of & Bosellini, A. 1987. Eoalpine and mesoalpine tectonics in the
Structural Geology, 8, 5, 581-583. Southern Alps. Geologische Rundschau, 76, 3, 735-754.
323
Thrust geometries, interferences and rotations in the Northwest
Himalaya
P. J. Treloar' 2, M. P. Coward', A. F. Chambers^, C. N. Izatt^ & K. C. Jackson'
Abstract: In North Pakistan the dominant transport direction throughout Himalayan collision has
been to the S or SSE. Southward propagation of thrusts within the thickened Indian Plate has,
however, been impeded by interference with SW-verging thrusts in Kashmir, at the western end
of the main Himalayan oroclinal chain. As a result of this interference, thrusts within both the
Pakistani and Kashmiri systems have become pinned at their lateral terminations. Lineation and
palaeomagnetic data document substantial rotations of whole thrust sheets, of up to 40° around the
pinned terminations, anticlockwise in Pakistan and clockwise in Kashmir. Although such
rotations are best seen within the Pliocene to Recent structures of the external zones, similar
rotations can be determined within Oligocene structures in the internal zones. The NW Himalayan
syntaxes are crustal scale folds which have grown within the zone of convergence between the two
thrust systems. The main Himalayan thrust system is interpreted as having been pinned within the
Himalayan chain, rather than at its western termination, due to early thickening of the northwestern
Indian Plate having acted as a mechanical impediment to the lateral propagation of the main
Himalayan thrusts.
The tectonic evolution of the Himalayan-Tibetan system of direction is everywhere approximately perpendicular to the
thickened continental crust is the result of the Tertiary colli- strike of the thrusts (Brunei 1986), movement directions
sion between India and Asia. Following breakup of within the main chain, as indicated by stretching lineations
Gondwanaland, India commenced its northward movement and fault plane solution data (Seeber et a/. 1981), are strongly
at about 80 Ma ago. Sea floor palaeomagnetic stripe data divergent along the length of the chain. As a consequence of
show this movement to have been at 15-20 cm/yr. Following the divergence in transport direction, together with the re-
continental collision during the early Tertiary, at about 52 Ma duction in the amount of transport towards the northwest, the
(Patriat & Achache 1984) or 45 Ma (Dewey et al. 1989) Ma, Indian Plate thrust sheets of the main Himalayan chain show
this rate decreased to about 5 cm/yr. As a result of the increasingly strong clockwise rotations towards their north-
cumulative post-coUisional northward drift, the Himalayan- western tips. Palaeomagnetic evidence for such clockwise
Tibet system must have accommodated over 2000 km of rotations have been described by Klootwijk etal. (1985) and
shortening through a combination of subduction of continental Bossart era/. (1990).
India under Asia (Argand 1924; Powell & Conaghan 1973; In North Pakistan the main Himalayan-age thrusts are
Butler & Coward 1989), thickening of the leading edge of ENE-striking with a dominantly southerly or SSE transport
India by thrusting (Coward & Butler 1985; Mattauer 1986), direction (Coward era/. 1988). This is in direct contrast to the
homogeneous thickening of Tibet and regions to the north NW-striking, SW-verging thrusts immediately to the the east
(England & Houseman 1986; Dewey et al. 1988) and some in NW India. Consequently there is a strong convergence of
form of lateral eastward extrusion of Tibet (Molnar & movement directions in the NW Himalaya which is marked,
Tapponier 1985; Tapponier et al. 1986; England & Molnar within the zone of convergence, by the development of the
1990). Here we deal with aspects of the geometry and timing large antiformal folds of the Northwest Himalayan syntaxes.
of the thrusting which accommodated thickening of the
This paper aims to summarize the tectonic evolution of the
northern part of the Indian Plate.
Pakistan Himalaya and to show how lineation and
The India-Asia collision is not a simple orthogonal one. palaeomagnetic data may be used to determine regional scale
Since collision India has rotated anti-clockwise with respect rotations. These data can then be used to constrain how the
toAsia(Klootwijkera/. 1985), with the result that some 1000 two converging thrust directions interfere and generate a
km more shortening has occurred in the eastern Himalaya range of regional structures.
than in the west (Dewey et al. 1989). The structural effects
of this westward decrease in continental subduction are
clearly seen. The main Himalayan chain has an oroclinal Geology of North Pakistan
form (Fig. 1) in which the strike of the major thrusts swings
from E-W in Nepal to NW-SE in Kashmir. As the transport In Pakistan, the Indian and Asian plates are separated by the
325
p. J. TRELOARCTAL.
Figure 2. Geological map of northern Pakistan showing the location of the major thrusts in the Indian Plate and of the Nanga Parbat and Hazara syntaxes,
and the separation of the Indian and Asian plates by the Kohistan island arc. R-Raikot; S-Sassi; PT - Panjal thrust; MBT - Main Boundary thrust.
326
THRUST GEOMETRIES, INTERFERENCES & ROTATIONS, NORTHWEST HIMALAYA
migration of the depocentre and the hiatus in sedimentation The earliest recorded deformations are within the internal
Babusar.
I ° J Banna Group
Karora Group ~ • -
Figure 3. Geological map of the northern part of the Indian Plate, south of the Main Mantle thrust, between the Swal and Kaghan valleys. The inset shows
how the region is divided into a series of large scale thrust nappes. Place names; B = Batagram, J = Jabori. Thrusts and shear zones: AT = Alpurai thrust;
BT = Batal thrust; KF = Khannian fault; MBT = Main Boundary thrust; MMT = Main Mantle thrust; MT ^ Mansehra thrust; PT = Panjal thrust; BSZ and
TSZ = Balakol and Thakot shear zones.
327
! • / . !
-.t'lrr'T'i P. J. TRELOAR ET AL.
zones within which we can identify two distinct periods of syntaxis. Ar-Ar hornblende ages of 39 + 2 Ma (Treloar & Rex
shortening. The first of these was synchronous with the 1990a & b; Lawrence et al. 1990) show that metamorphism
subduction of IndiaunderKohistan (Treloar £•/•(//. 1989b), and and early stages of post-metamorphic cooling through ca.
involved the early decoupling of cover and basement se- 500"C predated 40 Ma. , . , .:
quences associated with intense ductile shearing and The metamorphic pile was subsequently disrupted during
mylonitization. Porphyroblast-fabric relationships, includ- a second deformation event. During this, the decoupled cover
ing spectacularly spiralled garnets, show that metamorphism and basement sequences were re-imbricated within a series of
was synchronous with the ductile deformation {Treloar et al. lithologically distinct thrust nappes {Fig. 3) stacked within a
1989b & c). Within garnets in calcareous schists in the Swat S-verging crustal scale thrust stack (Coward etal. \ 988; Treloar
Valley, ilmenite inclusions in garnet cores and rutile inclu- et al. 1989c & d). Sharp metamorphic breaks mark the
sions in their rims and in the groundmass, show that prograde boundaries between these nappes, each of which are inter-
metamorphism, to a peak at about 625 + 50 C at 10 + 2 kb was nally imbricated by thrusts that placed already cooled but
along a path of increasing pressure. Gibbs' method analysis originally higher grade rocks on top of lower grade ones,
of samples from both the Hazara and Kaghan valley districts generating an overall tectonic inversion of the metamorphic
confirm the overall core to rim pressure increase within the sequence {Fig. 4). Mica K-Ar ages (Treloar & Rex 1990a&
Swat Valley. However, a reverse Ca zonation in near rim b) from shear zones that stack these nappes show that thrust
regions of garnet profiles is consistent with a late stage stacking pre-dated 30 Ma, although when it commenced is
pressure release. Evidence for this is also indicated from uncertain.
samples elsewhere in the crystalline zones. In one sample Post-metamorphic cooling histories show a period of
from the Kaghan Valley the mantling of rutile by ilmenite in rapid cooling of about 300°C in < 10 Ma during the Oligocene
a gamet-plagioclase-kyanite bearing rock clearly demonstrates to early Miocene which equates to a phase of rapid unroofing
this. and exhumation through a combination of erosion and
The P-T-t paths for the internal zones are thus in accord extensional normal faulting (Treloar & Rex 1990a). Erosion
with prograde metamorphism during active subduction and is recorded within the molasse sediments of the foreland
thickening of the leading edge of the Indian Plate under basins within which the main period of sedimentation com-
Kohistan. The late stage pressure decrease records early menced at the base of the Miocene. There is evidence for
unroofmg,probably through erosion of the overthrust Kohistan major crustal extension in the upper parts of the Indian Plate
block, synchronous with the uplift of the internal zone se- and the MMT zone itself {Fig. 5). Northward {hinterland)
quences in the hangingwall of a new thrust located to the verging shear zones with movement senses clearly defined by
south. This early erosion is recorded in the Eocene-aged extensional S-C and S-C fabrics, and normal faults disrupt
molasse sequence exposed within the core of the Hazara and telescope the metamorphic pile {Treloar et al. in press, a).
'•^"•
Banna Fm 1 ^ ^ ^ ^ Abbotlabafl Fm Karora Gp (covet)
Besnam Gp.
a & Cv ^ (Basemen gneisses)
fl A i A Kofiislan ! • - + + + *l Manwhra firafiilB ^S —-•-
Figure 4. Section across the Hazara nappe showing the internal imbrication within the nappe and its structural relationships with the underlying Besham
nappe and overlying low grade rocks of ihe Banna nappe. MMT - Main Mantle thrust. The Panjal thrust forms the effective sole thrust for the two internal
zone deformations. Note the upward and northward increase in metaniorphic grade, the result of post-metamorphic stacking and disruption of the
metamorphic pile.
328
THRUST GEOMETRIES, INTERFERENCES & ROTATIONS, NORTHWEST HIMALAYA
North ' '•'* '•''.' •? Ji'J-i;,'-' -"'• Jl.^^i' •':'• Souih The chlorite-bearing greenschist facies rocks of the Banna
BANNA Gp nappe (Treloar et al. 1989b} are separated from the structur-
towgfaOechl - be melasedimenlB
ally underlying sillimanite gneisses of the Hazara nappe by
one such shear zone (Figs 3 Sc 4). The MMT itself, formerly
a S-vergent thrust, became a N-side down extensional fault
with blueschist and greenschist rocks on its hangingwall and
amphibolite grade metasediments on its footwall. Fission
track data (Zeitler 1985) date this extension. In the Swat
valley zircon ages at >45 Ma are some 20 Ma older on the
MMT hangingwall than on the footwall where they are at <25
Ma. By contrast there is no significant difference in apatite
ages across the MMT. On both hangingwall and footwall
these are at about 16-20 Ma, implying that extension contin-
ued until after 23 Ma but ended before ca. 20 Ma.
Considerable tectonic uplift must have accompanied the
exhumation, during the course of which up to 20 km of
material was removed. Although some of this uplift may
have been isostatic in nature, much must have been through
uplift in the hangingwall of a major S-verging thrust. The
obvious candidate is the Panjal Thrust, which already formed
SWAT nappe
the effective sole thrust to the post-metamorphic thrust stack.
kyanile-bearing melasedirneots
Movement along the Panjal thrust must therefore have been
:i . e n '• {dis?)-continuous for over 20 Ma until about 20 Ma ago. This
pioriri-side down sbear bands
continued southward thrusting maintained the critical wedge
dimensions of the mountain front during erosion. This model
is analogous to those proposed for the Main Central thrust
B) West ot Ihe Indus River zone in Nepal and India (Royden & Burchfiel 1987; Searle &
Rex 1989) where high level extension is synchronous with S-
Figure 5. Sections across the Main Mantle thrust zone (a) east of the Indus
verging thrusting. ••
river and (b) west of the river showing the disposition of low grade rocks in
the hanging walls of late north-verging extensional normal faults each of
which have high grade rocks in their hanging walls.
Fig 13;
Km 50
Pre-Slwalik Siwalik and Quarternary
Figure 6. SimplifiedmapofthegeologyoftheforelandbasinsofnorthemPakistan, after Butler c? a/. (1987); and Baker era/. (1988). (B)Bhaun. (J) Eastern
Salt Ranges. MBT - Main Boundary thrust. NPDZ - North Potwar deformaiion zone.
329
p. J. TRELOAR ET AL.
rfv
b) The external zone deformation
Deformation within the external zones (Figs 2 & 6) can be
a)
divided into three phases: pre-, syn- and post-movement
along the MBT. Deformation which predated movement
along the MBT is restricted to the Lesser Himalayan se-
quences on the MBT hangingwall. This deformation is
characterized by the S-verging imbrication of the Cambrian
to Eocene sedimentary succession beneath a blanket of
Miocene molasse. Shortening is estimated to be at least 20
b)
km (Izatt 1990), with the molasse decoupled from the under-
lying shelf sequence by a roof thrust located in the Upper
Eocene sediments (Fig. 7a).
The MBT is normally viewed as the tectonic contact
between the already internally imbricated Lesser Himalayan
sequences on the hangingwall and the Tertiary molasse
sequences on the footwall (Fig. 7b). Although this is prob-
ably the case along the eastern margin of the Hazara syntaxis c)
(Burbank 1983; Burbank et al. 1986), to the west the thrust
cuts down into the molasse and emerges within the molasse
basins themselves. Displacement was at least 120 km, as may
be judged by how much of the thrust is exposed along the
Molasse
western flank of the antiformal Hazara syntaxis.
Cambrian to Eocene
Movement along the MBT was followed by deformation
within the North Potwar deformation zone (NPDZ) - a large Eocambrian salt
duplex structure developed on the MBT footwall. Deforma- Basement
tion within this duplex resulted in the folding of the structur-
ally overlying thrust sequence. Much of the backsteepening Figure 8. Generation of structures south of the Soan syncline. (a) early salt
of thrusts and apparent backthrusting within the Margalla cored pillows, (b) pop-ups developed above the salt pillows (e.g. the Qazian
Hills (Coward et al. 1988) is the result of this folding of fold), (c) late through going thrusting that uplifts the pop-up folds (e.g. the
Domeli thrust) and generated saU cored antiformal structures al the surface.
330
CROSS-SECTION ACROSS THE NORTHERN POTWAR PLATEAU
10 30 Km
i ": KEY
— n
MURREE (MIOCENE) ' ' i , i - T
- ^ o" -*'
'"' -^
H . -
•Si
-•3
:a. L_
-;, -,
MESOZOIC - CENOZOIC !^ j - } ^ I- :• -: •:•• g U ^ -
1 .. ]
LT Z
c?
.:i ^ ::
•,
^L
t ?7 L: 'Q. ^
k^ •-*
r: w
Zi O
s. « -- '^
/ - tf 1
Figure 9. North-south section across the external zones of North Pakistan. Note the folding of the Main Boundary thrust and structures on its hangingwall by the North Potwar deformation zone and the major
detachment within the foreland basin molasse.
p . J. TRELOAR£r/lL.
external zones was steep is demonstrated by the cooling and ductile internal zones. Within the external zones, however, it
uplift histories of the internal zones where rocks, having is clear that shortening rates have varied from >4 cm/yr to < 1
cooled to 100°C, were within 3 to 4 km of the then topo- cm/yr within the last 10 Ma. These variations may reflect
graphic surface by 20 Ma, or before the external zone defor- differences through time in the way that total Himalayan
mation commenced. Consequently the Main Boundary and shortening was partitioned between thickening of the Indian
Salt Range thrusts and associated structures can have generated crust, thickening of the Tibetan crust and the Asiatic crust to
little hinterland uplift and subsequent erosion. The MBT in the north, and the lateral extrusion of Tibet.
particular, with a minimum of 120 km displacement, have All of the internal zone thrusts and shears, the Panjal thrust
been a sub-horizontal detachment. and the MBT have been folded by the development of the
Timing of the deformation events, and the age of indi- antiformal Hazara Syntaxis. The large N-trending antiforms
vidual structures within the foreland basins, is tightly con- and synforms within the Hazara, Besham and Swat districts
strained by magnetostratigraphic and fission track analysis of relate to E-W shortening associated with the development of
the molasse sequences (Johnson et al. 1979, 1982, 1986). the syntaxis. A time chart of events within both the external
Structural relationships between the Chinji Formation and and internal zones of the Pakistan Himalaya is shown in
the MBT suggest that movement along the thrust was later Figure 10.
than 10 Ma (Izatt 1990). That structures related to the NPDZ
fold the MBT, implies not only that movement along that TIME STRUCTURE INTERNAL ZONE SEDIMENTATION
(Ma) METAMORPHISM
thrust predated deformation within the duplex, but that, as the
MBT had been folded, there can have been no further Soan syncline
SRT Base Siwalik ;
movement along it after the duplex had formed. As the first NPDZ
M. Boundary T.
phase movement along the SRT is dated at about 5 Ma,
displacement along the MBT followed by deformation within 100° C
the NPDZ must both have occurred within the time interval
20 200° C
of 10 to 5 Ma. This inferred age for movement along the MBT 300° C Base Murrees ::
as pre-5 and probably pre-8 Ma is earlier than similar move- Internal
Zone
ment along the MBT in Kashmir to the east of the Hazara Thnist
Stack
syntaxis where major pulses of movement are dated at about 500° C '^
5 and 2 Ma (Burbank 1983). The Soan syncline is precisely 40
650° C
dated at between 2.1 and 1.9 Ma. Deformation south of the
Soan syncline post-dates deposition of the late Pliocene Dhok 50 Heating
Pathan Formation, and thus dates the late stage movements on ?? MMT ?? COLLISION
the SRT as younger than 2 Ma. 60
I I
I ? I ?
Amounts of shortening have been estimated for many of
^^
the external zone structures (Table 2). The 70
^ ^ Ophiolite
Obduction
magnetostratigraphic data permit estimates of when these
structures were active. By combining these two estimates,
approximate rates of shortening can be determined for the North
332
THRUST GEOMETRIES, INTERFERENCES & ROTATIONS, NORTHWEST HIMALAYA
Figure 11. (a). Map to show the orientation of stretching lineations within
the internal zones of the Indian Plate MMT - Main Mantle thrust; (b). Map
to show movement trajectories within the internal zones based on the
lineation data in Figure 10a.
333
p . J. TRELOARCT4Z..
1.5 Ma, that is probably coeval with movements on the a short distance to the east, has not rotated. It has, however,
Chambal and Jogi Tilla thrusts (Fig. 13) both of which form undergone some deformation relative to the fault tip, produc-
part of the SRT system. ing oblique to lateral folds.
A palaeomagnetic investigation was undertaken in order Any analysis of regional scale rotations deduced from
to study the complex deformation near the eastern termina- lineation data must take into account the fact that not all the
tion of the Salt Range thrust. To the east, thrusts dip lineations may be of the same age and that some sequences
northeastward and folds and thrusts have northwesterly strikes may carry linear fabrics of more than one age. Within the
typical of the Kashmiri part of the main Himalayan chain. To ductile rocks of the internal zones this problem is particularly
the west, thrusts dip northwards and folds within the fold and acute as it is rarely possible to distinguish between early and
thrust belt have an ENE-trend more typical of the Pakistani late stage linear fabrics. The strains recorded within indi-
thrust belt. In this region, (Figs 6 & 13), the SRT transfers into vidual crystalline nappes relate to deformation prior to 20 Ma
the Jogi Tilla thrust to the north via a backthrust (Butler et al. ago, essentially in the hangingwall of the Panjal thrust.
1987), the Chambal thrust, within which beds show a near Subsequently, deformation has propagated southward and
180° change in strike. structurally downward initially into the imbricates north of
Specimens (236 hand samples) were collected from seven the MBT, thence onto the MBT itself, and then into the
sites in red mudstone and sandstone of the Miocene Chinji foreland basins. From, at the latest, 15 Ma ago the crystalline
Group of the Lower Siwaliks, which contain a primary (syn- nappes have been transported passively towards the S or SSE
depositional) magnetisation as evidenced by the presence of on the hangingwall of the late Miocene to Recent structures.
reversals. Inclinations of the mean magnetic vector do not The present day orientation of the main ductile stretching
vary and are in accord with the Miocene palaeofield. Figure lineations in the crystalline internal zones will, in part at least,
13 compares the unfolded declination of the mean magnetic be due to any passive rotation developed on the hangingwalls
vector from each site with that of the Miocene palaeofield. of the later thrusts.
Within the back-thrust (sites 4 - 6), the large changes in strike It can be assumed, on the basis of structural trends within
are not reflected in the magnetic declinations, showing that no the most northerly exposed part of the Indian Plate, around
rotation other than simple tilting around the present strike has Nanga Parbat, that the original stacking direction within the
occurred. Furthermore, the similarity between the declina- post metamorphic thrust stack was towards the south. A
tions and the palaeofield rules out any significant rotation of deviation in lineation trend from that orientation must reflect
the back thrust as a whole around a steep axis. In contrast, the either localized variations during stacking or significant post-
SRT west of the Chambal thrust has undergone an anticlock- deformation passive rotation. It may be argued that, although
wise rotation with respect to the back thrust and palaeofield, there has been a subsequent passive rotation on the hangingwall
the estimated rotation being 56 + 10° at section lA and IB of the later thrusts, lineation curvatures provide evidence for
(Fig. 13), although this decreases eastward towards site 4.
334
THRUST GEOMETRIES, INTERFERENCES & ROTATIONS, NORTHWEST HIMALAYA
Interference structures in the Northwest The Nanga Parbat syntaxis marks the most northerly exposed
Himalaya part of the Indian Plate internal zones. Here Indian Plate
gneisses outcrop within a structural half window flanked on
The NW Himalaya are thus clearly a zone of convergence either side by the basic rocks of the Kohistan Island Arc. The
between the SW-verging thrusts in Kashmir and the S or SSE- MMT is folded up and around the syntaxis which has the
verging thrusts in Pakistan. The changes in movement overall form of a crustal scale antiform. Within the syntaxis
direction in both systems (clockwise towards the western tips the main gneissic fabrics form a penetrative S-verging, L-S
of the Kashmir-Himalayan thrust system and anticlockwise fabric developed during southward thrusting of Kohistan
towards the eastern tips of the Pakistan thrust system) are over India. These early fabrics have subsequently undergone
predicted by models in which each system is pinned by the an E-W shortening that resulted in the formation of a series of
other with rotation developing around that pinning point and N-plunging folds with amplitude and half wavelengths in
increasing towards it. Structures developed in the convergence excess of 10 km. Along the western margin of the syntaxis
zone are naturally complex and have long been recognized. uplift is accommodated within a major fault zone by a
The Hazara and Nanga Parbat syntaxes (Fig. 2), two large combination of NW-verging thrusting and dextral strike-slip
scale antiformal folds, were recognized by Wadia (1931, faulting with the latter dominant within the northern part of
1932). Here we summarize their essential structural geometries the fault zone (Butler & Prior 1988a, b; Butler et al. 1989;
and show how differences in their evolutionary history can be Madin era/. 1989). The respective roles offolding within the
explained by different sequences of events within the inter- syntaxis and faulting on the western margin in accommodat-
fering thrust system. ing total uplift is discussed by Treloar et al. (in press, b). Ar-
Ar, K-Ar and fission track data (Treloar etal.l 989a; in press,
b; Zeitler 1985; Zeitier etal. 1989) demonstrate that syntaxial
335
p . J. TRELOAK ET AL.
late
syntaxial margins. Growth of the Hazara syntaxis occurred
during the early Pliocene (Burbank 1983). Bossart et al.
dextral tear,
propagates
south crustal scale
t (1988) suggested that the syntaxis grew as a result of an
"cross" fold
anticlockwise rotation of the movement direction from SW-
< > to SSE-directed thrusting. Two supporting lines of evidence
were presented for this. Firstly, quartz fibres showed anti-
clockwise curvatures and, secondly, late stage sinistral strike-
slip faulting along the western margin was attributed to the
late SSE-directed movements. Note that the sinistral ductile
mylonite zone described by Bossart et al. (1988) as being
related to the later stages of their anticlockwise rotation, is an
earlier structure interpreted as one of the ductile shears
involved in the post-metamorphic thrust stacking of the
internal zones (Treloar et al. 1989c & d). Along with other
similar aged shears this mylonite was subsequently passively
folded by growth of the Hazara syntaxis. The palaeomagnetic
documentation of a clockwise rotation of 45° within the core
of the syntaxis (Bossart et al. 1990) is apparently contradic-
tory to the anticlockwise rotation model.
Pinning Point
Figure 16. Model for the formation of the NangaParbat syntaxis. Rotation
of the movement vector towards the thrust tip resulted in NW- verging
thrusting along the western margin. These thrusts have been overprinted by
dextral strike-slip structures that propagated southward along the western
margin and which are related to tear faulting near the thrust tip. As strain a) Early Pliocene
became distributed throughout the hangingwall of the thrust, the rotation of
the shortening direction generated large scale folds within the syntaxis.
The structure and geometry of the northern part of the Hazara Figure 17. ModelfortheformationoftheHazarasyntaxis. (a) Early Pliocene
syntaxis has been described in detail by Bossart et al. (1988, deformation along the Main Boundary thrust (MBT) in the east and the Salt
1990). Essentially the structure is a large, N-plunging Range thrust (SRT) in the west, (note how movement had ceased already on
antiformal fold with curvilinear axes giving it a sheath-type the MBT in the west) together with the initiation of the unexposed SW-
verging Kotli thrust, pinned at its northwest termination; (b) Late Pliocene
geometry. Within the core of the syntaxis, foreland basin
to Recent deformation in the hangingwall of the pinned Kotli thrust generates
molasse sequences of Palaeocene to Miocene age have been clockwise rotation in the footwall of the MBT and folding of the MBT. Note
shortened and domed, and the structurally overlying Main anticlockwise rotation at the eastern tip of the Salt Range thrust and
Boundary and Panjal thrusts have been upfolded around the neotectonic uplift along the Kotli thrust - Indus Kohistan seismic zone.
336
THRUST GEOMETRIES, INTERFERENCES & ROTATIONS, NORTHWEST HIMALAYA
In the light of field observations by the authors, it is structure was the MBT. On the western margin of the
possible to interpret the totality of the data presented by syntaxis, however, the simultaneously propagating struc-
Bossart and his co-workers as reflecting the interference tures were those in the footwall of the MBT, such as the SRT
between two active and converging thrust wedges. The SSE- and related breakback structures.
vergent Pakistan thrust wedge propagated rapidly south-
wards during the Pliocene due to the presence of Eocambrian
c) Structures west of the Hazara syntaxis
evaporites within the Potwar region. The southwestward
advance of the Kashmiri thrust wedge was thus inhibited by West of the Hazara syntaxis, within the region of the Indus
the rapidly thickening Pakistani wedge, particularly in the and Swat valleys, the S-verging thrusts of the post-metamor-
north. This interference caused pinning and clockwise rota- phic thrust stack have been deformed by a series of large scale
tion in the core of the syntaxis as blind thrusts attempted to N-trending antiformal and synformal folds, with half wave-
propagate beneath the Pakistani thrust system (Fig. 17). The lengths of up to 40 km. Among these is the domal Besham
emergent Kotli thrust represents such a structure, the north- antiform, which folds the MMT and within the core of which
west prolongation of which terminates as a large tip fold that basement gneisses of the Besham nappe are exposed. These
uplifts the western limb of the Hazara syntaxis and folds the structures represent a change in the regional shortening and
MBT. Present day SW-verging thrusting along this margin is transport direction late in the tectonic history, from early S-
indicated by seismicity studies (Seeber & Armbruster 1979). directed thrusting to late stage E-W directed folding and
Thus the syntaxial bend of the MBT is not the result of an shortening. The age of this deformation is uncertain, but it
anticlockwise rotation of movement direction, rather it re- does deform the N-directed 20 Ma old extensional structures
sults from the folding, uplift and erosion of the MBT by developed in the upper part of the Indian Plate. Apatite ages
structures propagating simultaneously, but convergently, in from two samples (Zeitler 1985) from the core of the Besham
its footwall. There is no need to invoke significant Pliocene Dome, at about 4 Ma. are younger than ages of 16-20 Ma in
to Recent sinistral strike-slip movements along the western the surrounding areas. These young ages indicate that the
margin of the syntaxis (cf. Bossart et al. 1988), as both the growth of the Besham fold is Recent, possibly within the last
steepness and the N-S strike of the thrust reflect the magni- 5 Ma. As such, it is probable that these structures are related
tude of the folding of the MBT during syntaxial growth. It to the growth of the Nanga Parbat and Hazara syntaxes.
must be noted that the Pliocene thrust systems to the east of Indeed the Besham dome may be a proto-syntaxis itself.
the syntaxis have different geometries from those to the west. An important structure in this regard is the Indus Kohistan
On the eastern margin of the syntaxis, the Pliocene-aged seismic zone (IKSZ), a NE-dipping fault zone imaged
Figure 18. Sketch map showing how a zone of neotectonic uplift is located within the hangingwall of the Indus Kohistan seismic zone (IKSZ). This
topographic high, and its noilhward extension into the Nanga Parbat region marks the western end of the main Himalayan inountain front characterized by
elevations of greater than 4 km.
337
p . J. TRELOARCT/lt.
PLEISTOCENE (1.0 - 0 Ma). dextral strike-slip faulting indicate that the earlier S-verging
Frontal Himalayan thrusts thrust system has been completely overprinted by the SW-
Uplift of Muree-Pattan zone in IKSZ hangingwall verging one. Towards the southwest, in the Hazara syntaxis,
Explosive amplification of Nanga Parbat crustal scale
there was Pliocene-aged interference with folding of the
fold
PLIO-PLEISTOCENE (2.0-0.5 Ma).
MBT and clockwise rotations developed on the MBT footwall.
Salt Range thrusts (>25 km displacement) To the west of the Hazara syntaxis, in the Besham region, the
30-40° anticlockwise rotations in eastern Salt Ranges early S-verging thrust structures were domed during the
PLIOCENE (ca. 2.0 Ma) Pliocene by E-W shortening developed in response to newly
Folding on Potwar Plateau initiated SW-verging thrusting. To the south, in the Sah
PLIOCENE (5.0 - 2.5 Ma)
Growth of Hazara syntaxis and of Besham antiform
Ranges, anticlockwise rotations are developing within the
Accelerating growth of Nanga Parbat Fold SSE-verging thrust wedge as the thrusts are pinned by the
Early rotations (ca. 35°) on Salt Range thrusts interfering SW-verging structures.
UPPER MIOCENE (9.0 - 5.0 Ma)
Imbrication of Siwaliks in northen Potwar in MBT
footwall d) Timing
Incipient development of Nanga Parbat fold
MIDDLE to UPPER MIOCENE (15.0 - 9.0 Ma) The time intervals over which thrusting, thrust interference
SSE displacement of external thrust sheets (Nathia and rotation, and syntaxial growth within the convergent
Gali and MBT) zone have developed is indicated in Table 3. Within Pakistan
Pinning of thrust sheets (e.g. Panjal sheet in west) the main S-verging thrust system has been operative, since
Clockwise rotation in Kashmir
LATE OLIGOCENE TO EARLY MIOCENE (30.0 - 20.0 Ma)
collision, for over 50 Ma. Interference between the two thrust
Extension in Upper Parts of Indian Plate. systems with pinning of thrusts and associated rotations
Beginning of erosion and of foreland basin molasse around the pinned thrust tips has been ongoing for at least 30
sedimentation. Ma. The earliest evidence for such rotation is in the curved
Anticlockwise rotation on Panjal thrust (=sole thrust) lineation trajectories within some of the internal zone crystal-
LATE EOCENE to EARLY OLIGOCENE (40.0 - 30.0 Ma)
Post-metamorphic stacking in internal zones
line nappes, implying that the imbricating thrusts were inter-
Rotation due to early stages of thrust pinning. fering across the zone of convergence within 20 Ma of
EARLY to MID EOCENE (50.0 - 40.0 Ma) collision. The spectacular rotations of over 40°, and the
Collision and thickening/subduction of Indian Plate growth and explosive amplification of the crustal scale
Main phase metamorphism syntaxial folds are, however, features developed within the
Early erosion and Eocene age molasse sediments.
Pliocene to Recent.
Table 3. Time chan of events on the Indian Plate in the Pakistan
Himalaya.
Early
seismically (Seeber et al. 1981) and located west of the collision
Hazara syntaxis. Focal mechanism solutions show this to be
a SW-verging thrust fault, movement on which may have
been responsible for the destructive 1974 Pattan earthquake
(Jackson & Yielding 1983). Ifthis blind thrust detaches in the
mid to lower crust, it should structurally underlie both the
Hazara syntaxis and the Besham antiform. It may thus
represent either a completely new. Recent structure, or a
Recent reactivation of thrusts to which those folds were
hangingwall antiforms or tip folds. Although the seismic
zone is unexposed. Recent uplift associated with it has
generated a marked topographic high within the Murree and
Abbottabad region along the western margin of the Hazara
syntaxis (Fig. 18)
The southwesterly vergence on the IKSZ implies that it is
part of the SW-verging Kashmir thrust system. As such that
system can clearly be seen to be actively propagating toward
the SW, with the frontal thrusts cutting up through the
Pakistan thrust stack. Were this to continue the whole SSE-
verging Pakistan system should ultimately be overthrust by,
and incorporated within, the dominant SW-verging system.
In the interim a whole series of complex interference struc-
Figure 19. Diagram (modified after Klootwijk Si Bingham 1980, and
tures may be expected to develop, the locus of which should Klootwijk etal. 1985) to show the shape of pre-coUisional India, in particular
migrate towards the southwest. In the northeast, at Nanga the two oblique ramps along the northern margin and the predicted rotation
Parbat, SW directed movements, clockwise rotations and of thrust sheets emplaced onto those ramps.
338
THRUST GEOMETRIES, INTERFERENCES & ROTATIONS, NORTHWEST HIMALAYA
339
p . J. T R E L O A R £7 AL.
Conclusions
340
THRUST GEOMETRIES, INTERFERENCES & ROTATIONS, NORTHWEST HIMALAYA
wise rotations at the eastern terminations of the Pakistani mechanical impediment to the propagation of the slightly
thrust system. These rotations are the converse of the clock- younger thrusts of the main Himalayan system. Although
wise rotations recorded at the western terminations of the early collision and crustal thickening in the northwest
thrusts of the main oroclinal Himalayan thrust system. Both Himalaya probably controlled the pinning of the main oroclinal
sets of rotations are the result of the pinning of each thrust thrusts within the length of the Himalayan chain, strike-slip
system due to interference by the other. The two interfering faulting near the western edge of the Indian Plate margin
thrust systems meet in a major zone of convergence, charac- could have had a significant effect in controlling the exact
terized by the growth of the crustal scale northwest Himalayan location of the convergent zone. Once the zone of conver-
syntaxes. Continuing syntaxial growth, rotation, and the gence was initiated it would rapidly have become an increas-
southwestward propagation of blind thrusts of the Kashmiri ingly important feature due to the effects of oblique conver-
system into the Pakistani system demonstrate that interfer- gence, together with the continued anticlockwise rotation of
ence and convergence are active to the present day. the Indian Plate, with greater shortening in the east than the
Although the oroclinal shape of the main Himalayan thrust west.
system can be predicted by experimental modelling, the
pinning of the major Himalayan thrusts within the Himalayan
chain, rather than at its western termination, is probably the The Imperial College Himalayan Research Group has been largely sup-
ported by NERC grant GR3/61I3 to MPC and PJT. In addition we
result of oblique convergence. As a result of this, collision
acknowledge a DENI Research studentship to AFC, a BP Research studentship
and associated crustal thickening, enhanced by the obduction to CNI, and a BP funded postdoctoral fellowship to KCJ. We acknowledge
of Kohistan onto the Indian Plate, was earlier in the northwest discussions with many colleagues, especially R.W.H. Butler, M. P.Williams,
than the east. This thickened crust would have acted as a and R. J. Lillie, and reviews from R. A. Gayer and J. F. Dewey.
References
Argand, E. 1924. La tectonique de I'Asie. ISih International Geoiofiical & Prior, D. J. 1988b. Anatomy of a continental subduction zone.
Congress, 1, 170-372. Geologische Rundschau, 77, 239-255.
Baker, D. M., Lillie, R. J., Yeats, R. S., Johnson, G. D., Yousuf, M. & Zamin, , Prior, D. J. & Knipe, R. J. 1989. Neotectonics of the Nanga Parbat
MA.S.H. 1988. Development of the Himalayan frontal thrust zone: SaU syntaxis. Pakistan, and crustal stacking in the northwest Himalaya.
Range, Pakistan. Geology. 16, 3-7. Earth and Planetary Science Letters, 94, 329-343.
Bale, P. 1986. Tectonique Cadomienne en Bretagne Nord: interaction Coward, M. P. 1986. A section through the Nanga Parbat syntaxis, Indus
decrochement chevauchment: champ,s de deformation et modelisations valley, Kohistan. Geological Bulletin of the University of Peshawar, 18.
experimentales. Unpublished Ph.D. thesis, University of Rennes, 361p. 147-152.
Besse, J. & Courtillot, V. 1988. Palaeogeographic maps of the continents & Butler, R.W.H. 1985. Thrust tectonics and the deep structure of the
bordering the Indian Ocean since the early Jurassic. Journal of Geo- Pakistan Himalaya. Geology, 13, 417-420.
physical Research,9i, 11791-11808, , Butler, R.W.H., Chambers. A. F., Graham, R. H., Izatt, C. N., Khan,
Bossart. P., Dietrich, D„ Greco, A., Ottiger, R. & Ramsay, J. G. 1988. The M. A., Knipe, R. J.. Prior, D. J., Treloar, P. J. & Williams, M. P. 1988.
tectonic structure of the Hazara-Kashmir Syntaxis, southern Himalayas, Folding and imbrication of the Indian crust during Himalayan collision.
Pakistan. Tectonics, 1, 273-297. PhilosophicalTransactions of the Royal Society of London,Series A,i26,
& Ottiger, R. 1989. Rocks of the Muiree Formation in northern 377-391.
Pakistan: indicators of a descending foreland basin of late Paleocene to & Kim, J. H. 1981. Strain within thrust sheets. In: McClay. K. R. &
middle Eocene age. Eclogae Geologicae Helvetiae, 82, 133-16.^. Price. N. J. (eds) Thrust and Nappe Tectonics. Geological Society of
Bossart. P., Ottiger, R. & Heller. F. 1990. Rock magnetic properties and London Special Publication 9, 275-292.
structural developments in the core of the Hazara-Kashmir Syntaxis, NE Dewey, J. F., Cande, S. & Pitman, W. C. 1989. Tectonic evolution of the
Pakistan. Tectonics,'), 103-121. Indiii/Eurasiacollision zone. Eclogae Geologicae Helvetiae,%2,717-734.
Brunei, M. 1986. Ductile thrusting in the Himalayas: ductile thrusting and , Shackleton, R. M., Chang Chengfa & Sun Yujin. 1988. The tectonic
shear sense criteria. Tectonics, 5, 247-265. evolution of the Tibetan plateau. Philosophical Transactions of the
Burbank, D. W. 1983. The chronology of intermontane basin development Royal Society of London. Series A, 327, 379-413.
in the northwestern Himalaya and the evolution of the Northwest , Hempton, M. R., Kidd, W.S.F., Saroglu, F. & Sengor, A.M.C. 1986.
syntaxis. Earth and Planetary Science Letters, 64, 77-92. Shoiteningofcontinentallitho.sphere: the neotectonics of Eastern Anatolia
& Beck, R. A. 1989. Early Pliocene uplift of the Salt Range; temporal - a young collision zone. In: Coward, M. P. & Ries, A. C. (eds) Collision
constraints on thrust wedge development, northwest Himalaya, Paki- Tectonics. Geological Society of London Special Publication, 19, 3-36.
stan. Geological Society of America Special Paper, 232, 113-128. England, P. C. & Houseman, G. A. 1986. Finite strain calculations of
, Raynolds, R.G.H. & Johnson, G. D. 1986. Late Cenozoic tectonics and continentaldeformation.il. Application to the India-Asia plate collision.
sedimentation in the northwestern Himalayan foredeep; II, eastern limb Journal of Geophysical Research. 91, 3664-3676.
of the northwest syntaxis and regional synthesis. In: Allen, P. & & Molnar. P. 1990. Right lateral shear and rotation as the explanation
Homewood. P. (eds) Foreland Basins. International Association of for strike slip fauhing in eastern Tibet. Nature. 344, 140-142.
Sedimentologists Special Publication, 8, Blackwells, 293-306. Fischer, M. W. & Coward, M. P. 1982. Strain within thrust sheets: the Heilam
Butler, R.W.H. & Coward, M. P. 1989. Crustal scale thrusting and continen- sheet of NW Scotland. Tectonophysics, 88, 291-312.
tal subduction during Himalayan collision tectonics of the NW Indian Izatt, C. N. 1990. Variations in thrust front geometry across the Potwar
Plate. In: Sengor, A.M.C. (ed) Tectonic evolution of the Telhxan region Plateau and Hazara/Kalachitta hill ranges, northern Pakistan. Unpub-
NATO ASl Series. C, 259, 387-413. lished Ph.D. thesis, University of London, 353p.
, Coward, M. P., Harwood, G. M. & Knipe, R. J. 1987. Salt: its control Jackson, J. & Yielding, G. 1983. The seismicity of Kohistan, Pakistan:
on thrust geometry, structural style and gravitational collapse along the source studies of the Haniran (1972.9.3), Darel (1981.9.12) and Patan
Himalayan mountain front in the Salt Range of northern Pakistan. In: 3. (1974.12.28) earthquakes. Tectonophysics. 91. 15-28.
J. O'Brien & I. Lerche (eds) Dynamical geology of .salt and related Johnson. G. D.. Johnson, N. M., Opdyke.N. D. & Tahirkelli, R.A.K. 1979.
structures. Academic Press, 399-418. Magnetic reversal stratigraphy and sedimentary tectonics of the Upper
& Prior, D. J. 1988a. Tectonic controls on the uplift of Nanga Parbat, SiwalikGroup, eastern Salt Ranges and southwestern Kashmir./«: Farah,
Pakistan Himalayas. Nature, 333, 247-250. A. & De Jong, K. A. (eds) Geodynamics of Pakistan. Geological Survey
ofPakistan.Quetta, 149-165. '
341
p . J. TRELOAR ET AL.
, Raynolds, R.G.H. and Burbank, D. W. 1986. Late Cenozoic tectonics Searle, M. P. & Rex, A. J. 1989. Thermal model for the Zanskar Himalaya.
and sedimentation in the northwestern Himalayan foredeep; I, Thrust Journal of Metamorphic Geology, 1, 127-134.
ramping and associated deformation in the Potwar region. In: Allen, P. Seeber, L. & Armbruster, J.G.I 979. Seismicity in the Hazara arc in northern
& Homewood, P. (eds) Foreland Basins, International Association of Pakistan: decollement versus basement faulting. In: Farah, A. & De Jong,
Sedimentologists Special Publication, 8, Blackwells 273-291. K. A. (eds) Geodynamics of Pakistan, Geological Survey of Pakistan,
, Zeitler, P. K., Naeser, C. W., Johnson, N.M., Summers, D. M.. Frost, Quetta, 131-142."
C, D., Opdyke, N. D. & Tahirkelli, R.A.K. 1982. The occurence and , Armbruster, J. G. & Quittmeyer, R. C. 1981. Seismicity and continen-
fission track ages of late Neogene and Quartemary volcanic sediments, tal subduction in the Himalayan arc. American Geophysical
Siwalik Group, northern Pakistan. Palaeogeography, PalaeocUmatology Union.Geodynamics Series, 5, 215-242.
and Palaeoecology, il. 67-93. Tapponier, P., Peltzer, G. & Armijo, R. 1986. On the mechanics of the
Klootwijk, C. T. & Bingham, D. K. 1980 The extent of greater Asia. III. collision between India and Asia. In: Coward, M. P. & Ries, A. C. (eds)
Palaeomagnetic data from the Tibetan sedimentary series. Hakkhola Collision Tectonics. Geological Society of London Special Publication,
region, Nepal Himalaya. Earth and Planetary Science Letters, 51, 381- 19,115-157.
405. Treloar, P. J., Rex, D. C , Guise, P. G., Coward, M. P., Searle, M. P.,
, Conaghan P. J. & Powell, C.McA. 1985. The Himalayan arc; large Petterson, M. G., Windley, B. F., Jan, M. Q. & Luff, I. W. 1989a. K-Ar
scale continental subduction, oroclinal bending and backarc spreading. and Ar-Ar geochronology of the Himalayan collision in NW Pakistan:
Earth and Planetary Science Letters, 75, 167-183 constraints on the timing of collision, deformation, metamorphism and
Lawrence, R. D., Kazmi, A. H. & Snee, L. W. 1990. Geological setting of the uplift. Tectonics, 8, 881-909.
emerald deposits of Swat, North Pakistan. In: Kazmi A. H. & Snee, L. W. . Coward, M. P., Williams, M. P. & Khan, M. A. 1989b. Basement cover
(eds) Emeralds of Pakistan: Geology, gemmology, and genesis. Van imbrication south of the Main Mantle Thrust, North Pakistan. Gfo/o^/ra/
Nostrand Reinhold, New York. 1-25. Society of America Special Paper, 232, 137-152.
Lillie, R. J., Johnson, G. D., Yousuf, M., Seamin, A.S.H. & Yeats, R. S. 1987. , Broughton, R. D., Williams, M. P., Coward, M. P. & Windley, B. F.
Structural development within the Himalayan fold-and-thrust belt of 1989c. Deformation, metamorphism and imbrication of the Indian Plate
Pakistan. In: Beaumont, C. & Tankard, A. J. (eds) Sedimentary basins south of the Main Mantle Thrust, North Pakistan. Journal of Metamor-
and basin forming mechanisms', Canadian Society of Petroleum Geolo- phic Geology, 7, 111-125.
gists Memoir, 12, 379-392, , Williams, M. P. & Coward, M. P. I989d. Metamorphism and crustal
Madin, I. P., Lawrence, R. D. & Ur-Rehman, S. 1989. The northwestern stacking in the north Indian Plate, North Pakistan. Tectonophysics, 165,
Nanga Parbat - Haramosh massif: evidence for crustal uplift at the 167-184.
northwestern comer of the Indian craton. Geological Society of America, & Rex, D. C. 1990a Cooling and uplift histories of the crystalline thrust
Special Paper, 2i2, 169-182. stack of the Indian Plate internal zones west of Nanga Parbat, Pakistan
Mattauer, M. 1986. Intracontinental subduction, crust-mantle decollement Himalaya. Tectonophysics, 180, 323-349.
and crustal-stacking wedge in the Himalayas and other collision belts. In: & Rex, D. C , 1990b. Post-metamorphic cooling history of the Indian
Coward, M. P. & Ries, A. C. (eds) Collision Tectonics, Geological So- Plate crystalline thrust stack, Pakistan Himalaya. Journal of the Geo-
ciety of London Special Publication, 19, 37-50. logical Society of London, 147, 735-738.
Molnar P. & Tapponier, P. 1985. Cenozoic tectonics of Asia: effects of a . Potts, G. J. & Wheeler, J. Structural evolution and asymmetric uplift
continental collision. Science, 189, 419-426. of the Nanga Parbat syntaxis, Pakistan Himalaya. Geologische Rundschau,
Opdyke. N. D.. Lindsay, E. H., Johnson, G. D., Johnson, N. M., Tahirkelli, (in press, b).
R.A.K. & Mirza, M.A. 1979. Magnetic polarity stratigraphy and verte- , Rex, D. C. & Williams, M. P. The role of erosion and extension in
brate palaeontology of the Upper Siwalik Subgroup of northern Pakistan. unroofing the Indian Plate thrust stack, Pakistan Himalaya. Geological
Palaeogeography, Palaeoclimatolgy, Palaeocology, 27, 1-34. Magazine, (in press a).
, Johnson, N. M., Johnson, G. D., Lindsay, E. H. & Tahirkelli, R.A.K. Wadia, D. N. 1931. The syntaxis of the northwest Himalaya: its rocks,
1982. Palaeomagnetism of the Middle Siwalik Formations of northern tectonics and orogeny. Records of the Geological Survey of India,65,189-
Pakistan and rotation of the Salt Range decollement. Palaeogeography, 220
Palaeoclimatolgy. Palaeocology, 37, 1-15. 1932. Note on the geology of Nanga Parbat (Mt.Diamir) and adjoining
Patriat, P. & Achache, J. 1984. India-Eurasia collision chronology has portions of Chilas, Gilgit district, Kashmir. Records of the Geological
implications for crustal shortening and driving mechanisms of plates. Survey of India, 66, 212-234.
/Varure, 311, 615-621. Zeitler, P. K. 1985. Cooling history of the NW Himalaya. Tectonics, 4, 127-
Powell, C. M. & Conaghan, P. J. 1973. Plate tectonics and the Himalayas. 151.
Earth and Planetary Science Letters, 20, 1-12. , Chamberlain, C. P. & Jan, M. Q. 1989. Geochronology and tempera-
Pudsey, C. J., Coward, M. P., Luff, I. W., Shackleton, R. M., Windley, B. F. ture history of the Nanga Parbat-Haramosh massif, Pakistan. Geological
& Jan, M. Q. 1986. The collision zone between the Kohistan arc and the Society of America Special Paper, 232, 1 -22.
Asian Plate in N W Pakistan. Transactions of the Royal Society of Edin-
burgh, Earth Sciences, 76, 463-479.
Royden, L. H. & Burchfiel, B. C. 1987. Thin skinned N-S extension within
the convergent Himalayan region: gravitational collapse of a Miocene
topographic front. In: Coward, M. P., Dewey, J. F. & Hancock, P. L. (eds)
Extension Tectonics, Geological Society of London Special Publication,
28,611-619.
342
Balanced and retrodeformed geological cross-section from the frontal
Sulaiman Lobe, Pakistan: Duplex development in thick strata along
the western margin of the Indian Plate
I.A.K. Jadoon'*, R. D. Lawrence^ & R. J. Lillie^
Abstract: A balanced structural cross-section has been constructed integrating seismic reflection
profiles, drillhole, surface geology, and Landsat data across the tectonically active frontal
Sulaiman fold belt in the western Himalayas. Restoration of the section provides information
regarding the chronology of structures, structural style, sequence of thrusting, and the amount of
shortening. General structural form evidenced by gentle topography and a broad fold belt is similar
to that of other mountain belts underlain by a weak detachment. A sequence of about 10 km of
dominantly platform (>7 km) and molasse strata thickens tectonically to about 15 km, 129 km
north of the southwards verging deformation front. Nearly all of the 10 km thick stratigraphic
sequence has been detached at the deformation front. Structural style is that of a foreland-verging
duplex separated from the roof sequence by a passive-roofthrust in thick Cretaceous shale. This
structure is expressed at the surface by fault-related folds. Toward the northerly hinterland,
progressively older rocks are present at the surface in the hinge zones of the anticlines. They have
been uplifted by duplexing several kilometres higher than their regional stratigraphic level. The
passive-roofthrust has not been cross-cut by backthrusts, and it is present over a distance of 60 km
along the line of section. Progressive deformation reveals a series of structural and geometric
features including: (1) broad concentric folding at the fault tip; (2) development of a passive roof
and duplex sequence by forward propagation of floor and roof thrusts; (3) forward propagation of
the duplex as critical taper is achieved; and (4) tear faults and extensional normal faults within the
overthrust wedge. A retrodeformed cross-section shows that about 76 km of orogenic contraction
in the cover sequence has occurred across the frontal 129 km of the Sulaiman fold belt.
The Himalayan mountain system represents an active conti- the frontal part of collision zones (Rich 1934; Dahlstrom
nent-continentcollisionzoneextendingwestwardfromBurma 1969a, 1970; Suppe 1980,1983; Laubscher 1981; Acharyya
through northern India and Nepal into Pakistan (Gansser & Ray 1982; Bachman et al. 1982; Jones 1982; Davis etal.
1981). The broad Sulaiman fold belt is developed by 1983; Davis & Engelder 1985; Banks & Warburton 1986;
transpression as a result of the left-lateral strike-slip motion Boyer 1986; Mitra 1986; Lillie et al. 1987; Jaume & Lillie
along the Chaman fault zone and southward thrusting along 1988; McDougal 1989; Izatt 1990). Studies of active moun-
the western boundary of the Indian subcontinent (Figs 1 & 2; tain belts (i.e. Himalaya and Taiwan) are important because
Sarwar & De Jong 1979; Lawrence e? a/. 1981; Farah e/a/, they provide constraints on collisional processes that are
1984;Quittmeyere?a/. 1984). The frontal part of the Sulaiman unavailable in ancient mountain belts. In this study seismic
fold belt is seismically active (Quittmeyer etal. 1979,1984); reflection and well data, available from the frontal part of the
however, the stratigraphy is not disrupted by any thrust faults active Sulaiman fold belt are integrated with surface geology
that break the surface (Fig. 3). The style of deformation from and Landsat data (Figs 3 & 4) in order to: (1) determine the
the western Sulaiman fold-and-thrust belt is reported to be of thickness and nature of the overthrust wedge; (2) analyse the
forward propagating duplexes developed in a piggy-back structural style and nature of deformation; (3) study progres-
fashion (Banks & Warburton 1986). Southward migration of sive deformation and its effects on the basin geometry; and
the deformation front shed erosion products into the active (4) estimate the amount of compressional shortening in the
Sulaiman foredeep, where 7 km of molasse strata are cur- cover of Phanerozoic sediments,
rently present in the Sibi molasse basin (Banks & Warburton
1986).
Recent studies constrained by seismic reflection and Tectonic setting and stratigraphy
borehole data in the North American Cordillera, Appalachi-
ans, Alps, Himalayas, and Taiwan have provided insight into The Himalayan mountain belt changes trend from northwest-
the mechanism of deformation and geometry of structures in southeast in India to northeast-southwest in Pakistan (Fig. 1).
* Present address: Department of Geosciences, Oregon State University, Corvallis, Oregon, 97331-5506, USA
343
I.A.K. JADOON ET M.
IHoPHIOLITES WELLS
Figure 2. Generalized tectonic map of the Sulaiman fold belt in Pakistan (after Kazmi & Rana 1982). Area of investigation (Fig. 3) is shown by rectangle.
Line AA' is part of this study, while line BB' is constructed by Humayun etal. (1990). Cross-sections C C and DD' are shown in Figure 5 (after Banks &
Warburton 1986). Line BE' shows Bouguer gravity profile (Khurshid pers. comm.).
344
DUPLEX DEVELOPMENT ALONG THE WESTERN MARGIN OF THE INDIAN PLATE
Range (Hunting Survey Corporation 1961). Renewed south- etal. 1983; Johnson ef a/. 1985; Raynolds & Johnson 1985).
ward thrusting since Late Oligocene-Early Miocene con- The main structural elements in the Sulaiman fold belt are
stantly reworked the molasse strata migrating the Indus basin east-west trending arcuate folds and faults which rotate
farther east and south (Banks & Warburton 1986; Waheed et rapidly to a north-south direction along the margin of the
al. 1988). This is similar to the southward migration of the active fold belt (Fig. 2). Imbricate faults are visible at the
active foredeep basins of the Ganges plain in India and the surface only in the north (Hunting Survey Corporation 1961;
Jhelum plain in Pakistan (Acharyya & Ray 1982; Raiverman Kazmi & Rana 1982). They gradually disappear toward the
30' N
29 N
UNDIFFERENTIATED
FAULTS
NORMAL FAULTS
THRUST FAULTS
SYNCLINE
ANTICLINE
40 KM
Figure 3. Generalized geological map of southern Sulaiman lobe. Mapping is compiled from the unpublished maps of the Oil and Gas Development
Corporation (OGDC), the Geological Survey of Pakistan (GSP), the Hunting Survey Corporation (1961), and from Landsat images (1:125,000) supplied
by Earth Satellite Corporation. Available seismic reflection data is shown in Figure 4. Deformed and retrodeformed section (AA') is shown in Figure 9.
345
I.A.K. JADOON ET AL.
30°N
Seismicity from the frontal part of the Sulaiman fold belt
(Quittmeyer et al. 1984), along with multiple unconformities
in the Siwalik molasse (Iqbal & Shah 1980; Banks &
Warburton 1986) illustrate the ongoing deformation that is
taken up by broad folds along the southern Sulaiman front.
346
DUPLEX DEVELOPMENT ALONG THE WESTERN MARGIN OF THE INDIAN PLATE
1 1 1 1 1 1 1
DUPLEX
10 ' • ' • ' • ' • ' • ' • • • • • • • • • I '
;ii:i:i:i:i:i;i:iaxo :i:i:i:i:i:i:i:i;i;i;i;i;i3
5 10 KM
B J I
Figure 5. Structural cross-sections from the western Sulaiman fold belt C C and nonhem Kirthar Ranges DD' (modified from Banks & Warburton 1986).
(a) Passive-roof duplex geometry with a floor thrust at the base of the sedimentary section and a passive-roof thrust in the Ghazij (Eocene) and Goru
(Cretaceous) Formations, (b) Anantiformal stack duplex and the associated foredeep with? km of molas.se sediments. A forward facing monocline is the
surface expression of the duplex. See Figure 2 for location of cross-sections. Sand and clear pattern represents Molasse (Neogene) and Platform strata
(Eocene to Cretaceous) that are a part of the roof-sequence. Brick pattern representing Jurassic to Precambrian rocks is a part of the duplex-sequence.
Random lines show crystalline basement.
thrust faults and duplexes at depth. Seismic reflection pro- of the style of deformation and deposition in the foredeep
files also show that northwards from the Bugti syncline rocks basin, progressive deformation and crustal shortening.
exposed at the surface are structurally elevated by the
overthickened, active wedge. The resultant structural relief
is 4 to 8 km from south to north. Balanced structural cross-section
(5) Overall structural style is of forward-verging duplexes
bounded between a floor thrust near the base of the sedimen- Drillhole data from the frontal part of the Sulaiman fold belt
tary section and a passive-roof thrust in thick Cretaceous and adjacent foredeep (Fig. 4), have been provided to Oregon
shales. Fault related folds are exposed at the surface. Frontal State University by the Hydrocarbon Development Institute
broad and gentle folds (Sui and Loti), wavelength about 20 of Pakistan (HDIP), Texaco and Amoco. A composite
km, may be primarily formed as a result of ductile flow of seismic line (bold lines in Fig. 4) has been constructed to
material in the core of the anticlines at a depth of about 10 km project subsurface data onto a 174 km balanced structural
(seismic line 81-LO-2 in Fig. 8). cross-section (A-A' in Figs 3 & 9). A balanced structural
(6) Total shortening parallel to the direction of tectonic cross-section is one that can be retrodeformed; thus it pro-
transport along the duplex structures and the broad frontal vides an opportunity to evaluate if the solution is geologically
folds is estimated as 76 km. Only a fraction of shortening (< reasonable (Bally c^fif/. 1966;Dahlstrom 1969a, 1970:Gwinn
1 km) is accommodated by the surficial frontal folds (Sui and 1970; Woodward era/. 1989). A line length balancing tech-
Loti), over a distance of about 55 km. nique is applied to the section except at the base of the broad
The details of these structures are discussed below in the frontal folds where the technique is invalid due to the ductile
context of seismic reflection profiles and the structural cross- flow of the rocks in the core of the gentle anticline. The
section A-A' (Figs 7, 8 & 9). This is followed by discussion section was balanced in this basal zone using an area balanc-
347
I.A.K. JADOON ET AL.
ing technique (Dahlstrom 1969a; Woodward et al. 1989). a normal stratigraphic sequences of Cretaceous and Jurassic
rocks. Deeper wells would have demonstrated repeated
stratigraphy or elevated basement if a thin-skinned model is
Surface and seismic expression of the frontal region
incoirect. Molasse strata thin toward the hinterland, with
Most of the seismic lines in Figure 4 include data of 5 seconds sporadic outcrops in the cores of synclines (Fig. 9). This
two way travel time, yet basement could only be seen in implies that the molasse sediments are continually reworked
OGDC line SAJ-22 from the Sulaiman foredeep (Fig. 7). The and redeposited toward the foreland. In effect the molasse
thickness of the wedge in the frontal Sulaiman foredeep is sediments migrate toward the foreland in response to south-
about 8-10 km. Stratigraphic thickness about 60 km south- ward translation of thrust sheets. 745 m of molasse strata
east of the deformation front is about 6 km (Figs 3, 7 & 9). were encountered in the Kandkot-2 and 593 m in the Mari-2
Stratigraphic thickness of 8-10 km at the deformation front of gas fields in the foredeep. Total thickness of the molasse
the Sulaiman fold belt contrasts with the 2-3 km stratigraphic strata from the frontal Sulaiman foredeep is about 2400 m
thickness of the wedge in front of the Salt Range/Potwar (Fig. 9). Thickness of the molasse strata in the Sibi molasse
Plateau (Lillie ef a/. 1987). basin along the western Sulaiman and northern Kirthar Ranges
It is important to locate the basement on the seismic lines is about 7000 m (Banks & Warburton 1986; Fig. 5). In the
from the frontal Sulaiman fold belt and adjacent foredeep in eastern Sulaiman foredeep, about 3500 m of molasse sediments
order to evaluate; (a) the total thickness of the sedimentary are present (Humayone? a/, 1990). At the western and eastern
wedge above the crystalline basement; (b) the basement margins of the Sulaiman Lobe, the surface expression of the
slope, which is important for inferring the mechanism of defomiation front is a foreland-dipping monocline above a
thrusting (Davis etal. 1983; Davis & Engelder 1985); and (c) foreland-verging duplex sequence (Banks & Warburton 1986;
basement structures, their genesis, and effect on thrusting. Humayon etal. 1990).
The top of the basement can be seen only on OGDC line The surface expression of structures from the frontal part
SAJ-22 within the foredeep. Just south of the Sulaiman front, of the Sulaiman fold belt (Sui and Loti) is two broad and very
the basement reaches depths beyond the 5 seconds two-way gentle, doubly-plunging anticlines with wavelengths of about
travel time of the other available seismic lines. However, the 20 km. The wavelengths of intervening synclines is 6-8 km.
basement configuration for the frontal Sulaiman Ranges (Fig.
9) has been interpreted by extrapolating the layercake
stratigraphy into the thrust belt from the frontal regions and
adjascent foredeep using the seismic lines (ZU-10, ZU-7/7E,
LO-14 in Figs 4,7 & 8). Given an average basement slope (6)
of 2.5°, PreCambrian to Quaternary rocks that are about 6 km
thick in the foredeep at Mari (Fig. 9) thickens strati graphically
to about 10 km at the deformation front. It attains a thickness
of about 15 km below the Tadri structure 114 km north of the
deformation front. Planar stratigraphy and gentle and broad
structures (Sui and Loti folds), as far north as the Bugti
syncline are inferred to reflect planar basement. However,
the presence of rift related features of the Tethyan margin is
not precluded, as transitional crust of about 25 km is inferred
by Bouguer gravity modelling at the deformation front (Jadoon
et al. unpubl. data).
Prominent reflections from the sedimentary wedge come
from Eocene and Palaeocene limestones. Cretaceous sand-
stone and limestone, and from the top and base of the massive
Jurassic limestone. At the surface progressively older strata
are exposed towards the hinterland in the cores of doubly
plunging anticlines (Fig. 3). Boreholes through the frontal
folds, including Pirkoh anticline (Figs 2, 4 & 9), penetrate a
normal stratigraphic sequence for about 3000 m and reach
upper Cretaceous rocks. In the Mari gas field in the foredeep,
anormal stratigraphic sequence of about 3300 m from Siwalik
to Cretaceous (lower Goru) has been drilled. This sequence
includes about 2100 m of Cretaceous, 800 m of Eocene and
Palaeocene, and about 500 m of molasse strata (Kamran &
Ranki 1987). All units of the carbonate-dominated sequence Figure 6. Simplified stratigraphic column of frontal Sulaiman fold belt.
Approximate seismic velocities are estimates based on thicknesses from the
thicken to the north on the seismic lines except the Creta- well data, sonic logs, and converting stacking velocities from seismic lines
ceous. North of Pirkoh anticline, two wells drilled by Amoco to interval velocities. Detachment horizons are shown with a duplex
on the Tadri and Jandran structures respectively, penetrated sequence below and a roof sequence above Cretaceous shales.
348
D U P L E X DEVELOPMENT ALONG THE WESTERN MARGIN OF THE INDIAN PLATE
834-SAJ-22 (Migrated) 4 km
- 0
Figure 7. Seismic line about 30 km south of the deformation front showing crystalline basement at a depth of 3-4 seconds on two way travel time (about
6-8 km depth). See Figure 4 for location. Line SAJ-22 is 8-40 Hz, migrated vibroseis source, recorded in 1983 by OGDC and processed by Petty-Ray
Geophysical Company.
Surface expression of the southerly limb of the third folded is reported from the western (Banks & Warburton 1986; Fig.
structure (Pirkoh) is a foreland-dipping monocline with dips 5) and eastern (Humayon et al. 1990) Sulaiman Range, the
between 35° and 70°. The top of this structure has almost Kohat Plateau in the Trans-Indus Salt Range (Ahmed 1989;
horizontal strata over a distance of about 16 km (Figs 8 & 9). McDougal 1989), the northern Potwar Plateau (Lillie et al.
Further north, less open folds of smaller wavelength exist. 1987; Jaswal et al. 1990) and from other foreland fold and
Only Siwalik, Eocene and Palaeocene rocks are exposed thrustbelts(i.e. Canadian Cordillera, Price 1981,1986; Jones
along the line of the section. From south to north, progres- 1982; Appalachians, Boyer & Elliot 1982; Mitra 1986; the
sively older strata occupy the hinge zones ofthe anticlines as Scottish Highlands, Elliot & Johnson 1980; the Pyrenees,
structurally deeper levels are exposed. In addition to folds, Uossacketal. 1984; Williams 1985; the Alps, Boyer & Elliot
normal and tear(?) faults are exposed at the surface. The 1982; Papua New Guinea, Hobson 1986; and the Taiwan
Ridge, Saini Mund, and Pezbogi Nala faults on the Loti thrust belt, Suppe 1980, 1983).
structure have a dip-slip offset of about 20 m and possibly The deformed section is about 129 km long and restores to
displace the axis of the Loti syncline several hundred metres an undeformed length of about 205 km which gives a short-
by a strike-slip component (Fig. 3). ening of about 76 km (Fig. 9). This is very unevenly divided
The seismic expression of structures along the line of between the duplexes (75 km shortening), and broad Sui and
cross-section is of broad concentric anticlines (Sui and Loti) Loti frontal anticlines (<1 km shortening). The central
and more complex structures to the north (Fig. 8). The Sui Sulaiman shortening determined here is similar to the 95 km
and Loti folds maintain their layer parallel thickness and of shortening found by Banks & Warburton (1986) for the
wavelength on the seismic lines, unlike typical concentric equivalent portion of the western Sulaiman Lobe and also to
folds (Dahlstrom 1969b), where anticlines become tighter the 70 km of shortening in the Kohat Plateau south of the
and synclines become broader at depth. It is inferred that the Main Boundary thrust (McDougal 1989).
space in the cores of these anticlines is occupied by the ductile
flow of material as a result of tectonic compression of the
southward-propagating thrust sheet at depths of 10 to 15 km.
Structural style and development
Concentric folding is seen as deep as 5 seconds of two way
Style of deformation
travel time data on the seismic lines across the Sui and Loti
structures (Figs 8 & 9). Basement is expected between 5 and The Sulaiman lobe is an actively deforming fold belt that
6 seconds. This implies that virtually all the Phanerozoic thickens northwestwards over a basement slope of 2.5°.
section is detached from the crystalline basement, with the About 10 km of undeformed platform and molasse strata, as
decollement near the base of the wedge. North of the Bugti measured at the deformation front, are thickened in a thrust
syncline, exposed rocks have a structural relief of about 4 km wedge to about 15 km, 129 km north ofthe deformation front.
in the Pirkoh and about 8 km in the Tadri structures (Figs 8 & This elevation is interpreted in this paper as due to thin-
9). The style of deformation above a detachment in Creta- skinned structural duplication. However, the major thrust
ceous shales is of passive folding in the roof thrust sheet; faults that are responsible for this thickening ofthe wedge do
below, it is a duplex sequence of Jurassic and older rocks. A not crop out at the surface. Balanced and retrodefonned
passive-roof backthrust in the Cretaceous Sembar shale ac- cross-sections based on seismic control show that the style of
commodates forward movement of this duplex sequence deformation is a duplex sequence of massive Jurassic lime-
(Figs 6, 8 & 9). Consequently, there is not a thrust fault at the stone and older rocks probably detached from the crystalline
surface in the tectonically thickened wedge from the frontal basement along a sole thrust, and a roof-sequence of thick
part ofthe Sulaiman fold belt. A similar style of deformation Cretaceous shales overlying these duplexes on a passive
349
I.A.K. JADOON ET AL.
350
• . ' ' I \Jlilr\l:K (.:c, i i j ^ ^ . i , •- '• :<;••••, .••,•;..'',,-•. ••' "••"(; , i i . ' , -• <•' ,1 ..- , • • " : . i^ i . T
-•w- •'•-•>
L'-k 'T 'if.'
APPROXIMATE
• ^''^^^\2'^'l. TADRI ''S'';.'i^3l".r KURDAN DANDA FOR
1 SEC = 1 2 0 0 m
>*^^*J-.*^^- .** .
4—
Q
Z 2-
o
o
Ul SSIC? ^^
en 4 - "^f^fe—. ^
fSIC^ AND
^-' -V^mASSICANDOC.E^sSS
V* WOB! *<'i^:-:'
Figure 8. Composite uninterpreted and interpreted seismic line from Chauki syncline to Tadri. Tertiary shallow marine rocks and Siwaliks are exposed at the surface. Basement is below 5 seconds on 2-way travel time.
The section shows duplex bounded by a passive-roof thrust in Cretaceous shales and a floor thrust probably just above crystalline basement. The tip of the blind thrust extends below the Loti anticline (Fig. 9). Note that
concentric folding is the structural style of the broad Loti anticline; the space in the core of the broad folds (Loti and Sui) maybe filled by ductile flow or fine structures (small scale imbricates and duplex) within a decollement
zone at a depth of more than 10 km. Line 81 -LO-2 is 24-fold, migrated dynamite source, recorded and processed in 1981 by OGDC. Line 816-PRK-3 is 24 fold, migrated dynamite source, recorded by OGDC in December
1980 to January 1981 and processed by Geophysical Service Inc. Azaiba, Oman. Line W16-EUis 10-40Hz, migrated vibroseissource, recorded and processed by Western Geophysical Company of America in 1975. Lines
are tied along strike. See Figure 8 for geological details of seismic data gap between lines W16-EU and 816-PRK-3. Note that the horizontal scale differs between all the lines.
FRONTAL SULAIMAN FOLD BELT
TADRI KURDAN DANDA PIRKOH BUGTI LOT! SUI
76 km TOTAL SHORTENING
^ S ^ EOCENE - PALEOCENE
I I CRETACEOUS
^ 3 JURASSIC
I ^ T R I A S S I C - EOCAMBRIAN
150 km 100 km 50 km
|>V\|PRECAMBRIAN BASEMENT
Figure 9. Actual and restored, NNW-SSE geological cross-section of the frontal Sulaiman foreland fold belt of the active Himalayan Mountain system in central Pakistan, (a) Cross-section based on seismic reflection
profiles, surface geology, borehole, and LANDSAT data, (b) Balanced cross-section involving further interpretation of the lower figure, (c) Retrodeformed cross-section based on AA'. Roof-sequence extends continually
for a distance of about 150 km north of the tip of the duplex and is not cut by a major backthrust. An equal amount of the roof-sequence must be removed primarally by erosion in the Loralai valley in the hinterland. In
the current balanced section (top figure) only the shortening associated with the folds in the roof-sequence is shown. Seismic data has been projected from the bold lines in Figure 4 on to the cross-section AA' in Figure
3. Letters identifying the individual horses in the duplex sequence are from the individual mountains (shown on Fig. 3), formed by the duplex propagation. From south to north these mountains are L, Loti; PK, Pirkoh;
D, Danda Range; K, Kurdan Range; and T, Tadri.
DUPLEX DEVELOPMENT ALONG THE WESTERN MARGIN OF THE INDIAN PLATE
(Jadoon unpubl. data) shows that roof sequence extends over the Eocene Kirthar limestone. When sufficient topographic
a distance of about 150 km northwards from the tip of the slope was developed (4° in this section), the frictional resist-
duplex in the Bugti syncline and no major cross-cutting ance at the base and top of the thrust wedge was overcome and
backthrust has yet been discovered. Finally, base of the roof the roofthrust propagated southward (Fig. 10c).
sequence (thick Cretaceous shale) is exposed at the surface in (4) Within the duplex sequence, wedging of an intraplate
the broad Loralai valley where majority of the roof sequence thrust (between K and D and below the passive-roofthrust)
is removed by erosion. This example suggests that in the produced the hinterland verging Kurdan fault-bend fold (Fig.
initial stages of its evolution, a passive roof sequence may lOd). The fault in the core of the Kurdan anticline remained
extend over several duplex horses without overstep backthrusts passive during 3.75 km of foreland directed displacement of
cross-cutting the roof sequence. This idea is based on the the Kurdan intraplate thrust sheet (K) along the basal detach-
absence of a major backthrust and presence of hinterland and ment. Uplift and translation of the Kurdan structure shifted
foreland verging faults with minor displacement which the deformation front farther south. A modern analogue to
propagate out of synclines in the roof sequence north of this this situation is the Sibi molasse basin in front of the western
line (Jadoon work in progress). Sulaiman/Kirthar ranges (Fig. 5), and the foredeep in front of
the eastern Sulaiman ranges.
(5) Propagation of the sole detachment continued in front
Duplex development, orogenic contraction, and deposition
of the antiformal stack. The length of the horse (K -i- D) being
in the foredeep molasse basin
about 42 km. This detached sequence stepped upsection and
Sequential restoration of the balanced structural cross-section was translated along a flat for a distance of about 25 km (Fig.
from the active Sulaiman fold-and-thrust belt provides an lOd-e). Successively older rocks were exposed in the cores
opportunity to unravel the progressive deformation and pro- of the fault-related anticlines. The top of Cretaceous sand-
vide information on the deposition, uplift, and forward mi- stone in the Tadri structure was raised 8 km from its regional
gration of the foredeep basin. The following is a chronological stratigraphic position. The decollement in all these stages
description of the evolution along the cross-section A-A' in remains at a depth of 12-15 km.
Figure 10. (6) The Pirkoh thrust slice (PK) stepped-up southwards in
(1) Erosion of molasse and platform sediments from front oftheDanda(D) monocline (Fig. lOe-f). Thishorsewas
southward migrating thrust sheets in the Loralai and Kohlu translated for a distance of about 20 km to form the broad,
areas north of the studied section (Fig. 2) developed a molasse fault-bend Pirkoh anticline. The surface expression of the
basin that thinned toward the foreland. Deposition continued duplex was that of a broad monocline. The topographic slope
until depth to the basement became sufficient for the was about 1 ° to the south and the total displacement within the
decollement to propagate southward. This initiated a thrust duplex sequence was about 75 km. The depositional axis of
sheet of massive Jurassic limestone and older strata, bounded the molasse basin migrated farther toward the foreland
between a floor and a roof thrust, that stepped up-section and throughout compression.
slid along thick Cretaceous shale at the base of the roof (7) The present day geometry developed by very gentle
sequence for 16.25 km (Fig. lOa-b). concentric folding (L and S) in front of Pirkoh anticline (Fig.
(2) Surface expression of the duplex became a foreland 1 Og). We suggest that the space in the cores of these anticlines
dipping monocline. Location of the ramp in this case is is occupied by ductile flow of fine-grained sedimentary rocks
arbitrary, positioned only for balancing purposes. The dis- at a depth of more than 10 km and probably involving
placement of 16.25 km (10-b) within this duplex is the substantial pressure solution. Extensional normal faults with
amount of shortening along line A-A'. With uplift, the a component of strike-slip displacement are the dominant
foredeep basin migrated further south, and reworking of the surface structures along the hinge zones of the Pirkoh and
molasse sediments thickened the foredeep wedge. A topo- Loti anticlines. The deformed section of 129 km length
graphic slope of 2.8° was produced on the section and may within the duplex sequence restores to an undeformed length
have provided critical taper. The second thrust sheet (T) of 205 km (Fig. 9). Total displacement is 76 km at the present
stepped upsection below the tip of the first duplex and day. As discussed earlier, an equal amount of roof sequence
flattened along the shale horizon at the base of the Cretaceous must have been removed, primarily by erosion, in the broad
(Fig. lOb-c). Loralai valley. In this paper only the part of the shortening
(3) A displacement of 10 km of thrust slice T produced an that was taken-up by folds in the roof sequence is balanced
antiformal stack, Tadri, and a 6.5 km deep molasse basin (Fig. 9c).
filled with reworked molasse eroded from structures north of
the section. A modem example of this geometry exists in the
northern Kirthar and western Sulaiman ranges, where Jurassic Conclusions
limestone is exposed 9 km above its stratigraphic level with
a foredeep Sibi basin that contains 7 km of molasse strata Surface and subsurface data from the southern Sulaiman
(Banks & Warburton 1986; Fig. 5). Development of the Tadri Lobe have been integrated to look into the structural evolu-
antiformal stack and extreme steepening of the passive- tion of the active Himalayan fold belt along the western
roofthrust impeded backthrust motion. Continuous uplift margin of the Indian subcontinent. The important conclu-
allowed erosion through the deformed molasse strata and into sions are summarized as follows:
353
I.A.K. JADOON ETAL.
^^^^^^^^MM^^^^^^^M^S^M^iMi^MMiMiMMi
m^ mwmmmmmmmmrmrmyrmsmms^Jsmmm-SimSmm
fFr.^-7rT7T-/.-!^y?T77-^I~T7.77I)^/^>.>\/s>7'>7/7y^
a = 2°
0
13 = 2.5°
30 km SHORTENING 5 5
1-10.
15
- , ' i < / ^ v i ^ >•
150 km 50 km
;8 = 2.5
?|-io .
15
w^^^mmmmmmmm^mmmmm^i^m^^m<^-^^mm
Figure 10. Palinspastic restoration of duplex development in the frontal part of the active Sulaiman fold belt along the line of cross-section AA' (Fig. 3).
Area balancing is done below the Triassic over the two frontal broad folds (Fig. 9). The current basement slope (P) 2.5° is considered to remain constant
in the reconstruction. Topographic slope (a) changes at each step to create the suitable taper to overcome the frictional resistance at the base of the wedge
allowing the duplex to propagate towards the foreland. At each stage the deformation front progressively moves towards the foreland and continental molasse
strata are constantly reworked to thicken the foredeep wedge. See text (points 1 - 7) for discussion of A to G .
354
DUPLEX DEVELOPMENT ALONG THE WESTERN MARGIN OF THE INDIAN PLATE
(1) The gentler surface topography (< 1 °) and broad width in the duplex sequence and 76 km in the entire deformed 129
(>250 km) of the Sulaiman fold belt is similar to other km section. The restored section has a length of 205 km.
mountain belts underlain by a weak decollement. (6) Only a fraction of the shortening, < 1 km, is taken up by
(2) The compressive deformation along the frontal part of the broad, frontal Sui and Loti anticlines. These folds,
the Sulaiman fold belt is accomodated by a duplex whose extending over a distance of 55 km in front of the main
floor thrust is above the crystalline basement and roof thrust mountain belt, are concentric in the seismic lines as deep as
is in thick Cretaceous shales. The surface expression of 5 seconds two way travel time on seismic data. The basement
deformation in the duplex is fault-related folds (Pirkoh, is expected between five and six seconds on two way travel
Danda, Kurdan, and Tadri), where exposed rocks at the time data. This suggests that the space in the core of these
surface are structurally uplifted 4-8 km above their regional anticlines is primarily filled by ductile flow of material at
stratigraphic level. depths of about 10 km, as a result of tectonic compression;
(3) The roof sequence is not breached along the cross- implying that tip-line folding preceeds faulting in the southern
section, suggesting that a major passive-roof thrust extends Sulaiman fold belt.
over a 60 km length along the cross-section.
(4) Progressive structural development is as follows; (a) This work in the Sulaiman fold belt is part of a cooperative project between
concentric folding behind the fault tip (Sui and Loti anti- Oregon State University (OSU) and Hydrocarbon Development Institute of
clines); (b) the development of a passive-roof duplex at the Pakistan (HDIP) and is supported by NSF grants INT-86-09914 and EAR-
8816962. Additional support was provided by Amoco, Texaco Overseas
deformation front; (c) forward propagation of the duplex to
Petroleum Company, and Mobil Oil Company. Ishtiaq Jadoon at OSU is
produce a variety of structural geometries. From south to supported by USAID. We would like to thank the Oil and Gas Development
north, these features include a fault-bend fold (Pirkoh), a Corporation (OGDC) and Amoco for releasing seismic and well data for this
leading-edge ramp-overlap anticline (Danda), an intraplate project. Cooperation from Geological Survey of Pakistan (GSP) is also
fold (Kurdan), and an anticlinal stack (Tadri). Molasse acknowledged. We thank to various individuals for logistical help and
stimulating discusssion during the course of this study, particularly Hilal
sediments have been continually reworked and the depositional
Raza, Riaz Ahmed, Manshoor Ali, Dan Davis, Ghazanfer Abbas, Mansoor
axis of the foredeep basin migrated southward due to south- Humayun, Jalil Ahmed, Amjid Cheema, Gary Huftile, Mirza Baig, and
ward migration of the deformation front. Mazhar Qayum. Critical review by J. Warburton and C. Izattt greatly
(5) Planispastic restoration indicates a shortening of 75 km improved the original manuscript.
References
Abbas, G. & Ahmad. 1979. The Muslimbagh Ophiolites. In: Farah, A. & De 1970. Structural geology in the eastern margin of the Canadian Rocky
Jong, K.A. (eds) Geodynamics of Pakistan. Geological Survey of Paki- Mountains. Bulletin of Canadian Petroleum Geology. 18, 332-406.
stan, Quetta, 243-250. Davis, D., Suppe, J. & Dahlen, F. A. 1983. Mechanics of fold-and-thrust
Abdul-Gawad, M. 1971. Wrench movement in the Baluchistan and relation belts and accretionary wedges. Journal of Geophysical Re-
to Himalayan-Indian Ocean Tectonics. Geological Society of America search, 8S,l\53-\\72.
Bulletin,»2, 1235-1250. & Engelder, T. 1985. The role of salt in fold-and-thrust belts.
Acharyya, S. K. & Ray, K. K. 1982. Hydrocarbon possibilities of concealed Tectonophysics, 119, 67-88.
Mesozoic-Paleogene sediments below Himalayan nappes, reappraisal. Elliot, D. & Johnson, M. R. W. 1980. Structural evolution in the northern part
American Association of Petroleum Geologists Bulletin, 66, 57-70. of the Moine thrust belt, NW Scotland. Royal Society of Edinburgh
Ahmed, I. 1989. Sedimentology and structure of the south-west Kohat Transactions,, 71, 69-96.
Plateau. Ph.D. thesis, Department of Earth Sciences, University of Farah, A., Abbas, G., De Jong, K. A. & Lawrence, R. D. 1984. Evolution of
Cambridge (unpubl.). the lithosphere in Pakistan. Tectonophysics, 105, 207- 227.
Allemann, F. 1979. Time of emplacement of the Zhob valley ophiolites and Gansser, A. 1981. The geodynamic history of the Himalaya./«; Gupta, H. K,
Bela ophiolites. In: Farah, A. & De Jong, K. A. (eds) Geodynamics of & Delany, F. (eds) Zagros, Hindu Kush, Himalaya: Geodynamic evolu-
Pakistan. Geological Survey of Pakistan, Quetta, 215-242. tion. American Geophysical Union, Geodynamic Series, 3, 111-121.
Bachmann, G. H. Dohr, G. & Muller, M. 1982. Exploration in a classic thrust Gwinn, V. E. 1970. Kinematic patterns and estimates of lateral shortening,
beh and its foreland: Bavarian Alps Germany. American Association of Valley and Ridge and Great Valley Provinces, central Appalachians,
Petroleum Geologists Bulletin, 66, 2529-2542. south-central Pennsylvania. In: Fisher, G. W. et al. (eds) Studies in
Bally, A. W., Gordy, P. L. & Stewart, G. A. 1966. Structure, seismic data, Appalachian Geology, Central and Southern, Interscience, New York,
and orogenic evolution of southern Canadian Rocky Mountains. B«//e?/n 127-146.
of Canadian Petroleum Geology, 14, 337-381. Harris, L. D. & Milici, R. C. 1977. Characteristics of thin-skinned style of
Banks, C. J. & Warburton, J. 1986. 'Passive-roof duplex geometry in the deformation in the southern Appalachian and potential hydrocarbon
frontal structures of the Kirthar and Sulaiman mountain belt, Pakistan. traps. United States Geological Survey, professional paper, 1018, 40p.
Journal of Structural Geology, 8, 229-237. Hobson, D. M. 1986. A thin skinned model for the Papuan thrust belt and
Boyer, S. E. & Elliot, D. 1982. Thrust Systems. American Association of some implications for hydrocarbon exploration. Australian Petroleum
Petroleum Geologists Bulletin, 66, 1196-1230. Exploration Association Journal, 26, 214-224.
1986. Styles of folding within thrust sheets: Examples from the Hossack, J. R., Deramond, J. & Graham, R. H. 1984. The geological structure
Appalachian and Rocky Mountains of the USA and Canada. Journal of and development of the Pyrenees (abst). Toulouse Conference on
Structural Geology, 8, 325-340. Thrusting and Deformation, Abstracts of Programs, 46-47.
Butler, R. W.H. 1982. The terminology of structures in thrust belts. Journal Humayon, M., Lillie, R. J. & Lawrence, R. D. 1990. Structural interpretation
of Structural Geology, 4, 239-245. of eastern Sulaiman foldbelt and foredeep, Pakistan. Tectonics (in press).
Dahlstrom, C.D.A. 1969a. Balanced cross-sections. Canadian Journal of Hunting Survey Corporation. 1961. Reconnaissance Geology ofpart of West
Earth Science, 6, 743-757. Pakistan, AColumbo PlanCooperative Project.Go\emmenlof Canada,
1969b. The upper detachment in concentric folding. Bulletin of Ca- Geodynamic, Toronto, 550p.
nadian Petroleum Geology. 17, 326-346. Iqbal, M.W.A. & Shah, S.M.I. 1980. Records of the Geological Survey of
Pakistan: A guide to the stratigraphy of Pakistan. 37p.
355
I.A.K. JADOON ET AL.
Izattt, C. N. 1990. Variation in thrust front geometry across the Potwar Mitra, S. 1986. Duplex structures and imbricate thrust systems: Geometry,
Plateau and Hazara/Kalachitta hill ranges, northern Pakistan. Ph.D structural position, and hydrocarbon potential. American Association of
thesis. Imperial College of Science, Technology and Medicine, Univer- Petroleum Geologists Bulletin, 70,1087-1112.
sity of London, 353p (unpubl.). Price, R. A. 1981. The Cordilleran foreland thrust and fold belt in the
Jadoon, I.A.K., Lillie, R. J., Humayon, M., Lawrence, R. D., Ali, S. M. & southern Canadian Rocky Mountains. In: McClay, K. R. & Price, N. J.
Cheema, A. 1989. Mechanism of deformation and the nature of the crust (eds) Thrust and Nappe Tectonics, Geological Society of London Special
underneath the Himalayan foreland fold-and-thrust belts in Pakistan. Publication, 9, 427-448.
EOS. Transaction, American Geophysical Union, 70, 1372-1373. 1986. The southeastern Canadian Cordillera: thrust faulting, tectonic
Jaswal, T., Lillie, R. J. & Lawrence, R. D. 1990. Structure and evolution of wedging, and delamination of the lithosphere. Journal of Structural
the Dumal oil field, Nothem Potwar Deformed Zone, Pakistan./ime/(ca/j Geology, 8, 239-254.
Association of Petroleum Geologists Bulletin (in press). Quittmeyer, R. C , Farah, A. & Jacob, K. H. 1979. The seismicity of Pakistan
Jaume, S. C. & Lillie, R. J. 1988. Mechanics of the Sah Range-Potwar and its relation to surface faults. In: Farah, A. & De Jong, K.
Plateau, Pakistan. A fold and thrust belt underlain by evaporites. Tectonics, A. (eds) Geodynamics ofPakistan, GeologiealSurveyofPakistan, Quetta,
7,57-71. 351-358.
Johnson, N. M., Opdyke, N. D., Johnson, G. D., Lindsay, E. H. & Tahirkheli, , Kaffa, A. A. & Armbruster, J. G. 1984. Focal mechanism and depths
R.A.K. 1982. Magnetic polarity stratigraphy and ages of the Siwalik of earthquakes in Central Pakistan: A tectonic interpretation. Journal of
Group rocks of the Potwar Plateau, Pakistan. Paleogeography, Geophysical Research, 89, 2459-2470.
Paleoclimalology, Paleoecology, 37, 17-42. Raiverman, V., Kunte, S. V. & Mudherjea, A. 1983. Basin geometry,
, Stix, J., Tauxe, L., Cerveny, P. F. & Tahirkheli, R.A.J. 1985. Cenozoic sedimentation, and hydrocarbon prospects in northwestern
Paleoclimatic chronology, fluvial processses, and tectonic implication of Himalaya and Indo-Gangetic plains. Petroleum Asia Journal, 6, 67-92.
the Siwalik deposits near Chinji Village, Pakistan. Journal of Geology, Raynolds, R.G.H. & Johnson, G. D. 1985. Rates of Neogene depositional and
93, 27-40. deformational processes, northwest Himalayan foredeep margin, Paki-
Jones, P. B. 1982. Oil and gas beneath east-dipping thrust faults in the stan. In: Snelling, N. J. (ed.) The chronology of the Geological Record.
Alberta Foothills. In: Rocky Mountain Association of Geologists Geological Society of London Memoir, 10, 297-311.
Guidebook, Denver, Colorado, 61-74. Rich, J. L. 1934. Mechanics of low-angle overthrust faulting as illustrated by
Kamran, M. & Ranki, U. 1987. Pakistan well data. Hydrocarbon Develop- the Cumberland thrust block, Virginia, Kentucky, Tennessee. American
ment Institute of Pakistan, unpublished report. Association of Petroleum Geologists Bulletin, 18, 1584-1596.
Kazmi, A. H. & Rana, R. A. 1982. Tectonic map of Pakistan, 1:2,000,000. Rowlands, D. 1978. The structure and seismicity of a portion of southern
Geological Survey of Pakistan, Quetta. Sulaiman Ranges, Pakistan. Tectonophysics, 51, 41-56.
Laubscher, H. P. 1981. The 3D propagation of decollement in the Jura. In: Sarwar, G., & De Jong, K. A. 1979. Arcs, oroclines, syntaxes: The curvature
McClay, K. R. & Price, N. J. (eds) Thrust and Nappe Tectonics. Geo- of mountain belts in Pakistan. In: Farah, A. & De Jong, K. A. (eds)
logical Society of London Special Publication, 9, 311-318. Geodynamics of Pakistan, Geological Survey of Pakistan, Quetta, 351-
Lawrence, R. D., Khan, S. H., De Jong, K. A., Farah, A. & Yeats, R. S. 1981. 358.
Thrust and strike-slip fault interaction along the Chaman fault zone, Seeber, L., Armbruster, J. G. & Quittmeyer, R. C. 1981. Seismicity and
Pakistan. In: McClay, K. R. & Price, N. J. (eds) Thrust and Nappe continental subduction in the Himalayan Arc. In: Gupta, H. K. & Delany,
Tectonics, Geological Society of London Special Publication, 9, 363-370. F. M. (eds) Zagros, Hindukush, Himalaya, Geodynamic Evolution,
& 1990. Structural reconnaissance of Khojak flysch, Pakistan and American Geophysical Union, Geodynamic Series, 3, 215-242.
Afghanistan (in press). Suppe, J. 1980. Imbricated structure of western foothills belt, south central
Lillie, R. J. & Yousaf, M. 1986. Modem analogs for some midcrustal Taiwan. Petroleum Geology of Taiwan, 17, 1-16.
reflections observed beneath collisional mountain belts. In: 1983. Geometry and kinematics of fault bend folding. American
Barazangi, M. & Brown, L. (eds) Reflection Seismology: The Conti- Journal of Science, 283, 684-721.
nental Crust, American Geophysical Union, Geodynamic Series, 14,55- Waheed, A., Wells, N. A., Ahmad, N. & Tabutt, K. D. 1988. Paleocurrents
65. along an obliquely convergent plate boundary, Sulaiman fold belt, south
, Johnson, G. D., Yousaf, M., Zamin, A. S. H. & Yeats, R. S. west Himalayas, West central Pakistan. Abs. American Association of
1987. Structural development within the Himalayan foreland fold-and- Petroleum Geologists Bulletin, 72, 255-25.
thrust belt of Pakistan, In: Beaumont, C. & Tankand, A. J. (eds) Sedi- Williams, G. D.1985. Thrust Tectonics in the south central Pyrenees.
mentary basins and basin-forming mechanisms. Canadian Society of Journal of Structural Geology, 7, 11-17.
Petroleum Geologists Memoir, 12, 379-392. Woodward, N. B., Boyer, S. E. & Suppe, J. 1989. Balanced geological cross-
, Lawrence, R. D., Humayon, M. & Jadoon, l.A.K. 1989. The Sulaiman sections: An essential technique in geological research and exploration.
thrust lobe of Pakistan; Early stage underthrusting of the Mesozoic rifted American Geophysical Union, Short course in geology, 6, 132p.
margin of the Indian subcontinent. Geological Society of Yeats, R. S., Khan, S. H. & Akhtar. M. 1984. Late Quaternary deformation
America, Abstract with program, 318. of the Salt Range of Pakistan. Geological Society of America Bulletin, 95,
& Davis, D. M. 1990. Structure and mechanics of the foldbelts of 958-966.
Pakistan. Thrust Tectonics 1990, Programme with Abstracts, Royal & Hussain, A. 1987. Timing of structural events in the Himalayan
Holloway and Bedford New College, University of London (4-7 April foothills of northwestern Pakistan. Geological Society of America Bul-
1990), 69. letin, 99, 161-176.-176.
McDougal, J. 1989. Geology and Geophysics of the foreland fold-thrust belt 1989. Folds and blind thrusts in the sub-Himalaya and 1905 Kangra
of northwestern Pakistan. Ph.D. thesis, Oregon State University, 140p earthquake (M=8). EOS Transactions, American Geophysical Union,
(unpubl.). 70, 1368.
356
PART SEVEN
NW American Cordillera
The Monashee decollement of the southern Canadian Cordillera: a
crustal-scale shear zone linking the Rocky Mountain Foreland belt to
lower crust beneath accreted terranes
Richard L. Brown,' Sharon D. Carr,' Bradford J. Johnson,' Vicki J. Coleman,'
Frederick A. Cook^ & John L. Varsek^
Abstract: The Monashee decollement, a crustal-scale shear zone in the hinterland (Omineca belt)
of the southern Canadian Cordillera is interpreted as correlating with the sole thrust of the Rocky
Mountain Foreland belt. LITHOPROBE seismic reflection data indicate that the shear zone is
rooted in the lower crust beneath the Intermontane Superterrane. The decollement is exposed on
the margins of the Monashee complex, an elongate domal culmination within the Omineca belt.
This complex includes Early Proterozoic gneisses of North American basement. The hangingwall
of the decollement consists of deformed and metamorphosed North American continental margin
and oceanic accreted terranes of the Selkirk allochthon. Layered reflections observed in
LITHOPROBE seismic reflection data correlate with sheared and transposed rocks in both the
hangingwall and the footwall of the Monashee decollement; the decollement is imaged as
reflections, cutoffs and correlatable reflection segments. Field relations, kinematics and
geochronology indicate that the shear zone has experienced a complex history that included
protracted easterh' directed shear of hangingwall rocks and imbrication of basement rocks in its
footwall. U-Pb zircon dating of syn- and post-kinematic leucogranites in the decollement zone
demonstrates that the final stages of thrusting occurred in the Late Palaeocene. The cessation of
thrusting and the onset of crustal extension in the southern Omineca belt correspond to the end of
thrusting in the Foreland belt.
Easterly directed thrust faults in the Rocky Mountain Fore- mainder is thought to have been transferred eastward by
land belt merge with a sole thrust above Precambrian base- displacement of cover rocks on a structurally higher detach-
ment rocks of the North American craton (Bally et al. 1966; ment surface called the Monashee decollement.
Price & Mountjoy 1970; Cooke? a/. 1988). These basement This paper briefly reviews the structural history of the
rocks remained rigid and uninvolved in the thin-skinned Monashee decollement and offers an interpretation of its
deformation of the cover rocks as far west as the southern subsurface extent that is consistent with seismic reflection
Rocky Mountain Trench (Fig. 1). Supracrustal shortening in data from the LITHOPROBE southern Canadian Cordillera
the Foreland belt at latitude 52°N is approximately 50%, transect. The purpose is to demonstrate that the Monashee
amounting to an easterly displacement of 200 km relative to decollement is an important part of a thrust system that can be
a fixed reference in the underlying basement (Price & traced in the subsurface as far east as the Rocky Mountain
Mountjoy 1970). Trench and at least as far west as the central Intermontane
Basement involvement in thrust deformation west of the belt. Cortelation of the sole thrust of the Foreland belt with
Rocky Mountain Trench is evident from surface geology in the Monashee decollement is explicit in this interpretation,
the Omineca belt, the metamorphic hinterland of the orogen The implications of this interpretation are examined and it is
(Fig. 1; Simony e? a/. 1980; McDonough & Simony 1988). suggested that North American basement extends much
Brown et al. (1986) and Monger et al. (1986) proposed farther west than previously recognized (Armstrong et al.
models that accommodate the thin-skinned shortening in the 1977; Armstrong 1988).
Foreland belt by development of a duplex in basement rocks
beneath the Omineca belt. Balanced sections have been
drawn in which shortening of the basement beneath the Geological setting
Omineca belt is assumed to have accommodated at least half
of the shortening in the Foreland belt (Brown et al. 1986; The Mesozoic and Tertiary tectonic history of the southern
Joumeay 1986; Murphy 1987; Parrish era/. 1988). There- Canadian Cordillera involved four interrelated processes:
357
R. L. BROWN ET AL.
Figure 1. Tectonic map of southern Omineca Belt, modified from Parrish er a/. (1988). Insets locate the morphogeological belts of the Cordillera and the
Intermontane (IM) and Insular (IS) superterranes. The Omineca Belt is bounded on the east by the southern Rocky Mountain Trench and on the west by
the Okanagan Valley-Eagle River normal fault system (OVF/ERF) and terrane accretion boundary (TAB). Intermontane Superterrane lies west of terrane
accretion boundary. Diagonal stripe pattern indicates the Shuswap complex. Basement rocks (dark grey pattern) are exposed in the Frenchman Cap (FC)
and Thor-Odin (TO) culminations of the Monashee complex and in the Malton complex (M). Other high-grade culminations within the Shuswap complex
are individually named the Kettle-Grand Forks (K), Okanagan (OC), Priest River (PR) and Valhalla (V) complexes. Mesozoic-Palaeocene thrust faults are
the Monashee decollement (MD) and Gwillim Creek shear zones (GCS). Late Mesozoic-Tertiary strike-slip faults are the Fraser-Straight Creek (FSF),
Yalakom (YF), Hozameen (HF) and Pasayten (PF) faults. Eocene normal faults are the Columbia River fault (CF), Granby fault (GF), Greenwood fault
system (GWF), Kettle fault (KF), Newport fault (NF), Okanagan Valley-Eagle River fault system (OVF/ERF), Purcell Trench fault (PTF), Slocan Lake fauU
(SLF) and Valkyr shear zone (VS). Numerals correspond to LITHOPROBE seismic reflection lines shown in Figure 4. Box labelled CA locates Cariboo
Alp (see Figs 3 & 5).
accretion of outboard terranes to the western margin of North large part been unroofed by tectonic removal of low-grade
America, protracted compression and crustal thickening, cover on crustal-scale Eocene extension faults. The
strike-slip faulting and crustal extension (Monger ef a/. 1982; Intermontane belt comprises weakly metamorphosed late
Gabrielse 1985; Brown era/. 1986; Parrish era/. 1988). This Palaeozoic to Early Jurassic oceanic and island arc terranes.
history is reflected in the distribution of the five The Coast belt primarily consists of Cretaceous to early
morphogeological belts that parallel the structural grain of Tertiary plutons with metamorphic pendants, while the Insular
the orogen (Monger et al. 1982; Gabrielse & Yorath 1989; belt consists of Palaeozoic to Jurassic volcanic, plutonic,
Fig. 1, inset). The Foreland belt consists of rocks of the metamorphic and sedimentary rocks.
Proterozoic to Palaeozoic North American pericratonic prism Arc and oceanic terranes were amalgamated to form the
and of the Mesozoic to early Tertiary foreland basin; these Intermontane Superterrane and the Insular Superterrane (Fig.
rocks were imbricated by Mesozoic to early Tertiary thin- 1, inset), which were accreted to the westem North American
skinned thrusts and folds. The Omineca belt is the metamor- margin in the Jurassic and Cretaceous (Monger et al. 1982;
phic and plutonic hinterland to the Foreland belt and has in Price et al. 1985). The accreted terranes lie predominantly
358
THE MONASHEE DECOLLEMENT, SOUTHERN CANADIAN CORDILLERA
outboard of the Omineca belt, although relatively thin slivers The Monashee decollement
are preserved at high structural levels within the Omineca
belt. The Monashee decollement is a crustal-scale easterly vergent
The present crustal structure of the southern Omineca belt and westerly rooted ductile compressional shear zone along
is a result of compressional crustal thickening and subsequent which cover rocks of the Selkirk allochthon are detached
extension. During the compressional phase, a thick compos- from parautochthonous basement. The decollement is ex-
ite sheet of internally deforming cover rocks, the Selkirk posed as a 1-2 km thick zone of intense ductile strain over a
allochthon (Read & Brown 1981), was transported eastward strike length of >200 km on the flanks of Frenchman Cap and
over basement on the Monashee decollement (Brown 1980; Thor-Odin culminations (Fig. 1). Precambrian basement and
Read & Brown 1981). The Selkirk allochthon includes (i) unconformably overlying Proterozoic to Cambrian cover
Proterozoic to late Palaeozoic metasedimentary rocks of the rocks in the footwall (Hoy & Godwin 1988; Parrish &
palaeocontinental margin, (ii) deformed and metamorphosed Scammell 1988 and references therein) are exposed in a
late Palaeozoic to Early Jurassic volcanic and sedimentary tectonic window, the Monashee complex (Read & Brown
rocks distal to North America, and (iii) obducted and com- 1981; Brown &Joumeay 1987). The Monashee decollement
plexly infolded late Palaeozoic to Early Jurassic remnants of is truncated on the eastern margin of the Monashee complex
accreted terranes that collectively form the Intermontane by the east-dipping early Tertiary Columbia River normal
Superterrane. fault (Fig. l,Lane 1984).
During Eocene crustal extension the Selkirk allochthon The Monashee decollement is part of a thrust system that
was dissected by generally north-trending low- to moderate- accommodated thick-skinned crustal shortening; it is the roof
angle plastic-brittle normal fault systems (Columbia River - thrust to an antiformal basement duplex (Brown et al. 1986,
Standfast Creek, Granby, Greenwood, Kettle, Newport, Fig. 2; Monger etal. 1986). Rocks of the Selkirk allochthon
Okanagan Valley-Eagle River, Purcell Trench and Valkyr- are inferred to be duplicated by blind thrusts in the hangingwall
Slocan Lake, Fig. 1). The Shuswap complex, a Cordilleran (Fig. 4). Displacements on these structures, which may be out
metamorphic core complex (Coney 1980; Armstrong 1982), of sequence, fed into the Monashee decollement.
consists of rocks that were buried during compression and
were exhumed by these normal fault systems. These high-
Kinematics
grade rocks are exposed as elongate domal culminations, the
Kettle-Grand Forks, Monashee, Okanagan, Priest River and Independent and consistent kinematic indicators are well
Valhalla complexes, which are collectively called the Shuswap preserved along the length of the decollement and have been
complex (Fig. 1). Beneath the normal fault systems that studied in outcrop, hand sample and thin section (Brown
bound these complexes, upper-amphibolite-facies rocks pre- 1980; Joumeay 1986; Joumeay & Brown 1986; Scammell
serve evidence of Mesozoic to Palaeocene compression, are 1986; Coleman 1987,1989; Bosdachin 1989; Harrap 1990).
overprinted by fabrics and mineral assemblages that were Criteria including C-S fabrics (Berthe et al. 1979), strain-
generated during extensional ductile deformation (Carr et al. insensitive foliation (Means 1981; Simpson & Schmid 1983),
1987; Parrish etal. 1988; Johnson 1988a, 1989b, 1989c) and asymmetrical extensional shears (Piatt & Vissers 1980; White
are characterized by an early Tertiary cooling history etal. 1980; Piatt 1984), rotated porphyroclasts (Passchier &
(Armstrong 1982; R. L. Armstrong, unpubl. data, 1984; Simpson 1986),micafish(Lister&Snoke 1984), asymmetrical
Parrish et al. 1988 and references therein). In contrast, the foliation boudinage and asymmetrical pull-aparts (Hanmer
hangingwall rocks consist of unmetamorphosed sedimentary 1984, 1986) consistently record a northeasterly directed
and volcanic strata and rocks that were metamorphosed in the sense of shear (Fig. 2).
Middle Jurassic during accretion of the Intermontane
Superterrane; these rocks were uplifted to high structural
Timing
levels and cooled before the Late Cretaceous (Read & Wheeler
1976; Parrish & Wheeler 1983; R. L. Armstrong, unpubl. Joumeay (1986) documented a zone of inverted low-pressure
data, 1984; Parrish et al. 1988 and references therein). metamorphic isograds in the footwall of the decollement that
Basement rocks are exposed in the Monashee complex overprint previously quenched high-pressure assemblages.
(Frenchman Cap and Thor-Odin culminations) and in the He interpreted these data as evidence that the Monashee
Malton complex (Fig. 1). These include Early Proterozoic decollement experienced a two-stage history of movement to
gneisses that are similar in age to autochthonous North which he tentatively assigned Middle Jurassic and Creta-
American basement (Duncan 1984; Evenchick et al. 1984; ceous-Palaeocene ages (see Brown & Joumeay 1987). The
Parrish & Ross 1989,1990; Parkinson 1990). The Monashee inverted metamorphism implies that the footwall must have
complex was overthrust by Late Proterozoic strata of the been cooler than the overthrust Selkirk allochthon during the
Windermere Supergroup, which are of North American final stages of movement and that thrusting was followed by
provenance and were deposited outboard (west) of the present rapid uplift and cooling to allow preservation of the inverted
exposure of the Monashee complex. These relationships lend isograds.
support to the interpretation that the basement within the At present, there is no direct geochronological evidence
Monashee complex was part of the North American craton. for mid-Mesozoic motion on the decollement. Carr (Carr
359
f >i:^! tr -i.i ) / / K i R. L. B R O W N ET AL.
Figure 2. (a) A 58 Ma posl-kinematic pegmatiie dyke in Carihoo duplex cross-cuts ihe sheared footwall rocks of the Monashee decollement. See text for
further explanation and Figure 3 for location, (b) Asymmetrical extensional shears and back-rotated asymmetrical boudins within sheared basement gneisses
of the Cariboo duplex indicate an upper-member-to-the-northeast sense of shear. See Figure I for location, (c) Sheared gneisses in footwall of Monashee
decollementon west flank ofPrenchman Cap culmination. Asymmeiricleucosomepods and shearfoliationindicatean upper-member-to-the-northeast sense
of shear. See Figure I for location, (d) Sheared gneisses in the hangingwall of the Monashee decollement on the west tlank of Thor-Odin culmination. C-
S fabrics and asymmetrical extensiona! shears indicate an upper-member-to-the-northeast sense of shear. Light-coloured layer is concordant sheared
pegmatite. See Figure 1 for location. ..' , -
Figure 3. View looking north through Cariboo Alp to Thor-Odin culmination (TO) of Monashee complex, showing imbricated slices of basement (B) and
darker grey cover of the Cariboo duplex. Note undeformed pegmatite dykes (P). ST - sole thrust.
360
LOWER ROCKY
OKANAGAN ARROW VALHALLA MOUNTAIN
LAKE LAKE COMPLEX TRENCH
tj-,..J>;-'-_''
^,^:\''i^,^'- • v^V^'""''^^r^^^' *v-i~-^
O
I
^^^^.b^j^^. >
^m^tM
^^^^'^^^^
•10
s
m
15 55
LINES
6 AND 7 S-N OFFSET
CROSS 100 km 60 km
Figure 4. Top section: LITHOPROBE seismic reflection data across the Omineca Belt (lines 1-9) and part of
the Intermontane Belt (line 10). Data are migrated and coherency-filtered and are shown with no vertical
exaggeration for a velocity of 6.0 km/s. Bottom panel: structural interpretation. Shaded pattern represents
crystalline basement interpreted as being of North American cratonic origin. Position of Moho in lines 1 and
2 is based on refraction data. NBR (near-basement reflector) is the sole thrust to the Foreland Belt, which lies
east of the Rocky Mountain Trench. NBR and Monashee decoUement (MD) are shown with bold line to
emphasise inferred correlation across Slocan I^ake normal fault (SLF). Eocene normal faults are shown as
alternating long and short dashes: BF=Beaven fault, CF=Cherryville fault, CRF=Columbia River fault, OVF
= Okanagan Valley fault, VS = Valkyr shear zone. For location of lines, see Figure 1. Note that arrows
representing movement to theNE point to the right on W-E cross sections and to the left on iheN-S cross section
for line 6.
THE MONASHEE DECOLLEMENT, SOUTHERN CANADIAN CORDILLERA
1989, 1990; Carr & Brown 1990) demonstrated that the the leading edge of blind thrust sheets; displacement on such
Monashee decollement on the southern flank of Thor-Odin thrusts would have fed into the Monashee decollement.
was active in the Late Palaeocene and that motion ceased by Exposed basement rocks beneath the Monashee
58 Ma. Timing constraints are based on U-Pb zircon dating decollement are imbricated and isoclinally folded, as described
of deformed sillimanite-bearing syn-kinematic leucogranites above for the Cariboo Alp locality (Fig. 5; Reesor & Moore
in the Monashee decollement (62 + 0.3 & 59 + 0.3 Ma) and 1971; Duncan 1984; Joumeay 1986). Reflections from
a post-kinematic cross-cutting pegmatite dyke (58 + 0.5 Ma) beneath the decollement in lines 6, 7 & 8 appear to have
(Figs 2 & 3). structural configurations that are similar to such imbrications.
Shortening in the foreland continued into the Palaeocene The crust between the inferred position of the Monashee
(Price & Mountjoy 1970; Donelick & Beaumont 1990). decollement and the reflection MOHO tapers westward from
Therefore, the kinematic and geochronological data from the the east end ofline 7 to the central part of line 10(Fig.4). The
Omineca belt lend strong support to the proposal that the Monashee decollement cannot be traced beyond the centre of
Monashee decollement was kinematically and temporally line 10 with confidence, because interpretation of
linked to the Foreland belt (Brown & Read 1983; Brown et al. subhorizontal reflections in the western part of line 10 is
1986). The following section reviews deep seismic reflection ambiguous. Displacement on Eocene normal faults, which
data that further corroborate this interpretation. penetrate to mid-crustal levels (Okanagan Valley-Eagle River
fault system. Figs 1 & 4), must be balanced somewhere by
stretching in the lower crust (Johnson & Brown 1990), and
LITHOPROBE seismic reflection data therefore these subhorizontal reflections may have been
produced by Eocene fabrics. In any case, if the crust beneath
LITHOPROBE deep seismic reflection data have been re- the Monashee decollement is North American basement,
corded in a series of profiles that extend from the Foreland then the reflection geometry suggests that the attenuated
belt to the Pacific Ocean (Clowes et al. 1987; 1988 transect margin of North America continues westward beneath the
data). The data presented in Figure 4 constitute only that part accreted terranes at least as far as the central part of line 10.
of the transect bearing directly on the nature of the reflection This implies that the Intermontane Superterrane must have
characteristics of the Monashee decollement and its extent to been emplaced as thin crustal slices.
the east and west. These data have been processed into The trace of the Monashee decollement in line 6 extends
stacked sections using standard procedures, which include southward down to approximately 7 s (-21 km depth), where
post-stack migrations and coherency filtering, and are plotted it is projected into line 5 east of Lower Arrow Lake (Fig. 4).
with no vertical exaggeration for a velocity of 6.0 km/s. The decollement is not as readily discernible in lines 5 & 4 as
The Monashee decollement is exposed ~10 km north of it is in line 6. However, it is clear that reflections are nearly
LITHOPROBE lines 6 & 7 on the southern flank of Thor- horizontal where the decollement continues into line 5 from
Odin culmination, where it is a highly strained zone up to 2 line 6, and these reflections arch over to become gently east-
km thick (Duncan 1984; Coleman 1989; Fig. 1). At Cariboo dipping beneath line 4. These reflection patterns are consist-
Alp (Figs 2 & 3), the decollement zone is a duplex with ent with the domal geometry of the Valhalla complex known
imbricated horses of basement and cover gneisses (Coleman from geological constraints, and it can be assumed that the
1989; Fig. 5). There is a strain gradient in the hangingwall decollement continues eastward beneath the Valhalla com-
where gneisses are transposed into parallelism with the plex and that it is truncated in the footwall of the Early Eocene
decollement in a zone approximately 5 km thick. The trace Slocan Lake normal fault at approximately 7 s (~21 km depth)
of the decollement has been projected, using well constrained (Fig. 4).
surface geology, into the intersection of lines 6 & 7 (Figs 1 & The Slocan Lake fault is a moderately east-dipping normal
4) at approximately 2 s (two way travel time), where it is fault with a brittle hangingwall and a mylonitic footwall
correlated with a zone of layered reflections (Fig. 4). Above (Parrish 1984; Carr et al. 1987). The fault has been inter-
and below this zone are domains of reflections with different preted on the basis of both geological and seismic data as a
apparent dip and characteristic spacing and intensity; trun- crustal-scale extensional shear zone (Cook er a/. 1987,1988;
cations of reflections are imaged in places along the bound- Parrish ef a/. 1988). Carr (1986) estimates displacement of
ary. Apparent southward dips on line 6 and westward dips on 10-20 km along the central part of the fault.
line 7 indicate a southwesterly dip for the decollement, It is proposed that the sole thrust of the Foreland belt was
consistent with surface geology. Reflections correlated with the eastern continuation of the Monashee decollement; the
the decollement dip moderately to the southwest across lines sole thrust coincides with the near-basement reflector (NBR,
7 & 8 to about 5 s and continue westward on lines 9 & 10, Fig. 4), which is typically 1-2 km above cratonic basement
where they become subhorizontal at about 25 km depth (~8 (Bally era/. 1966; Cook era/. 1988; Eaton & Cook 1988). At
s) on the central part of line 10 (Fig. 4). This suggests that the the west end of line 2 and at the east end of line 3, the sole
Monashee decollement and related structures are listric into thrust is cut by the Slocan Lake fault at -30 km depth (Cook
the lower crust. Antiformal geometries in the hangingwall of et al. 1988). Figure 4 shows that the amount of separation
the Monashee decollement that occur on lines 6 & 9 are between the inferred cutoffs of the sole thrust and the Monashee
juxtaposed against underlying throughgoing layers (Fig. 4). decollement along the projected trace of the Slocan Lake fault
These may be interpreted as hangingwall anticlines located at is approximately 17 km. The estimates of displacement on
361
R. L. BROWN ET AL.
STRUCTURES
THRUST FAULT
»28 STRETCHING LINEATION
7- FOLIATION So-,
;.-,-! V::-\ vV'-''-i;"'.V,;'^v--. ^t;->:,Vb>cr,'-!J'-,'.MT. SKADE SYNCLIN. ,„ „-, ^ F3 FOLD HINGE, Z=VERGENCE
F3 AXIAL PLANE
AXIAL SURFACE TRACE
-SYNCLINE
I I COVER GNEISSES
BASEMENT GNEISSES
Figure 5. Cariboo duplex; (a) map and (b) cross section illustrate basement slices in footwall of Monashee decollement. See Figure 3 for view of part of
Cariboo Alp and Figure 1 for location.
362
THE MONASHEE DECOLLEMENT, SOUTHERN CANADIAN CORDILLERA
the Slocan Lake fault, together with similar geochronological strong subhorizontal reflections near the base of the crust
and kinematic signatures are compelling evidence for cor- beneath the western Intermontane belt (Fig. 4, line 10) may be
relating the sole thrust of the Foreland belt in the hangingwall a result of superimposed extensional shearing and could be
of the Slocan Lake fault with the Monashee decollement. the locus for balancing extension observed at higher crustal
levels.
This paper has made use of seismic data produced by the Canadian
Conclusions LITHOPROBE project under the direction of Dr. R. Clowes. The interpre-
tation of the data presented in this paper has been carried out independently
The Monashee decollement is a crustal-scale compressional by the authors, but we readily acknowledge the contribution of other
shear zone of great areal extent. It carried middle and upper members of the interpretation team. In particular, we have benefited from
crustal rocks in its hangingwall, including strata of North discussions with Peter Carroll, Murray Joumeay, Ernie Kanasewich, Randy
Parrish, Ray Price and Carl Spencer. The seismic field work was carried out
American cratonic provenance and amalgamated slices of by Sonix Exploration of Calgary, with initial processing by Western Geo-
westerly derived accreted terranes. physical of Calgary. Final processing and data preparation were done at the
The Monashee decollement extends eastward in the sub- University of Calgary LITHOPROBE seismic processing facility.
surface to the Rocky Mountain Trench and is inferred to be
the westward continuation of the sole thrust of the Foreland Geological field work by RLB, SDC, BJJ and VJC has been funded by
Natural Sciences and Engineering Research Council of Canada (NSERC)
belt. It extends westward in the middle crust beneath the grant A2693, a LITHOPROBE supporting geoscience grant and Energy,
Intermontane Superterrane and can be traced into the lower Mines and Resources Canada research agreements to RLB. Also gratefully
crust, where its westward continuation becomes ambiguous. acknowledged are Geological Society of America research grants in 1987
Basement gneisses of probable North American cratonic and 1988 and Geological Survey of Canada (GSC) field support to SDC,
British Columbia geoscience research grant RG88-02 to BJJ and a Texaco
origin occur in the southern Omineca belt and are projected
Canada Resources geological research grant to VJC. SDC, BJJ and VJC
westward in the subsurface beneath the Intermontane belt. would like to thank Randy Parrish and the GSC Geochronology Section for
This implies that accreted rocks of the Intermontane the privilege of working in the geochronology laboratory. FAC and JLV are
Superterrane are thin crustal slices that have been transported supported through NSERC operating grant A2623 to FAC. Thanks to L.
northeasterly onto the attenuated margin of North America. Hardy for help with preparation of the manuscript. The paper is designated
LITHOPROBE contribution number 186.
Middle to lower crust beneath the Monashee decollement
in the Omineca belt does not appear to have been involved in
significant penetrative Tertiary extensional strain. However,
References
Armstrong, R. L. 1982. Cordilleran metamorphic core complexes-from 1990. Late Cretaceous-early Tertiary tectonic evolution of the southem
Arizona to southern Canada. Annual Reviews of Earth and Planetary Omineca belt, Canadian Cordillera. Ph.D. thesis, Carleton University
Science, 10, 129-154. (unpubl.).
1988. Mesozoic and early Cenozoic magmatic evolution of the Cana- & Brown, R. L. 1990. Southem Cordilleran LITHOPROBE transect:
dian Cordillera. Geological Society of America Special Paper, 218, 55- evidence for crustal thickening and collapse of the Omineca belt,
91. Valhalla complex to Okanagan Valley, British Columbia. Geological
, Taubeneck, W. H. & Hales, P. 0.1977. Rb-Srand K-Argeochronometry Association of Canada-Mineralogical Association of Canada. Annual
of Mesozoic granitic rocks and their Sr isotopic composition, Oregon, Meeting. Program with Abstracts, 15, A22.
Washington, and Idaho. Geological Society of America Bulletin, 88,397- , Parrish, R. & Brown, R. L. 1987. Eocene structural development of the
411. Valhalla complex, southeastem British Columbia. Tectonics, 6,175-196.
Bally, A. W., Gordy, P. L. & Stewart, G. A. 1966. Structure, seismic data, and Clowes, R. M., Brandon, M. T., Green, A. G., Yorath, C. J., Sutherland
orogenic evolution of southern Canadian Rockies. Bulletin of Canadian Brown, A., Kanasewich, E. R. & Spencer, C. 1987. LITHOPROBE -
Petroleum Geology, 14, 337-381. southem Vancouver Island: Cenozoic subduction complex imaged by
Berthe, D., Choukroune, P. & Jegouzo, P. 1979. Orthogneiss, mylonite and deep seismic reflections. Canadian Journal ofEarth Sciences. 24,31 -51.
non-coaxial deformation of granites: the example of the South Amorican Coleman, V. J. 1987. A metamorphic and kinematic study of the Monashee
Shear Zone. Journal of Structural Geology, 1,31 -42. decollement, southeastem British Columbia. B.Sc. thesis, Carleton Uni-
Bosdachin, R. 1989. Structure, stratigraphy, and tectonothermal evolution of versity (unpubl.).
the central western flank of the Monashee complex, southeastern British 1989. The Cariboo duplex at the southem boundary of the Monashee
Columbia. M.Sc. thesis, Carleton University (unpubl.). complex, Shuswap terrane, southem B.C. In: Current Research, Part A.
Brown, R. L. 1980. Frenchman Cap dome, Shuswap complex, British Geological Survey of Canada, Paper 89-lA, 89-93.
Columbia: a progress report. In: Current Research. Part A. Geological Coney, P. J. 1980. Cordilleran metamorphic core complexes: an overview.
Survey of Canada, Paper 80-1 A, 47-51. In: Crittenden, M. D., Jr., Coney, P. J. & Davis, G. H. (eds) Cordilleran
& Joumeay, J. M. 1987. Tectonic denudation of the Shuswap meta- Metamorphic Core Complexes. Geological Society of America Memoir,
morphic terrane of southeastern British Columbia. Geo/o^_y, 15,142-146. 153,7-31.
& Read, P. B. 1983. Shuswap terrane of British Columbia: a Mesozoic Cook, F. A., Green, A. G., Simony, P. S., Price, R. A., Parrish, R., Milkereit,
'core complex'. Geology, 11, 164-168. B., Gordy, P. L., Brown, R. L. Coflin, K. C. & Patenaude, C. 1987.
, Joumeay, J. M., Lane, L. S., Murphy, D. C. & Rees, C. J. 1986. LITHOPROBE southem Canadian Cordilleran transect: Rocky Moun-
Obduction, backfolding and piggyback thrusting in the metamorphic tain thrust belt to Valhalla gneiss complex. Geophysical Journal of the
hinterland of the southeastern Canadian Cordillera. Journal ofStructural Royal Astronomical Society, 89, 91 -98.
Geology. 8, 255-268. 1988. LITHOPROBE seismic reflection stmcture of the southeastem
Carr, S. D. 1986. The Valkyr shear zone and the Slocan Lake fault zone: Canadian Cordillera: initial results. Tectonics. 7, 157-180.
Eocene structures that bound the Valhalla gneiss complex, southeastern Donelick, R. A. & Beaumont, C. 1990. Late Cretaceous-early Tertiary
British Columbia. M.Se. thesis, Carleton University (unpubl.). cooling of the Brazeau thrust sheet. Central Foothills, Alberta: implica-
1989. Implications of Early Eocene Ladybird granite in the Thor-Odin- tions for the timing of the end of Laramide thrusting in the region.
Pinnacles area, southem British Columbia. In: Current Research,Part E. Geological Association ofCanada-Mineralogical Association ofCanada,
Geological Survey of Canada, Paper 89-lE, 69-77. Annual Meeting, Program with Abstracts, 15, A34.
363
R. L. BROWN ET AL.
Duncan, I. J. 1984. Structural evolution of the Thor-Odin gneiss dome. Murphy, D. C. 1987. Suprastructure/infrastructure transition, east-central
Tectonophysics, 101, 87-130. Cariboo Mountains, British Columbia: geometry, kinematics and tec-
Eaton, D. W. S. & Cook, F. A. 1988. LITHOPROBE seismic reflection tonic implications. Journal of Structural Geology, 9, 13-30.
imaging of Rocky Mountain structures east of Canal Flats, British Parkinson, D. 1990. Age and isotopic character of basement gneisses in the
Columbia. Canadian Journal of Earth Sciences, 25, 1339-1348. southern Monashee complex, southeastern B.C. Geological Association
Evenchick, C. A., Parrish, R. R. & Gabrielse, H. 1984. Precambrian gneiss of Canada - Mineralogical Association of Canada, Annual Meeting,
and Late Proterozoic sedimentation in north-central British Columbia. Program with Abstracts, 15, AlOl.
Geology, 12, 233-237. Parrish, R. 1984. Slocan Lake fault: a low angle fault zone bounding the
Gabrielse, H. 1985. Major dextral transcurrent displacements along the Valhalla gneiss complex. Nelson map area, southern British Columbia.
Northern Rocky Mountain Trench and related lineaments in north- In: Current Research, Part A. Geological Survey of Canada, Paper 84-
central British Columbia. Geological Society of America Bulletin, 96,1 - lA, 323-330.
14. Parrish, R. R. & Ross, G. 1989. Precambrian basement in the southern
& Yorath, C. J. 1989. DNAG #4. The Cordilleran orogen in Canada. Canadian Cordillera: the cratonic connection. LITHOPROBE Cordilleran
Geoscience Canada, 16, 67-83. Workshop Summary, 72-77.
Hanmer, S. K. 1984. The potential use of planar and elliptical structures as 1990. Review of the Precambrian basement of the Cordillera. Geological
indicators of strain regime and kinematics of tectonic flow. In: Current Association of Canada-Mineralogical Association of Canada, Annual
Research, PartB. Geological Survey of Canada, Paper 84-lB, 133-142. Meeting, Program with Abstracts, 15, AlOl.
1986. Asymmetric pull-aparts and foliation fish as kinematic indica- Parrish, R. & Scammell, R. 1988. The age of the Mount Copeland syenite
tors. Journal of Structural Geology, 8, 111-122. gneiss and its metamorphic zircons, Monashee complex, southeastern
Harrap, R. 1990. Stratigraphy and structure of the Monashee terrane in the British Columbia. Geological Survey of Canada, Paper 88-2, 21-28.
Mount English area, west of Revelstoke, B.C. M.Sc. thesis, Carleton Parrish, R. R. & Wheeler, J. O. 1983. U-Pb zircon age of the Kuskanax
University (unpubl.). batholith, southeastern British Columbia. Canadian Journal of Earth
Hoy, T. & Godwin, C.I.I 988. Significance of a Cambrian date from galena Sciences, 20, 1751-1756.
lead-isotope data for the stratiform Cottonbelt deposit in the Monashee , Carr, S. D. & Parkinson, D. L. 1988. Extensional tectonics of the
complex, southeastern British Columbia. Canadian Journal of Earth southern Omineca belt, British Columbia and Washington. Tectonics, 7,
Sciences, 2S, 1534-1541. 181-212.
Johnson, B. J. 1988a. Progress report: stratigraphy and structure of the Passchier, C. W. & Simpson, C. 1986. Porphyroclast systems as kinematic
Shuswap metamorphic complex in the Hunters Range, eastern Shuswap indicators. Journal of Structural Geology, 8, 831-843.
Highland (82L). In: Geological Fieldwork, 1987. British Columbia Piatt, J. P. 1984. Secondary cleavages in ductile shear zones. Journal of
Ministry of Energy, Mines and Petroleum Resources, Paper 1988-1,55- Structural Geology. 6,439-442.
58. & Vissers, R. L. M. 1980. Extensional structures in anisotropic rocks.
1989b. Geology of the west margin of the Shuswap terrane near Journal of Structural Geology. 2, 397-410.
Sicamous: implications for Tertiary extensional tectonics (82L, M). In: Price, R. A. & Mountjoy, E. W. 1970. Geologic structure of the Canadian
Geological Fieldwork, 1988. British Columbia Ministry of Energy, Mines Rocky Mountains between Bow and Athabaska rivers - a progress report.
and Petroleum Resources, Paper 1989-1, 49-54. Geological Association of Canada Special Paper, 6, 7-26.
1989b. Structural style of crustal extension in a cross-section through , Monger, J. W. H. & Roddick, J. A. 1985. Cordilleran cross-section,
the Shuswap complex, southeastern British Columbia. Geological So- Calgary to Vancouver. In: Tempelman-Kluit, D. J. (ed.) Field Guides to
ciety of America, Abstracts with Programs, 21, 98. Geology and Mineral Deposits in the Southern Canadian Cordillera.
& Brown, R. L. 1990. Crustal structure of the Omineca belt adjacent to Geological Society of America, Cordilleran Section Guidebook, Van-
line 19 of LITHOPROBE southern Canadian Cordillera transect. Geo- couver, British Columbia, 3-1 to 3-85.
logical Association of Canada-Mineralogical Association of Canada, Read, P. B. & Brown, R. L. 1981. Columbia River fault zone: southeastern
Annual Meeting, Program with Abstracts, 15, A66. margin of the Shuswap and Monashee complexes, southern British
Joumeay, J. M. 1986. Stratigraphy, internal strain and thermo-tectonic Columbia. Canadian Journal of Earth Sciences, 18, 1127-1145.
evolution of northern Frenchman Cap dome; an exhumed duplex structure, & Wheeler, J. O. 1976. Geology of Lardeau west-half map area.
Omineca hinterland, S.E. Canadian Cordillera. Ph.D. thesis. Queen's Geological Survey of Canada, Open File Map 288.
University (unpubl.). Reesor, J. E. & Moore, J. M., Jr. 1971. Petrology and structure of Thor-Odin
& Brown, R. L. 1986. Major tectonic boundaries of the Omineca belt gneiss dome, Shuswap metamorphic complex, British Columbia. Geo-
in southern British Columbia: a progress report. In: Current Research, logical Survey of Canada Bulletin 195.
Part A. Geological Survey of Canada, Paper 86-lA. Scammell, R.J. 1986. Stratigraphy, structure and metamorphism of the north
Lane, L. S. 1984. Brittle deformation in the Columbia River fault zone near flank of the Monashee complex, southeastern British Columbia: a record
Revelstoke, southeastern British Columbia. Canadian Journal of Earth of the Proterozoic crustal thickening. M.Sc. thesis, Carleton University
Sciences, 21, 584-598. (unpubl.).
Lister, G. S. & Snoke, A. W. 1984. S-C mylonites. Journal of Structural Simony, P. S., Ghent, E. D., Craw, D., Mitchell, W. & Robbins, D. B. 1980.
Geology, (,,t\l-6^?,. Structural and metamorphic evolution of northeast flank of Shuswap
McDonough, M. R. & Simony, P. S. 1988. Structural evolution of basement complex, southern Canoe River area, British Columbia. In: Crittenden,
gneisses and Hadrynian cover. Bulldog Creek area. Rocky Mountains, M. D., Jr., Coney, P. J. & Davis, G. H. (eds) Cordilleran Metamorphic
British Columbia. Canadian Journal of Earth Sciences, 25, 1687-1702. Core Complexes. Geological Society of America Memoir, 153,445-461.
Means, W. D. 1981. The concept of steady-state foliation. Tectonophysics, Simpson, C. & Schmid, S.M.I 983. An evaluation of criteria to deduce the
78, 179-199. sense of movement in sheared rocks. Geological Society of America
Monger, J.W.H., Price, R. A. & Tempelman-Kluit, D. J. 1982. Tectonic Bulletin,94, 1281-1288.
accretion and the origin of the two major metamorphic and plutonic welts White, W. H., Burrows, S. E., Carreras, J., Shaw, N. D. & Humphreys, F. J.
in the Canadian Cordillera. Geology, 10, 1261-1266. 1980. On mylonites in ductile shear zones. Journal of Structural Geol-
(principal compiler), Woodsworth, G. J., Price, R. A., Clowes, R. N., ogy,!, 175-187.
Riddihough, R. P., Currie, R., Hoy, T., Preto, V. A. G., Simony, P. S.,
Suavely, P. D. & Yorath, C. J. 1986. Transect B2, southern Canadian
Cordillera, North American continent-ocean transects. Decade of North
American Geology Program, Geological Society of America.
364
The Skeena fold belt: a link between the Coast Plutonic Complex, the
Omineca belt and the Rocky Mountain fold and thrust belt
Carol A. Evenchick
Geological Survey of Canada, 100 West Pender St., Vancouver, British Columbia,
Canada, V6B 1R8
Abstract: The northern Canadian Cordillera contains a second fold and thrust belt west of the
well-known Rocky Mountain fold and thrust belt and the metamorphic/plutonic Omineca belt.
The Skeena fold belt occupies one quarter of the width of the Cordillera, and merges to the west
with a second plutonic/metamorphic belt, the Coast Plutonic Complex. The Skeena fold belt has
many features common to thin-skinned fold and thrust belts, including: low-angle thrust faults
which sole into a detachment; a wide variety of fold styles depending on the rock type; a minimum
of 44% shortening; a foreland basin that was cannibalized by continued deformation; termination
in a frontal triangle zone; and a hinterland of metamorphic and plutonic rocks. The Skeena fold
belt is similar in many respects to the Rocky Mountain fold and thrust belt, but rather than involving
a continental terrace wedge, it deformed a terrane of Devonian to Lower Jurassic strata (Stikinia)
which accreted to North America in the early Mesozoic. Also deformed is the Jurassic to
Cretaceous clastic succession which overlies Stikinia, and the Cretaceous clastic succession of the
associated foreland basin.
The Skeena fold belt represents as much as 160 km of northeasterly shortening which occurred
between latest Jurassic (?) and early Tertiary time. The shortening was broadly contemporaneous
with crustal thickening and plutonism in the Coast Plutonic Complex and Omineca belt, dextral
strike-slip on faults east of the Skeena fold belt, and development of the Rocky Mountain fold and
thrust belt. Generally concurrent Cordillera-wide deformation suggests that there was a structural
link between all of these zones.
The Canadian Cordillera includes a collage of terranes which verging thrust faults and associated folds in the southern
accreted to the western margin of North America in Mesozoic Rockies accommodated a minimum of 170 km shortening
time (Fig. \\Concyetal. 1980; Monger 1984). Some terranes (Price 1981) between Late Jurassic and early Tertiary time.
were pericratonic, and merely slipped along the margin of The Intermontane belt (Fig. 1) is defined by low-lying
North America. Others, however, were exotic and assembled topography, sub-greenschist facies volcanic and sedimentary
into larger terranes prior to their accretion to North America. strata, and I-type plutonic rocks associated with Mesozoic
Their accretion as 'superterranes' had a profound tectonic and Cenozoic volcanism (Gabrielse et al. in press). In the
response. Another framework for Cordilleran geology is that terrane framework, the volcanic and oceanic strata comprise
of morphogeological belts (Gabrielse et al. in press) which, several accreted terranes. They are overlain by Mesozoic and
defined on stratigraphy, structure, metamorphic grade, and Cenozoic clastic successions which postdate amalgamation
physiography, expresses the cumulative geological history. of the terranes (overlap assemblages on Fig. 1). A large
While a terrane is defined by a characteristic stratigraphy, the proportion of the belt is underlain by lower Palaeozoic
process of superterrane accretion or tightening is inferred to platformal carbonates, mid-Palaeozoic to lower Middle
have produced the major metamorphic and plutonic belts, the Jurassic island-arc volcanic rocks, and related sedimentary
Coast and Omineca belts (Fig. 1; Monger et al. 1982). Ac- and plutonic rocks of Stikinia (Monger 1984). Except for
cretion of the Intermontane Superterrane (Fig. 1) to North structures alluded to in an overview by Wheeler et.al. (1972),
America in the mid-Mesozoic was followed in the Late the significance of structures in much of the northern
Cretaceous by either accretion of the Insular Superterrane Intermontane belt have been underestimated in regional
(Monger et al. 1982), or terrane tightening above a magmatic syntheses (e.g. Souther & Armstrong 1966; Monger & Price
arc (van der Heyden 1989). 1979; Mongers? a/. 1982). Most of Stikinia was assumed to
East (cratonward) of each metamorphic/plutonic belt is a have acted as arigid block during accretion of the superterranes.
belt oflower grade, stratified rocks with fewer intrusive rocks Recent mapping, however, has elucidated the nature and
(Foreland and Intermontane belts; Fig. 1). The well-known significance of a regional fold and thrust belt which occupies
Rocky Mountain fold and thrust belt (Foreland belt) consists most of the width of the northern Intermontane belt (Evenchick
of an eastward tapering wedge of miogeoclinal, platformal 1991a). This is one of three companion papers on the Skeena
and foreland basin successions which were deposited on fold belt. The first (Evenchick 199 laX describes its regional
cratonic North America between Precambrian and early extent, geometry, and evolution; the second (Evenchick
Tertiary time (e.g. Bally etal. 1966; Price 1981). Northeast 1991 b) presents arguments for the involvement of pre Middle
365
C. A. EVENCHICK
C Coatt BaKTarrana*
In Middle Jurassic time Stikinia was overlain by a wide-
1 Inaular S u p a r l a r r a n a spread clastic succession called the Bowser Lake Group. The
O Outer Tarranai
Overlap aaaamblagaa
group is Middle Jurassic to Early Cretaceous in age, and is
on s t l k l n l B
T Tertiary, Quaternary divided into a lower marine unit (Ashman Formation), and an
XK;:;Ci'et3Ceou$(Sustul Basin)
J Jurassic (Bowsw Basin) upper unit of shallow marine to nonmarine facies (Figs 3 &
4; Tipper & Richards 1976; Gabrielse & Tipper 1984;
Evenchick 1989; Cookenboo & Bustin 1989; MacLeod &
Hills 1990; Evenchick & Green 1990). There is no single
complete section of Bowser Lake Group, but mapping of
several partial sections suggests that it is at least 3500 m thick.
The Ashman Formation is characterized by incompetent
siltstone and fine grained sandstone, with competent medium
grained sandstone and conglomerate beds The lack of marker
beds prohibits large-scale structural analysis. In the undivided
shallow marine to nonmarine facies there is a wide variety of
proportions and thicknesses of sandstone and conglomerate
*- / / y • V beds They control both the style and scale of folds, and
U.S.A.~ ~*^~ provide a means of estimating the amount of shortening.
The Skeena fold belt is bounded on the west by the Coast belt,
on the north and east by a triangle zone at the front of
deformation, and on the south by the limit of strata which
display folds prominently {Evenchick 1991a). Although the
fold belt may continue to the south, structures are obscured by
poor exposure, the presence of massive volcanic rocks which
display folds poorly, and overlapping younger strata.
Strata involved in the Skeena fold belt are assemblages of
Figure 2. Geology of the Skeena fold belt, modified after Wheeler &
Stikinia, as well as overlying Jurassic and Cretaceous clastic
McFeely (1987). Structures are compiled from the following maps; Geo-
successions called Bowser Lake and Sustut Groups. Most logical Survey of Canada 1957; Richards 1975, 1980; Eisbacher 1974
Stikinian rocks at the northeast margin of the fold belt are Souther 1972; Tipper 1976; Read 1983; Woodsworthe/a/. 1985; Grove 1986
Lower to Middle Jurassic volcanic rocks of the Hazelton Evenchick 1988b, 1989; Britton 1989; Evenchick & Green 1990). ^t
366
THE SKEENA FOLD BELT
Northeast Skeena Fold Belt The northeast margin of the fold belt
1-
Structural style '^ -y
UJ P a l e o c e n e
present day
erosion surface
1- The scale and geometry of the fold belt are illustrated in maps
Maastrichtian •.'.Brothers''- and cross-sections of its northeast margin (Figs 4 & 5). The
: 'Peak Fm'.'.
o dominant structures are northwest trending folds that are
Campanian CD either upright, or have west dipping axial surfaces; many are
iiiiJTansrO^:::::
Santonian northeast verging (Figs 4, 6, 7 & 8). Folds plunge gently,
Turonian present day \-.\\\Creek\\\\\ except where they interfere with local folds and as a result
erosion surface
Cenomanian plunge up to 35". They are commonly hundreds of metres in
o Formai/of}': amplitude and wavelength, and their geometry varies from
UJ open to tight, with both angular and rounded hinges. Chev-
a Albian
ron, box and disharmonic folds are also common. Folds in the
<
volcanic rocks and immediately overlying strata are an order
UJ Aptian
or of magnitude larger in wavelength and amplitude than those
O Barremian deposited on in the Bowser Lake Group (Fig. 5).
folded and faulted
Hauterivian Bowser Lake, Klippen of volcanic rocks on Tango Creek Formation and
Spatsizi. and thrust faults with volcanic rocks in the hanging wall, and
Valanginian Hazelton groups
Spatsizi or Bowser Lake Group in the footwall (e.g. Fig. 6),
Berriasian show that the volcanic rocks (i.e. Stikinia) were intimately
Tithonian urtdiv/d0d involved in contractional deformation. The large amount of
Kimmeridgian
Oxfordian ^ * • 0 •"" 10 .
o Callovian
CO
Bathonian
CO
<
DC Bajocian
Spa tsiz^
-) Aalenian Grou,
' , '' J S - -,' '
_*< \ ^::::::::;:;^^i:::::;:/::::::::::;-:;;:^l^-^^^.;-V^
Toarcian
-.^i^' Fish
Pliensbachian volcanic s
-t;:::::;;::M
Sinemurian
Hettangian
O
Carnian
CO
•i'./
-<
E
^V.
sandstone, siltstone, conglomerate and tuff is divided into the ^•- ailal lurlace iFsca
- « - ' Ihtuat fault
Tango Creek and Brothers Peak Formations (Figs 3 & 4;
Eisbacher 1974). These units, which form the Sustut Group,
were derived from the southwest and deposited broadly
during formation of the Skeena fold belt. The Sustut Basin is Figure 4. Geology of east Spatsizi area, at the northeast margin of the
Skeena Fold Belt. Lines of cross-sections are shown by letters. Abbrevia-
therefore interpreted as the foreland basin (in part) to the
tions of structural features are: BF Bowsprit fault; CF - Crescent fault; CFF
evolving Skeena fold belt (Evenchick 1991a). - Cold Fish fault: DF - Denkladia fault; GF - Griffith fault; JLA - Joan Lake
Anticline; MCF - Mink Creek fault; MWF - Mount Will fault; SF - Spatsizi
fault; SPF - Sunday Pass fault; TF - Tualon fault.
367
rv
:J-.i
44% shorlening
Tuaton Lake section ?:>
•-awft NE
sw H I
CO
•»*•
ft!
. 3 . 3 .
Figure 5. Balanced and restored cross sections of the Skeena fold belt. Locations of the sections are marked on Figure 4 by bold lines and letters. Abbreviations of structural features are: BF Bowsprit fault; CF - Crescent
fault; CFF - Cold Fish fault; DF - Denkladia fault; JLA - Joan Lake Anticline; MCF - Mink Creek fault; MWF - Mount Will fault; SF - Spatsizi fault; SPF - Sunday Pass fault; TF - Tuaton fault. Patterns are the same
as for Figure 4. Lines in the volcanic rocks in ihe Joan Lake section are form lines.
jmi
THE SKEENA FOLD BELT
Figure 6. Tight fold in Ha7.ellon Group volcanics in the hanging wall of the Mount Will fault. Recessive strata in the saddle to the right
are Spatsizi Group (SG) overlying the volcanics. The thrust fault is marked by the bold dashed line, and the contact between the Hazelton
Group and the Spatsizi Group by the thin solid line. Viewed to the northwest.
Figure 7. Folds and hanging-wall cutoff on the Mount Will fault, in the Spalsizi Group. View to northwest.
369
C. A. EVENCHICK
Figure 8. Northeast verging chevron folds in sihstone and sandstone of the Bowser Lake Group. View to northwesl.
shortening, dominance of northeast verging and upright km across the Bowser Basin (Fig. 2).
structures, and similarity between structures in the volcanic The cross-sections only cover the northeast margin of the
rocks and typical fold and thrust belt structures suggest that fold belt, but a regional compilation (Fig. 2) illustrates that
thrust faults in the volcanics sole into a basal detachment in contractional structures continue far to the west. Structures
Stikinia, and thatthebeltis 'thin skinned' (Evenchick 1991a). in the volcanic rocks west of the basin are cryptic, but folds
Three structural levels of the Skeena fold belt are shown in overlying strata indicate that the Skeena fold belt continues
in Figure 5. The Joan Lake section (ABCDE) crosses the to the east margin of the Coast Plutonic Complex where it is
Sustut Basin, and the most extensive exposures of volcanic engulfed by post-tectonic. Tertiary plutons (Evenchick 1991 b).
rocks. The Griffith Creek section (FG) crosses the Sustut
Basin and a large region of structurally thickened Tango
Constraints on the timing of deformation
Creek Formation. The Tuaton Lake section (JKLMN) is
dominated by folded Bowser Lake Group, which has only Stratigraphic and palaeontological data for the Sustut Group,
local markers. summarized in Figure 9, constrain the timing of deformation.
The cross-sections also display three zones of common In one region, thrust faults and folds in pre-Sustut strata are
structural style. On its northeast side the Sustut Group unconformably overlain by Sustut Group (Fig. 9; Evenchick
unconformably overlies Stikinia and strata are flat lying or 1987). Because the youngest strata below the sub-Sustut
gently warped. The second zone is characterized by a gradual unconformity are Oxfordian in age (H. W. Tipper, pers.
southwestward steepening of bedding to form a northeast comm. 1989), and because the basal Sustut Group is likely to
dipping monocline of Brothers Peak Formation. Up to 200 m be Aptian or Albian in age (Sweet & Evenchick 1990; A.
of Tango Creek Formation conformably underlies the Broth- Sweet, pers. comm. 1990), the oldest structures of the
ers Peak Formation, and continues downwards into at least northeastern fold belt formed between Oxfordian and mid-
2000 m of structurally thickened Tango Creek Formation. Cretaceous time.
This zone is characterized by northeast verging thrust faults Palaeocurrent and provenance studies by Eisbacher (1974)
and folds which, in the northeast, dip northeast below the illustrate an association between the Sustut Basin and the
monocline, a geometry similar to that of the triangle zone at Skeena fold belt (Fig. 9). They show that rivers flowed from
the front of deformation in the Rocky Mountain fold and the east (Omineca belt), carrying the first miogeoclinal and
thrust belt (Jones 1982; Price 1986). Southwest of the metamorphic detritus to the northern Intermontane belt.
triangle zone is a third zone, a 20 km wide belt of folded and During deposition of the Tango.Creek Formation a western
thrust faulted volcanic rocks and Bowser Lake Group. To the source is indicated by a change in palaeocurrents and a
southwest, folded Bowser Lake Group extends another 170 prominence of chert clasts, assumed to have been eroded
^m
THE SKEENA FOLD BELT
from the Bowser Lake Group. This is inferred to be the first Shortening and variations in structural style
clastic record of the evolving Skeena fold belt (Eisbacher
1974; Evenchick 1991a). Clasts in the Brothers Peak For- Estimates of shortening have been made at the northeast
mation were derived from the east and west, and denote margin of the fold belt (Evenchick 1991a). They were
continued input from the fold belt. Deformation of the derived by constructing balanced cross-sections (following
Brothers Peak Formation indicates that shortening continued Dahlstrom 1969) and restored sections in regions where
to latest Cretaceous or early Tertiary time. several stratigraphic units provide control on the position of
Regional relationships at the south and west margins of the cut-offs of contacts by faults. Where all of the shortening is
fold belt place local constraints on the timing of deformation. within the Bowser Lake Group, it was estimated by measur-
For example, on the southwest margin of the fold belt, ing the bed lengths of prominent local markers around folds
widespread Tertiary plutons intrude contractional structures where constant bed thickness is preserved. Details of the
in the Bowser Lake Group and Middle Jurassic granite (54°N sections are given in Evenchick (1991a). In the absence of
to 55°N, Fig. 2; Woodsworth etal. 1985; age of granite from seismic and well data, the sections rely on a combination of
P. van der Heyden, pers. comm. 1990), indicating that folds down plunge projection and extrapolation of relationships
and thrust faults formed between Late Jurassic and Tertiary among the structural levels exposed by gently plunging
time. In the southeast (55°N to 56°N, 126°W to 128°W; Fig. structures, and large dip-slip faults (Fig. 4). Additional
2), Late Cretaceous to Eocene post kinematic plutons (Richards factors guiding construction of the sections include: (1) the
1980; T. Richards, pers. comm. 1989), show that some of the Sustut Group is pinned to Stikinia on the east side of the
folding apparently ended before the Late Cretaceous. How- Sustut Basin; (2) the southwest side of the Sustut Basin is the
ever, cross-cutting plutons only limit the age of those particular location of a triangle zone which is inferred to lie above the
structures, and do not preclude younger detachments carry- northeast limit of deformation; (3) the basal detachment is
ing the plutons. The regional relationships provide examples assumed to dip gently southwest from the triangle zone (see
which fall within the Late Jurassic (?) to latest Cretaceous or discussion in Evenchick 1991a); (4) fold geometry in the
early Tertiary age for the fold belt that was determined from volcanics and the position of the folds with respect to known
relationships in the Sustut Group. There is, however, no detachments are consistent with the hypothesis that they are
conclusive evidence to demonstrate or eliminate Late Jurassic a result of displacement along underlying thrust faults which
and earliest Cretaceous deformation. have a ramp-flat geometry.
In the Joan Lake section, the leading edge of the Denkladia
thrust sheet has been displaced 20 km northeastward relative
371
C. A. EVENCHICK
Hauterivian I
I
I
Valang.
cooling I1 slow
Berriasian ' V V V V V V v>v V y^Jry vvvvvvvvvvvvvv
J^^pi^p^rv^^^^^^v v v v v v v v v v v v v v v v ,^ Kootena/^
G.C. . VVVVVVVVVVVVVVVVVVVVVVVVVV cooling ^ °
Tithonian Femie •
I
Kimmeridg. jy\ >^
o Oxfordian rapid jjplift^
m
Dowser Lake Group
v> Callovian
major thickening
v>
< Bathonian 9?
cc
3 Bajocian
-J
Aalenian
SpSiSfZI _ _ _ ^ g / VVVVVVVVV
Group __ '~^^^vvv VVVVVVVVV ' obduction of
Toarcian -'Ouesneliia
^vv,o/camcsvvvv
Pliensbach. :'vvvvvvvvvvvvvv
^vvvvvvvvvvvvvvvvvv
Figure 10. Summary of evolution ofthe Skeena fold belt in relation to other tectonic elements in the Canadian Cordillera. Time scale from Palmer (1983).
Abbreviations: G.C. Gamsby Complex; m metamorphism; MDl Monashee decollement; MD2 Monashee decollement; OB Omineca belt; SFB Skeena fold
belt. Bars on the left side of Intermontane Belt reflect the span of ages for events shown, solid where well constrained and dashed where uncertain. Full
arrows are from sources of clastic detritus, and half arrows are active thrust faults.
to the northeast limit of deformation. In the Griffith Creek the volcanic rocks and overlying strata varies, both units
section, the leading edge ofthe Crescent thrust sheet has been display generally the same amount of shortening (Evenchick
displaced 22 km northeastward relative to the east limit of 1991 a). Therefore, the thrust faults observed in the volcanic
deformation. These estimates represent 36 and 60% shorten- rocks are assumed to feed upward and eastward into detach-
ing for the Joan Lake and Griffith Creek sections respec- ments below and above systems of folds in the Spatsizi,
tively. They are minimum estimates because not all pre- Bowser Lake, and Sustut Groups. The fold belt could be
Sustut structures can be restored. In the Tuaton Lake section, viewed as a regional system of blind thrust faults with folding
folding near Mount Umbach represents 44% shortening in being primarily a result of fault propagation (Suppe 1985).
the Bowser Lake Group (Fig. 5). Other estimates of the
shortening by folding in the Bowser Lake Group, combined
with the shortening in the Joan Lake and Griffith Creek Regional tectonic framework
section result in an average of 44% shortening for the north-
east margin of the fold belt (Evenchick 1991a). The evolution of the Skeena fold belt with respect to other
The difference in structural style between the volcanic belts is illustrated schematically in Figure 10. Structural
rocks and overlying strata is apparent in photographs, map elements ofthe northem and southern Omineca and Foreland
pattern and cross-sections of the northeast margin ofthe fold belts are enlisted here because hundreds of kilometres of
belt. The volcanic rocks appear to be deformed primarily by dextral strike slip displaced the Skeena fold belt northward
thrust faults, and form folds which are an order of magnitude relative to the Foreland belt and parts of the Omineca belt
larger than the numerous detached, disharmonic folds in the before, during, and after development ofthe fold belt (Gabrielse
Bowser Lake Group. Although the style of deformation of 1985). In detail, the nature of the belts varies widely along
372
THE S KEEN A FOLD BELT
base of lithosphere
Figure 11. Schematic diagram of the Late Cretaceous tectonic framework, illustrating a possible tectonic link between the Skeena fold belt, and the Coast,
Omineca and Foreland belts.
strike, but their general characteristics apply to this discus- Discussion and conclusions
sion.
Latest Jurassic folding in the Skeena fold belt, if present, The Skeena fold belt is a regional fold and thrust belt that
was coeval with early crustal thickening along the east side of spans most of the Intermontane belt. It is characterized by
the Coast Plutonic Complex (van der Heyden 1989), crustal structures typical of fold and thrust belts, and formed in Late
thickening, metamorphism, and uplift in the Omineca belt Jurassic (?) to earliest Tertiary time. From the surface
(Archibald e/a/. 1983; Browne? a/. 1986), and the beginning geometry of the Skeena fold belt, shortening is inferred to
of deposition of the Kootenay - Femie clastic wedge in the sole into a basal detachment that roots farther west in the
Foreland belt. The clastic wedge deposits in the Foreland belt Coast Plutonic Complex, where major crustal thickening was
are attributed to the load of thrust sheets in the Rocky contemporaneous with development of the fold belt. Moreo-
Mountain fold and thrust belt (Price 1973, 1981). Although ver, if the estimate of shortening at the northeast margin of the
Jurassic structures in the Skeena fold belt cannot be demon- fold belt can be applied across the fold belt, then the Skeena
strated, at least some shortening occurred before mid-Creta- fold belt represents about 160 km of shortening of the Bowser
ceous time. Basin and Stikinia. Furthermore, 160 km of Stikinia repre-
A mid-Cretaceous to earliest Tertiary age for the Skeena senting the footwall of the basal detachment must be found in
fold belt is indicated by many regional relationships. Mid- or under the Coast Plutonic Complex.
Cretaceous time was also a period of major crustal thickening The structural and temporal associations between the
in the Coast Plutonic Complex (Crawford etal. 1987; van der Skeena fold belt and the Coast Plutonic Complex suggest that
Heyden 1989). Crustal thickening by west verging thrust their evolution was kinematically linked. The spatial and
faults on the west side of the belt was accompanied by kinematic relationships between foreland and metamorphic/
amphibolite facies metamorphism and emplacement of plutonic hinterland are similar along many ancient conver-
synkinematic plutons. Later, east verging thrust faults were gent margins, including the Omineca belt and Rocky Moun-
active on the east side of the belt. Many plutons were tain fold and thrust belt (Price 1981; Archibald et al. 1984;
emplaced in the Omineca belt at this time. Regional anticlinoria Brown et al. 1986); the Asiak thrust-fold belt and Hepburn
in the northern Omineca belt are a result of Early to mid- metamorphic internal zone of Wopmay Orogen (e.g. Hoffman
Cretaceous shortening (Evenchick 1988a). The southern et al. 1988); Appalachian Orogen (e.g. Hatcher 1981); the
Omineca belt underwent slow cooling, uplift and erosion, Himalayas (e.g. Gansser 1964; Mattauer 1986) and the Alps
with deformation at the deepest levels (Archibald etal. 1983). (Trumpy 1980). In the Canadian Cordillera, however, short-
In the Foreland belt, deposition of the Blairmore clastic ening in the Skeena fold belt and its metamorphic/plutonic
wedge resulted from continued shortening in the Rockies. hinterland was contemporaneous with shortening in the
Crustal thickening, metamorphism, and uplift in the Coast Foreland belt and its metamorphic/plutonic hinterland. This
Plutonic Complex extended into earliest Tertiary time relationship invites speculation on the gross structure of the
(Crawford et al. 1987; van der Heyden 1989). To the east, Cordillera.
folded strata of the Brothers Peak Formation demonstrates The double pair of fold belt and hinterland permits pos-
that deformation in the Skeena fold belt lasted into latest tulation of a structural link between all belts of the Cordillera,
Cretaceous or early Tertiary time. The last displacement on illustrated schematically in Figure 11. Two systems of
contractional shear zones in the southern Omineca belt was detachments shown in Figure 11 provide a mechanism for
during the Palaeocene (Journeay 1989; Brown & Carr 1990). transferring displacement from the subduction zone to the
The shortening was transmitted eastward into the Rockies east side of the Coast Plutonic Complex, and to the relatively
(Archibald et al. 1984; Brown et al. 1986) at a time broadly high structural level of the Skeena fold belt. East verging
corresponding to deposition of the last foreland basin depos- structures observed on the east side of the complex (Fig. 10),
its (Belly River Formation) in the Foreland belt. and associated metamorphism and plutonism may be part of
that system. As well, a deeper detachment may have led from
373
C. A. EVENCHICK
the subduction zone to the Omineca belt and Rocky Mountain The Skeena fold belt and its position in the evolution of the
fold and thrust belt. The crustal duplex shown in the Omineca Canadian Cordillera is a good example that collision tectonics
belt, and its link with the Foreland belt is from Brown et al. cannot be viewed as rigid blocks welding to a continental
(1986), but structures above the duplex related to earlier margin.
terrane accretion are not shown.
Between the Skeena fold belt and the Rocky Mountain The field work for this project was undertaken as part of the regional mapping
fold and thrust belt is a major fault system which displays program of the Geological Survey of Canada. I am indebted to Hu Gabrielse
and Howard Tipper for their support, and for shared expertise on the geology
hundreds of kilometres of dextral strike slip (Gabrielse 1985).
of the northern Cordillera. During the first two years of the project I was
These faults may flatten into the deeper detachment as shown supported by a Natural Sciences and Engineering Research Council Post
in Figure 11. Oblique displacement on the lower detachment Docloral Fellowship. An early draft of the manuscript was read by Hu
is is shown in Figure 11 to be partitioned into both strike slip Gabrielse. 1 also appreciate the helpful comments of the critical reviewer,
and thrust components. R. L. Brown.
References
Anderson, R. G. &Thorkelson, D. J. 1990. Mesozoic stratigraphy and setting 1991 b. Structural relationships of the Skeena Fold Belt on the west side
for some mineral deposits in Iskut River map area, northwestern British of the Bowser Basin, northwest British Columbia. Canadian Journal of
Columbia. In: Current Research, Part E, Geological Survey of Canada, Earth Sciences (in press).
Paper90-1E, 131-139. & Green, G. M. 1990. Structural style and stratigraphy of southwest
Archibald, D. A., Glover, J. K., Price, R. A., Farrar, E. & Carmichael, D. M. Spatsizi map area. In: Current Research, Part F, Geological Survey of
1983. Geochronology and tectonic implications of magmatism and Canada, Paper 90-lF, 135-144.
metamorphism, southern Kootenay arc and neighbouring regions, south- Gabrielse, H. 1985. Major dextral transcurrent displacements along the
eastern British Columbia - I. Jurassic to mid-Cretaceous. Canadian Northern Rocky Mountain Trench and related lineaments in north-
Journal of Earth Sciences, 20, 1891-1913. central British Columbia. Geological Society ofAmerica Bulletin, 96,1 -
, Krogh, T. E., Armstrong. R. L. & Farrar, E. 1984. Geochronology and 14.
tectonic implications of magmatism and metamorphism, southern & Tipper, H. W. 1984. Bedrock geology of Spatsizi map area (104H).
Kootenay arc and neighbouring regions, southeastern British Columbia Geological Survey of Canada, Open File 1005.
- II. Mid-Cretaceous to Eocene. Canadian Journal ofEarth Sciences, 21, & Yorath, C. J. 1989. DNAG#4. The Cordilleran Orogen in Canada.
567-584. Geoscience Canada, 16, 61-S3.
Bally, A. W., Gordey, P. L. & Stewart, G. A. 1966. Structure, seismic data, , Monger, J.W.H., Wheeler, J. O. & Yorath, C. J. in press.
and orogenic evolution of southern Canadian Rocky Mountains. 6u//e«'/i Morphogeological Belts, Tectonic Assemblages and Terranes. In; Chapter
of Canadian Petroleum Geology, 14, 337-381. 2, Tectonic Framework. In: Gabrielse, H. & Yorath, C. J. (eds). The
Britton, J. M. (compiler) 1989. Geology and Mineral Deposits of the Unuk Cordilleran Grogen: Canada, Geological Survey of Canada, Geology of
Area. British Columbia Ministry of Energy, Mines, and Petroleum Canada no. 4.
Resources, Open File Map 1989-10. Gansser, A. 1964. Geology of the Himalayas. Wiley Interscience, London,
Brown, R. L., Joumeay, J. M., Lane, L. S.. Murphy, D. C. & Rees, C. J. 1986. 289p.
Obduction, backfolding and piggyback thrusting in the metamorphic Geological Survey of Canada, 1957. Stikine River Area, Cassiar District,
hinterland of the southeastern Canadian Cordillera. Journal ofStructural British Columbia. Geological Survey of Canada, Map 9-1957.
Geology, 8, 255-268. Grove, E. W. 1986. Geology and Mineral Deposits of the Unuk River -
& Carr, S. D. in press. Lithospheric thickening and orogenic collapse Salmon River - Anyox Area. British Columbia Ministry ofEnergy, Mines,
within the Canadian Cordillera. Pacific Rim 90 Congress. and Petroleum Resources, Bulletin 63.
Coney, P., Jones, D. L. & Monger, J.W.H. 1980. Cordilleran Suspect Hatcher, R. Jr., 1981. Thrusts and nappes in the North American Appala-
Terranes. Nature, 288, 329-333. chian Orogen. In: McClay, K. R. & Price, N. J. (eds) Thrust and Nappe
Cookenboo, H. & Bustin, R. M. 1989. Jura-Cretaceous (Oxfordian to Tectonics. Geological Society of London Special Publication, 9.491 -500.
Cenomanian) stratigraphy of the north-central Bowser Basin, northern Hoffman, P. F., Timil, R., King, J. E., St-Onge, M. R. & Lucas, S. B. 1988.
British Columbia. Canadian Journal of Earth Sciences, 26, 1001 -1012. Axial projections and modes of crustal thickening, eastern Wopmay
Crawford, M. L., Hollister, L. S. & Woodsworth, G. J. 1987. Crustal orogen, northwest Canadian shield. In: Clark Jr., S.P., Burchfiel, B.C.
deformation and regional metamorphism across a terrane boundary. & Suppe, J. (eds) Processes in Continental Lithospheric Deformation.
Coast Plutonic Complex, British Columbia. Tectonics, 6, 343-361. Geological Society of America, Special Publication, 218, 1-30.
Dahlstrom,C.D.A. 1969. Balanced cross-sections. Canai^/anyoHcna/o/fart/; Jones, P. B. 1982. Oil and gas beneath east-dipping underthrust faults in the
Sciences,6.143-151. Alberta Foothills, /n; Powers, R.B. (sA.) Studies of the Cordilleran Thrust
Eisbacher, G. H. 1974. Sedimentary and tectonic evolution of the Sustut and Belt. Rocky Mountain Association of Geologists. 61-74.
Sifton basins, north-central British Columbia. Geological Survey of Joumeay, J. M. 1989. The Shuswap Thrust: Sole fault to a metamorphic-
Canada, Paper 73-31. plutonic complex of Late Cretaceous-Early Tertiary age, southern
Evenchick, C. A. 1987. Stratigraphy and structure of the northeast margin of Omineca belt, B.C. In: Geological Society of America Abstracts with
the Bowser Basin, Spatsizi map area, north-central British Columbia. In: Programs, 21, 99.
Current Research, Part A, Geological Survey of Canada, Paper 87-1 A, MacLeod, S. & Hills, L. V. 1990. Sedimentology and palaeontology of Late
719-726. Jurassic (Oxfordian) to Early Cretaceous (Aptian) strata, northern Bowser
1988a. Stratigraphy, metamorphism, structure, and their tectonic im- Basin. In: Geological Association of Canada, Mineralogical Associa-
plications in the Sifton and Deserters ranges, Cassiar and northern tion of Canada, Program with Abstracts, IS, A80.
Rocky mountains, northern British Columbia. Geological Survey of Mattauer, J. 1986. Intracontinental subduction, crust-mantle decollemem
Canada, Bulletin 376. and crustal-stacking wedge in the Himalayas and other collision belts. In:
1988b. Structural style and stratigraphy in northeast Bowser and Sustut Coward, M. P. & Ries, A. C. (eds) Collision Tectonics, Geological So-
basins, north-central British Columbia. In: Current Research, Part E, ciety of London, Special Publication 19. 37-50.
Geological Survey of Canada, Paper 88-lE, 91-95. Monger, J.W.H. 1984. Cordilleran Tectonics: a Canadian perspective.
1989. Stratigraphy and structure in east Spatsizi map area, north-central Bulletin of the Geological Society of France, 7, 255-278.
British Columbia. In: Current Research, Part E, Geological Survey of & Price, R. A. 1979. Geodynamicevolutionof the Canadian Cordillera-
Canada, Paper 89-1E, 133-138. progress and problems. Canadian Journal of Earth Sciences, 16,770-791.
1991 a. Geometry, evolution, and tectonic framework of the Skeena , Price, R. A. & Tempelman-Kluit, D. J. 1982. Tectonic accretion and
Fold Belt, north-central British Columbia. Tectonics (in press). the origin of the two major metamorphic and plutonic welts in the
Canadian Cordillera. Geology, 10, 70-75.
374
THE SKEENA FOLD BELT
Palmer, A. R. (compiler). 1983. The Decade of North American Geology Tipper, H. W. (compiler). 1976. Smithers. B. C. 93L. Geological Survey of
1983 time scale. Geology, 11, 503-504. Canada, Open File 351.
Price, R. A. 1973. Large-scale gravitational flow of supracrustal rock.s, & Richards, T. A. 1976. Jurassic stratigraphy and history of north-
southern Canadian Rockies. In: De Jong, K. A. & Scholten, R. (eds) central British Columbia. Geological Survey of Canada Bulletin, 270.
Gravity and Tectonics. Wiley, New York, 491-502. Trumpy, R. 1980. An outline of the geology of Switzerland. In: Trumpy, R.
198L The Cordilleran foreland thrust and fold belt in the southern (ed.) Geology of Switzerland A Guide-Book 26th International Geo-
Canadian Rocky Mountains. In: McClay, K. R. & Price, N. J. (eds) logical Congress, Guidebook GIO, 7-102.
Thrust and Nappe Tectonics. Geological Society of London Special van der Heyden, P. 1989. U-Pb and K-Ar Geochronometry of the Coast
Publication, 9, 427-448. Plutonic Complex, 53°N to 54°N, British Columbia, and Implications for
1986. The southeastern Canadian Cordillera: thrust faulting, tectonic the Insular-Intermontane Superterrane boundary. Unpublished Ph.D.
wedging, and delamination of the lithosphere. Journal of Structural thesis. University of British Columbia.
Geolofiy. 8. 239-254. Wheeler, J. O., Aitken, J. D., Berry, M. J., Gabrielse, H., Hutchison. W. W.,
Read. P. B. 1983. Geology, Classy Creek (1()4J/2E) and Stikine Canyon Jacoby, W. R.. Monger, J.W.H.. Niblett, E. R., Norris, D. K., Price, R. A.
(I04J/1W). Geoloiiical Survey of Canada. Open File 940. & Stacey. 1972. The Cordilleran Structural Province. //;.• Price, R.A. &
Richards, T. A. 1975. Geology of McConnell Creek (94D\E) map area. Douglas, R.J.W. (eds) Variations in Tectonic Styles in Canada. Geo-
Geolos^ical Survey of Canada, Open File 342. logical Association of Canada Special Paper, 11, 1-82.
1980. Geology of Hazelton (93M) itiap area. Geological Survey of & McFeely, P. (compilers). 1987. Tectonic Assemblage Map of the
Canada, Open File 720. Canadian Cordillera and adjacent parts of the United States of America.
Sweet, A. R. & Evenchick, C.A.I 990. Ages and depositional environments Geological Survey of Canada, Open File 1565.
of the Sustut Group, north-central British Columbia. /;;. Geological , Brookfield, A.'j., Gabrielse, H„ Monger, J,W.H., Tipper, H. W. &
Association of Canada, Mineralogical Association of Canada, Program Woodswonh, G. J. (compilers). 1988. Terrane Map of the Canadian
with Abstracts, 15, A127. Cordillera. Geological Sun-ey of Canada, Open File 1894.
Souther. J. G. 1972. Telegraph Creek map-area, British Columbia. Geological Woodsworth, G. J., Hill, M, L. & van der Heyden, P. 1985. Preliminary
Survey of Canada, Paper 71-44. Geologic Map of Teirace (NTS 103 I East Half) Map Area, British
& Armstrong, J. E. 1966. North Central Belt of the Cordillera of British Columbia. Geological Survey of Canada, Open File 1136.
Columbia. In: Canadian Institute of Mining and Metallurgy, Special , Anderson, R. G. & Annstrong, R. L. 1989. Plutonic regimes in the
VV)//w)('8. 171-184. Canadian Cordillera. Geological Survey of Canada Open File, 1983.
Suppe, J. 1985. Principles of Structural Geology. Prentice-Hall Inc., 537p.
Thomson, R. C , Smith, P. L. & Tipper, H. W. 1986. Lower to Middle
Jurassic (Pliensbachian to Bajocian) stratigraphy of the northern Spatsizi
area, north-central British Columbia. Canadian Journal of Earth Sci-
ences,23, 1963-1973.
375
Geometric evidence for synchronous thrusting in the southern Alberta
and northwest Montana thrust belts
Steven E. Boyer
Department of Geological Sciences, University of Washington, Seattle,
Washington 98195, USA
Abstract: Any model of thrust kinematics or mechanics must account for the hitherto unexplained
presence of imbricate stacks on the forelimbs and crests of folds and structural culminations.
Although map-scale patterns of folded thrusts and the sedimentary record of foredeeps seem to
conclusively argue for a relatively simple hinterland-to-foreland Cpiggy-back') progression of
thrusting, such a simple sequence cannot produce such crestal imbricate stacks. The only model
compatible with existing structural and stratigraphic evidence is one in which thrusts at depth
develop in a cratonward progression but remain active during the initiation of younger thrusts.
Thus at any one time two or more thrusts may be simultaneously active. This model differs from
previous ones in which older thrusts were assumed to become deactivated as new thrusts
propagated toward the foreland. Synchronous thrusting in the subsurface produces 'out-of-
sequence' thrusting at the Earth's surface. The adoption of a synchronous thrust model over the
classic 'piggy-back' model has important implications for mechanical models of thrusting, cross-
section balancing, and hydrocarbon exploration in thrust belts and leads to more realistic models
of duplex fault zone development. First, while not proving the critically-tapered wedge model of
thrust mechanics, the observations demonstrate that the model is viable. Second, since most cross-
section balancing techniques include the assumption of piggy-back thrusting, "out-of-sequence'
and synchronous thrusts must be identified if a section is to be correctly balanced. Third, models
of hydrocarbon generation based on a cratonward progression of thrusting may lead to inaccurate
conclusions if applied to regions of synchronous thrusting; appraisal of basins and prospects based
on synchronous thrust models produce results considerably different from those generated
assuming simple sequential thrust development. Lastly, existing duplex models cannot adequately
explain the large variation in the geometry of duplex fault zones. However, synchronous thrusting,
if halted at various stages, can produce a large variety of typically observed duplex geometries.
There have been two principal schools of thought regarding reactivation and/or duplex underplating to maintain critical
the sequence of thrust imbrication within thrust and fold taper (Davis et al. 1983: Boyer & Geiser 1987; Piatt 1988;
belts; one maintaining that thrusts generally are younger Dahlen & Suppe 1988; Morley 1988) and (2) a piggy-back
toward the foreland (Bally e/a/. I966;DahIstrom 1970; Roy se sequence cannot produce many observed structural
et al. 1975) and a second holding that thrusts develop in a geometries. This paper, dealing with the latter, presents
foreland-to-hinterland sequence (Milici 1975;Mudge 1970). observations which indicate that synchronous thrusting in the
Until recently the theory of cratonward thrust imbrication, Canadian and US Rockies is more common than has previ-
with allowances for minor 'out-of-sequence' thrusting ously been recognized. Therefore simple models of piggy-
(Dahlstrom 1970) has become the most widely accepted, back thrusting may not have universal applicability. If these
having taken on the role of an axiom in the study of thrust conclusions are correct, they have considerable implications
kinematics. In hinterland-to-foreland progression of thrust- for the mechanics of thrust emplacement, the application of
ing, older thrusts become inactive and are carried passively, cross-section balancing techniques, and patterns of hydrocar-
in a so-called 'piggy-back' fashion, upon the backs of under- bon generation and entrapment in thrust belts.
lying younger thrusts. Many thrust models, including those
for duplex development (Boyer 1978; Boyer & Elliott 1982)
have been built upon this foundation, as the arguments for a Existing criteria for the determination of thrust
hinterland-to-foreland thrust progression are based on con- sequence
vincing stratigraphic and geometric evidence (see discussion
in the following section). However, much structural and Most patterns of synorogenic sedimentation suggest a
stratigraphic evidence has come to light which casts doubt on cratonward progression of thrusting. Armstrong & Oriel
the simple piggy-back thrust model (Ori & Friend 1984; (1965) postulated a west-to-east progression of thrusting in
Searle 1985; Butler 1987; Roeder 1988; Morley 1988). the Idaho-Wyoming-Utah thrust belt, commencing in latest
The conclusion that thrusts develop in a relatively straight Jurassic and terminating in early Eocene. They based their
forward hinterland-to-foreland, progression has at least two conclusions on observations that from west to east succes-
major shortcomings: (1) it is incompatible with compelling sively younger rocks were involved in thrusting and that
new mechanical models for the emplacement of thrust belts, during the Cretaceous period the axis of maximum deposition
which require synchronous, 'out-of-sequence' thrusting, thrust migrated eastward (Armstrong & Oriel 1965, p. 1854-6).
377
S. E . BOYER
Presumably this eastward migration of the depositional axis preserved in orogenic belts, many thrusts can be shown to be
accompanied the eastward advance or growth of the thrust either be late and out-of-sequence or to have been periodically
belt. Oriel & Armstrong (1966, p. 2618) further noted that for reactivated (Lamerson 1982; Hurst & Steidtmann 1986). In
Mesozoic to early Tertiary formations 'The westernmost agreement with the stratigraphic evidence, the use of balanced
known occurrence and coarsest facies of each (syntectonic) cross sections also indicates that imbricate stacks are often
unit lie in a belt east of the coarsest facies of the next older generated by 'out-of-sequence' thrusting (Delphia &
unit.' Although there is much debate concerning the absolute Bombolakis 1988). Many out-of-sequence imbricate stacks
ages of the various thrusts (Heller etal. 1986), most workers occur at the tops of ramps in thrust sheets and have been
appear to accept a west-to-east progression for the US Rocky produced in experiments (Morse 1977, Fig. 3). Also out-of-
Mountain thrust belts (Royse et al. 1975; Wiltschko & Dorr sequence thrusts may develop at the base of a ramp as a means
1983). In the Canadian foothills the eastward migration of of decreasing the ramp angle (Serra 1977; Knipe 1985).
Mesozoic to Palaeocene synorogenic foredeeps has led to It should be noted that the term 'out-of-sequence" thrust is
similar conclusions (Bally ef a/. 1966, p. 365-369; Price 1973, used by proponents of piggy-back thrusting to refer to those
1981). thrusts which disobey the 'rules' of thrust development and
Arguments for a hinterland-to-foreland progression of it has been suggested that the use of this term be discontinued
thrusts are fortified by geometric evidence, most importantly (Geiser, pers. comm. 1990). Such thrust imbricates are
patterns of folded thrusts (Verrall 1968; Jones 1971). The usually thought to be of minor significance (Dahlstrom 1970,
existence of folded thrusts was noted at least as early as the p. 354; Woodward 1987). This is probably true for portions
1930s by geologists working in the Canadian Rocky Moun- of the Canadian and US Rockies, including the southern
tains (Hake etal. 1935, 1942; Hume 1941; Hage 1942; and portion of the Utah-Wyoming thrust belt. However, patterns
Deiss, 1943). Scott (1951), attributed the folded thrusts to of imbrication, especially the locus of imbricate faults on the
two distinct phases of deformation: (1) low-angle thrust crests of folds, in the southern Alberta and northwest Mon-
faulting, followed by (2) regional-scale folding and accom- tana thrust belts, will be used in subsequent sections to argue
panying 'adjustment faulting' and reactivation of the steeper that such out-of-sequence motions are not incidental and that
portions of folded thrusts. However, extensive drilling and they are symptomatic of synchronous motion on two or more
seismic acquisition, associated with petroleum exploration in major thrusts at depth.
the Canadian Rockies, and field observations have led to the
conclusion that folded thrusts are best explained by a
cratonward progression of thrusting in which older thrusts are Imbricate stacks situated on the forelimbs of folds
carried piggy-back and folded over ramps in lower, more
frontal thrusts (Bally et al., 1966; Verrall 1968; Dahlstrom Stacks of imbricate thrusts are often found on the crests or
1970; Jones 1971). limbs of structural culminations (Figs 1, 4b, 5 & Table 1).
Despite the geometric and stratigraphic evidence for a Several examples gleaned from published sections are listed
foreland progession of thrusting there has long existed a in Table 1. All these structures share common features: (1)
school of thought maintaining that mechanical arguments the upper fault in the sequence emanates from the backlimb
and geometric evidence suggest the opposite sense of propa- of the underlying fold and is usually quite planar; (2) struc-
gation (Mudge 1970; Milici 1975). Mechanical theory turally lower faults on the forelimb display a decreasing
(Hubbert & Rubey, 1959) has indicated that for tabular thrust radius of curvature with depth (i.e. the lower faults in the
sheets to be emplaced by horizontal compression they must sequence are more tightly folded than the higher faults); and
be limited to lengths deemed by many to be unrealistically (3) strata on the backlimb of the fold display little or no
short. This has led to the suggestion that gravity is the only evidence of disruption or thrust repetition.
viable driving force for the emplacement of thrust sheets The geometry of crestal imbricate stacks have been de-
(Milici 1975; Mudge 1970; LeMoine 1973; Scholten 1973; scribed in the older thrust literature. Most commonly men-
Choukroune & Seguret 1973). In a surficial gravity slide the tioned are 'back-limb' imbricates, or imbricates which overlie
imbricates often break backward from the toe, and if such a the hinterland-dipping limbs of folds. It has been supposed
mechanism operated in the emplacement of thrust sheets it for some time that 'backlimb' imbricates often form
was thought that thrust belts must also develop in a foreland- sequentially from foreland to hinterland, or in a so-called
to-hinterland sequence. These conclusions have been sup- ' out-of-sequence' order, on the back limbs of existing structural
ported by patterns of thrust intersection which suggest that culminations (Douglas 1950; Bally et al. 1966). Douglas
more frontal thrusts have been truncated by younger thrusts (1950, p. 88-95) maintained that field observations and map
toward the hinterland (Milici 1975). patterns supported a model in which the backlimb faults
The mechanical arguments for foreland-to-hinterland thrust formed simultaneously with the underlying anticlines. Hume
sequences must now be considered less reliable in light of (1957, p. 408-409) followed Douglas (1950) in arguing for
recent work suggesting that thrust sheets can indeed be simultaneous folding and imbrication. Douglas (1950, p. 89)
emplaced by horizontal compression if a critical taper angle attributed back-limb out-of-sequence thrusts to greater
is maintained (Davis era/. 1983). However,the physical field interbed slip within the west limbs of folds, forming 'out-of-
evidence for hindward thrust sequences cannot be as easily the-syncline' thrusts (Hake et al. 1942; Dahlstrom 1970).
dismissed. Where the youngest synorogenic sediments are However, in most of the structures discussed in this paper, the
378
GEOMETRIC EVIDENCE FOR SYNCHRONOUS THRUSTING
r McConnell Thrust
!|!00/M*>r;l;l;]/ ?M2^/ y////
i^^'p^^iiiiS^
zwioosg Mtn.anejBt )
L_ _J
p6x v= 1
* Also see numerous additional examples suggested by Gilluly (1960, p. 76-77). This table refers to structures that resemble results produced by the
synchronous thrust model, but these structures, with the exception of the Goat Ridge window, were not usually interpreted as such.
379
S. E. BOYER
®(D,
®(D(D
^^kP:^^
yi!.|i!i|i!i|i|i|i!l[Im!.|i|i!i|i!.|i|i!i|i!.|i!l|i^|T|.!i|i|i|.!i|i!.|i!i|i!i|.!i!^^
Figure 3. Gilluly's model of simultaneous folding and thrusting (Fig. 2) provided the inspiration for this synchronous thrust model, (a) After formation
of the initial thrust ('A') anew thrust ('B', dashed) propagates toward the foreland, (b) Movement on thrust 'B' folds the upper flat of thrust 'A' deactivating
imbricate ' 1'. Renewed movement on thrust 'A' breaks across the crest of the frontal fold, forming an out-of-sequence imbricate, '2' (c & d) The process
continues with additional out-of-sequence imbricates, ' 3 ' and '4", being formed as motion alternates between thrusts 'A' and 'B' With each stage of
displacement on thrust 'B' the amplitude of the frontal ramp anticline increases and new out-of-sequence develop in conjunction with continued motion on
thrust 'A'. See text for additional discussion. Although the model invokes alternating motion between thrusts 'A' and 'B', the two thrusts will appear to
have been synchronous when viewed in the context of geological time.
similar structures throughout the greater Rocky Mountain disparate observations of the hindward and cratonward schools
region (Boyer 1986a). None of the examples that will be of thrust progression, amodel was prepared based on Gilluly's
discussed can easily be explained by folding of existing (1960) concept of synchronous folding and thrusting (Boyer,
imbricates as suggested by piggy-back thrust models 1986a). It was assumed that both schools of thought were
(Dahlstrom 1970; Boyer & Elliott 1982). correct in their observations but partially in error in their
conclusions.
The model was drawn using the kink domain geometry
Simultaneous thrust model and flat - ramp - flat thrust/fold geometries (Rich 1934;
Douglas 1950; Boyer & Elliott 1982 ) and the equations of
In order to explain the locus of imbricate stacks on the crests Suppe (1983). In Figure 3a an imbricate (dashed) is about to
and forelimbs of underlying culminations and to rectify the form after some initial movement on an older thrust. Move-
380
GEOMETRIC EVIDENCE FOR SYNCHRONOUS THRUSTING
ment on the younger frontal thrust forms a low-amplitude (1983) showed no detachment within the Jurassic of the west
ramp anticline which in turn folds the existing thrust. As it is limb. However, in order to allow restoration of the tectonic
folded, that portion of the older thrust which overlies the shortening on the east limb of the structure, such a detach-
frontal limb of the leading ramp anticline is inactivated (Figs ment is certainly required.
3b & 3c). However, continued movement on the higher thrust If the model (Fig. 3) applies to the origin of the northern
produces an out-of-sequence imbricate which breaks across Sawtooth structure, the numerous major thrusts within the
the crest of the frontal anticline. As this process proceeds, the Sawtooths must have been synchronously active. The pres-
amplitude of the frontal anticline grows and additional out- ence of the forelimb imbricate stack suggests that the upper
of-sequence imbricates propagate from the crest of the detachment in the Jurassic section must have been active
structure. Out-of-sequence imbrication continues until the during the growth of the culmination. Since this detachment
frontal ramp anticline attains a maximum amplitude, a con- roots westward in Palaeozoic carbonates, thrusting on western
dition which is met as the rear kink plane of the ramp anticline imbricates must have occurred synchronously with faults of
has climbed unto the upper flat (Fig. 3d). From this stage Palaeozoic strata coring the frontal culmination to the east.
forward, simultaneous motion on the two thrusts will proceed
without the introduction of additional crestal imbrication and
the youngest and highest imbricate will remain active. All Southern Alherta, Canada
imbricate faults occur on the forelimb of the structure; compare A section through the Lewis thrust sheet and the Waterton gas
the model (Fig. 3d) with actual examples (Fig. 1 & Table 1). field in Alberta, Canada (Fig. 5), displays geometries strik-
Note that in this model, movement alternated between the ingly similar to those mapped by Mudge & Earhart (1983) in
two major thrusts, so the thrusts are not strictly synchronous. the northern Sawtooth Range. The Lewis thrust sheet of
However, when averaged over geological time the motions of Middle Proterozoic metasedimentary strata is folded into a
the two thrusts will appear to be simultaneous. Such alternating broad syncline, the west and east limbs of the syncline having
stick-slip motion probably also applies to many other thrusts been uplifted by imbrication of Palaeozoic carbonates within
that have been described as being synchronous. two underlying duplex fault zones (Bally et al. 1966; Gordy
Dahlstrom (1970, p. 354) has noted that regional and local et al. 1977, pp. 25 & 31). The two duplex structures are on
thrust sequences are not entirely compatible. However, out- trend with the two antiformal features that emerge southward
of-sequence imbrications may not be incompatible with the from beneath the Lewis sheet to form the northern Sawtooths.
propagation patterns of regional thrusts when viewed in the In order to understand the kinematic development of the
light of the simultaneous thrust model just described. Note two culminations sequentially restored cross sections were
that 'in-sequence' initiation of thrusts at depth, combined constructed by reversing the process utilized in the model
with continued motion on all thrusts, leads to 'out-of-se- (Fig. 3). Assuming that the least curved faults emanating
quence' imbrication at shallow or surface levels (Fig. 3). In from the backlimb of the frontal culmination were the last to
this model hinterland-to-foreland thrust propagation, syn- form, these faults were restored first. Using balancing
chronous thrusting, and out-of-sequence imbrication are all arguments, Dahlstrom (1969, p. 750) demonstrated that im-
compatible. bricates of Mesozoic rocks overlying Waterton field must be
separated from the underlying duplex of Palaeozoic carbonates
by a detachment; i.e. shortening within the Mesozoic strata is
Application of the synchronous thrust model to far in excess of that within the Palaeozoic sequence. Since the
the southern Alberta & northwest Montana thrust Mesozoic imbricates branch from a regional detachment
belts within the Jurassic section, and this detachment in turn roots
westward into the trailing duplex, restoration of the frontal
As summarized in Table 1, crestal and forelimb imbricate imbricates required partial restoration of imbricates within
fans commonly overlie folds and thrust culminations in many the trailing duplex (Figs 5a & 5b). At the same time as the
thrust belts. These features are especially common in the highest imbricates of the frontal structure were restored
Rocky Mountains of southern Alberta and the northwest several of the underlying imbricates of Palaeozoic carbonates
Montana Sawtooth Ranges. Applying the synchronous thrust were also restored. This process decreased the amplitude of
model (Fig. 3) to these thrust terranes one can draw conclu- the frontal structure so that in the next stage (Figs 5b & 5c) the
sions concerning their probable kinematic development highest imbricate of the frontal duplex and an additional
imbricate of the trailing duplex could be restored. The next
sequence (from Figs 5c to 5d) was performed by continuing
Northern Sawtooth Range, Montana, USA the process of removing the highest forelimb imbricates
At the northern termination of the Sawtooth Range in north- while restoring footwall imbricates to decrease the amplitude
west Montana a crestal imbricate stack overlies the frontal ofthe frontal duplex. This restoration is a simplification; to
limb of a fault-cored culmination (Fig. 4). On the west limb show the complete evolution ofthe section one would need to
of the structure is a normal stratigraphic sequence of Missis- construct as many sequential sections as there are imbricates
sippian, and unconformably overlying Jurassic and lower overlying the frontal structure, approximately twelve.
Cretaceous strata. Indeed, owing to the lack of duplicated Note that this procedure reversed the process utilized in
strata on the west flank of the culmination, Mudge & Earhart the model (Fig. 3). Although the 'real' example is considerably
381
S. E. BOYER
a ..•./.-v114°/_
:::::::,:,:::,:,: J^2^.
Mudge&Earhart(1983)
Figure 4. Imbricate stacks overlying the frontal culmination of the northern Sawtooth Range. The location of the map 'b' is indicated by the black rectangle
on map 'a'. Application of the synchronous thrust model (Fig. 3) suggests that the folded detachment separating the Mesozoic (Jurassic and Cretaceous)
sequence (Mz) from underlying Mississippian carbonates (M) was active in conjunction thrusts of Mississippian and Cambro-Devonian (C-D) which fill
the core of the north-plunging structure. Movement on imbricates within the frontal anticlinorium (Palaeozoic strata) increased the amplitude of the
anticlinorium. As the anticliorium grew portions of the overlying Jurassic-level detachment were deactivated and abandoned as the imbricate faults which
overlie the northeast limb of the structure. Since the upper detachment roots westward in Palaeozoic carbonates this implies synchronous motion on faults
of the eastern and western Sawtooth Ranges. The structure of the northern Sawtooths is similar to that at Waterton gas field ( ' W in Fig. 4a). The sequential
evolution depicted in Figure 5 (Waterton Field) would also apply to the Sawtooths.
382
GEOMETRIC EVIDENCE FOR SYNCHRONOUS THRUSTING
Figure 5. (a) The Lewis thrust sheet is folded by two duplexes of Palaeozoic carbonates, Waterton gas field to the east and the Flathead duplex to the west
beneath the trailing edge of the Lewis thrust. These two structures lie on strike with the two culminations at the northern termination of the Sawtooth Range
(Fig. 4). The position of the section is shown in Figure 4a. Waterton field and the frontal culmination of the Sawtooths (Fig. 4b) are similar in form, both
with imbricate stacks overlying their forelimbs; the Sawtooth culmination can be considered a surface analogue to the Waterton field duplex, (b) The section
was restored by reversing the process used in producing the model of Figure 3. The structurally highest and presumably the youngest imbricates,' I' were
removed first. Then the amplitude of the leading duplex was decreased by restoring 4 frontal imbricates of the duplex. This process unfolded one or more
overlying imbricates. Since the imbricates in Mesozoic section merge at depth with a Jurassic-level detachment, which roots westward into the Flathead
duplex, imbricates of the Flathead duplex must be restored simultaneously with the imbricates overlying the frontal duplex, (c) The process is continued
by removing the displacement on imbricate '2', while simlultaneously restoring an additional imbricate of the Flathead duplex. Also the amplitude of the
Waterton duplex was further decreased by partially restoring imbricates ' 3 \ (d) Between stages ' C and 'D', the remaining duplex imbricates of Waterton
structure were restored and the displacements on imbricates '4' were removed. Sequential restoration of this section using the synchronous thrust model
indicates that the two duplex fault zones grew simultaneously.
more complex that the model, one can see that structure is The same conclusions are reached by analysis of a section
broadly similar to the model which was produced by syn- through the Bow Valley structure immediately northwest of
chronous thrusting. The presence of the imbricate stacks on Calgary (Fig. 1). Here all imbricates lie upon the crest and
the forelimbs of the Waterton and frontal Sawtooth structures forelimb of the structure. The upper detachment within the
is thus interpreted to mean that major thrusts within the Jurassic must have been active during emplacement of the
Canadian Front Ranges (on trend with the trailing duplex) Moose Mountain sheets. Since the upper detachment joins
and beneath the Foothills (on strike with the leading duplex) with the deeper portions of the McConnell thrust to the west,
were simultaneously active. Although the southern Canadian application of the synchronous thrust model (Fig. 3) requires
Rocky Mountain and northwest Montana thrust belts may the McConnell remained active or was reactivated during
have initiated by a hinterland-to-foreland propagation of growth of the Bow Valley structure.
thrusting, the model requires that the earliest formed thrusts
remain active, rather than being deactivated as previously
inferred.
383
S. E. BOYER u~)
u.
S
1-
u. s
WATERTON -" PINCHER CREEK ft.
FIELD
3,000 10,000
S.L - -S.L.
^ ^ ^ ^ —-"—^^""""^
3,000 10,000
6,000 Precambrian Basement 20,000
.,
^ LEWIS
9 THRUST
FLATHEAD
DUPLEX
#
WATERTON
'- A DUPLEX
_ H ^km^
PINCHER CK.
9 STRUCTURE
10 0 83 63 41 1 2) 0
k(km)
Figure 6. (a) Cross section through Walerton field (from Bally f/<j/. 1966). There are four major thrust structures: 1) underlying Pincher Creek field is
a frontal thrust with displacement of of less than 1 km, 2) Waterton field duplex whose imbricates have a combined shortening of approximately 11 km, 3)
the Flathead duplex accounting for 20 km shortening, and 4) the Lewis thrust with a minimum displacement of 42 km. The amount of shortening was estimated
fromthesectionsof Bally fr(i/.( 1966). (b). The graph shows the amount of displacement or shortening on each of these four structures, plotted versus distance
L^,. L^jis the distance to each structure, measured in the restored state from the right end of the section. The systematic hindward increase in displacement
suggests that existing western thrusts, such as the Lewis, remained active as new thrusts propagated toward the foreland.
384
GEOMETRIC EVIDENCE FOR SYNCHRONOUS THRUSTING
Discussion
The synchronous thrust model (Fig. 3) explains many of the
conflicts that have arisen among various schools of thought
concerning thrust sequence. In the model, a hinterland-to-
foreland sequence of propagation at depth, combined with
continuing motion of existing thrusts, produces 'out-of-se-
FRONT quence' movements at more shallow levels. Working prima-
RANGES rily with surface data Milici (1975) was most likely to observe
'out-of-sequence' geometries, the shallow manifestation of
synchronous thrusting, whereas Dahlstrom (1970) and Bally
et al. (1966), utilizing abundant subsurface control (seismic
and well data), were more likely to be influenced by the
geometry of deeper structures which suggest a foreland
progression. Thus, both the cratonward and 'hindward'
schools of thrust kinematics may have been in part correct in
their observations and conclusions, as they were dealing with
different parts of the same puzzle. Synchronous thrusting is
compatible with observations that have been incorporated
into the existing model of simple hinterland-to-foreland
thrusting: folded thrusts, 'backlimb' thrusts, interaction of
laterally propagating thrusts, multiple-stage thrusting at two
structural levels, and the 'incidental' reactivation of existing
thrusts.
385
S. E. BOYER
marked by a multitude of Mesozoic imbrications indicate the underlying anticlines, thrust ramps, or the leading edges of
presence of carbonate sheets stacked on top of each other emergent thrusts, are thought to be of minor magnitude (Bally
without intervening Mesozoic clastic strata.' In their inter- et al. 1966, p. 371; Dahlstrom 1970; Woodward 1987).
pretation of the Bow Valley structure they appealed to a two- However, the ubiquity of these structures (Table 1) indicates
stage thrusting wherein the detachment at within the Jurassic that cumulative shortening of such imbricate faults may be
was active first, imbricating the Mesozoic sequence from greater than previously supposed even though the displace-
footwall Palaeozoic carbonates (Bally etal. 1966, p.350-351, ment on single faults may be quite minor. According to the
Fig. 13, phases I & II). Subsequently faults broke from a sole mechanical models of Davis et al. (1983) and Dahlen &
thrust at the base of the Palaeozoic sequence and cut up Suppe (1988) thrust wedges undergo continuous deformation
section through the pre-existing upper detachment (Bally et to maintain a critical taper. Erosion and the accretion of
al. 1966, p.350-351, Fig.l3, phase III). frontal imbricate thrust sheets decreases thrust-belt taper,
As an alternative to the sequential model of Bally et which in turn drives internal deformation required to main-
a/. (1966) this paper argues that the imbricated Mesozoic tain taper. Synchronous thrusting and associated out-of-
elastics are found in conjunction with underlying multiply sequence imbrication provides one means of continuously
imbricated Palaeozoic carbonates because the Jurassic-level maintaining taper.
detachment was active during footwall imbrication as illus-
trated in the synchronous thrust model (Fig. 3). Furthermore,
Geometry and kinematics of duplex fault zones
if the so-called 'back-limb thrusts' and the imbricates on the
forelimbs of the structures are part of the same system, as The ideal model for the development of duplex fault zones
proposed here, the combined displacement on these systems (Boyer 1978; Boyer & Elliott 1982) has failed to explain
may not be as insignificant as previously supposed. many structural complexities found within duplex fault zones
(Davis & Jardine 1984; Fermor & Price 1987). A common
criticism of the existing model is that in order to produce a
Thrust reactivation
'flat-topped' duplex the spacing and displacement of the
One's intuition suggests that it is unrealistic to expect older imbricate faults must have a precise and fixed relationship.
thrusts to be deactivated as movement is transferred to Observed duplexes usually fail to match the model in this
younger thrusts: 'That the pre-existent movement planes in regard. The size and spacing of imbricates is often extremely
the deformed mass would remain totally inert during this irregular (see the section of the Waterton Lakes duplex as
transportation is inherently unlikely: some unsystematic in- adapted from Douglas (1950), by Boyer & Elliott 1982, p.
cidental movement is to be expected.' (Dahlstrom 1970; p. 1206) and duplex geometry may vary from spaced ramp
355). Periodic reactivation would likely result in out-of- anticlines of the 'bumpy-roofed' duplex to the overlapping
sequence and backlimb thrusting. The thrust model of this imbricates of an antiformal stack (Mitra 1986). The mechani-
paper invokes a continuous thrust reactivation, but the peri- cal properties of the rock packages involved within duplexes
ods of reactivation are so closely spaced in time as to give the may explain some of the variability in duplex geometry
appearance of synchronous thrusting. Although the model (Mitra & Boyer 1986).
involves alternating motion between two thrusts, when viewed Alternatively, these various geometries can be produced
in the context of geological time the motions can be thought as a continuous process during synchronous thrusting (Fig.
of as simultaneous. 8). If thrust displacement ceases early in the development of
synchronous thrust duplexes, a bumpy-roof results. Should
shortening continue or be resumed a flat-roof duplex may be
Implications produced. Additional shortening would pile the imbricates to
form an antiformal stack. Note that synchronous-thrust
'Rules' of thrusting, the understanding of thrust-belt duplexes can easily be distinguished from' standard' duplexes
geometries and the hydrocarbon exploration strategies over produced by a simple hinterland-to-foreland sequence of
the past 30 years have generally been based on the hinterland- imbrication; the former will have imbricate faults in the
to-foreland or 'piggy-back' model of thrusting. However, if hanging wall of the roof thrust (Fig. 8) whereas the latter do
the forelimb imbricates in the Rocky Mountains were formed not (compare Fig. 8 of this paper with Fig. 19 of Boyer &
in a systematic fashion indicative of synchronous thrusting, Elliott 1982).
and if such structures are common in other orogens, various
implications arise for the mechanical models of thrust sheet
Sequence of thrusting in other thrust belts
emplacement, the kinematic development of duplex fault
zones, the interpretation of thrust sequence in other thrust The simultaneous thrust/fold model of Gilluly (1960; see
belts, the balancing of cross sections, and the modelling of Figs 2 & 3 of this paper) is substantiated by the existence of
hydrocarbon generation and entrapment in thrust belts. eye-lid windows. Oriel's (1951) description of the Hot
Springs eye-lid window (North Carolina) indicates that it
could be interpreted using Gilluly's model, and Rodgers
Mechanical models of thrust emplacement
(1970, p. 170) argued for a break-back sequence above
Out-of-sequence imbricates, whether they be associated with possible eye-lid windows of the southern Appalachians, such
386
GEOMETRIC EVIDENCE FOR SYNCHRONOUS THRUSTING
®
iJ^iit—Jiii^^^i^l^^lii^i^^^^^i^iii-LLij^^i^i^mi^^^
. I . I . I . I . I . '• • \ ^ • ^ • . • I
J2)
.® ® ®
Figure 8. Formation of a duplex fault zone by synchronous thrusting. The model of Figure 3 has been applied to four imbricate faults. With continuous
movement on the four thrusts the amplitude of each ramp anticline increases and younger out-of-sequence imbricate faults break across the crest of each
structure. Note that in the initial stages of displacement (b) ramp anticlines are spaced, but with increased shortening synclines tighten (c) and eventually
the synclines disappear as a flat-topped duplex is formed (d). With additional displacement the flat-topped duplex of section d would be transformed into
an antiformal stack (Boyer & Elliott 1982, Fig. 12; Mitra 1986, Fig. 4). The numbers appended to the out-of-sequence imbricates indicate the relative ages
of these faults,' 1' being the oldest and ' 3 ' the youngest. Note that the imbricates comprise three distinct packages, each of which contains imbricates that
become younger toward the hinterland. A 'piggy-back' duplex can be distinguished from a 'synchronous-thrust' duplex by tfie lack or presence of minor
imbricate faults in the hangingwall of the roof thrust. Compare the evolution of the synchronous thrust duplex with that of a 'normal' duplex (Fig. 19 of
Boyer & Elliott 1982).
387
S. E. BOYER
as the Mountain City window. Rodgers (1970, p. 173) also between hydrocarbon generation and entrapment than will
noted that the synchronous thrust-fold model of Gilluly sequential thrusting. In sequential thrusting, hydrocarbons
(1960) might be applied to the Grandfather Mountain win- are likely to be expulsed ahead of the deformation front and
dow but felt that it was not required. Boyer & Elliott (1982) are unlikely to be trapped within the thrust belt. In this case
ignored the 'eye-lid' nature of the upper thrusts surrounding most hydrocarbon-bearing structures will be found outside
these two structures. A more realistic model incorporating the thrust belt, perhaps on forebulges associated with the
synchronous duplex models (Fig. 8) rather than the sequential thrust load. Synchronous thrusting produces traps concurrent
model (Boyer & Elliott 1982, Fig. 19), would be compatible with generation so that most hydrocarbons will be trapped
with the observations and conclusions of Oriel (1951) and intemal to the thrust-belt (Boyer 1990). It is important that
Rodgers (1970). the sequence of thrusting be well understood, so that the
The synchronous thrust model which has been presented correct model can be applied.
for southern Alberta and northwest Montana is also compat-
ible with additional structural and stratigraphic evidence
indicating synchronous and out-of-sequence thrusting in Conclusions
several other thrust belts (Ori & Friend 1984; Searle 1985;
Butler 1987; Morley 1988; Roeder 1988). The occurrence of imbricate faults sited predominantly on the
crests and forelimbs of structural culminations in the thrust
belts of southern Alberta and northwest Montana suggest that
Cross-section balancing
numerous thrust faults have been synchronously active. This
Inherent in many cross-section balancing techniques is the contradicts existing models of piggyback thrusting. Kin-
assumption that thrusts have developed in 'piggy-back' se- ematic arguments for synchronous thrusting are fortified by
quence, one of the standard 'rules of thrusting'. Based on this patterns of a cross-strike decrease in thrust displacement and
assumption, oversteepened faults in a restored cross section along-strike displacement transfer. When existing thrusts
may either reflect out-of-sequence imbrication (Dahlstrom remain active rather as new thrusts propagate toward the
1970, see his discussion of the Moose Mountain culmination, foreland, older thrusts near the hinterland display greater
p. 353, and his section of the structure. Fig. 16) or indicate that displacement.
the section is not balanced (Woodward et al. 1985; 1989). The synchronous thrust model (Fig. 3) appears compatible
Therefore, to properly balance a cross section, out-of-se- with most observations of the geometry of Rocky Mountain
quence thrusts must be identified. If they are not, unrealistic thrust belts (Bally et al. 1966; Dahlstrom 1970; Royse et al.
fault trajectories produced in restorations will be attributed to 1975) and many of the interpretations concerning thrust
error, rather than the true cause, out-of-sequence thrusting, sequence (Douglas 1950; Hume 1957). Furthermore, it
and attempts will be made to revise the section which is reconciles many of the outstanding differences between the
already correct. Should synchronous thrusting prove to be majority school of the hinterland-to-foreland sequence and
more common than piggy-back thrusting, many 'balanced' the much maligned adherents of the hind ward thrust progres-
sections may actually be incorrect. Indeed, if synchronous sion. With a growing body of evidence that several thrusts
and out-of-sequence thrusting are mechanical requirements, may be synchronously active within thrust belts, we can no
Morley (1988, p. 557) has even suggested that 'restored longer accept various piggy-back thrust models (Bally et al.
sections that do not contain OOST's (out-of-sequence thrusts 1966; Dahlstrom 1970; Boyer & Elliott 1982) as constituting
which appear as zig-zag patten in restored sections) may be 'rules'(Royse ef a/. 1975) ofthrust belt behaviour. Although
the inadmissible sections!' the piggyback model may still apply in a number of settings
it can no longer be assumed to have universal applicability.
These conclusions have several implications. They sug-
Hydrocarbon generation and entrapment
gest that in many instances synchronous thrust models should
Utilization of burial history profiles and time-temperature be applied to the development of duplex fault zones, as they
indexing (Lopatin 1971; Waples 1980; Waples 1981, p. 95- provide an explanation for variation in duplex geometries
106) to predict the relative timing of hydrocarbon generation (spaced ramps, flat-roof, and antiformal stacks). Evidence
and trap development has been a useful technique in under- for synchronous thrusting is compatible with mechanical
standing existing hydrocarbon-bearing structures and in the models of thrust emplacement (Davis et al. 1983), which
exploration for new fields (Warner 1983). However, the require that thrust belts shorten internally to balance de-
accuracy of the technique is dependent upon a correct knowl- creased taper resulting from erosion and frontal accretion.
edge of the sequence of thrusting (Boyer 1990). Most models There are economic implications of this model as well. Most
of hydrocarbon generation and entrapment in thrust belts basin modelling and thermal maturation profiling routines in
assume a simple hinterland-to-foreland sequence of thrust- thrust belts are based on the piggyback model and therefore
ing, a model that has been reinforced by more than 30 years need to be modified when applied to regions of synchronous
of thrust belt exploration. However, all other factors being thrusting.
equal, synchronous thrusting will produce a different timing
388
GEOMETRIC EVIDENCE FOR SYNCHRONOUS THRUSTING
The observations which led to this paper resulted from discussions with I wish to acknowledge my gratitude to all my industry friends and colleagues
Ricardo Presnell during the early 1980s, when he and I were employed by with whom I worked during this period, as much of my past and present
SOHIO (Standard Oil Production Co., now HP Exploration) in Denver. thrust-belt research has benefited from this collaboration. Observations made
While we were attempting to interpret a number of seismic lines and tie in this paper served as the nucleus for research directed toward evaluating the
geological maps to subsurface data, I was struck by the odd appearance of a applicability of the 'critically-tapered-wedge' model to thrust tectonics in
'folded' imbricate stack (Fig. 5, as mapped by Mudge & Earhart 1983) which the Rocky Mountains. I gratefully acknowledge the National Science
brought to mind the work of Gilluly (1960) in Nevada. Those observations Foundation which is supporting this research under grants EAR-88-03623
led to this paper. I am grateful for nine years of employment in the petroleum and EAR-90-04303. I wish to thank Bob Holdsworth and Rob Butler, who
industry which permitted me to work throughout the thrust belts of the reviewed the manuscript and provided much constructive criticism. I've
western US. Most beneficial has been the fieldwork conducted while attempted to incorporate as many of their suggestions as possible and hope
employed by ARCO during 1978 and 1979 and the seismic interpretation and that the result is a more comprehensible paper.
cross-section construction performed with SOHIO from 1981 to 1987.
References
Armstrong, F. C. & Oriel, S. S. 1965. Tectonic development of Idaho- Fermor, P. R. & Price, R. A. 1987. Multiduplex structure along the base of
Wyoming thrust belt. American Association of Petroleum Geologists the Lewis thrust sheet in the southern Canadian Rockies. Bulletin of
Bulletin, 49, 1847-1866. Canadian Petroleum Geology, 35, 159-185.
Bally, A. W., Gordy, P. L. & Stewart, G. A. 1966. Structure, seismic data, and Fox, F. G. 1959. Structure and accumulation of hydrocarbons in southern
orogenic evolution of southern Canadian Rocky Mountains. Bulletin of foothills. Alberta, Canada. Bulletin of the American Association ofPe-
Canadian Petroleum Geology, 14, 337-381. troleum Geologists, 43, 992-1025.
Boyer, S. E. 1978. Structure and origin of the Grandfather Mountain Gallup, W. B. 1955. Geology of the Pincher Creek gas and naptha field and
window. North Carolina. Ph.D. thesis, Johns Hopkins University. its regional implications. Billings Geological Society 6th Annual Field
1986a. Geometric evidence for simultaneous, rather than sequential, Conference, 150-159.
movement on major thrusts; implications for duplex development. Gordy, P. L., Frey, F. R. & Norris, D. K. 1977. Geological field guide for the
Geological Society of America Abstracts with Programs, 18, 549. CSPG 1977 Waterton-Glacier Park Field Conference. Canadian Society
1986b. Styles of folding within thrust sheets: examples from the of Petroleum Geologists, Calgary.
Appalachian and Rocky Mountains of the USA and Canada. Journal of Gilluly, J. 1960. A folded thrust in Nevada - inferences as to time relations
Structural Geology, 8, 325-339. between folding and faulting. American Journal of Science, 258-A, 68-
1990. New models for thrust kinematics and implications of the 79.
generation, migration, and entrapment of hydrocarbons in fold and thrust Hage, C. O. 1942. Folded thrust faults in Alberta Foothills west of Turner
be\ts(sibst.). American Association ofPetroleum Geologists Bulletin,14, Valley. Royal Society of Canada Transactions. 3rd series, section 4,36,
617. 67-78.
& Elliott, D. 1982. Thrust systems. /Imer/con Association ofPetroleum Hake, B. F., Willis, R. & Addison, C. C. 1942. Folded thrust faults in the
Geologists, 66, 1196-1230. Foothills of Alberta. Geological Society of America Bulletin, S3,291 -334.
& Geiser, P. 1987. Sequential development of thrust belts: implications Heller, P. L., Bowdler, S. S., Chambers, H. P., Coogan, J. C , Hagen, E. S.,
for mechanics and cross section balancing. Geological Society of Shuster, M. W., Winslow, N. S. & Lawton, T. F. 1986. Time of initial
America Abstracts with Programs, 19, 597. thrusting in the Sevier orogenic belt, Idaho-Wyoming and Utah. Geol-
Butler, R.W.H. 1987. Thrust sequences. Journal of the Geological Society, ogy, 14, 3SS-391.
London, 144, 619-634. Hennessey, W.J.I 975. A brief history of Savanna Creek gas field. Canadian
Choquette, A. L. 1959. Theoretical approach to foothills and mountain Society ofPetroleum Geology Guidebook, Exploration Up-date '75, 18-
deformation of western Canada. Alberta Society ofPetroleum Geologists 21.
Journal, 1, 234-237. Hubbert, M. K. & Rubey, W. W. 1959. Role of fluid pressure in mechanics
Choukroune, P. & Seguret, M. 1973. Tectonics in the Pyrenees: role of of overthrust faulting: I. Mechanics of fluid-filled porous solids and its
compression and gravity. /«: De Jong, K. A. & Scholten, R. (eds) Gravity applications to overthrust faulting. Geological Society of America Bul-
and Tectonics. Wiley, New York, 141-156. letin,10, 115-166.
Dahlen, F. A. & Suppe, J. 1988. Mechanics, growth, and erosion ;of Hume, G. S. 1957. Fault structure in the Foothills and eastern Rocky
mountain belts. Geological Society of America Special Paper, 218,161- Mountains of southern Alberta. Geological Society of America Bulletin,
178. 68,395-412.
Dahlstrom,C.D.A. 1969.Balanced crof,% section?,. Canadian Journal ofEarthHunt, C. W. 1956. Panther dome: a minor orogen of the Canadian Cordillera.
Sciences, 6, 743-757. Alberta Society of Petroleum Geologists 6th Annual Field Conference
1970. Structural geology in the eastern margin of the Canadian Rocky Guidebook, 44-55.
Mountains. Bulletin of Canadian Petroleum Geology, 18, 332-406. Hurst, D. J. & Steidtmann, J. R. 1986. Stratigraphy and tectonic significance
Davis, D., Suppe, J. & Dahlen, F. A. 1983. Mechanics of fold-and-thrust belts of the Tunp conglomerate in the Fossil Basin, southwest Wyoming. The
and accretionary wedges. Journal of Geophysical Research, 88, 1153- Mountain Geologist, 23, 6-13.
1172. Jones, P. B. 1971. Folded faults and sequence of thrusting in Alberta
Davis, G. A. & Jardine, E. A. 1984. Preliminary studies of the geometry and foothills. American Association ofPetroleum Geologists Bulletin, 55,292-
kinematics of the Lewis allochthon, Saint Mary Lake to Yellow Moun- 306.
tain, Glacier National Park, Montana. Montana Geological Society Field Knipe, R. J. 1985. Footwall geometry and the rheology of thrust sheets.
Conference & Symposium .Northwest Montana & Adjacent Canada, 201 - Journal of Structural Geology, 7, 1-10.
209. Lageson, D. R. 1984. Structural geology of Stewart Peakculmination, Idaho-
Deiss, C. F. 1943. Stratigraphy and structure of southwest Saypo quadrangle, Wyoming thrust belt. American Association of Petroleum Geologists
Montana. Geological Society of America Bulletin, 54, 205-262. Bulletin, 6S, 401-416.
Delphia, J. G. & Bombolakis, E. G. 1988. Sequential development of a Lamerson, P. R. 1982. The Fossil basin and its relationship to the Absaroka
frontal ramp, imbricates, and a major fold in the Kemmerer region of the thrust system, Wyoming and Utah. Rocky Mountain Association of Ge-
Wyoming thrust belt: Geological Society of America Special Paper,222, ologists, Geologic Studies of the Cordilleran thrust belt, 279-340.
207-222. LeMoine M. 1973. About gravity gliding tectonics in the western Alps: In:
Douglas, R.J.W. 1950. Galium Creek, Langford Creek, and Gap map areas. De Jong, K. A. & Scholten, R. (eds) Gravity and Tectonics. Wiley, New
Alberta. Geological Survey of Canada Memoir 255, 124p. York, 201-216.
1958. Mount Head map-area. Alberta. Geological Survey of Canada Lopatin, N. V. 1971. Temperature and geologic tirne as factors in coalification
Memoir 291. (in Russian). Izv. Akad. Nauk SSSR, Seriya geologicheskaya, 3,95-106.
Erdman, O. A., Belot, R. E. & Slemko, W. 1953. Pincher Creek area. Alberta. Milici, R. C. 1975. Structural patterns in the southern Appalachians: evi-
Alberta Society ofPetroleum Geologists 3rd Annual Field Conference dence for a gravity slide mechanism for the Alleghanian deformation.
and Symposium, 139-157. Geological Society of America Bulletin, 86, 1316-1320.
389
S. E. BOYER
Mitra, G. & Boyer, S. E. Energy balance and deformation mechanisms of Rodgers, J. 1970. The tectonics of the Appalachians. John Wiley & Sons,
duplexes. Journal of Structural Geology, 8, 291 -304. New York, 271 p.
Mitra, S. 1986. Duplex structures and imbricate thrust systems: geometry, Roeder, D. 1988. Andean-age structure of Eastern Cordillera (Province of La
structural position, and hydrocarbon potential. American Association of Paz, Bolivia). Tectonics, 1, 23-39.
Petroleum Geologists Bulletin, 70, 1087-1112. Royse, F., Jr., Warner, M. & Reese, D. L. 1975. Thrust belt structural
Morley, C. K. 1988. Out-of-sequence thrusts. Tectonics, 7, 539-561. geometry and related stratigraphic problems, Wyoming-Idaho-northem
Morse, J. 1977. Deformation in ramp regions of overthrust faults: experi- Utah: In: Bolyard, D.W. (ed.) Deep Drilling Frontiers of the Central
ments with small-scale rock models. Wyoming Geological Association Rocky Mountains. Rocky Mountain Association of Geologists, 41-54.
29th Annual Field Conference Guidebook, 457-470. Scholten, R. 1973. Gravitational mechanisms in the northern Rocky Moun-
Mudge, M. R. 1970. Origin of the Disturbed Belt in northwestern Montana. tains of the United States: In: De Jong, K. A. & Scholten, R. (eds) Gravity
Geological Society of America Bulletin, 81, 377-392. and Tectonics. Wiley, New York, 473-489.
& Earhart, R. L. 1980. The Lewis thrust fault and related structures in Scott, J. C. 1951. Folded faults in Rocky Foothills of Alberta, Canada.
the Disturbed Belt, northwestern Montana. US Geological Survey Pro- American Association of Petroleum Geologists Bulletin, 35,2316-2347.
fessional Paper 1174 , 18p. Searle, M. P. 1985. Sequence of thrusting and origin of culminations in the
Mudge, M. & Earhart, R. L. 1983. Bedrock geologic map of part of the northern and central Oman Mountains. Jouraa/ ofStructural Geology, 7,
northern Disturbed Belt, Lewis and Clark, Teton, Pondera, Glacier, 129-143.
Flathead, Cascade, and Powell Counties, Montana. US Geological Sur- Serra, S. 1977. Styles of deformation in ramp regions of overthrust faults.
vey Map 1-1375, scale 1:125,000. Wyoming Geological Association 29th Annual Field Conference Guide-
Nickelsen, R. P. 1988. Structural evolution of folded thrusts and duplexes on book, 487-498.
a first-order anticlinorium in the Valley and Ridge Province of Penn- Stockmal, G. S. 1979. Structural geology of the northern termination of the
sylvania. Geological Society of America Special Paper, 222, 89-106. Lewis thrust. Front Ranges, southern Canadian Rocky Mountains. M.Sc.
Ori, G. G. & P. F Friend. 1984. Sedimentary basins formed and carried thesis. The University of Calgary, Alberta (unpubl.).
piggyback on active thrust sheets. Geology, 12, 475-478. Suppe, J. 1983. Geometry and kinematics of fault-bend folding. American
Oriel, S. S. 1951. Structure of the Hot Springs window, Madison County, Journal of Science, 283, 684-721.
North Carolina. American Journal of Science, 249, 1-30. Verrall, P. 1968. Observations on geological structure between the Bow and
& Armstrong, F. C. 1966. Times of thrusting in Idaho-Wyoming thrust North Saskatchewan rivers. Alberta Society of Petroleum Geologists
belt: Reply. American Association ofPetroleum Geologists Bulletin, 50, 16th Annual Field Conference Guidebook, 106-118.
2614-2621. Waples, D. 1980. Time and temperature in petroleum exploration: applica-
Piatt, J. P. 1988. The mechanics of frontal imbrication: a first-order analysis. tion of Lopatin's method to petroleum exploration. American Associa-
Geologische Rundschau, 11, 577-589. tion of Petroleum Geologists Bulletin, 64, 916-926.
Price, R. A. 1965. Flathead map-area, British Columbia and Alberta. 1981. Organic geochemistry for exploration geologists. Burgess
Geological Survey of Canada Memoir 336, 221 p. Publishing Co., 95-106.
1973. Large-scale gravitational flow of supracrustal rocks, southern Warner, M. A. 1983. Source and time of generation of hydrocarbons in the
Canadian Rockies: In: De Jong, K. A. & Scholten, R. (eds) Gravity and Fossil Basin, western Wyoming thrust belt. Rocky Mountain Association
Tectonics. Wiley, New York, 491-502. of Geologists Geologic Studies of the Cordilleran thrust belt, 805-815.
1981. The southern foreland thrust and fold belt in the southern Wiltschko, D. V. & Dorr, J. A., Jr. 1983. Timing of deformation in overthrust
Canadian Rocky Mountains: In: McClay, K. R. & Price, N. J. (eds) belt and foreland of Idaho, Wyoming, and Utah. American Association
Thrust and Nappe Tectonics, Geological Society of London Special of Petroleum Geologists Bulletin, 67, 1304-1322.
Publication, 9, 427- 448. Woodward, N. B. 1987. Geological applicability of critical-wedge thrust-
et al. 1972. The Canadian Rockies and tectonic evolution of the belt models. Geological Society of America Bulletin, 99, 827-832.
southeastern Canadian Cordillera. XXIV International Geological Con- , Boyer, S. E. & Suppe, J. 1985. An outline of balanced cross-sections.
gress, Excursion AC 15. University of Tennessee Studies in Geology, 11, 2nd edition, 170p.
Rich, J. L. 1934. Mechanics of low-angle overthrust faulting illustrated by , & . 1989. Balanced geological cross-sections. American
Cumberland thrust block, Virginia, Kentucky and Tennessee. American Geophysical Union Short Course in Geology, 6, 132p.
Association of Petroleum Geologists Bulletin, 18, 1584-1596.
390
The analysis of fracture systems in subsurface thrust structures from
the Foothills of the Canadian Rockies
Mark Cooper
BP Resources Canada Limited, #2100, 855-2nd Street SW, Calgary, Alberta, Canada
T2P 4J9
Abstract: The Foothills of the Canadian Rocky Mountains are one of the classic examples of a
thin-skinned fold-thrust belt. A regionally significant detachment in the shales of the Jurassic
Femie Formation separates the intensely folded and imbricated elastics of the Jurassic and
Cretaceous from the thrust structures in the prospective dolomite reservoirs of the Triassic and
Upper Palaeozoic. As the foothills are traced northwards from Alberta to British Columbia the
amount of displacement on the thrusts in the foothills decreases. Occasionally, far-travelled thrust
sheets achieve sufficient elevation above regional to cause outcrop of the reservoir units. In these
structures, which are analogues of the prospective structures at depth, the mesoscopic structures
and fabrics can be studied; the opportunities to examine the structures in the subsurface by remote
techniques has been much more limited. The advent of new well-logging technology such as the
Formation Microscanner allows the geometry of subsurface foothills structures to be analysed in
considerable detail. The exact geometric relationships of both bedding and fracture systems can
be elucidated in-situ and correlated with core data. Two case studies using these techniques are
presented; a tip fold from the British Columbia Foothills and a ramp anticline from the Alberta
Foothills. The analysis reveals a systematic relationship between fracture set orientation and
density to both structural geometry and lithological variations.
The Foothills of the Canadian Rocky Mountains have been nated by numerous folds and small displacement imbricate
explored for oil and gas since the early part of this century. thrusts within the Mesozoic elastics. The intensity of the
This has resulted in an extensive literature on the fold and folding and the magnitude of the thrust displacements gener-
thrust belt which includes some of the classic papers on thrust ally decreases to the north.
geometry and techniques of analysis (e.g. Douglas 1950, In Alberta the closely spaced folds and imbricates in the
1958; Bally et al. 1966; Dahlstrom 1969, 1970; Price & Mesozoic clastic rocks obscure the simpler geometry of the
Mountjoy 1970 and Thompson 1979). The Canadian Rockies Palaeozoic carbonate thrust sheets in the subsurface. The
rarely exceed 200 km in width and form part of the 5000 km abundance of seismic and well data from gas exploration
long North American Cordilleran fold and thrust belt that activities provides the constraints to develop quite detailed
extends from SE California to Alaska. Development of the geometric models of these structures. The Palaeozoic car-
fold and thrust belt commenced in the Late Jurassic and bonate thrust sheets tend to be widely spaced, relatively
terminated in the Eocene, and is conventionally regarded as simple structures developed above shallow angle ramps and
the result of the accretion of a series of exotic terranes to the often have significant displacements. This is in contrast to the
Pacific Margin of North America ( Monger et al. 1982). A closely spaced, steeply dipping ramp anticlines and tip folds
good, recent review of the fold and thrust belt is given by of the Mesozoic and implies a detachment in the Jurassic
McMechan & Thompson (1989). Femie Formation between these two structural regimes.
The Foothills belt lies between the Front Ranges, a series As the foothills are traced northwards into British Colum-
of large thrust sheets of Palaeozoic rocks to the west, and the bia (Fig. 2), the amount of displacement on the thrusts
largely undeformed Western Canadian Sedimentary Basin to gradually decreases, thrusts at surface are less common and
the east (Figs 1 & 2). The eastern boundary is characterized folds predominate (McMechan & Thompson 1989). The
by a frontal monocline and/or a triangle zone (Jones 1982) change in style is gradual and is concomitant with a more
which becomes a less reliable indicator of the limit of de- diffuse eastern limit of deformation. Small displacement
formation as the belt is traced northwards. The boundary with thrusts at depth persist much further to the east but have little
the Front Ranges is characterized by a major displacement expression at the surface. In the southern foothills of NE
thrust with Palaeozoic and/or Proterozoic strata in the British Columbia, the primary exploration targets are the
hangingwall (e.g. the McConnell and Lewis thrusts. Fig. 1). carbonates of the Upper Triassic which are deformed into a
Within the Foothills belt thrust sheets with outcrops of Upper series of northeast vergent folds with steeply dipping north-
Palaeozoic rocks are locally observed (e.g. the Livingstone east limbs that are locally overtumed (Barss & Montandon
and Brazeau Thrusts, Fig. 1) but surface structure is domi- 1981).
391
M. COOPER
392
ANALYSIS OF FRACTURE SYSTEMS IN THRUST STRUCTURES, FOOTHILLS CANADIAN ROCKIES
compression. Wiltschko et al. (1985) measured strain by Example 1: Fracture development in a tip fold
calcite twin lamellae in the Mississippian Newman Limestone
of the Pine Mountain block which shows early layer parallel Regional setting
shortening and subsequent inter-bed shear. Spang & Groshong
This example is located in the southern portion of the NE
(1981), also working in the Appalachians, concluded that the
British Columbia sector of the foothills in the Monkman area
maximum compressive strain was always sub-parallel to the
(Fig. 2) where the prolific gas fields of Sukunka and Bullmoose
layering and perpendicular to the fold axis with the maximum
occur (Barss & Montandon 1981). The reservoir units in this
extension sub-parallel to the fold axis. A very similar
area are the dolomites of the Upper Triassic Pardonet and
observation was made by Spang et al. (1981) above thrust
Baldonnel Formations (Fig. 3). The Triassic is deformed by
ramps in Devonian and Mississippian carbonates of the Front
a series of thrust imbricates, which detach in the anhydrite
Ranges in the Canadian Rockies where the maximum prin-
beds of the Charlie Lake Formation, and tip (fault propaga-
cipal strain (averaging 6%) is layer parallel or at a small angle
tion) folds; typical sections through these structures are given
to bedding. This provides particularly useful and relevant by Barss & Montandon (1981). The tip folds have up to 800
data on the state of strain in structures analogous to those in m of amplitude and usually involve the Halfway Formation
the foothills described in this paper. (Fig. 3) which is penetrated in structure in some of the wells.
The geometric constraints provided by the well and seismic
data suggest that the detachment is within the shales of the
Toad-Grayling Formation (Fig. 3). The seismic mapping of
the Triassic structures is complicated by the presence of
detachments within the numerous shale units of the Upper
Cretaceous section (McMechan 1985, and Fig. 3) and the
strongly deformed sands and shales of the Nikanassin Group.
The detailed 1:50000 scale maps of the area produced by the
British Columbia Department of Mines (Kilby & Wrightson
1987) illustrate the low amplitude, short wavelength folds
developed in the Cretaceous above some of these detach-
ments. The shallower deformation makes it difficult to
extrapolate surface structure to depth with any confidence; it
also distorts the seismic ray paths thus making seismic
imaging difficult.
ZOHl 1
ZONE 2
ZONE 3
ZONE 4
ZONE 5
ZONE 6
ZONE 7
. ZONE 8
' ZONE 9
ZONE 10
ZONE 11
Figure 3. Mesozoic and Upper Palaeozoic stratigraphy of the Monkman Figure 4. Detailed structural cross-section through the lower portion of the
area. The shading within the map column shows the stratigraphic subdivi- Example 1 well showing dip data, faults and stratigraphic boundaries. The
sions used in Figure 2. subdivision of the structure into the 11 dip domains (structural zones)
discussed in the text is also indicated.
393
M. COOPER
394
ANALYSIS OF FRACTURE SYSTEMS IN THRUST STRUCTURES, FOOTHILLS CANADIAN ROCKIES
-iVUv. ;[
' r.t:
^^t;.; ^'j.-. :
395
M. COOPER
396
ANALYSIS OF FRACTURE SYSTEMS IN THRUST STRUCTURES, FOOTHILLS CANADIAN ROCKIES
fractures are orthogonal to bedding but have a strike parallel is derived from the spherical variance of the data set and
to the fold axis. The fracture data from the core (in zone 2) provides a measure of the dispersion of the data about the
are also included and show a similar distribution of fractures mean vector, a high value indicates low dispersion. With the
to the FMS data. The raw data are difficult to interpret and to exception of the Set 1 fractures in Zones 1 and 3 all the Rbar
clarify geometric relationships the mean orientations of bed- values exceed the critical values at a 99% level of signifi-
ding and of each fracture set have been calculated using the cance; this indicates that, with these exceptions, all of the data
statistical methods of Mardia (1972) and Woodcock (1977); sets are not uniformly distributed but show clustering about
the results are summarized in Table 1. In each of the dip the mean (Davis 1983). The relationships of the fracture sets
domain zones the mean azimuth and dip are given with the to the fold are conveniently illustrated by plotting the mean
number ofreadings and the 'Rbar' value. The Rbar parameter orientations of the fracture sets as great circles together with
the best fit great circle to the bedding poles (the theoretical ac
plane of the fold) and the calculated fold axis (Fig. 9). The
ZONE W ZONE n CORE
stereonet derived by this method from Zone 2 illustrates how
the Set 3 fractures tend to be orthogonal to bedding (parallel
to the fold axis) and how the Set 1 fractures are closely parallel
to the theoretical ac plane. In Zones 5-7 the Set 1 fractures
appear to diverge further from the theoretical ac plane of the
structure (see Fig. 8). As these zones cover most of the steeply
ZONE 7 ZONE 8 ZONE 9 dipping part of the structure this may be due to increased
strain on the steep limb. However, the apparent differences
in the mean orientations are probably caused by the small
IJ number ofreadings for which the means are being calculated
(Table 1).
The Set 3 fractures in Zone 2 are dispersed over approxi-
mately 60° of arc (Fig. 8) which suggests that they are the
ZONE 4 ZONE 5 ZONE 6
'j\
Figure 8. Equal-area stereographic projections for the 11 structural zones Figure 9. Simplified equal-area stereographic projection of the data from
in the well. Bedding poles; open squares. Set 1 fracture s; open circles. Set Zone 2. The mean orientations of the bedding. Set 1 fractures and Set 2
2 fractures;crosses. Set 3 fractures; filled squares. The data from the core cut fractures are plotted as great circles together with the fold axis and the
in Zone 2 is shown as a separate plot. theoretical ac plane of the fold.
397
M. COOPER
398
ANALYSIS OF FRACTURE SYSTEMS IN THRUST STRUCTURES, FOOTHILLS CANADIAN ROCKIES
399
M. COOPER
SOUTH-WEST
MISSISSIPPIAN
Figure 12. Generic cross-section through the Alberta Foothills to illustrate the various structural types in the Upper Palaeozoic carbonates. All the wells
shown are west of the 5th meridian. The location of the section is shown on Figure 1 (based on a section by Ray Widdowson).
• * ^
1^/
Figure 13. Typical stratigraphic column for the Alberta Foothills. The
shading in the right hand column of the figure shows the stratigraphic
subdivisions used in Figure 1.
/ K eiACK.-ONE
has shown much higher fracture density on the steeply
dipping west limb than on the gently dipping east limb. _-— -- -- \ ^^
f£BNPE * fiOBI>E.gg — - -z^^^^^C^
^__-;;;;ij-^^gjrVTOBD£K^__
"u~~iiJ''"'?--''-'-'"----""""~~~ ~ ~
'~~H SHUN"*
Regional setting
Figure 14. Detailed structural cross-section through the Example 2 well
The second example is located in the Alberta Foothills to the showing dip data, stratigraphic tops and major faults. The structure of the
west of Calgary (Fig. 1). The productive Palaeozoic carbonate Mesozoic rocks in the well is discussed in the text.
400
ANALYSIS OF FRACTURE SYSTEMS IN THRUST STRUCTURES, FOOTHILLS CANADIAN ROCKIES
Figure 15. Equal area stereographic projection of 120 poles to bedding from Figure 17. Simplified equal area stereographic projection of the data from
the Mississippian section of the Example 2 well. The spherical variance of Zone 3. The mean orientations of the bedding and the three fracture sets are
the data is 0.0233 reflecting the strongly unimodal distribution of the data. plotted as great circles together with the fold axis and the theoretical ac plane
A best-fit great circle can be fitted to the data and indicates a fold axis of the fold.
plunging 3° to 334°.
401
M. COOPER
the relationships of the fractures to the fold structure (Figs 16 also summarizes the response of the array sonic tool to the
& 17). Three sets of fractures can be seen in the well; Set 1 fractured intervals. Fractures within the rock will produce
has a strike slightly oblique to the ac plane of the fold but dips characteristic effects on the source waveform that can be
moderately to steeply northwest and are thus approximately detected and analysed thus providing a method for fracture
perpendicular to the fold axis (Fig. 17). The other two sets are identification. The main effects of a fracture are to slow the
approximately symmetrical about the plane perpendicular to source wave, to reduce the amplitude of the wave and to
bedding and a strike sub-parallel to orogenic strike and the attenuate the energy of the wave. The standard Schlumberger
fold axis (Fig. 17). The subdivision of these two sets was processing presents the data in 2 alternative ways. The first
arbitrarily based on the orientation data; the data could as shows the phase of the wave detected by Receiver 1 to a
easily have been subdivided into three sets. The scatter recording time of 10000 microseconds; this is a long enough
suggests strongly that these fracture sets are Steams (1968) recording time to see the Stoneley waves (Fig. 19). Fractures
Type 2 fractures; a conjugate set of fractures with an extension tend to produce a characteristic chevron pattern due to reflec-
fracture set along the acute bisectrix which is sub-parallel to tion at the fracture interface. The energy attenuation can also
the fold axis. be calculated and is displayed in Figure 19; this is expressed
in decibels per metre and the attenuation effect decreases with
time as the slower waves are detected. The other display
Fracture distribution and origin shows phase and amplitude plotted against recording time,
The distribution of the fractures in the Mississippian rocks (Fig. 19) fractured zones cause a phase shift to the right due
penetrated by the well is summarized in Figure 18, which to the slowing of the waveform and hence a later recording
shows that the fractures occur throughout the well but be- time at the receiver. The amplitude display shows attenuation
come less frequent in the lower part of the well. This diagram of the amplitude in fractured zones. The more fractured the
rock the more attenuated the wave energy becomes. This tool
correlates quite well with the FMS fracture logs (Figs 10 &
18).
From the perspective of hydrocarbon production the ac-
tual number of fractures that intersect the wellbore is the most
important factor. However, this does not yield the true
fracture spacing as this is dependent on the angular relation-
ships between the wellbore and the fractures; a fracture with
a dip azimuth parallel to the wellbore has much less chance
of being intersected by the well than one which is at a high
angle to the wellbore. The true fracture spacing will thus be
a simple function of apparent fracture spacing identical to the
relationship between true and apparent dip.
The relationship of the conjugate fracture set to the ramp
anticline is illustrated by a block diagram (Fig. 20). The Set
1 fractures are omitted from the diagram for clarity. The
conjugate fractures form in orientations that conform closely
with the Type 2 conjugate shear fractures of Stearns (1968)
which are considered to form later in the folding history
(Stearns & Friedman 1972) when the bending strains cause
extension in bedding perpendicular to the fold axis. Similar
conjugate sets have been described in other Laramide folds in
the Foothills of the Canadian Rockies (Muecke &
Charlesworth 1966; Price 1967;Norris 1971) and in Montana
(Burger & Thompson 1970; Steams 1968). The Set 1
fractures may be extensional fractures related to the sharp
south-southeast plunge of the fold known from seismic data.
If so, they must form at a different time to the conjugate set
as the stress regimes are mutually exclusive. As in the case
Figure 18. Summary log of the Example 2 well showing the distribution of of Example 1 the development of the fractures is considered
the three fracture sets in relation to lithology and bedding dip. The to be enhanced by high pore fluid pressure levels during
subdivision of the Turner Valley Formation follows the scheme of Rupp Laramide deformation.
(1969). The location of the structural zone boundaries and the portion of the
well illustrated by Figure 19 are indicated in the zone column. The number
of fractures of each set are recorded as histograms with a 5 m sample interval;
the number of fractures measured in the cored intervals are indicated by
cross-hatching.
402
ANALYSIS OF FRACTURE SYSTEMS IN THRUST STRUCTURES, FOOTHILLS CANADIAN ROCKIES
tooo A,
0)
B
m
H
0. 0«0
40 00 0. 071 C
0. (H*t O
0. OS*
0. 04»
o. o-^o
0. 033
0. 014
a. Dtt
0, 008
••••"^^wPffl^^ffS^
Figure 19. An example of the data output from the array sonic togging tool showing the effect of the fractures on the phase, amplitude and energy attenuation
of the source wavelet. The right hand phase column is recorded for a longer time and shows the chevron pattern effect on the phase of the Stonely waves.
403
M. COOPER
References
Allmendinger, R. W. 1982. Analysis of microstructures in the Meade Plate 1970. Structural Geology in the eastern margin of the Canadian Rocky
of the Idaho-Wyoming Foreland Thrust belt, US A. Tectonophysks, 85, Mountains. Bulletin of Canadian Petroleum Geology, 18, 332-406.
221-251. Davis, D. C. 1983. Statistics andData Analysis in Geology. Wiley, New York,
Bally, A. W. Gordy, P. L. & Stewart, G. A. 1966. Structure, seismic data and 646p.
orogenic evolution of southern Canadian Rocky Mountains. Bulletin of Dennis, B., Standen, E., Georgi, D. T. & Calow, G. O. 1987. Fracture
Canadian Petroleum Geology, 14, 337-381. Identification and Productivity Predictions in a Carbonate Reef Com-
Barss, D. L. & Montandon, F. A. 1981. Sukunka-Bullmoose gas fields: plex. 62nd Society of Petroleum Engineers Annual Technical Confer-
models for a developing trend in the southern Foothills of northeast ence and Exhibition, Dallas. Paper SPE 16808.
British Columbia. Bulletin ofCanadian Petroleum Geology, 29,293-333. Douglas, R.J.W. 1950. Callum Creek, Langford Creek and Gap map-areas.
Berger, P. & Johnson, A. M. 1980. First-order analysis of deformation of a Alberta. Geological Survey of Canada Memoir, 255, 124p.
thrust sheet moving over a ramp. Tectonophysics, 70, T9-T24. 1958. Mount Head map-area. Alberta. Geological Survey of Canada
Burger, H. R. & Thompson, M. D. 1970. Fracture analysis of the Carmichael Memoir, 291, 24lp.
Peak Anticline, Madison County, Montana. Geological Society of Engelder, T. 1985. Loading paths to joint propagation during a tectonic
America Bulletin, 81, 1831-1836. cycle: an example from the Appalachian Plateau, USA Journal of
Cant, D. J. & Stockmal, G. S. 1989. The Alberta foreland basin: relationship Structural Geology, 7, 459-476.
between stratiraphy and Cordilleran terrane-accretion events. Canadian Fissinger, M. R. & Gyllenstein, A. 1986. Fracture Detection in North Sea
Journal of Earth Sciences, 26, 1964-1975 Reservoirs.Transactions of the Society ofPetroleumWell Log Analysts,
Dahlstrom,C.D.A. 1969. Balancedcross-sections. Canarf/ao/ou/Tia/o/fart/; J0th European Formation Evaluation Symposium, Aberdeen, Scotland,
Sciences, 6, 743-757. Paper Q.
404
ANALYSIS OF FRACTURE SYSTEMS IN THRUST STRUCTURES, FOOTHILLS CANADIAN ROCKIES
Friedman, M. 1969. Structural analysis of fractures in cores from Saticoy Narr, W. & Currie, J. B. 1982. Origin of fracture porosity - Example from
Field, Ventura County, California. American Association of Petroleum Altamont field, Utah. American Association of Petroleum Geologists
Geologists Bulletin, 53, 367-389. Bulletin,66, 1231-1247.
Hossack, J. R . 1978. The correction of stratigraphic sections for tectonic Norris, D. K. 1971. Comparative study of the Castle River and other folds in
finite strain in the Bygdin area, Norway. Journal of the Geological So- the Eastern Cordillera of Canada. Geological Survey of Canada Bulletin,
ciety of London, 135, 229-242. 205, 58p.
Jones, P. B. 1982. Oil and gas beneath east-dipping underthrust faults in the Price, R. A. 1967. The tectonic significance of mesoscopic subfabrics in the
Alberta Foothills. In: Powers, R.B. (td.),Studies of the Cordilleran Thrust southern Rocky Mountains of Alberta and British Columbia. Canadian
Beh. Rocky Mountain Association of Geologists, 61-74. Journal of Earth Sciences, 4, 39-70.
Kalkeuth, W. & McMechan, M. E.,1988. Burial history and thermal matu- & Mountjoy, E. W. 1970. Geologic structure of the Canadian Rocky
rity. Rocky Mountain Front Ranges, Foothills and Foreland, east-central Mountains between Bow and Athabasca river, a progress report. In:
British Columbia and adjacent Alberta, Canada. American Association Wheeler, J. O. (ed.). Structure of the Southern Canadian Cordillera.
of Petroleum Geologists Bulletin, 72, 1395-1410. Geological Association of Canada, Special Publication, 6, 7-25.
Kilby, W. E. & Wrightson, C.B.I 987. Bullmoose mapping and compilation Rupp, A. W. 1969. Turner Valley Formation of the Jumping Pound area.
project (93P/3,4). British Columbia Ministry of Energy, Mines and Foothills, southern Alberta. Bulletin ofCanadian Petroleum Geology, 17,
Petroleum Resources, Geological Fieldwork 1986, Paper 1987-1,373-378. 460-485.
Mardia, K. V. 1972. Statistics ofDirectional Data. Academic Press, London, Sanderson, D. J. 1982. Models of strain variation in nappes and thrust sheets
357p. a review. Tectonophysics, 88, 201-233.
McMechan, M.E. 1985. Low-taper triangle zone geometry: an interpretation Serra,0.1989. Formation Microscanner Image Interpretation.Schlnmberger
for the Rocky Mountain Foothills, Pine Pass-Peace River area, British Education Services. 117p.
Columbia. Bulletin of Canadian Petroleum Geology, 33, 31-38. Sheriff, R. E. 1973. Encyclopedic Dictionary of Exploration Geophysics.
& Thompson, R.I. 1989. Structural style and history of the Rocky Society of Exploration Geophysicists, Tulsa Oklahoma, 266p.
Mountain fold and thrust belt. In: Ricketts, B. D. (ed.) Western Canada Spang, J. H. & Groshong Jr., R. H., 1981. Deformation mechanisms and
Sedimentary Basin: A Case History. Canadian Society of Petroleum strain history of a minor fold from the Appalachian Valley and Ridge
Geologists, 47-72. Province. Tectonophysics, 72, 323-342.
McQuillan, H. 1973. Small-Scale Fracture Density in the Asmari Formation , Wolcott, T. L. & Serra, S. 1981. Strain in the ramp regions of two minor
of Southwest Iran and its Relation to Bed Thickness and Structural thrusts, southern Canadian Rocky Mountains. In: Carter, N. L. etal. (eds)
Setling. American AssociationofPetroleumGeologists Bulletin,51,2361- Mechanical Behaviour ofcrustal rocks. The Handin Volume, American
2385. Geophysical Union, Geophysical Monograph, 24, 243-250.
1974. Fracture Patterns on Kuh-e Asmari Anticline, Southwest Iran. Steams, D. W. 1968. Certain aspects of fracture in naturally deformed rocks.
American Association of Petroleum Geologists Bulletin, 58, 236-246. In: Rieker, R. E. (ed.) National Science Foundation Advanced Science
Mitra, S. 1987. Regional variations in deformation mechanisms and struc- Seminar in Rock Mechanics. Special Report, Air Force Cambridge Re-
tural styles in the central Appalachian orogenic belt. Geological Society search Laboratories, Bedford, Mass., AD66993751, 97-118.
of America Bulletin, 98, 596-590. & Friedman, M. 1972. Reservoirs in Fractured Rock. American As-
1988. Effects of Deformation Mechanisms on Reservoir Potential in sociation of Petroleum Geologists Memoir,l6, 82-100.
Central Appalachian Overthrust Belt. American Association of Petro- Suppe, J. 1983. Geometry and kinematics of fauh bend folding. American
leum Geologists Bulletin, 72, 536-554. Journal of Science, 283, 684-721.
Monger, J.W.H., Price, R. A. & Tempelman-Kluit, D. J. 1982. Tectonic Thompson, R. I. 1979. A structural interpretation across part of the northern
accretion and the origin of two major metamorphic and plutonic welts in Rocky Mountains, British Columbia, Canada. Canadian Journal ofEarth
the Canadian Cordillera. Geology, 10, 70-75. Sciences, U, 1228-1241.
Morris, C. F., Little, T. M., & Letton, W. III. 1984. A New Sonic Array Tool Wiltschko, D. v., Medwedeff, D. A. & Millson, H. E. 1985. Distribution and
for Full Waveform Logging. Society of Petroleum Engineers Annual mechanisms of strain within rocks on the Northwest ramp of Pine
Technical Conference and Exhibition, Houston, Paper SPE 13285. Mountain block, southern Appalachian foreland: a field test of Theory.
Muecke, G. K. & Charlesworth, H.A.K. 1966. Jointing in folded Cardium Geological Society of America Bulletin, 96, 426-435.
Sandstones along the Bow River, Alberta. Canadian Journal of Earth Woodcock, N. H. 1977. Specification of fabric shapes using the eigenvalue
Sciences, 3, 579-596. method. Geological Society of America Bulletin, 88, 1231-1236.
405
Thrust tectonics and Cretaceous intracontinental
shortening in southeast Alaska
C. M. Rubin & J. B. Saleeby
Abstract: An imbricate thrust belt extending along strike for more than 2000 km overprints the
tectonic boundary between two of the largest allochthonous crustal fragments (Intermontane and
Insular superterranes) in the North American Cordillera, and affects rocks west of the Coast
Plutonic Complex in southeast Alaska and western British Columbia. Deformation was broadly
coeval with mid-Cretaceous magmatism and involved the emplacement of west-directed thrust
nappes over a structurally intact and relatively unmetamorphosed basement. The Palaeozoic and
lower Mesozoic Alexander terrane forms structural basement for much of the thrust belt along a
moderately northeast-dipping decollement.
There were two main episodes of mid-Cretaceous deformation, which were contemporaneous with
the emplacement of tabular plutonic bodies. Older structures record ductile southwest-vergent
folding and faulting, and regional metamorphism, associated with a well-developed axial-planar
foliation. Second-generation structures include southwest-directed thrust faults that juxtapose
contrasting metamorphic grades and refold earlier structures.
Structural, stratigraphical and geochronologic data suggest that regional-scale deformation in
southeast Alaska occurred between 113 Ma and 89 Ma. Deformation involved the imbrication of
marginal basin(s) and a magmatic arc, overprinting the older tectonic boundary between the
Insular superterrane and the late Mesozoic western margin of North America (i.e. the Intermontane
superterrane). Contractional deformation along the length of the thrust belt was broadly coeval
with arc magmatism, and thus records intra-arc tectonism. Late Palaeocene to Early Eocene
deformation and uplift may mark the transition from contractional to extensional tectonism, and
perhaps records the collapse of tectonically thickened crust.
The western margin of the Coast Plutonic Complex of the this case Alexander terrane and Wrangellia, against the
North American Cordillera comprises a northwest-trending western margin of Mesozoic North America; and (2) intra-arc
zone from western British Columbia to northern southeast deformation (Rubin ef a/. 1990a). Most workers agree, how-
Alaska (latitude 49-59°; Fig. 1). The belt is characterized by ever, that the mid-Cretaceous and younger calc-alkaline
widespread Cretaceous crustal shortening and metamorphism, plutons originated in a convergent-margin setting (Armstrong
and deep erosion (Crawford et al. 1976; Rubin et al. 1990a). 1988). This paper presents an overview of mid-Cretaceous
Thrust sheets in the belt show extensive basement involve- structural history and propose a model for the deformational
ment, and include juvenile ensimatic elements of the Alexan- framework of the western metamorphic belt of the Coast
der terrane and continent-derived North American slope- Plutonic Complex in southern southeast Alaska, between
and-rise deposits. The involvement of crystalline basement latitudes 55° and 56° N (Fig. 1).
in thrust sheets and large thrust displacements imply that
deformation affected the upper and middle crust, and not just
a thin, detached sedimentary cover. This paper develops Tectonic framework
evidence of significant crustal thickening in the late Mesozoic,
and draw analogies to similar orogenic belts such as the The western North American Cordillera has been subdivided
Andes (Suarez etal. 1983), where an anomalously thick crust into two N-NW-trending lithotectonic belts, the Intermontane
existed by late Mesozoic time. and Insular superterranes (Figs 1 & 2; terranes I and II of
The tectonic evolution of the western margin of the Coast Monger etal. 1982). The Intermontane superterrane consists
Plutonic Complex in western Canada and southeast Alaska of lower Palaeozoic continent-derived slope- and- rise depos-
has been discussed by numerous workers (Hollister 1982; its, and upper Palaeozoic to lower Mesozoic ensimatic arc
Monger ef a/. 1982; Crawford e/a/. 1987; Rubin e/a/. 1990a). assemblages that probably formed adjacent to North America
Two principal tectonic interpretations have been proposed to (see summaries in Miller 1987; Rubin et al. 1990b). The
explain the extensive late Mesozoic deformation that extends superterrane was accreted to North America in Early to
for over 2000 km, mostly along the western side of the Coast Middle Jurassic time (Monger et al. 1982), and thus formed
Plutonic Complex: (1) collision of allochthonous terranes, in the western margin of the North American continent during
407
C. M. RUBIN & J. B. SALEEBY
jTJjlntermontane Superterran^
stikine Insular Superterrane
Cache Creek
Alexander terrane
Quesnel
Wrangellia
Coast Plutonic Complex
Fragments in the San Juan
NW Cascades System
Insular Superterrane
( j n Cretaceous oceanic
1111 accretionary complexes
[MI] Upper Pz - lower Mz primitive
UllliJ arc & rift assemblages
p^ Lower Pz, lower Mz primitive
S island arc assemblages
100 km
Figure 1. Generalized distribution of the Intermontane and Insular superterranes and the miogeocline of the northern Cordillera (modified after Monger
etal. 1982; Monger & Berg 1987; Wheeler etal., 1988.; Miller 1987). Zone of mid-Cretaceous and younger deformation shown in wavy pattern. Also shown
are some large-strike-slip faults; SC-FF = Straight Creek-Frazer River fault system. AT = Alexander terrane; Wr = Wrangellia. Inset shows superterranes
and intervening Coast Plutonic Complex (CPC); TT-NRMT: Tintina-northern Rocky Mountain trench; QC-F = Queen Charlotte -Fairweather fault system.
Cretaceous time. The Insular superterrane, consisting of The tectonic boundary between the two superterranes
juvenile, mantle-derived volcanic arc and rift assemblages, is generally coincides with the the Coast Plutonic Complex
separated from the Intermontane superterrane to the west by (Fig. 1). A regionally extensive mid- to Late Cretaceous
the Mesozoic and early Cenozoic Coast Plutonic Complex thrust belt places rocks of the Intermontane superterrane
(Fig. 1). The Insular superterrane has a strike length of 2000 above the eastern margin of the Insular superterrane. Differ-
km; apparently it had no palaeogeographic relation to North ent structural levels are exposed throughout the belt, from
America until the Cretaceous (Monger et al. 1982; Saleeby high-grade gneiss on the east, to low-grade phyllite on the
1983; Gehrels & Saleeby 1987a). west, making this an excellent area to study thrust tectonics.
408
CRETACEOUS INTERCONTINENTAL SHORTENING IN SOUTHEAST ALASKA
Geological and stratigraphic setting of the Alava sequence are obscured by mid-Cretaceous and
younger deformation, making stratigraphical correlation with
The Alexander terrane is structural basement to the thrust adjacent terranes uncertain. Lithostratigraphy, age, and tec-
belt; it consists of a structurally intact lower Palaeozoic tonic setting suggest that the Alava sequence is similar to the
ensimatic arc sequence overlain by middle Palaeozoic clastic upper Palaeozoic and lower Mesozoic Stikine terrane
and carbonate strata (Fig. 2, Gehrels & Saleeby 1987a). The (Intermontane superterrane; Rubin & Saleeby 1988). The
Palaeozoic rocks are capped by an Upper Triassic rift assem- presence of quartzite precludes a stratigraphical correlation
blage (Fig. 2, Gehrels & Saleeby 1987b). In most areas, rocks with juvenile, age-equivalent elements of Wrangellia.
of the Alexander terrane are only slightly deformed and are The Kah Shakes sequence occupies even higher structural
not highly metamorphosed (Gehrels & Saleeby 1987), except levels in the northern portions of the thrust belt and consists
near its eastern boundary where the Alexander terrane is of lower Palaeozoic quartzite, marble, amphibolite, calc-
affected by late Mesozoic deformational. Upper Jurassic and silicate rocks, and mid-Palaeozoic orthogneiss (Fig. 2). De-
Lower Cretaceous marine pyroclastic and basinal strata of the trital zircon from quartz-rich meta-psammitic rocks yield
Gravina sequence depositionally overlie the Alexander terrane Proterozoic upper intercept U-Pb ages (Saleeby & Rubin
are (Fig. 3, Berg et al. 1988), and constitute the remnants of 1989;Gehrelsefa/. 1990). Precambrian inheritance in zircon
an oceanic island arc that was constructed on the Alexander collected from orthogneiss implies an old continental source
terrane (Rubin & McClelland 1989; Rubin & Saleeby unpubl. for some of the terrigenous rocks in the Kah Shakes sequence.
data). Middle Palaeozoic and lower Mesozoic rocks of the The older portions of the Kah Shakes sequence represent
Alava sequence structurally overlie Alexander terrane and outer shelf- and continental-rise deposits derived from the
Gravina sequence rocks (Fig. 2). Locally, turbidite and North American continent. This sequence may have been the
channel-fill deposits of the Gravina sequence unconformably depositional basement to the Alava sequence. Negative e^^
overlie the Alava sequence and thus form an overlap between values and high initial Sr values for correlative rocks to the
the Alexander terrane and the Alava sequence. The Alava north support the interpretation that the Kah Shakes sequence
sequence consists of three lithotectonic units: (1) is derived, in part, from old continental crust (Samson et al.
Pennsylvanian and Permian mafic metavolcanic tuff and 1989) and may be stratigraphicalally equivalent to the Yu-
flows, and massive crinoidal marble interbedded with black kon-Tanana terrane (Fig. 2).
phyllite; (2) localized sequences of quartzite, mafic Rocks in the thrust belt were intruded by mafic-ultramafic
metavolcanic rocks, and marble; and (3) upper Middle Triassic complexes, epidote-bearing tonalite, quartz diorite and
carbonaceous marble, mafic metavolcanic tuff, breccia, and granodiorite plutons, sills and dikes that range in age from =
flows. The original stratigraphical relations and distribution 110 to 89 Ma (U-Pb ages on zircon, Rubin & Saleeby 1987a,
Figure 2. Petrotectonic and structural histories of the Intermontane and Insular superterranes of British Columbia and southern Alaska (modified from
Saleeby 1982; Gardner etal. 1986; Miller 1987; Anderson 1989).
409
C. M. RUBIN & J. B. SALEEBY
NE
o o
56,000 .
Aeromagnetic Profile
(after U.S.G.S., 1977)
Figure 3. Geological domains, aeromagnetic, and gravity profiles across the western metamorphic belt of the Coast Plutonic Complex showing the mid-
Cretaceous structural domains. Late Paleocene-Eocene fabrics do not strongly overprint this area. AT = Alexander terrane; GS = Gravina sequence; AS
= Alava sequence; KSS = Kah Shakes sequence.
b; Arth et al. 1988). These =100 Ma or younger plutons Kah Shakes sequence.
crosscut the earlier regional metamorphic fabric; 95-100 Ma Structural styles and the stacking order of thrust sheets are
plutons are cut by thrust faults, whereas 90 Ma plutons different in the southern portion of the thrust belt. There,
crosscut thrust faults. A belt of Early Palaeocene to Early structurally lowest thrust sheets consist of lower Paleozoic
Eocene plutons occurs east of the thrust belt (Smith & Diggles supracrustal and meta-igneous rocks of the Alexander terrane
1981); fabrics in these younger rocks reflect Tertiary defor- and are structurally overlain by siliceous schist, orthogneiss,
mation (Saleeby & Rubin 1989). Geochemical and isotopic and marble of the middle to lower Palaeozoic Kah Shakes
data indicate that all the intrusive rocks have continental sequence. Upper Jurassic to Lower Cretaceous metaturbidites
magmatic affinities (Arth et al. 1988). of the Gravina sequence structurally overlie the Kah Shakes
sequence. These rocks are overthrust by lower amphibolite
facies rocks of the upper Palaeozoic and lower Mesozoic (?)
Internal structure of the thrust belt Alava sequence.
Non-penetrative foliation, lower greenschist facies min-
The northern part of the thrust belt consists of four structural eral assemblages, and mesoscopic folding and thrust faulting
domains (Fig. 3), three of which are described here. The characterize domain I (Figs 4 & 5). Thrust faults strike NW
structurally lowest thrust sheets (domain I) consist of lower and dip moderately to the NE. Rocks in the thrust sheets
greenschist fades supracrustal and meta-igneous rocks of the display a dominant northeast-striking foliation (Fig. 6) that is
lower Palaeozoic Alexander terrane and the Upper Jurassic parallel or subparallel with original bedding surfaces. Foliation
and Lower Cretaceous Gravina sequence. These rocks are is axial-planar to west-vergent asymmetric folds that are cut
overthrust by upper greenschist and, in metamorphic aureoles, by thrust faults. These west-vergent folds are interpreted to
locally developed lower amphibolite facies rocks of the be kinematically linked to the thrust faults and to have formed
middle to lower Palaeozoic Kah Shakes sequence and the contemporaneously with faulting (Figs 4 & 5). Domain I is
upper Palaeozoic and lower Mesozoic (?) Alava sequence characterized by low-temperature and low-pressure meta-
(domain II). The structurally higher allochthonous unit morphic mineral assemblages, andalusite-bearing contact
consists of amphibolite grade rocks of the Kah Shakes se- metamorphic aureoles, unstrained fossils, and relict phenocryst
quence, which is cut by numerous tonalite sills and dikes assemblages. Preliminary finite-strain studies on deformed
(domain III). Metamorphic grade increases significantly pebbles in coarse-grained meta-conglomerates of the Gravina
with structural level. Structural basement to domain I and II sequence indicate the rocks are comparatively weakly de-
is the Alexander terrane; higher nappes contain slices of formed; Rj^^ ratios are 5:1 or smaller (C. Rubin unpubl. data).
basement consisting of the Alexander terrane rocks and the Based on these data, rocks in domain I show little finite strain
410
CRETACEOUS INTERCONTINENTAL SHORTENING IN SOUTHEAST ALASKA
and did not experience temperatures greater than 400° C or imbricate series of thrust sheets with intervening thrust slabs
pressures above 3 kb. of Kah Shakes (Figs 4 & 5b). Stratigraphic control is based
Domain II lies structurally above domain I and is com- on isotopically dated (Pb-U zircon) granitic clasts in laterally
posed largely of imbricate thrust sheets consisting of lower continuous Gravina sequence metaturbidites, distinctive
Palaeozoic and lower Mesozoic supracrustal rocks of the Permian fossiliferous marble and Middle Triassic
Alava sequence, Upper Jurassic (?) and lower Cretaceous carbonaceous marble horizons in the Alava sequence
metasedimentary rocks of the Gravina sequence, and local (Silberling et al. 1982). The southern Revillagigedo thrust
lower Palaeozoic orthogneiss of the Alexanderterrane. Some separates domains I and II (Figs 3 & 5), ramps to deeper
of these rocks (i.e. Alava sequence) may be thrust slices of structural and stratigraphical levels, and dips moderately
Intermontane superterrane rocks (Fig. 3). The Alava and towards the east. Meta-igneous pebble- to cobble-conglom-
Gravina sequences on central Revillagigedo Island form an erate in the Gravina sequence shows flattening fabrics that
Younger Intrusive Rocks Upper Jurassic & Lower Cretaceous Gravina Sequence
_ _ _ Metamorphosed greywacke, argilltte, conglomerate,
Middle Cretaceous Intrusive Rocks \l'J^ basalt-andeslte tuff, breccia & pillow flows, & hypabyssal
" intrusive rocks
Tonalite, granodiorite, diorite, & gabbro Lower Palaezoic Kah Shakes Sequence
I I Devonian orttiogneiss, lower Paleozoic quartz-bearing
Zoned ultramafic complexes / Gabbro |:::::| psammitic rocks, silicic metavolcanic rocks, amphibolite,
metapelite, quartzite & marble
Upper Palaezoic & Lower Mesozoic Alava Sequence
Metamorphosed mafic pillow flows, tuff & Palaezoic & Lower Mesozoic Alexander Terrane
breccia, argillite, marble, & quartzite
Triassic conglomerate, limestone, basalt, & rhyolite
\ Geologic contact
^ (dashed where Inferred
& dotted where covered)
Thrust Fault
(dashed where Inferred
> | & dotted where covered)
BP = Busily Point
pluton
MB = Motli Bay
pluton
TA = Thome Arm
CI = Carol Inlet
CI ^ George Inlet
Gi = Gravina island
Al = Annette Island
Figure 4. Distribution of geological units and major structures in southern southeast Alaslca. Geological map modified from Berg et al. (1988).
411
C. M. RUBIN & J. B. SALEEBY
Cleveland Peninsula
O Og S § i
sw NE
Gravlna sequence I Alava sequence Kah Shakes sequence Bell Is. Pluton
Alexander
terrene
Gravina conglomerates
Figure 5. Geological cross-section across the southern Cleveland Penin,sula and Carol Inlet. No vertical exaggeration. Locations of cross-sections shown
on Figure 4.
parallel foliation surfaces. Numerous deformed tonalite and ramp-flat geometries are formed. Class IC and similar-style
quartz diorite sills and dikes are present in domain II and were class 2 folds (fold classification after Ramsay & Huber 1987),
probably emplaced during thrust faulting (Hollister & with axial planes dipping moderately towards the northeast,
Crawford 1986). The contemporaneity of Ar*'/Ar^'hornblende are common in domain II (Fig. 7). The folds verge to the
cooling ages from metamorphic rocks (Sutter & Crawford northwest. Bedding-cleavage lineations dip moderately to
1985) and U-Pb zircon ages in the tonalitic plutons and sills the northeast (Fig. 7a). A wide range of rock compositions
(Saleeby 1988) suggest that metamorphism, thrust faulting, has produced a range of mineral assemblages: white mica-
and igneous activity were synchronous. Thrust faults com- biotite-quartz-plagioclase-chlorite ± garnet ± calcite are
monly break along weak units that are locally composed of common in psammitic rocks; in metabasalts, green amphibole-
marble phaccoids in a penetratively cleaved phyllonite. Where biotite-plagioclase-epidote-chlorite assemblages are com-
thrust faults cut through relatively competent marble, flat- mon. In contact aureoles, staurolite and kyanite are present.
Figure 6. Lower-hemisphere, equal-area plot showing poles to foliation surfaces domain I, (a) northwest Cleveland Peninsula, (b) Annette Island, (c) Gravina
Island.
412
CRETACEOUS INTERCONTINENTAL SHORTENING IN SOUTHEAST ALASKA
Figure 7. Lower-hemisphere, equal-area plot showing, (a) poles to foliation and cleavage in domain II, (b) poles to foliation and trend and plunge of small-
scale folds in domain II, and (c) poles to foliation and trend and plunge of elongation lineation in domain III.
413
C. M. RUBIN & J. B. SALEEBY
and 850° C, the temperature range estimated from the solidus occurred prior to the Late Cretaceous (see Brandon et al.
temperature of an undersaturated tonalite (Huang & Wyllie 1989). Similarly, the Queen Charlotte Group consists of
1986). Hornblende from the same intrusive body in the Albian to probably Campanian (=113 Ma to 74.5 Ma) marine
western part of the thrust belt yields Ar^7Ar'' cooling ages of fine- to very coarse-grained clastic strata that is approxi-
97 Ma (Sutter & Crawford 1985), and are presumed to have mately 3 km thick (Sutherland-Brown 1968), although sedi-
retained argon at 530° + 40° C (Harrison & McDougall 1980). mentation may have been continuous from the Hauterivian
These data yield cooling rates between 112°C/Ma and 75 °C/ (Haggart 1989; Haggart et al. 1989). Metamorphic and ig-
Ma Hornblende and biotite from plutons in the eastern part neous clasts in conglomerates from the upper part of the
of the thrust belt yield K/Ar cooling ages of 89 Ma and 82 Ma, Queen Charlotte Group were derived from the Coast Plutonic
respectively (Smith & Diggles 1981), and yield cooling rates Complex (Yagishita 1985; Higgs 1990); perhaps recording
of 70°C/m.y to 30°C/Ma loading due to the emplacement of thrust nappes. These
Apparent uplift rates can be inferred using K/Ar cooling synorogenic deposits may represent the westward dispersal
ages for hornblende and biotite (Smith & Diggles 1982), and of sediments that were derived from the uplifted thrust belt
using U-Pb zircon ages (Rubin & Saleeby 1987a,b) from (Yagishita 1985; Higgs 1990); however, a local source for the
tonalitic plutons that have well-constrained geobarometry. A sediments can not be ruled out (Gamba et al. 1990). The
tabular tonalite pluton (Bushy Point pluton) yielding U-Pb presence of strong lithosphere beneath the Insular superterrane
ages of 95 Ma is inferred to have crystallized at a depth of = may have prevented the development of deep basins; thus
25 km. The pressure estimate is based on AF (total alu- shallow foredeep basins were formed as the crust was flexed
minium) content in hornblende (Hammarstrom & Zen 1986) by the weight of the thrust sheets.
and the experimental calibration of the AF in the hornblende
geobarometer (Johnson & Rutherford 1989). Hornblende
and biotite from the surrounding country rock yield K/Ar Implications for basement tectonics and
cooling ages of 83.0 Ma ± 2.5 and 80.2 ± 4 Ma, and are pre- convergent margin tectonics
sumed to have retained argon at 280° ± 40° C and 530° ± 40°
C, respectively (Harrison 1981; Harrison & McDougall 1980). Major differences between the structural styles present in
Assuming a mid-Cretaceous geothermal gradient of =25°C/ southeast Alaska and mountain belts characterized by thin-
km, these data imply an apparent uplift rate of=0.9 km/Ma To skin deformation (e.g. fold and thrust belt of the Canadian
the south, the inferred mid-Cretaceous average uplift rate is Rockies) reflect fundamental differences in basement defor-
=0.58mm/yr (Zen 1988), which is comparable with the uplift mation, level of crustal exposure during deformation, and
rates inferred for the Ketchikan area. These rates are com- subsequent tectonic evolution. Mid-Cretaceous thrust faults
parable with modem orogenic uplift and erosion rates (see in southeast Alaska dip between 40° and 70°, and affect
Dahlen & Suppe 1988). predominantly crystalline basement, and are perhaps analo-
Synorogenic to post-orogenic marine and nonmarine gous to present-day deformation in the high Andes (Stauder
foredeep sequences may be represented by Upper Cretaceous 1975; Suarez et al. 1983). Steeply dipping thrust faults and
parts of the Queen Charlotte and Nanaimo Groups in western focal depths between 15 and 25 km characterize the central
British Columbia. The Nanaimo Group consists of Turonian Andes (Chinn & Isacks 1983; Stauder 1975; Suarez et al.
to Maastrichtian (=90 Ma to 65 Ma) marine and non-marine 1983). Most earthquakes within the central Andes occur
sandstone, conglomerate, and mudstone, approximately 2.5 within basement and show little evidence of thin-skin defor-
km thick, that unconformably overlies rocks of Wrangellia mation. Locally, however, some steeply dipping recent thrust
(Muller & Jeletsky 1970; England 1989). The Upper Creta- faults recognized in the high Andes (e.g. Pariahuana fault)
ceous Nanaimo conglomerates contain metamorphic clasts have shallower seismisity, with seismic activity no deeper
that were probably derived from thrust sheets exposed to the than 20 km (Philip & Megard 1977). In places, contempora-
east, indicating that both metamorphism and thrust faulting neous minor folds formed parallel to the fault trace (Philip &
20
Figure 8. Simplified cross-section across the western Cordillera at the end of the Cretaceous compressive phase. Topography is current present day erosion-
level. SRMT = southern Rocky Mountain trench; MD = Monashee decollement. Strike-slip faults normal to cross-section not included. Rocky Mountain
thrust belt simplified from Price & Mountjoy (1970); metamorphic hinterland of the Rocky Mountain thrust belt simplified from Brown etal. (1976); Skeena
fold belt modified from Evenchick (this volume). Crustal thickness in southeast Alaska extrapolated from Barnes (1972a, b, 1977).
414
CRETACEOUS INTERCONTINENTAL SHORTENING IN SOUTHEAST ALASKA
Megard 1977). Contractional deformation may be due to a Complex, where thrust-related fabrics are truncated by east-
combination of factors which include, high convergence dipping Tertiary normal faults (McClelland et al. 1990).
rates (Pardo-Casas & Molnar 1987), the age of the subducted Abundant mid-Cretaceous and younger sills and plutons, and
ocean floor (Molnar & Atwater 1983), the angle of the a moderate geothermal gradient resulted in hot, thermally
subducting slab (Wdowinski et al. 1989), and the mechanical weakened crust which effectively reduced the overall strength
coupling between the South American and Nazca plates of the crust. By analogy with regions of active extension in
(Uyeda & Kanamori 1979). Such faulting is analogous to the high Andes, tectonically thickened mid-Cretaceous crust
mid-Cretaceous thrust faulting in southeast Alaska, where may have collapsed and subsided due the gravitational poten-
thrust faults dip steeply and extend into crystalline basement. tial of a high mountain belt.
Mid-Cretaceous thrust faults, such as the Black Mountain Locally, the emplacement of arc-derived calc-alkaline
thrust (Fig. 4), probably formed in a comparable tectonic magmas was coeval with mid-Cretaceous deformation. Thus,
setting, possibly recording coupled subduction between the mid-Cretaceous contraction occurred in an intra-arc setting.
North American and Pacific/Farallon plates. The high Andes Addition of voluminous early- to mid-Tertiary plutons coin-
have also undergone very different styles of deformation cided with and outlasted erosion, uplift, and inferred exten-
during their tectonic evolution, in which back-folding or sion of the tectonically thickened and thermally weakened
'retrocharriage' occurred during the later stages of thrust crust. The mid-Cretaceous to Early Tertiary tectonic history
faulting. In this context, the westward dipping faults and reflects contractional and extensional events that occurred in
foliation surfaces in the eastern portion of the thrust belt in a convergent-margin setting. Changes in overall plate con-
southeast Alaska may be analogous to the modem day vergence rates and the rate of subduction may explain differing
structural reversals seen in the Andes. Although the analogy structural styles (see discussion in Royden & Burchfiel
between the mid-Cretaceous thrust system and the Andes is 1989). High convergence rates combined with strongly
imperfect, the mid-Cretaceous thrust system involves base- coupled subduction are characterized by compressional arc
ment deformation, whereas much of the deformation to the and back-arc regimes, and are typical of continental convergent
east (i.e. east-vergent thrust faulting in the Bowser basin) may margins (Uyeda & Kanamori 1979), whereas extension may
have occurred aseismically along low-angle decollements. have resulted from decoupling of the down-going slab from
By analogy with the Andes, mid-Cretaceous thrust faulting the overriding plate. Slower subduction rates reduce horizontal
was driven by strongly coupled subduction in which defor- compressive stresses required to support high topography,
mation is superposed on either an older collisional (Saleeby and may result in topographic collapse by normal faulting
& Rubin 1990), a strike-slip boundary (McClelland & Gehrels (Dalmayrac & Molnar 1983). Mid-Cretaceous tectonism
in press) or a combination of the two. The presence of coeval coincides with an abrupt increase of convergence rates (>100
igneous activity, convergence between continental fragments, km/Ma) between the Farallon and North American plates
and the transport of crystalline slabs along thrust faults in the (Engebretson et al. 1985); the intense deformation in the
mid-Cretaceous support such an analogy with the Andes. northwestern Cordillera may have been related to rapid
Beneath most of the western wall rocks of the Coast subduction. The correlation of rapid subduction with
Plutonic Complex, the present-day crustal thickness is less intracontinental tectonism has also been proposed in the
than 30 km (Barnes 1972a, b, 1977), and locally may be as Andes (see discussion in Pardo-Casas & Molnar 1987).
thin as 25 km (Yorath ef a/. 1985). Emplacement pressure of Controversy exists concerning the role of collision of
=6-7 kb for 100 Ma plutons (Hammarstrom & Zen 1986; allochthonous crustal elements (e.g. the Alexander terrane
Rubin & Saleeby 1988) suggest the crust was at least 25 km and Wrangellia) in late Mesozoic deformation of the north-
thick at this time. Thrust faulting during the mid-Cretaceous west Cordillera (Monger et al. 1982; Crawford et al. 1987;
(90 Ma) may have thickened the crust to 50 or 60 km. This Pavlis 1989; Rubin et al. 1990a). The absence of an accre-
estimate for mid-Cretaceous crustal thickness is consistent tionary wedge and high-pressure metamorphic terranes,
with the present-day thickness of 25 to 30 km, plus the combined with the lack of evidence for late Mesozoic oceanic
amount of crust that must have been removed to expose crust between the Insular and Intermontane superterranes at
Cretaceous mid-crustal rocks. This is in accord with crustal these latitudes, are important constraints that argue against
thicknesses in regions which have undergone similar styles of collisional tectonic models. There is no general agreement
crustal thickening, such as the hinterland of the Canadian concerning plate tectonic models that explain mid-Creta-
Rockies (Monger & Price 1979; Monger et al. 1985). ceous and younger deformation, yet deformation and igneous
Uplift and erosion of the thrust belt in an extensional activity must have occurred in a magmatic arc setting. Thus,
setting during Palaeocene and Eocene time subsequently the collision of exotic tectonic elements and the concurrent
thinned the crust; the available constraints on present crustal closure of an late Mesozoic oceanic basin is not required to
structure and magnitude of Tertiary extension, however, are explain the stratigraphical, structural, and metamorphic rela-
poorly known. Rapid early Tertiary uplift rates of about 2 km/ tions seen in southeast Alaska and western British Columbia.
ma have been inferred for the western flank of the Coast Terrane accretion did not play a major role in mid-Cretaceous
Plutonic Complex (Hollister 1982; Harrison et al. 1979), deformation and mountain building. Mid-Cretaceous de-
although the uplift has also been ascribed to contraction formation was broadly coeval with crustal shortening to the
(Hollister 1982;Crawfordera/. 1987). Extensionaltectonism east in Bowser Basin (Skeena fold belt of Evenchick 1991,
is documented on the west flank of the Coast Plutonic this volume) and to the south in the southern portions of the
415
C. M. RUBIN & J. B. SALEEBY
Coast Plutonic Complex (Joumeay 1990). The coincidence mation formed in response to east-west compression over a
in timing of deformation in the western Canadian Cordillera distance of at least 600 km across the Northern Cordillera, and
suggest that there is a detachment which underlies all of these developed in a structural setting much like that of the modem
zones and which may continue as far east as the Rocky Andes.
Mountain fold and thrust belt (Fig. 8). The fault that underlies
C.M.R, is grateful for long discussions with Bill McClelland, Meghan
continental slope-and-rise deposits west of the Coast Plutonic
Miller, Jim Monger, and Margi Rusmore on Cordilleran tectonics. Field and
Complex may be the western continuation of the Monashee laboratory work for southeast Alaska regional studies has been supported by
decollement, which separates North American-type crust US National Science Foundation Grants EAR 86-05386 and EAR 88-03834
with Palaezoic to Mesozoic terranes (Fig. 8). This interpre- (to JBS); additional support (to CMR) was provided by Geological Society
tation is supported by layered seismic reflectors near the of America Penrose Grants, a Sigma-Xi grant-in-aid, and by the US Geological
southern boundary of the Coast Plutonic Complex which Survey, Alaska Branch. Some of the work reported here was part of a
doctoral thesis by C.M.R. at Caltech. The stereonet program was graciously
have been correlated to the Monashee decollement (Brown et provided by Rick Allmendinger. Informal reviews by Meghan Miller were
al. 1991, this volume). The remarkable similarity of timing quite helpful. Hugh Gabrielse, Ken McClay, and Bob Thompson provided
of deformation across the western Cordillera suggests very helpful and critical reviews.
tectonism was a widespread intracontinental event. Defor-
References
Anderson, R. G. 1989. A stratigraphical, plutonic, and structural framework Gamer, M. C , Bergman. S. C , Gushing, G. W., MacKevett, E. M., Plafker,
for the Iskut River map area, northwestern British Columbia. In: Current G., Campbell, R. B., Dodds, C. J., McClelland, W. C. & Mueller, P. A.
research, PartE. GeologicalSurvey of Canada Paper, 89-lE, 145-154. 1988. Pennsylvanian pluton stitching of Wrangellia and the Alexander
Armstrong, R. L. 1988. Mesozoic and early Cenozoic magmatic evolution of terrane, Wrangell Mountains, Alaska. Geology, 16, 967-971,
the Canadian Cordillera. Geological Society of America Special Paper, Gehrels, G. E. & Saleeby, J. B. 1987a. Geological framework, tectonic
218,55-91. evolution, and displacement history of the Alexander terrane. Tectonics,
Arth, J. G., Barker, F. & Stem, T. W. 1988. Coast Batholith and Taku plutons 6, 151-173.
near Ketchikan, Alaska: Petrography, geochronology, geochemistry, & Saleeby J.B.I 987b. Geology of Prince of Wales Island, southeastem
and isotopic character. American Journal of Science, 288-A, 461-489. Alaska. Geological Society of America Bulletin, 98, 123-137.
Barnes, D. F. 1972a Simple bouguer gravity anomaly map of Prince Rupert , McClelland, W. C , Samson, S. D., Patchett, P. J. & Jackson, J. L. 1990.
1:250,000 quadrangle, southeastern Alaska, showing station locations, Ancient continental margin assemblage in the northem Coast Moun-
anomaly values, and generalized 10-milligal contours. US Geological tains, southeast Alaska and northwest Canada. Geology, 18, 208-211.
Survey Geophysical Open-File Map, 1 sheet. Harland, W. B., Cox, A. V., Lewellyn, P. G., Pickton, C.A.G., Smith, A. G.
1972b. Simple bouguer gravity anomaly map of Ketchikan 1:250,000 & Walters, R. 1982. /I Geological time scale. Cambridge University Press,
quadrangle, southeastern Alaska, showing station locations, anomaly Cambridge, England, 13Ip.
values, and generalized 10-milligal contours. US Geological Survey Haggart, J. W. 1989. Reconnaissance stratigraphy and biochronology of the
Geophysical Open-File Map, 1 sheet. Lower Cretaceous Longarm Formation, Queen Charlotte Islands, British
1977. Bouguer gravity map of Alaska: US Geological Survey Geo- Columbia. Geological Survey of Canada Paper, 89-lH, 39-46.
physical Investigations Map GP-913, 1 sheet, scale 1:2,500,000. , Lewis, P. D. & Hickson, C. J. 1989. Stratigraphy and structure of
Berg, H. C , Jones, D. L. & Richter, D. H. 1972. Gravina-Nutzotin belt: Cretaceous strata. Long Inlet, Queen Charlotte Islands, British Colum-
Tectonic significance of an upper Mesozoic sedimentary and volcanic bia. Geological Survey of Canada Paper, 89-lH, 65-72.
sequence in southern and southeastern Alaska. US Geological Survey Hammarstrom, J. M. & Zen, E. 1986. Aluminum in hornblende: An empiri-
Professional Paper ^m-D, 1-24. cal igneous geobarometer. American Mineralogist, 71, 1297-1313.
, Elliott, R. L. & Koch, R. D. 1988a. Geological map of the Ketchikan Harrison, T. M. 1981. The diffusion of''"Ar in hornblende. Contributions to
and Prince Rupert quadrangles, Alaska. US Geological Survey Miscel- Mineralogy and Petrology, 78, 324-331.
laneous Investigations Series, Map 1-1807, I sheet, scale 1:250,000. & McDougall I. 1980. Investigations of an intrusive contact, northwest
Brandon, M. T., Cowan, D. S. & Vance, J. A. 1988. The Late Cretaceous San Nelson, New Zealand-I. Thermal, chronological, and isotopic con-
Juan thrust system, San Juan Islands, Washington. Geological Society of straints. Geochemica et Cosmochimica Acta, 44, 1985-2003.
America Special Paper, 221, 81 p. , Armstrong, R. L., Naeser, C. W., & Harakal, J. E. 1979. Geochronology
Brown, R. L., Joumeay, J. M., Lane, J. M., Murphy, D. C. & Rees, C. J. 1986. and thermal history of the Coast Plutonic Complex, near Prince Rupert,
Obduction, backfolding and piggyback thmsting in the metamorphic British Columbia. Canadian Journal of Earth Sciences, 16, 400-410.
hinterland of the southeastern Canadian Cordillera. Journal ofStructural Higgs, R. 1990. Sedimentology and tectonic implications of Cretaceous fan-
Geology, 8, 255-258. delta conglomerates. Queen Charlotte Islands, Canada. Sedimentology,
Chinn, D. S. & Isacks, B. L. 1983. Accurate source depths and focal mecha- 37,83-103.
nisms of shallow earthquakes in western South America and the New Hollister, L. S. 1982. Metamorphic evidence for rapid uplift of a portion of
Hebrides Island arc. Tectonics, 2, 529-653. the central gneiss complex. Coast Mountains, B.C. Canadian Miner-
Crawford, M. L., Hollister, L. S. & Woodsworth, G. J. 1987. Crustal alogist, 20, 319-332.
deformation and regional metamorphism across a terrane boundary. & Crawford, M. L. 1986. Melt-enhanced deformation: A major
Coast Plutonic Complex, British Columbia. Tectonics, 6, 343-361. tectonic process. Geology, 14, 358-561.
Engebretson, D. C , Cox, A. & Gordon, R. G. 1985. Relative motions Huang, W. L. & Wyllie, P. J. 1986. Phase relationships of gabbro-tonalite-
between oceanic and continental plates in the Pacific basin. Geological granite-water at 15 kb with applications to differentiation and anatexis.
Society of America Special Paper, 206, 59p. American Mineralogist. 71, 301-316.
England, T.D.J. 1989. Lithostratigraphy of the Nanaimo Group, Georgia Johnson, M. C. & Rutherford, M. J. 1989. Experimental calibration of
Basin, southwestern British Columbia. In: Current Research, Part E, aluminium-in-homblende geobarometer with application to Long Val-
Geological Survey of Canada, 89-lE, 197-206. ley caldera (California) volcanic rocks. Geology, 17, 837-841.
Dahlen, F. A. & Suppe, J. 1988. Mechanics, growth, anderosion of mountain Jordan, T., Isacks, B., Allmendinger, R., Brewer, J., Ramos, V. & Ando, C.
belts. Geological Society of America Special Paper, 218, 161-178. 1983. Andean tectonics related to the geometry of the subducted plate.
Dalmayrac, B. & Molnar, P. 1981. Parallel thrast and normal faulting in Peru Geological Society of America Bulletin, 94, 341-346.
and constraints on the state of stress. Earth and Planetary Science Let- Joumeay, J. M. 1990. A progress report on the structural and tectonic
ters, SS,47i-4»l. framework of the southern Coast Belt, British Columbia. Geological
Gamba, C. A., Indrelid, J. & Taite, S. 1990 Sedimentology of the Upper Survey of Canada, 90-lE, 183-195.
Cretaceous Queen Charlotte Group, with special reference to the Honna McClelland, W. C , & Gehrels, G. E. 1990. Geology of the Duncan Canal
Formation, Queen Charlotte Islands, British Columbia. Geological shear zone and its bearing on the accretionary history of the Alexander
Survey of Canada, 90-lF, 67-73. ten:ane,southeastem Alaska. Geo/o^?(ca/Soa>fyo/4mOTCflB«//ef/>((in press).
416
CRETACEOUS INTERCONTINENTAL SHORTENING IN SOUTHEAST ALASKA
, Gehrels, G. E., Samson, S. D. & Patchett, P.J . 1990. Geological and Saleeby, J. B. 1983 Accretionary tectonics ofthe North American Cordillera:
structural relations along the western flank of the Coast Mountains Annual. Reviews in Earth and Planetary. Sciences, 15, 45-73.
batholith: Stikine River to Cape Fanshaw, central SE Alaska. Geological 1988. The inner boundary of the Alexander terrane in southern SE
Society of Canada Abstract with Programs, IS, 116. Alaska: part II southern Revillagigedo Island (RI) to Cape Fox (CI).
Miller, M. M. 1987. Dispersed remnants of a northeast Pacific fringing arc Geological Society of America Abstracts with Programs, 19. 828.
- Upper Paleozoic island arc terrane of Permian McCloud faunal affinity, & Rubin, C. M. 1989. The western margin of the Coast Plutonic
western US. Tectonics, 6, 807-830. Complex (CPC) in southernmost SE Alaska. Geological Society of
Monger, J.W.H. & Berg, H. 1987. Lithotectonic terrane map of western America Abstracts with Programs, 21, 139.
Canada and southeastern Alaska. US Geological Survey Miscellaneous & Rubin, C. M. 1990. Tectonic intercalation of crystalline nappes and
Field Studies Map MF 1874B, scale 1; 2,500,000. basinal overlap relations in southern SE Alaska: Implications for the
& Price, R. A. 1979.Geodynamic evolution ofthe Canadian Cordillera— initial stages of Insular superterrane accretion. Eos Transactions in the
progress and problems. Canadian Journal of Earth Science, 16,770-791. American Geophysical Union (in press).
, Price, R. A. & Tempelman-Kluit, J. D. 1982. Tectonic accretion and Samson, S. D., McClelland, W. C , Patchett, P. J. & Gehrels, G. E. 1989. Nd
the origin of the two major metamorphic and plutonic wehs in the isotopes and Phanerozoic crustal genesis in the Canadian Cordillera.
Canadian Cordillera. Geology, 10, 70-75. Nature, 337, 705-709.
, Clowes, R. M., Price, R. A., Simony, P. S., Riddihough, R. P. & Silberling, N. J., Wardlaw, B. R. & Berg, H. C. 1982. New paleontologic age
Woodsworth, G. J. 1985 B-2 Juan de Fuca plate to the Alberta plains; determinations form the Taku terrane, Ketchikan area, southeastern
Geological Society of America Decade of North American Geology Alaska. US Geological Survey Circular 844, 117-119.
Centennial Continent/Ocean Transect 7, scale 1:1,000,000, 21 p. Smith, J. G. & Diggles, M. F.1981. Potassium-argon determinations in the
Muller, J. E. & Jeletsky, J. A. 1970. Geology of the Upper Cretaceous Ketchikan and Prince Rupert quadrangles, southeastern Alaska. US
Nanaimo Group, Vancouver Island and Gulf Islands, British Columbia. Geological Survey Open-File Report 78-73N, 1-6.
Geological Survey of Canada Paper, 69-27, 77p. Stauder, W. 1975. Subduction of the Nazca plate under Peru as evidenced by
Pardo-Casas. F. & Molnar, P. 1987. Relative motion ofthe Nazca (Farallon) focal mechanisms and by seismicity. Journal of Geophyical Research,
and South America plates since the Late Cretaceous time. Tectonics, 6, 80, 1053-1064.
233-248. Suarez, G., Molnar, P. & Burchfiel, B. C. 1983. Seismicity, fault plane
Pavlis, T. 1989. Middle Cretaceous orogenesis in the northern Cordillera: A solutions, depth of faulting, and active tectonics of the Andes of Peru,
Mediterranean analog. Geology, 17, 947-950. Ecuador, and southern Columbia. Journal of Geophysical Research, 88,
Philip, H. & Megard, F. 1977. Structural analysis ofthe superficial deformation 10403-10428.
of the 1969 Pariahuanca earthquake (central Peru). Tectonophysics, 38, Sutherland-Brown, A 1968. Geology ofthe Queen Charlotte Islands British
259-278. Columbia. British Columbia Department of Mines and Petroleum Re-
Price, R. A. & Mountjoy, E. W. 1970. Geological structure ofthe Canadian sources Bulletin, 54, 226pp.
Rocky Mountains between Bow and Athabaska rivers—A progress Sutter, J. F. & Crawford, M. L. 1985. Timing ofmetamorphism and uplift in
report. Geological Association of Canada Special Paper, 6, 7-26. the vicinity of Prince Rupert, British Columbia and Ketchikan, Alaska.
Ramsay, J. G. & Huber, M. I. 1987. Techniques of modem structural Geological Society of America Abstracts with Programs,!!, 411.
geology, volume 2, Folds and Fractures. Academic Press, 700p. United States Geological Survey. 1977. Aeromagnetic map ofthe Ketchikan,
Royden, L, & Burchfiel, B.C.I 989. Arc systematic variations in ;thrust belt Prince Rupert, and northeastern Craig quadrangles, Alaska. US Geo-
style related to plate boundary processes? (The western Alps versus the logical Survey Open-File Report Map, 77-359.
Carpathians). Tectonics, 8, 51-61. Uyeda, S.& Kanamori, H. 1979. Back, arc opening and the mode of
Rubin, C. M & Saleeby, J. B. 1987a. The inner boundary zone of the subduction. Journal of Geophysical Research, 84, 1049-1061.
Alexander terrane in southern SE Alaska: A newly discovered thrust belt. Yagishita, K. 1985. Evolution of a provenance as revealed by petrographic
Geological Society of America Abstracts with Programs, 19, 455, analyses of Cretaceous formations in the Queen Charlotte Islands,
& 1987b. The inner boundary zone of the Alexander terrane in British Columbia, Canada. Sedimentology, 32, 671 -684.
southern SE Alaska, Part 1: Cleveland Peninsula to southern Revillagigedo Yorath, C. J., Woodsworth, G. J., Riddihough, R. P., Currie, R. G., Hyndman,
Island. Geological Society of America Abstracts with Programs, 19,826. R. D., Rogers, G. C , Seemann, D. A. & Collins, A. D. 1985. Continent-
& 1988. A new perspective on what is the Taku terrane in southern Ocean transect B1: Intermontane Belt (Skeena Mountains) to Insular
SE Alaska. Geological Society of America Abstracts with Programs, 20, Belt (Queen Charlotte Islands): Geological Society of America Decade
226. of North American Geology Centennial Continent/Ocean Transect 8,
& McClelland, W. C. 1989. The Gravina Belt: Remnants of a mid- scale 1:1,000,000, 7p.
Mesozoic oceanic arc in southern southeast Alaska. Eos Transactions in Wdowinski, S., O'Connell, J. & England, P. 1989. A continuum model of
the American Geophysical Union, 1308. continental deformation above subduction zones; Application to the
, Saleeby, J. B.,Cowan, D. S., McGroderM. F. & Brandon,M. T. 1990a. Andes and the Aegean. Journal of Geophysical Research, 94, 10331-
Late Mesozoic compressional tectonism: Development of a west-vergent 10346.
thrust system in the northwestern Cordillera. Geology, 18, 276-280. Wheeler, J. O., Brookfield, A. J., Gabrielse, H., Monger, J.W.H., Tipper, H.
, Miller, M. M. & Smith, G. M. 1990b. Tectonic development of W. & Woodsworth, G. J. 1988. Terrane map of the Canadian Cordillera.
Cordilleran mid-Paleozoic volcano-plutonic complexes. In: Harwood, Canadian Geological Survey Open-File Report 1894, scale 1 ;2,000,000
D. S. & Miller, M. M. (eds) The paleogeography of the Klamath Zen, E. 1988. Tectonic significance of high-pressure plutonic rocks in the
Mountains, Sierra Nevada, and adjacent areas. Geological Society of westerns Cordillera of North America. In: Ernst, W. G. (ed.) Metamor-
America Special Paper (in press). phism and cruslal evolution ofthe western United States, Rubey volume
VII, Prentice Hall, Englewood Cliffs, New Jersey, 41-68.
417
Glossary of thrust tectonics terms ^-rt
K.R.McClay '
This glossary aims to illustrate, and to define where possible, recognise the difficulty in precisely defining many of the
some of the more widely used terms in thrust tectonics. It is terms used in thrust tectonics as individual usages and pref-
presented on a thematic basis - individual thrust faults and erences vary widely.
related structures, thrust systems, thrust fault related folds, 3-
D thrust geometries, thrust sequences, models of thrust sys-
tems, and thrusts in inversion tectonics. Fundamental terms Thrust faults ^ -
are defined first, followed by an alphabetical listing of related
structures. Where appropriate key references are given. Thrust fault: A contraction fault that shortens a datum
surface, usually bedding in upper crustal rocks or a regional
Since some of the best studied thrust terranes such as the foliation surface in more highly metamorphosed rocks.
Canadian Rocky Mountains, the Appalachians, the Pyrenees,
and the Moine thrust zone are relatively high level foreland This section of the glossary defines terms applied to indi-
fold and thrust belts it is inevitable that much of thrust vidual thrust faults (after McClay 1981;Butler 1982; Boyer
tectonics terminology is concerned with structures found in & Elliott 1982; Diegel 1986).
the externa! zones of these belts. These thrustbelts character-
istically consist of platform sediments deformed by thrust Backthrusl: A thrust fault which has an opposite vergence
faults which have a ramp-flat trajectories (Bally el at. 1966; to that of the main thrust system or thrust belt (Fig. 2).
Dahlstrom 1969, 1970; Price 1981; Rich 1934; and others). Backthrusts are commonly hinterland-vergent thrusts.
Steps in the thrust surface generate geometrically necessary
FOHELAND-VERGENT
folds in the hangingwall above (Fig. I). Therefore much THRUST SYSTEM
EROSION LEVEL
\
Figure 1. Idealised thrust fault showing kink band style folding in the
hangingwall and no deformation in the footwall.
419
K.R. MCCLAY
rt \u'' .'A -
420
GLOSSARY OF THRUST TECTONICS TERMS
Footwall cut-off lines are the lines of intersection between Frontal ramp: A ramp in the thrust surface that is
a thrust surface and a stratigraphic horizon in the footwall of perpendicular to the direction of transport of the thrust sheet
the thrust (Fig. 6) (Diegel 1986). (Fig. 10). Ramp angles are commonly between 10° and 30°.
FOOTWALL SHORTCUT
Hinterland-vergent thrust: A thrust fault that verges
THflUST
towards the hinterland of the orogen (i.e away from the
B ^
f^^^^^p"!^ undeformed foreland). , j .. - _ . , . . . .
^t^iSTH."."*;
LEWIS THRUST SHEET
Figure 8. Footwall shortcut thrust, (a) Fault propagation fold with trajectory
of incipient footwall shoncui ihrust. (b) Final configuration after displace-
ment on the footwali shortcut thrust.
HINTERLAND FORELAND
Figure 11. Klippe (adapted after McClay & Insley 1986).
421
K.R. MCCLAY
Listric thrust fault: A concave upwards thrust fault such Pop -up: A section of hangingwall strata that has been
that the upper section is a steep high angle contraction fault, uplifted by the combination of a foreland vergent thrust and
the middle section is a medium angle contraction fault and the a hinterland vergent thrust (Fig. 15).
sole is a bedding plane parallel fault (Fig. 12).
Regional Regional
f / jr ^ f f / f *• y ^ y y ^ y ^ ^
^ v ^ ^ . ^ . ^ N ^ ^ . \ ^ \ N ^ \ >. \ V
V . l i V i V . 1 . V . > . S \ S V . \ J' y / jf f ^ jf
Figure 16. The concept of "regional' for bed A in a simple thrust structure.
Out of the syncline thrust: A thrust fault that nucleates and
propagates out from the core of a syncline (Fig. 14). Out of the
syncline thrusts are generated by the space problem in the Smooth trajectory thrust: A thrust fault with a trajectory
cores of tight synclines and may not necessarily be linked to that is smoothly varying and does not have a staircase form
other thrusts. (Fig. 17). Smooth trajectory thrusts are found in higher grade
metamorphic rocks (Cooper & Trayner 1986; McClay 1987)
where ductile penetrative strains are developed within the
thrust sheet. ' • <-
422
GLOSSARY OF THRUST TECTONICS TERMS
Sole thrust: The lowermost thrust common to a thrust tip point is the 2D equivalent of a tip line (Fig. 5).
system (may also be termed a floor thrust - see thrust systems
below). rt'-i ..A~ tif • - 1 rifJif^r''^-'
Thrust systems
Splay: A secondary thrust fault (i.e. smaller in size and
displacement) that emerges from a main thrust fault. Boyer Thrust system: A zone of closely related thrusts that are
& Elliott (1982) define four types of splays (Fig. 4). geometrically, kinematically and mechanically linked.
Rejoining splay (Fig. 4a), Connecting splay (Fig. 4b),
Isolated splay (Fig. 4c), and Divergent splay (Fig. 4d). This section of the glossary deals with linked thrust faults
that form thrust systems. The terminology for thrust systems
Thrust nappe: A large thrust sheet which may have been stems from Dahlstrom (1970), Boyer & Elliott (1982), Mitra
generated from a recumbent fold in which the lower limb has (1986) and modified by Woodward et al. (1989). Thrust
been faulted out to form the sole thrust of the nappe (Fig. 18). systems include duplexes, imbricate thrust systems and trian-
Thrust nappes may also be generated from detachment thrust- gle zones (Fig. 20 - next page).
ing and from inversion structures (cf. from inversion of ramp-
flat extensional fault systems - see McClay & Buchanan Duplexes '" " ' ^ '•'"
1991, this volume).
Duplex: An array of thrust horses bounded by n floor
thrust (i.e. sole thrust) at the base and by a roof thrust at the
top (Figs 20 & 21).
423
•t'-'.l: ? K.R. MCCLAY Z-^i M f ;
A DUPLEXES '•'' *••''-' ''••'''' ""'f'-^''- "''^ ••' inn:'\ c;! '•'.i-..i' i. ^ ' : . . . ; } u i " • ,-•
. •!*)'•,':;' i:-!'i;'tltrtr'ni
Ilia. OVERLAPPING RAMP ANTICLINES , lllb. FORELAND DIPPING DUPLEX
'if': i;
B. IMBRICATE SYSTEMS
III. TRAILING IMBRICATE FAN IV. BLIND IMBRICATE COMPLEX ' j ^ '- ^! Ki l^illtiU'!
EROSION LEVEL
\
• •••ms,; i ; , • • i i i V f
C. TRIANGLE ZONES
' - I ! , , •.
I. TRIANGLE ZONE
Figure 20. Thrust systems, (a) Duplexes, (b) Imbricate systems (schematic), (c) Triangle zones, (adapted after Boyer& Elliott 1982; Mitra 1986;
and Woodward fr u/- 1989).
424
GLOSSARY OF THRUST TECTONICS TERMS
SECOND
> ORDER •
DUPLEXES
INCREASED
DISPLACEMEffT
Figure 21. Duplex classification (modified after Mitra 1986). (a) Independent ramp anticlines and hinterland dipping duplexes, (b) True duplexes with
second order duplexes, (c) Overlapping ramp anticlines which produce antiformal stacks and, with increased displacement, foreland dipping duplexes.
Antiformal Stack: A duplex formed by overlapping ramp Breached duplex: A duplex in which 'out of sequence
anticlines which have coincident trailing branch lines (Fig. movement' on the link thrusts have breached or cut through
22). The individual horses are stacked up on top of each other the roof thrust (Fig. 23). Butler (1987) discusses breaching of
such that they form an antiform. duplex structures.
BREACHING
THRUSTS
425
K.R. MCCLAY
Corrugated or humpy roof duplex: A duplex in which the Link thrusts: Imbricate thrusts that link the floor thrust to
roof thrust is corrugated or folded (Fig. 24). the roof thrust of the duplex (Fig. 28). Link thrusts are
CORRUGATED commonly sigmoidal in shape (McClay & Insley 1986).
ROOF THRUST
Foreland dipping duplex: A duplex in which both the link Figure 28. Duplex link thrusts (adapted after McClay & Insley 1986).
thrusts and the bedding (or reference datum surfaces) dip
towards the foreland of the thrust belt (Fig. 25). Passive roof duplex: A duplex in which the roof thrust is
sipassive roof thrust (Fig. 29) such that the roof sequence has
not been displaced towards the foreland but has been
underthrust by the duplex (Banks & Warburton 1986).
Figure 29, Passive roof duplex (adapted after Banks & Warburton 1986).
Horse: Avolumeofrock completely surrounded (bounded) Planar roof duplex: A duplex in which the roof thrust is
by thrust faults (Fig. 27). planar except where it is folded over the trailing ramp and
over the leading ramp (Fig. 30). Groshong & Usdansky
(1988) demonstrate that such a geometry is a result of a
special combination of duplex thrust spacing and displace-
ment.
FLOOR THRUST
Figure 30. Planar roof duplex (true duplex model of Mitra 1986).
426
GLOSSARY OF THRUST TECTONICS TERMS
FLOOR THRUST
Figure 31. Smooth roof duplex where the roof thrust varies smoothly
without folding by the underlying link thrusts (adapted after McClay &
Insley 1986).
Triangle zones
Figure 32. Truncated duplex in which the upper section (leading branch
The term Triangle zone' was first used to describe the
lines) has been removed by an'out-of-sequence'thrust overriding theduplex.
thrustbelt termination in the southern Canadian Rocky Moun-
tains (e.g. Price 1981). There it is a zone of opposed thrust
Imbricate thrust systems dips, at theextemalmarginof the thrust bek and often with
a duplex or antiformal stack in the axial part. This is more
Imbricate thrust system: A closely related branching array correctiy described as a passive roof duplex (Fig. 29). Such
of thrusts such that the thrust sheets overlap like roof tiles ' triangle zones' are basically intercutaneous wedges (Price
(Fig. 20). 1986). Asecondusdgeof the term triangle zone refers to a
combination of two thrusts with the same basal detachment
Imbricate thrust systems may be formed a system of and with opposing vergence such that they form a triangular
overlapping fault propagation folds (tip line folds - see fault zone (see below).
related folds and folding below) as shown in Figure 33.
Imbricate fans may also form from duplexes which have the Intercutaneous thrust wedge: A thrust bounded wedge
leading branch lines eroded (Fig. 20). Boyer & Elliott (1982) bounded by a sole or floor thrust at the base and by a passive
point out the difficulty in distinguishing between imbricate roof thrust at the top (see Price 1986)
systems formed from duplexes which have had the leading
branch lines eroded and those imbricate systems formed from
a branching array of thrusts that die out into tip lines and
which have been subsequently eroded (Fig. 20).
427
K.R. MCCLAY
TRAILING / LEADING
ANTICUNE/SYNCLINE^ ANTICLINE/SYNCLINE
PAIR PAIR
Figure 40. Growth fault-bend fold (after Zoetemeijer & Sassi 1991, this
Figure 37. Fault-bend fold with leading and trailing anticline-syncline pairs. volume).
Fault propagation fold: A fold generated by propagation Growth strata: Strata that are deposited on a growth fold
of a thrust tip up a ramp into undeformed strata (Fig. 38). Also system as it develops (Fig. 40) and hence they record the
known as a tip line fold. evolution of the fold (see Suppe et al. 1991, this volume).
428
GLOSSARY OF THRUST TECTONICS TERMS
Lift-off fold: Lift-off folds (Mitra & Namson 1989) are Transported fault-propagation fold: A fault propagation
detachment folds whereby the beds and the detachment are fold that has been transported by thrust that has broken
isoclinally folded in the core of the anticline (Fig. 41). Lift- through onto an upper flat (Fig. 42).
off folds require a ductile decollement layer such as salt or
shale which can flow from the core of the fold.
/\
A \ • \ V-
1/
• <
\ /
•i
•>•-.
•,•;-,
\ f-V, /
^^^^ .•;•.-.•.•
"•.••••.••" S: ••••f •?.
\ /
DETACHMENT .,„^^ •.V 'A
•••,••••
B
Translated (or transported) detachment fold: A detach-
ment fold that is transported by a thoroughgoing thrust such
that it is displaced from its point of formation (e.g. up a
'\ .: ;(fW to
footwail ramp) (Fig. 43). Mitra (1990) discusses the differ-
ences between translated (transported) fault-propagation
folds and translated (transported) detachment folds.
••''.•/.*Vi': \ v > U ^ n A - \
Figure 41. Lift-off folds in which the detachment is isoclinally folded in ihe
core of the fold. A ductile detachment layer is needed in order to permit flow
of ductile material from the collapsed fold core, (a) Box lift-off fold, (b)
Chevron lift-off fold, (adapted after Mitra & Namson 1989)
Figure 43. Transported detachment fold (adapted after Mitra 1990).
•i". /
Overthrust shear: A term used for thrust or fold nappes
:f
that have been subjected to bulk shear strain generally in the Thrust structures in 3D
direction of nappe transport. See Ramsay et al. (1983),
Dietrich & Casey (1989), and Rowan & Kligfield (1991, this The movement of a thrust sheet over a corrugated surface
volume) for discussions of overthrust shear and nappe em- will generate flat topped anticlines and domes. These are
placement. termed culminations (Dahlstrom 1970; Butler 1982) and the
limbs of these structures are termed culmination walls.
Ramp anticline: An anticline in the hangingwal! of a
thrust generated by movement of the thrust sheet up and over Culmination: An anticline or dome with four way closure
a ramp in the footwall (Fig. 30). generated by movement of the thrust sheet over underlying
ramps.
Ramp syncline: A syncline in the hangingwall of a thrust
generated by movement of the thrust sheet up and over a ramp Displacement transfer zone: The zone where displace-
in the footwall (Fig. 37). Ramp anticlines and ramp synclines ment is transferred from one thrust to another (Dahlstrom
occur in geometrically and kinematically linked pairs - a 1970). Displacement transfer zones may occur by simple en-
leading anticline-syncline pair and a trailing anticline- echelon overlap of thrust faults (Fig. 44) or by tear faults
syncline pair (Fig. 37). parallel to the direction of tectonic transport (see tear faults
- below). ':• ' •
. . •>"
^ . _ « ) #
429
K.R. MCCLAY
Figure 47. Tear fault parallel lo the thrust transpon direction and separating
two parts of the thrust sheel each of which have different displacements.
Thrust sequences
430
GLOSSARY OF THRUST TECTONICS TERMS
of balanced and restored sections (Boyer and Elliott 1982; In sequence thrusting: A thrust sequence that has formed
Boyer 1991, this volume; Butler 1987; Morley 1988; Suppe progressively and in order in one direction (i.e either a
1985; Woodwardera/. 1989). A long accepted paradigm for forward-breaking sequence or a break-back sequence). Fig-
thrust tectonics is thai, in foreland fold and thrust belts, thrust ures 49 & 50 show in-sequence thrusting. i^i^w
faults develop sequentially in a sequence that both nucleates
in a forward-breaking sequence and verges towards the Out-of-sequence thrusting The opposite to 'in-sequence
foreland (Dahlstrom 1970; Bally <?M/. 1966; Boyer & Elliott thrusting'. Thrust faulting which develops in a sequence
1982; Butler 1982). Recently these basic 'rules' of thrust other than in sequence (Fig. 51). Break-back sequences of
tectonics have been challenged (see Boyer 1991, this volume; thrusts have commonly been called out-of sequence thrusts
and Tanner 1991, this volume). but the term should be more appropriately used to describe
thrust sequences which do not conform to a either a progres-
Breaching: Breaching occurs where an early formed sive forward-breaking or break-back sequence (Fig. 51). Out
thrust is cut by later thrusts (Butler 1987) (Fig. 48). It is a term of sequence thrusts commonly cut through and displace pre-
describing the local geometric relationships between thrusts. existing thrusts. Morley (1988) and Butler (1987) discuss out
The term may be applied to a breached duplex where the link of sequence thrusting.
thrusts do not join or anastomose with the roof thrust but cut
and displace it (Fig. 23).
•OUT OF SEQUENCE THRUST
BREACHING THRUST
Figure 48. Breached thrust - an early formed thrusi cut by later thrust.
Numbers indicate sequence of faulting.
Piggy-hack thrusi sequence: Piggy-back thrust sequence
occurs when topographically higher but older thrusts are
carried by lower younger thrusts (Fig. 50). This is essentially
Break-hack sequence: The sequence of thrusting where the same as a forward-breaking thrust sequence.
new (younger) thrusts nucleate in the hangingwalls of older
thrusts and verge in the same direction as the older thrusts Synchronous thrusting: Synchronous thrusting occurs
(Fig. 49). when two or more thrusts move together (i.e. not in a piggy-
back sequence). Boyer (1991, this volume) postulates that
synchronous thrust movement may be significant in the
Rocky Mountains of western North America. ' - —
431
K.R. MCCLAY
that subsequent deformation involves transport of the whole Self-similar thrust models J i.':''OJ^:n '•
wedge along the basal decollement or the wedge continues to
grow in a self-similar fashion by addition of new material at Self-similar model: It is commonly assumed that spatial
the wedge toe but maintaining the critical taper. Critical variations in thrust system geometries reflect temporal vari-
wedge models have used to explain the development of ations in the evolution of the structure (Fischer & Woodward
accretionary prisms and foreland fold and thrust belts (e.g. 1991, this volume). This is the basis for the self-similar
Dahlen & Suppe 1988; Dahlen 1990; Willett 1991, this growth model for thrust structures such as fault-bend folds or
volume; Liu et al. 1991, this volume) but have been criticised fault-propagation folds where, through time, the structure
by Woodward (1987). grows in amplitude as a simple geometric progression of the
- t d L,!ij(Ji'i,- first formed structural feature (Fig. 53). This model assumes
< ->;• t>i,. that the structure does not change form with time (e.g. evolve
from a detachment fold to a fault propagation fold) - an
assumption challenged by Fischer & Woodward (1991, this
volume). .
Inversion ; h /!hU-''-'.>
'l ^k ^H ^
\ ^ • ^ ^ ^ • ^ ^ "
^^^^^v^
.•.••.•.••.•••.•••.•••.•.••.•.••.•.••.•.•I;>^ ;-.^;-V> VV\v V/-;'^ V,v\v V'/\v\v V
\^•f.^;^
:^••V-^^;.^::^iC^;.V•^VV> ^^::-:
^ ^ ^^^I^^^B
Figure 54 Conceptual model of thrust faulting developed in an inverted
listric extensional fault system (after McClay & Buchanan 1991, this
volume).
This glossary was compiled with the help of Tim Dooley and Joaquina
Alvarez-Marron who are thanked lor assistance with the illustrations. I.
Figure 53. Self-similar growth model for the development of a fault-bend Davison, J. Alvarez-Marron, R. Knipe and C. Banks are thanked for critical
fold. reviews and comments.
432
GLOSSARY OF THRUST TECTONICS TERMS
References
Bally, A.W., Gordy, P.L., & Stewart, G.A., 1966, Structure, seismic data, McClay, K.R. 1981. What is a thrust? What is a nappe? In: McClay, K.R. &
and orogenic evolution of southern Canadian Rocky Mountains. Bulletin Price, N.J. (eds). Thrust and Nappe Tectonics. Geological Society of
of Canadian Petroleum Geology. 14, 337-381. London Special Publication, 9, 7-12.
Banks, C.J. & Warburton, J. 1986. 'Passive-roof duplex geometry in the McClay, K.R. 1987. The mapping of geological structures. Geological
frontal structures of the Kirthar and Sulaiman mountain belts, Pakistan. Society of London Handbook Series, Open University Press, 164p.
Journal of Structural Geology, 8, 229-238. McClay, K.R. & Insley, M.W. 1986. Duplex structures in the Lewis thrust
Boyer, S.E. 1991. Geometric evidence for synchronous thrusting in the sheet, Crowsnest Pass, Rocky Mountains, Alberta, Canada. Journal of
southern Alberta and northwest Montana thrust belts, this volume. Structural Geology, 8, 911-922.
Boyer, S. E., & Elliott, D. 1982. Thrust systems. American Association of McClay, K.R. & Buchanan, P.G. 1991. Thrust faults in inverted extensional
Petroleum Geologists Bulletin, 66, 1196-1230. basins, this volume.
Butler, R.W.H. 1982. The terminology of structures in thrust belts. Journal Mitra, S. 1986. Duplex structures and imbricate thrust systems: geometry,
of Structural Geology. 4, 239-246. structural position, and hydrocarbon potential. American Association of
Butler, R.W.H. 1987. Thrust sequences. Journal of the Geological Society, Petroleum Geologists Bulletin, 70, 1087-1112.
London, 144, 619-634. Mitra, S. 1990. Fault-propagation folds: geometry, kinematic evolution, and
Chappie, W.M. 1978. Mechanics of thin-skinned fold-and-thrust belts. hydrocarbon traps. American Association of Petroleum Geologists Bui-
Geological Society of America Bulletin. 89, 1189-1198. letin,14,92l-945,
Cooper, M.A. & Trayner, P.M. 1986. Thrust-surface geometry: implications Mitra, S. & Namson, J. 1989. Equal-area balancing. American Journal of
for thrust-belt evolution and section-balancing techniques. Journal of Science, 289, 563-599.
Structural Geology, 8, 305-312. Morley, C.K. 1988. Out-of-sequence thru.sts. Tectonics. 7, 539-561.
Cooper, M. A. & Williams, G. D. (eds), 1989. Inversion Tectonics, Geologi- Mosar, J. & Suppe, J. 1991. Role of shear in fault-propagation folding, this
cal Society of London Special Publication, 44, 376 p. volume.
Dahlen, F.A. 1990. Critical taper model of fold-and-thrust belts and accre- Price, R. A. 1981. The Cordilleran foreland thrust and fold belt in the southern
tionary wedges. Annual Reviews of Earth and Planetary Sciences, 18, Canadian Rocky Mountains. In: McClay, K.R. & Price, N.J. (eds). Thrust
55-99. and Nappe Tectonics, Geological Society of London Special Publica-
Dahlen, F.A., & Suppe, J. 1988. Mechanics, growth, and erosion ;of tion, 9, 427-448.
mountain belts. Geological Society of America Special Paper. 218, 161- Price, R.A. 1986. The southeastern Canadian Cordillera: thrust faulting,
178. tectonic wedging and delamination of the lithosphere. Journal of Struc-
Dahlstrom, C.D.A. 1969. Balanced cross sections. Canadian Journal of tural Geology, 8, 239-254.
Earth Sciences, 6, 743-757. Ram.say, J.G. 1981. Tectonics of the Helvetic Nappes. In: McClay, K.R. &
Dahlstrom, C.D.A. 1970. Structural geology in the eastern margin of the Price, N.J. (eds). Thrust and Nappe Tectonics, Geological Society of
Canadian Rocky Mountains. Bulletin of Canadian Petroleum Geology, London Special Publication, 9, 293-310.
18, 332-406. Ramsay, J.G. 1991. Some geometric problems of ramp-flat thrust models,
Davis, D., Suppe, J. & Dahlen, F.A. 1983. Mechanics of fold-and-thrust belts this volume.
andaccretionary wedges. J oional of Geophysical Research, 94,10,347- Ramsay, J.G. Casey, M. & Kligfield.R. 1983. Role of shear in development
54. of the Helvetic fold-thrust belt of Switzerland. Geology, 11, 439-442.
Diegel, F.A. 1986. Topological constraints on imbricate thrust networks, Rich, J.L. 1934. Mechanics of low-angle overthrust faulting illustrated by
examples from the Mountain City window, Tennessee, U.S.A. Journal Cumberland thrust block, Virginia, Kentucky and Tennessee. American
of Structural Geology, 8, 269-280. Association of Petroleum Geologists Bulletin. 18, 1584-1596.
Dietrich, D. & Casey, M. 1989. A new tectonic model for the Helvetic Rowan, M.G. & Kligfield. R. 1991. Kinematics of large-scale asymmetric
nappes. In: Coward, M. P., Dietrich, D. & Park, R. G. (eds). Alpine buckle folds in overthrust shear: an example from the Helvetic nappes,
Tectonics. Geological Society of London Special Publication, 45,47-63. this volume.
Elliott, D.E. 1976. The energy balance and deformation mechanisms of Suppe, J. 1983. Geometry and kinematics of fault-bend folding. American
thrust sheets. Philosophical Transactions of the Royal Society of London, Journal of Science , 283, 684-721.
283,289-312. Suppe, J. 1985. Principles of structural geology. Prentice Hall. New Jersey.
Fischer, M. P. & Woodward, N. B. 1991. The geometric evolution of 537p.
foreland thrust systems, this volume. Suppe, J., Chou, G. T. & Hook, S. C. 1991. Rates of folding and faulting
Geiser, P. A. 1988. The role of kinematics in the construction and analysis determined from growth strata, this volume.
of geological cross sections in deformed terranes. In: Mitra, G. & Wojtal, Tanner, P.W.G. 1991. The duplex model: Implications from a study of
S. (eds) Geometries and mechanisms of thrusting with special reference flexural-slip duplexes, this volume.
to the Appalachians, Geological Society ofAmerica. Special paper, 222, Thompson, R.l. 1981. The nature and significance of large 'blind' thrusts
47-76. within the northern Rocky Mountains of Canada. In: McClay, K.R. &
Groshong, R. H. & Usdansky, S.I. 1988. Kinematic models of plane-roofed Price, N.J. (eds). Thrust and Nappe Tectonics. Geological Society of
duplex styles. In: Mitra, G. & Wojtal, S. (eds) Geometries and mecha- London Special Publication. 9, 449-462.
nisms of thrusting with special reference to the Appalachians. Geologi- Willett, S. D. 1991. Dynamic and kinematic growth and change of a
cal Society of America. Special paper. 222, 197-206. Coulomb wedge, this volume.
Jamison, W.R. 1987. Geometric analysis of fold development in overthrust Woodward, N.B. 1987. Geological applicability of critical-wedge thrust-
terranes. Journal of Structural Geology, 9, 207-219. belt models. Geological Society of America Bulletin, 99, 827-832.
Knipe, R.J. 1985. Footwall geometry and the rheology of thrust sheets. Woodward, N.B.. Boyer, S.E., & Suppe, J. 1989. Balanced geological cross-
Journal of Structural Geology, 7. 1-10. sections. American Geophysical Union Short Course in Geology, 6,
Liu Huiqi, McClay, K. R. & Powell, D. 1991. Physical models of thrust I32p.
wedges, this volume.
433
Index
435
INDEX
Bending fold model 143 of Appalachians 156, i, 160, 183, 184, 224, Comiols oil well, Pyrenees 237
Benton fault, South Wales coalfield 96 264 Competence contrasts 188, 191
Bergamasc Alps 300 decollement 222 Compressing and extending flows 48, 309
Bernese Oberland, Switzerland 165-73 Canadian Cordillera 93, 225, 366 Compression 322, 358, 359, 378, 416
Besham antiform and nappe 332, 335, 337, 338 cross-sections 373, 414 along strike 309
Beuda evaporites. Southern Pyrenees 255, 257, Front Ranges 383, 385, 385. 391, 393 layer-parallel 53, 392-3
258, 261, 263 Monashee decollement 357-63 timing of, in Southern Alps 320
Biostratigraphy 261 northern 365-74 Compressional regime 19, 407
Birg fold, Wildhom nappe 166,167 Canadian Rocky Mountains 36, 80, 96 Consortium for Continental Reflection Profiling
BIRPS seismic profile 220, 243 Cate Creek duplex 207 (COCORP) 224
Blind thrusts see thrust faults Foothills 378. 391-404 Contact aureoles 412
Blue Ridge-Piedmont sheet 221, 222, 223, 224, Haig Brook duplex 207 Continent-continent coUision 217, 221-2, 225,
226,227-9,231 Lewis thrust zone 205 235-44, 247, 325, 343
Boixols thrust sheet 236-7, 241, 243, 248-9, McConnell thrust 42 Convergence 414
273 synchronous thrusting in 377-88 rates 415
palaeomagnetics of 265-7,269, 269-70 triangle zone 427 Copper Creek thrust zone 43, 45, 207
Bouguer gravity profile, Canavese line 280 Corbel thrust 293,294, 295-6, 297
Coast Plutonic Complex 410 Cantabrian Mountains 175-80 Cordillera,
Sulaiman Lobe 344, 346, 348 Carbonates346, 385-6, 391 Canadian see Canadian
Boundary conditions 9, 19, 23 see also limestone, dolomite North American see North American
Bow Valley structure 379, 379-80, 383, 386 Carboniferous, Coulomb,
Bowser Basin 373, 415 of Cantabria 175 behaviour 1, 20, 30. 86, 221-2, 230
Bowser Lake Group 366, 366-7,370. 370, palaeomagnetics 213, 214 see also Navier-Coulomb
371-2 southwest England 201 failure 19,71
Box folding 105, 128 Cardona, Coulomb wedge 1,2,11,217
Branch lines 379, 419, see also thrust sheets anticline 260. 261,262 critical see critical
Branchpoints 127,420 decollement 263 growth and change of 19-32
of axial surfaces 123 evaporites 255, 257,25S, 260-3 models 23, 23-8. 59, 71-80, 229-30,431-2,
definition of 420 Cariboo duplex, Canadian Rockies 360, 361, 432
Breaching 431 362 constant basal strength 24-5, 24-5, 28, 29
Break-back, Carol Inlet 4//-72 varying basal strength 26-7, 28,29
sequences see thrust sequences Carpathian Mountains 84, 210 in tension 28
thrust structures 207, 236, 337 Cataclasis 42.44, 277 theory 54, 77
Bresse-Rhone fault system 297 Cataclasites 217, 220-1,304 application to crystalline sheets 230, 231
Brevard fault 222, 223, 224, 229 banded 42-4, 43-4,45, 45, 46. 49 non-cohesive 20-1
British Columbia, matrix 44-5 viscous theology 30
Coast Plutonic Complex 407 see also mylonites Coulomb yield stress 19, 20, 21, 23, 28
Foothills 391, 393 Cate Creek duplex, Canadian Rockies 207 Crack tip process zone 181
see also Foothills Cathodoluminescence (CL) 44, 44 Crack-seal,
Brittle deformation 123, 130, 161, 164, 187, Caunton oilfield, England 95 calcite 199
219,281 Cebu, Philippines 108, 109 fibres 45
Brittle-ductile transition see ductile-brittle Central Alps see Alps inclusion bands and trails 202
Brittle-frictional behaviour 2, 8 Central Gneiss Belt Domain 226, 229,230 mechanism 198.208
Brooks Range, Alaska 350 Centrifugal modelling 53-68 Cracks,
Brunhes normal polarity interval 270 see also modelling effect of inclusions on 187-8
Buckle folding 57, 59-61, 165-72, 191 Chaman fault zone 339,340, 343,344, 344 see also fractures
Buckling 53, 68, 163, 165 Chambal thrust 334 Crawford thrust sheet 142
elastic 162 Charlie Lake Formation 393, 396 Creep,
and faulting 54, competition between 155, Chartreuse, French Alps 287-97, 288-96 diffusional 44
161 pre-existing structures 297 power law 219
Bude, England 201, 202. 205, Maer Cliff 206 stratigraphy and structure 288-90 Crenellations217,222
Buena Vista thrust, California 34 thrust zone kinematics 294-6 Crestal imbricate stacks 378-9, 381
Bullmoose gas field 393 Chauga belt, Appalachians 226, 228-9 Crestal-collapse grabens 93
Burial depths 155, 163-4. 172 Cleavage 169, 194, 237-8, 243, 263, 290, 297, Cretaceous 165
Buttressing 142, 165, 210, 265 195 Absaroka thrust sheet 156,/5S, 159
Byerlee's law 8. 230 bedding-parallel 168 of Alps 289,290, 297,300
in fault-propagation folds 175, 177 deformation 407
oblique 142 Pyrenees 238,241
orientations and timing 168 extensional tectonics 235, 237, 241-4, 247-8,
pressure solution 320 249, 290
in shales 45 Helvetic nappes /66
solution 165, 166-8, 170 Pyrenean 175, 235, 247-9, 253
Cadi thrust sheet 255, 257, 262, 265, 267, 271. in thrust zone 42 shale 116,//,S, 349, 353, 355
274 Climate coupling 15-16 Wyoming-Idaho-Utah thrust belt 156
California 105 Clusone fault 304 Critical Coulomb wedge 80, 229
Buena Vista thrust 34 Coalinga earthquake 33, 34 departure from 28
Los Angeles Basin 107.109 Coast Belt, Canadian Cordillera 358, 358 growth of 71 -2
Lost Hills anticline 113 Coast Plutonic Complex 365, 370, 372. 373, model 2, 20, 27, 431-2
Oak Ridge anticline 114 407,408,414,415-16 taper 54, 75, 77, 80, 90, 343, 353, 386
Owl Ridge thrust fault 392 Coast Ranges, California 33 taper angle 378
San Andreas fault 34 COCORP data 224 theory 19-32
San Joaquin growth structure 110 Cohesion 19 Cross-sections,
Santa Barbara Channel 107, HI Cohesive strength 73 analysis of 133-40
Transverse Ranges 33. 116 Collision 339, 340 balanced 10, 33, 184, 215, 236, 377
Winters Pass thrust 222 belts 235-44, 247, 325, 343 for Alps 299, 320, 321
Camarade basin 239 tectonics 217. 221. 277, 344, 374, 415 Chartreuse area 287, 290, 291-2, 292-4
Cambrian 241 Coltignone thrust 301 extended regimes 102
436
INDEX
437
INDEX
Ductile-brittle transition (DBT) 217-20, 221, Erosion 1, 2, 9, 11-17, 71, 156,252,388 brittle 71, 219
222, 229, 277 in Alps 281, 287 surfaces 161, 163
Ductility 158, 159, 163-4 and fault-bend folding 113 Fan structures 102,103
matrix 44 fluvial 6-7 Fan thrust system 240
Duplexes 133, 142, 193. 237, 311, 332, 355 Himalayan 328, 350, 353 Farallon/Pacific plate 415
antiformal stack 224, 239, 265, 320, 350, 353, in modelling 137-40 Fault breccias 198
359 process, in fault-bend fold model 133 Fault creep 33, 34, 38
Pyrenean 247, 252, 253 in Pyrenees 263 field example 35-7
and synchronous thrusting 386, 388 rate of 251 mechanisms of 34-5
breached 425 in Cordillera 413-14 Fault fabrics 226
bumpy roof 386, 426 Skeena belt 373 Fault gouge 230
classification of 423, 424-5, 425-7 and tectonics, coupled model 8 Fault plane solution data 325
cleavage 201 of thrust fault 152 Fault rocks 198, 217, 220-1, 226, 230
crustal217, 220, 229, 231,374 and uplift 399 cataclasite 220
development of 54, J93. 343-55, 343, 349-53, and wedge deformation 19, 20, 29, 30 studies 38
377, 386, iS7, 388 Erosional control, Fault slip 33, 34-5, 37, 38, 42, 89
flat-roof 386. 388 active compressional orogens 1-18 Hogsback sheet 36
flexural-slip 201-8 model for the Southern Alps, New Zealand 7- rates of 121
foreland-verging 343, 346, 347, 348, 350, 426 17 rates, and fold limb lengths 105
geometry of 202-i, 353. 355, 386, 388 tectonic model 2-3 styles in active terranes 33
hinterland-dipping 204, 426 Erosional unroofing 49 Fault splays 179
horses of 203, 260 Escape structures, continental 339 Fault steps, effect on later ramps 94
models for 61, 62. 67, 137,/JS, 139,201-8, Esia nappe, Cantabria 175, 179 Fault-bend folding see folding
206-8, 377 Eulerian theory 6, 19, 21, 162, 181, 188 Fault-bends 142, 147
passive roof i47, 350,351, 355, 426 European plate 235, 238-9, 240, 243-4, 247, Fault-bends see also ramps
planar roof 207, 386, 388, definition 426 253 Fault-propagation folding see folding
propagation of thrust faults in 53-68 in Alpine tectonics 225, 280, 281-2 Fault-ramp geometry, high angle 133, 134, 134-
Pyrenean 239, 244, 247, 260, 261, 263 Eustatic sea level changes 252, 321 5, 136, 137
smooth roof 201,203, definition 427 Evaporites 220, 292, 337, 346 Fault-related folding 53, 343, 346-7, 355
stacked 203 anhydrite 393 anticlinal cores 353
tipof205,207,208 depocentres in Pyrenees 258, 263 see also folding
trailing 381, 383 Himalayan 330 Faults,
truncated 427 in Pyrenees 236, 251-2, 258, 263 antithetic 97, 100
Duplexing 20 role in thrust detachments 127, 236, 239, 241 arcuate 345
Duplication, structural 229, 349 role in thrust geometry 255-63 brittle 221
of Southern Alps 300 and buckling, competition between 155, 161
Zagros 34 coseismic 33-4
Exhumation 19,20,30 and damping 33
Extending/compressing flows 48, 50, 309 domino93, 94, 95, 213-14
E Extension 30, 89, 93-104, 222, 243-4, 281, 306 extensional 237, 240, 251, 299, 328, 343, 350,
along-strike 399 353,358
Eakring oilfield, England 95 at oblique ramp 141 Himalayan 343, 350, 353
Earthquakes 34, 37, 38, 338 at ramp intersections 150-1 inherited 304, 308-9
along decollements 33 crustal see crustal syntectonic 38
Andean 414 ductile 225 imbricate see imbricate
deep-focus 219 hangingwall 151-2 listric see listric
focus320, 338, 350,414 and hangingwall collapse 142 mesoscopic 43
hazards, assessing 105, 116, 121 lateral 284, 309 normal 277, 281, 287, 289, 290, 307, 322,
in Himalaya,s 91 layer-parallel 404 .346,359,415
Ebro basin 247, 250, 255, 256, 257-61, 266 structures 199,301 hangingwall 150-1
Ebro foreland 236, 261, 262, 263 in thrust belts 19, 322, 323, 329, 407. 415 in Himalayan belt 349
Echaillon anticlines 292, 292-3 Extensional collapse 19 inherited 319
Eclogites 244, 283, 284 of wedge 29 plastic-brittle 359
ECORS seismic profile 135, 235-44, 236-40, Extensional faulting see faults relationship of, to ramps 299
242-3, 247, 248-9, 249, 265, External basins, inversion in 93-104 see also extensional
266-7,274,281 External Massifs. Alps 281. 282 oblique,
El Asnam earthquake, Algeria 33, 34, 37 inherited 309
Elastic behaviour 231 slip normal 152
Elastic deformation 162, 164 out-of-sequence (backstepping) 19, 29
Elastic rebound theory 35, 37-8 overlhrust 53
Elastic stiffness 35, 37 splay 295, 399
Elastic-plastic material 21 strike-slip 83. 180, 222, 255, 281, 282, 302,
Electric logging, Fabrics 152, 195,226,391 309,358
application to thrust systems 404 C-S359, 413 flower structures 302
Array sonic tool 398, 402, 403 diagenetic 199 Himalayan 335, 336, 338, 341
FMS tool 394-5, 396-8 extensional 328, 359 see also strike-slip
Elk Horn anticline, Montana thrust belt 350 flattening 411-12 synthetic 97
Emplacement of thrusts 41, 47, 224, 386 gneissic 335 tear 141, 142, 150-2, 262, 299, 306, 308-9,
Engadine Window, Central Alps 379 linear 199,217,226 430
England, poi-phyroblast 328 Himalayan 343, 344, 346, 349, 350
Eakring oilfield 95 Facies 50, 142, 202 tip of 118, 123, 124, 127, 128, 180. 187,343
southwest 165, 191,201 distribution 267, 290 transfer3l9, 322, 323
Eocene 225 effect on developing thrust 247, 249,250. transpressional 309
extension, Rockies 359, 361, 363 251 Femie Formation, Jurassic 391, 401
palaeodirection 271, 272 variations 188, 247, 297, 319, 323 Fibres 45, 202, 205
thrust emplacement, Pyrenees 266 Faggio line 302, 304, 306-8, 307, 309 bedding-plane 166, 170, 17
Enia block 304 Failure 187 calcite 195, 196, 196, 198, 199, 204. 205, 206
438
INDEX
lineations 198 Rocky Mountains see Rocky Mountains Fontana fault 304,307, 309
lineations 205 shear in 88 Foothills of Canadian Rockies,
Finite element, shortening and thickening 19 analysis of fracture systems 391-404
methods 20 Skeena, Canadian Cordillera 365-74 cross-section 400
and non-linear plastic rheologies 21-3 South Pyrenean 255-63, map of 256 Footwall,
model. South Island, New Zealand 7-9, 8 styles 155-64 anticlines 141
Finite rotation pole 210-11, 2 / / Venetian Alps 319 collapse 207
Finite-difference model 83, 91 Fold-thrust interaction 156. 161 cutoff 112
Fission track analysis 332 Folding 53, 58, 123, 125, 242, 355, 377 point 350,421
Flathead duplex 383-4 antiform/synform pair 191,192 downbending of 193, 197
Flats 141, 149,213,215,300,314 antiformal 287, 330, 335, 336, 337, 347, 350, flats 293, 421
and coseismic faulting 33-4 355 folds 191-2
footwall 150, 183-4,313,315,317-18,421 axial surfaces 105, 113-14, 116, 123, 124-5, geometry 144, 146, 146-7, 191
hangingwall 183. 421 127, 177.178 ramps 184,202
Flexuralflow 144,221 box 105, 128 and fault-propagation folds 180
component in buckle folding 172 buckle 57, 59-61, 165-72,169-70, 191 shortcut thrusts 97, 98, 70/, 102,103
Flexural slip 144, 166, 172, 297, 320 chevron 184, 186, 201, 202, 320 synclines 141, 191, 199, 320, 321, 428
bedding parallel 170 concentric 343, 349 Foredeeps 320, 343, 345, 353, 355, 377, 413-14
folding 143,202,300 crests of 346, 377 Foreland 53, 54
seismically induced 37 cylindrical 217, 226 basin 287, 289, 340, 373
Flexural structures 350, 414 detachment 53, 59, 68, 155-7,158, 158-9, Ebro 257-60
Floor thrusts see thrust faulting 163-4,260,428 Himalayan 328
Flow, in overthrust shear 165-72 molasse 282
analysis of, in external thrust zones 48-50 disharmonic 372 Pyrenean 247, 249, 257-8, 259, 260-3
by diffusive mass transfer 44-5 and evaporites 340 sedimentation in 257
compressing, fault-bend 33, 59, 68, 105, 141, 155-64, 175, Skeena fold belt 365
and extending 47 184,231,315,318 flexure 137
Stress vs. Depth plot 48 Alpine examples 299, 301, 304 fold-thru.st belts 71, 218, 229
ductile 45, 217, 218, 219, 220, 226, 229, 231, definition 428 development of 33-8
347, 349, 353 growth 111-14 evolution of 181-8
formation of Type F crystalline sheets 225 Himalayan examples 350, 353, 355 ramp-flat trajectories 93
flexural see flexural imbricate 138 Rocky Mountains see Rocky Mountains
heat 219 limb dip versus limb length 119 thin-skinned 102
laminar 48 models 191, 192-3.196,199,287-8.293,297 Foreland-vergent thrusts, 53, 71, 75, 80, 421
solutions for stresses and velocities 41 shear generated 747, 147-9 Forelimb 127, 157-9, 160-1, 165. 293, 299,
law 219 strain extrusion models 178-9 311,317,377
and curve shape of fold-thrust belts 84 theory 53. 150-1 of buckle folds 169, 170
linear 48 fault-propagation 33, 105, 123-31, 136, 191, deformation 168, 169, 175, 177-9, 180, 396
plastic 21 193,225.350 Alpine structures 290, 296, 297
steady-state creep 38 of Central Pyrenees 175.776-7, 177-80, 178 dip 147, 149, 297
Levy-Mises plastic 2 definition 428 geometry, in fault-propagation folds 180
matrix 45 forelimb deformation in 175-80 and growth strata 113-14, 116, 118
models 83-4, 84 growth 101.114-17, 118-19,720, 121 imbricate stacks on 377
viscous fluid 19-20 limitations of models of 172 overturned 130, 166
Newtonian and non-Newtonian 84 models 53, 59, 68 rotation 172
passive 221 see also tip folds shear strain 143
paths 19 shear effects in 129-30, 130, 131-2 of thrust anticlines 184, 287
plastic 19,219 Wyoming 155-64, 757, 181-88 Formation Microscanner 391
power-law, plane and non-plane 41-50 tlexuralslip37,201,2l7 Alberta Foothills 401-2, 404
solutions, for stress and velocity 48-50 growth 101,232,321,322 analysis of tip fold structure 396-8,398
viscous 2-3, 21 definition of 428 image for Pardonet Fm., Rockies 394-5
Flower structures 102, 302 hangingwall 61, see also hangingwall Forward-breaking back thrusts 102
Fluid flow, isoclinal 168,226,361,413 Forward-breaking sequences 102, 103
during thrusting 215 kink-band 128, 133-4, 175, 177,4/9,421 Fracture geometry 404
models 84 lift-off 128, 429 for ramp anticline 400-2,402
Fluid pressure 29, 163, 219, 220, 221 limb length versus fault slip 105, 116, 118-19, tip fold 396-8
abnormally high 399 J20, 111 Fracture systems, in Canadian Rockies 391-404
elevated 398-9 nappe see nappes Fractures 125, 142, 161, 169, 187, 231
high pore 402 open 290, 292 conjugate 392. 399, 402,404
hydrostatic pore 398-9 overthrust shear 429 crack tip process zone of 181
influence on basal .strength 21 overturned 166, 184 curving 199
Fluvial transport model 4, 5-6 passive 217 extension 392, 398. 402, 404
Flysch 322 propagation along strike 67-8 in fold cores 172
sediments 300, 304 ramp anticlines see ramp anticlines gas-filled 404
Himalayan 346 ramp synclines 429 hydraulic 399
Pyrenees 238, 239 rates 105,706,709, 109, 111, 116 microscopic 43-4
Venetian Alps 322 recumbent 165, 326 mineral-filled 43
Foinaven duplex, Moine thrust zone 203, 203 self-similar growth 123,124. 125 mode(l) 168
Fold theory 37 sheath 217. 225 open 398, 399
Fold-thrust belts 53-4. 57, 164, 287, 410 syn-cleavage 242 oilhogonal 397, 399, 404
backthrusting in 77 tip folds see folding, fault-propagation shear 392, 398, 402
Chartreuse 287-97 tip-line 175, 186-7,261,355 see also cracks
curved 83-91 transported detachment 429 Fracturing 163, 295
Helvetic nappes 277 transported fault-propagation 429 in hangingwall 150, 151
imbricate, Alaskan 407-16 Foliation220, 410. 411,413 penetrative 290
inherited structures 297 axial plane 407 Friction 265
model 53-68 strain-insensitive 359 basal71-2, 73,77, 80
439
INDEX
variation in 78-9 H
coefficient of 23, 71, 73, 77, 217, 229-30
internal, of wedge 77 Haig Brook duplex, Canadian Rockies 207 Iberian plate 211, 270, 271, 272
Frictional resistance 162, 353 Halfway Formation, British Columbia 393, 396, and Pyrenean evolution 235, 238-9, 240, 243,
Frictional sliding 71 399 247,253
Friuli Platform 319, ill, ^11-^ Hangingwall, Idaho-Wyoming thrust belt 142, 350, 377-8,
Front Ranges, Canada 383, 385, 391, 393 vergent backthrusts 102 399
Frontal ramp anticlines 149 vergent thrust 100 Imbricate,
anticlines 93, 94, 149, 151, 155, 175,776, fans 71, 77, 80,103, 266, 377, 379, 427
182,237,361,400 of Absaroka thrust sheet 183
Chartreuse 293, 296, 297 of Central Pyrenees 175
and ramp-flat thrust models 191 -200 geometry 319, 323
bypass thrust /03 leading 427
collapse 142 trailing 427
Gallinera thrust 300 cutoff 112, 116, 118, 119, 186,237 roof sequences 350, 353
Ganges plain 345 angle 178 stacks 244. 265, 326, 377, 378
Garumnian Formation 248-9, 257, 267 map /83 break-back 262
Generoso basin 304 map for Rocky Valley 187 crestal 377, 383, 388
Geometric models, fault-related folding 175, point 350 faultcrest378, 380, 381
191 deformation 143,419 fold forelimb 378, 379, 380, 381, 383
Geothermal gradients 172, 219, 414, 415 flat 293 Pyrenean 274
assumed in models 9 geometry 149 Rocky Mountain example 382
Gerola well 304, 308 ramp 178-9 thrust systems 33, 241, 377, 386, 391, 411
Giffre line and zone 277, 282-4 cutoff 396 blind complex 427
Gilluly model for thrusting and folding 379-80, ramp-anticline 97 definition 424, 427
379, 386 shortcut faults 400 Pyrenean 239, 247
Giudicarie system 300 synclines 98, 151 thrusts 68. 157, 236, 345-6
Glaciers, analogy with 46, 48 truncation 159-60 development of 59
Glide horizons 83 Harpoon structures 93, 95, 100, 102, 103 Imbrication 204, 205, 330, 332, 340, 407
Global sea-level changes 322 Hanford dyke 38 of basement 357, 361
Gneiss 226, 239, 241, 329, 335, 337, 359, 361, Hartland Quay, Southwest England 201, 202, footwall 142,207,386
408,409,410,413 in Hogsback sheet 37
204-5
basement 363 Haystack Peak anticline 155, 156-8, 161. 163, of lower crust and mantle 281, 284
Gotthard sublid 282, 283 164 multiple 133, 137
Grabens 152 Hazara nappe 328, 328-9, 335 out-of-sequence 386
crestal collapse 93, 97, 98, 100, 102 Hazara syntaxis 327-8, 330, 332-3, 335, 336, Inclusions 187-8
influence of. on thrusting 309 336-7, 338-9 bands, in calcite fibres 198
inversion of 100 Hazelton Group, Skeena fold belt 366, 369 of ductile material 311, 312
reactivation of crestal-collapse 100 Helena Salient 210 ilmenite/rutile 328
Grain contacts, pressure solution at 170 Indenters 277
Helvetic nappes 165-72, 191, 277, 279, 280,
Grand Valley fauU 182. 183. 183 Adriatic279, 281,283, 284
281-3
Granite 320 in fold-thrust belt development 83
see also Alps
Carboniferous 319 rigid 339
Hercynian,
c lasts 411 India see Himalayas
basememt 243
leuco-357, 361 Indian Ocean, Owen Fracture Zone 339
basement 235, 236,2i7, 237-8, 239, 241, 257
in Pyrenean zone 238 Indian plate 7-9, 11, 13,210,325-41
of Southern Alps 299-300 duplex development in 343-55
Granulite facies 217, 225, 238 folding 211,213
Gravina sequence, British Columbia and Alaska geology 326-7
orogeny241,242, 242. 243 Indus Kohistan seismic zone 337, 337-8
409,410,411-12,413 palaeomagnetics 213
Gravitational, Indus valley, structures of 337-8, 345
Himalayas 16 Inherited tensional structures 320, 323
collapse 30 coseismic fauhing 33
in modelling 55 Instability envelopes 155, 161,162, 162-4
curve fitting 84,91 Insubric line 224, 277, 278, 279, 283, 284, 299
effects 1,71,90, 193,378,415 duplexes 229, 349
Gravity data 224, 240, 287, 348, 4/0 displacement along 281-2
foreland belt 33 timing of strike-slip on 279-81, 279
Gravity slides 165,378 earthquakes 34
Greenschist facies 217, 224, 225, 226, 239, 241, Insubric-Helvetic lid 277, 278, 279, 280, 281
Nahan thrust 34 Insular Superterrane 358, 358-9, 365, 407-8,
319,328-9,410,413 Northwest 325-41
Grenville orogeny 222 Oligo-Miocene unroofing 328 408. 409, 4\5
Grenville Province 226, 229, 230 Pakistan, time chart 338 Intercutaneous thrust wedge 427
Griffith Creek section, Skeena fold belt 368. palaeomagnetics of 210 Interference structures 325,326, 338
370,371-2 Sulaiman Lobe 343-55 Intermontane beh 358-9, 370, 372, 373
Grigna-Valtorta segments 301, 301-2, 303. 304, thrust detachment 244 Intermontane Superterrane 357, 361, 363. 365,
306, 306 Hinterland-to-foreland thrusting see piggy-back i66, 407-8.405,409, 410,
Grooves, on fault surfaces 290, 294 Hogsback thrust sheet, Wyoming 30, 35, 36, 415
Grooving 205 "37, 38 Internal zones 46, 327-8, 338, 340
Growth faults 98 Hot Springs eye-lid window 386 Himalayas 332, 334
Growth folds 101, 321, see also folding Hubbert-Rubey fault slip model 34 of thrust systems 325
Growth strata 707, 113-14 Hunter Creek anticline 155, 159-61, 760, 163-4 Intemides 218
rales of folding and faulting determined from Hydrocarbons 386, 402 and evolution of thrust sheets 217-31
105-21 entrapment of 142, 158-9, 388 Venetian Alps 319
Growth structures 101, 105 exploration for 377 Inversion 93-104, 328,4.32
Growth triangles 108, 112 use of palaeomagnetism in trap assessment of extensional fault systems 237, 239, 248-9,
Guidicarie line 279 215 269
Gulf of Mexico 105, 1/3 in half-graben system 93, 94, 94
Perdidofoldbeh775 of listric fault systems, 97, 97
Inversion structures 93, 94, 94, 95, 97
440
INDEX
441
INDEX
442
INDEX
443
INDEX
444
INDEX
heterogeneous 175, 178, 179 Subsidence, thermal 248, 249, 253 Tethys 290
internal 187,212 Sui fold 346-7, 348-9, 350, 355 opening of 300
lateral 141, 150,75/ Sukunka gas field 393 Tethys Ocean 278, 290, 300, 348
layer-parallel 169, 393 Sulaiman fold belt 343-55 Thermal subsidence 290
longitudinal 165, 169, 172, 177 borehole data 347, 348 Thick-skinned tectonics 217
non-plane, in surge zones 47-8 geology and tectonic map 344-5, 346,348 model for 240
penetrative 172, 213 seismic cross sections 347, 349, 351-2, 354 Thickening, crustal 325, 326
extensional 363 Sulaiman foredeep 343, 346, 348,349, 349 Thickness,
plane I, 150, 178, 195,313 Sunda Arc 95 effect on fold wavelength 191
plastic 21-2 Superstition Hills earthquake, California 37 in modelling 72-3, 89-90
rotational 334 Surface mass transport 15 stratigraphic,
seismic, partitioning of 38 Surface processes 3 effect on trajectory 94
Strain factorisation 142, 152, 175 model 9-10 variation in 75, 76,11, 90, 91, 297, 319,
for forelimb deformation 177-9 Surface transport calculation 9 320,323
Strain gradient 187 Surge zones 47 Thin-skinned fold-thrust belts 365, 391-404
Strain hardening 80, 231 Sustut Basin and Group, Skeena fold belt 367, Thin-skinned tectonics 165, 217-18, 221-2, 261,
Strain partitioning 59-61, 60-7, 68 367,370-1,372,373 357,358,414
Strain rate 19, 46, 50, 86-7, 164, 172, 217, 219, Sutures 222, 224 model for 97, 140, 141,240
221 Swat Valley, deformation in 328-9, 332 Thor-Odin culmination 359, 361
in deforming wedge 25, 28-9 Swat Valley, Himalayas, structures in 328-9, Thrust faulting,
distribution 1 332,337-8 break-back 241
extensional 29 Swiss foredeep basin 287 forward-breaking 271, 388
internal 20 Syn-inversion sequences 93 lateral propagation and 385
Strain softening 231 Syn-inversion strata 96, 97, 101,/O/, 102 layer-parallel 38
Stratigraphic pinch-out 83 Syn-rift sequences 94 multiple-stage 385-6
Stratigraphic separation, and oblique ramps 152 Syn-rift strata 93, 98, 100, 707, 102, 248, 290 out-of-sequence 184, 377
Stratigraphy, Synchronous, piggy-back 343, 377, 388
effect of, on thrust sheet geometry 187 sedirhentation 71 related to basal decollement, kinematic
mechanical 155-6, 161, 163,366 thrust faulting 377-88 sequence for 314
variation in, across thrusts 241, 287 Syntaxes 328, 341 sequence of see thrust sequences
Strawberry Creek drainage, Wyoming 157 Hazara 330. 332, 336-7 synchronous 377-88
Stress 1, 19,38,48 Himalayan 325, 327 Thrust faults 53, 80, 217, 325, 422
depth, diagram 2/9 Syntectonic deposition 138-9, 140, 322, 377, backthrusts, definition of 419
compressive 282, 415 378,414 backthrusts see backthrusts
controls on fold thrust style 155-64 Pyrenean 235, 236, 247,249, 263, 266 blind 67, 123, 159, 165, 202, 263, 266, 287,
critical 162 320,321,337,338,341,
deviatoric 2, 22, 46, 161, 221 359
differential 50, 57, 161 blind see also buried
failure 161 branching 293
internal, buried 237, see also blind
in critical wedge 71 conjugate sets 199
in orogen 1 Tabas-e-Golshan mainshock, Iran 34 cutoffs 295-6
mean 161 Tadri anticline 346, 348, 349, 350, 353, 355 'deactivated' 377, 383, 386
Mohr circles for, Cumberland Plateau 49 Taiwan 343, 350 definition of 419
plane flow 47 accretionary wedge 229 dip domain geometry 318
principal 49, 161, 164 fold 20 displacement 59, 63, 65, 181, 292
regional 321 thrust belt 133 emergent 420
yield 28 Taperl9, 20,253, 263 flat 353, definition 420
Stress field, critical 25, 377, 388 floor 201, 202, 203, 247,255, 261, 343, 347,
coupling to internal strain rate 20 Tauem Window 224-5 355
inhomogeneous 90 Tear faults 150-2,262 footwall,
Stress paths 155, 163 see also faults -vergent 97
Stress profile 41, 48, 50 Tectonic delamination 239, 243 imbricates 308
Stress regime 404 Tectonic loading see thrust sheets shortcut 93, 94, 100,421
Stress space 155, 161, 163 Tectonic slides 225 footwall-vergent 98. 100
Stress-strain rate relation 44 Tectonic thickening see thickening foreland-vergent 53, 75, 76, 77, 80,421
Striae 287, 294, 296, 302 Tectonic transport 333 frontal 19,240
stereonets for 293-4 Tectonic velocities 9 hangingwall-flat421
Strike-slip 415 Tectonics, hangingwall-vergent (backthrusts) 93, 97
fault systems 235 intra-arc 407 hinterland-vergent. definition 421
faulting 90, 91, 175, 239, 240, 277, 353, 365, and sedimentation 247, 251, 255, 261 imbricate 296, see also imbricate
374 Tejerina fold, Cantabria 175, 777, 177-80,178, Iink201,202, 203, 205-8
Pyrenees 238 779 out of graben 97, 98, definition 422
see also faults Tell Atlas, Algeria 33, 34 out-of-sequence 217, 222, 230, 241, 257, 299,
structures 102 Temperature, and thrust emplacement 211, 272 302, 304, 309, 377, 378,
tectonics 339-40, 343, 344, 349 Tennessee thrust system 181,184, 184-6, 7S5- 381
Alpine 278-81 7, 187-8 out-of-the-syncline 378, 422
terrane 213-14 Tertiary, overlapping 61,429
Stylolites 43, 125, 177,399 extension, Rockies 415 passive roof 343, 346, 347,351, 353, 355
Subalpine chain 282 molasse 287, 289, 293 backthrust 252
Subalpine chain, France, Chartreuse fold 287- Himalayas 327, 327, 328, 330, 335, 336, propagation of 53-68, 57, 59-61,65,67
97, 288-96 346, 348 ramp-flat 371
Subduction 25, 217, 221, 231, 235 rocks and palaeomagnetics 209 roof 251, 330, 343, 350, 355, 359,401
Alpine 277, 283, 284 of Southern Pyrenees 247, 249-52,250, 253, flexural-slip duplex 201-3, 205-6, 207
American 373-4, 415 255, 257, 258 'sharp frontal shear zone-type' 314, 316,317,
crustal 243, 244, 247, 252, 325, 326 turbidite facies (Pakistan) 205 318
Himalayan 328, 339 Tethyan margin 348 sole311-l8, 357, 365, 386,423
445
INDEX
Himalayan 329, 349, 353 Trailing-edge anticlines 184 role of crustal duplexes 229
Pyrenean247,261,262, 271 Transfer zone 213,322, 322, 323, 401 Sulaiman fold belt 353
see also sole Transform, continental 339 of thrust fault 152
thick ductile 318 Transport directions 143, 152, 240, 262, 263, zones 75
spacingof57, 75,76, 77, 80 265,274,281,287,300,306 US Rocky Mountains 377-88 see also Rocky
splay 295, 423 divergent 281, 284 Mountains
terminations 187, 335-6, 339 Himalayan 332-5, 337, 347 Utah, Uinta Basin 398-9
Thrust ramps see ramps lateral escape 299, 308
Thrust sequences 19, 54, 57, 93, 100,314, 315, material, in folds 125
343 and mesofabric 226
backthrust 350, 353 of oblique ramp 141
break-back 97, 98, 102, 103, 207, 337, 386, regional 151
388 Transpression 152, 277, 279, 280, 284, 343,
Pyrenean 236, 247, 252, 253, 266, 271 344 Valhalla complex, Canadian Cordillera 359,
critical Coulomb wedge model 73, 74, 75 in Southern Alps 299-309, 319-23 361
definition 430-1 in Southern Alps, New Zealand 9 Valley and Ridge Province, Appalachians 379,
forward-breaking 97, 431 Transtension 306 392
and hydrocarbon exploration 388 Transverse Ranges. California 33, 37. 116 Valsugana line 319-20, i20-2, 321-3, 323
in-sequence 388, 431 Transverse zone 277. 282-3 Valtorta segment 299, 30J-3, 302, 306-7, 306,
inverted extensional regime 102 chronological development of a 304, 306-8 309
out-of-sequence 251, 252, 431 cross section of 309 Variscan,
passive roof 350 kinematics of 299-309 basemem224, 226, 319
piggy-back (forward-breaking) 137, 265, 266, Tremp basin 236, 251, 267, 269, 211,273 orogeny 201
431 Trento Platform 319, 322, 322-3 Veins 43, 44, 125, 169, 170, 198,295
re-imbrication of 328 Triangle zones 391, 401 bedding-normal 168
roof 349, 350 Alpine 319, 320-1,323 calcite/quartz fibre 201, 202
wedge models 80 definition 427 dilatant 165, 166-8,769, 172
Thrust sheets 48. 71, 77, 186, 23\,23I, 326, Skeena fold belt 365, 366, 370, 371 en echelon 168, 170
353 Triassic, extensional (tensile) 45, 168
basement 241 Ballabio-Barzio zone 302, 302-3, 304 fibre 208
-involved 236 cataclasites 304 on fold forelimbs 172, 177
blind 361 detachment horizons. Chartreuse 292 mineral-filled 43, 44, 166
carbonate 391 evaporites 136 orientations and timing 168
crystalline 217-31, 326-7 Nugget sandstone, Wyoming 36, 37 patterns, cross-cutting 45
deformation in 209 palaeomagnetic work on 2 / / , 211-12, 214 quartz 202, 205
emplacement of 41, 47, 224, 386 Pyrenean 237, 238 shear 168
flow of 48 Southern Alps, Italy 299, 304, 306-8 Velocity,
loading 249. 251, 252, 373, 388,414 stratigraphy of 300, 301-2,302 field 48
molasse-bearing 327 Tuaton Lake fault, Skeena fold belt 307-8, 370, particle 19
movement of 41, 46-7, 50 372 profiles 41
strength of 28, 29, 46, 49 Turbidites 201, 205, 239, 409, 410, 411 in laminar flow 47
Type C megathrust 221-2, 223, 224-5 deposition of 247. 248, 249, 251, 253 Velocity distribution 1
Type F 221, 225-6, 227, 229 Turkey, North Anatolian fault in 34 Venetian Alps 319-23
types of 2J7, 231 cross-section across 321
Thrust systems 61, 181 inherited structures of 319
cratonward-progressing 377 Venezuela 105,113, 113
duplexes see duplexes Ventura Basin,Califomia 114
evolution of 133-40,540 Verkhoyansk Range, USSR 85, 87, 91
modelled 73, 74, 75 Villalvemia-Varzi-Levamo line 279, 279, 281
foreland-vergent 71 Ultracataclasites 42-3 Virginia 53, 55 see also Appalachians
geometry 77, 78-9, 80, J39, 187 Unconformities 289, 320-1, 321-2, 323, 344-5, Visco-elasticity theory 37
horses 201, 205, 350, 353, 426 346 Viscosity 2, 37, 84
imbricate see imbricate angular H I , 113, 115-16, 139,322 in finite element methods 23
intercutaneous thrust wedge 427 in anticlinal limbs 261 Viscous,
kinematics of 181 in Pyrenees 237, 247, 248, 271 creep 45, 46
out-of-sequence 377, 378, 385 Unconsolidated rocks 144 fluid, shear of 45
piggy-back 133, 249, 257, 377, 378, 380, 386 Underplating 20, 30, 71-2, 231, 2S0, 281-2, 284 model for compression 1, 2
stacked 340, 410 duplex 377 Volcanic rocks 409
synchronous 379-80, 383-8 sediment 25 deformation style 372
triangle zones see triangle tectonic 277 von Mises yield criterion 19, 21, 22, 23
Thrust vergence, Underthrust cushions 312 Voreppe thrust. Chartreuse 292-3, 292-3, 294-5
foreland-vergent 260, 261, 353 Underthrusting29. 312
hinterland-vergent 260, 261, 353 in coupled tectonic-erosion model 9
Tibetan thrust system 325. 332, 339 plates, slip velocity on 19
Tie-lines, structural 282 of sediment 19
Tintina fault 225 Unroofing 1,224,328,340 W
Tip, due to extension 20
folds 181-2, 188, 263, 295, 337-8 Uplift 1,2,9, 11-15,80,98, 100 Wales,
fracture development in 391, 393-5, 396- in Alps 277,280, 281, 282. 287. 321 palaeomagnetics of 212
400, 404 control on stratigraphic patterns 139 South Wales coalfield 96
see also folding footwall 289 Walhalla nappe 226,227-9
lines 175, 181,261,293 and fracture development 399 Wall rock, 43-4, 45, 198
definition of 423 Himalayan 329, 335-6, 338, 340, 343, 350 Waterton Field, Canadian Rockies 379, 381,
structures 63, 181-2, 183.201 isostatic 281 382-4, 383
thrust 49, 50, 152, 187, 191,294 lower crust and mantle 240 Wedges.
deformation 186, 188 rates 108,415 clastic 373
Topographic slope, of wedge 75,77 in Cordillera 413-14 Coulomb see Coulomb
446
INDEX
critical state 1 Wyoming-Idaho thrust belt 155, 163-4, 181, Yield point 36, 38
taper 71,377 265,350 of rigid-plastic material 2
internal deformation of 229 palaeomagnetics of 210 Young's modulus 37
modelsof41.71,75,77, 80, 85 see also Idaho Yukon-Tanana sheet, Canadian Cordillera 225,
otogenic 252 409
taper of 19
Whitney Canyon-Carter Creek anticline 155,
158-9,/59, 161, 163, 164
Widemouth Bay, Southwest England 201, 202,
205,207
Wildhom nappe, Switzerland 165, 166, 168 Xenoliths413
Windows 224, 430 Zagros Mountains,
eye-lid 386, 388 evaporites of 34
tectonic 238, 359 fracture development 392
Windows phase, Alps 279, 281
Work hardening 38
material 36-7
Work softening 36-7, 38 Yield,
Wrangellia407, 414, 415 envelope of Coulomb materials 86-7
Wyoming Salient 33 strength, of Coulomb material I
slip lineations 91 stress 19,21,22,23,49
thrust sheet 35, 36 plastic 47
thrusts of 36, 37 of wedges 41
447