Regional Geology and Tectonics: Principles of Geologic Analysis: Volume 1: Principles of Geologic Analysis [1, 2 ed.] 0444641343, 9780444641342

Table of contents :
Cover
Regional Geology and Tectonics: Volume 1: Principles of
Geologic Analysis
Copyright
Contents
List of Contributors
Foreword
Introduction
1 Regional geology and tectonics of sedimentary basins
References
2 The Earth: core, mantle and crust
Overview
Methods of investigation
The lithosphere
The oceanic crust
The continental crust
The subcrustal lithosphere
The mantle
The core
References
Further reading
3 Age of the oceans
Introduction
Ocean basin research – historical perspective
Dating the oceanic lithosphere
Magnetic anomaly data
Geomagnetic timescales
Limitations and uncertainty in magnetic anomaly data
Drill-core data
Limitations and uncertainty in dated samples
Age proxies
Present-day oceanic lithosphere
Volcanic additions and extended continental margin features
Mapping seafloor age
Gridding methods
Uncertainty quantification
Final remarks
References
Further reading
4 Plate boundaries and driving mechanisms
Introduction
Boundaries
Divergent/constructive
In oceanic lithosphere
In continental lithosphere
Convergent/destructive
Involving subduction
In continental lithosphere
Conservative/strike-slip
In oceanic lithosphere
In continental lithosphere
Between oceanic and continental lithosphere
Oblique, partitioned-strain, and diffuse plate boundaries
Driving mechanisms of plate motion
The torque balance
Negative buoyancy of slabs (slab pull and slab-normal forces)
Slab resistance and collision resistance
Slab suction
Transform shear
Gravitational body forces/rifts and ‘ridge push’
Basal shear force
Summary
References
5 Plate kinematic reconstructions
Introduction
Making plate kinematic reconstructions
The history and workspace of plate kinematic reconstructions
Rotations
Requirements for reliable reconstructions
Choices of markers
Apparent paleomagnetic poles
Intraplate volcanic chains/hotspots
Magnetic reversal isochrons
Fracture zones
Continent–ocean boundaries
Piercing points from intracontinental structures
Diffuse and regional markers
Mantle constraints – subducted slabs in tomography
Model production and assessment
Interactive visual forward modelling
Statistical modelling
Regional/global models with multiple plates
Absolute plate motion and reference frames for reconstructions
Hotspot reference frame
Paleomagnetic reference frame
Lower mantle slabs reference frame
Large Low-shear-velocity provinces
Using plate reconstructions
As context for regional geological and tectonic studies
The Alps as a previously extended continental margin
The collision of India with Eurasia and shortening of the Indian continent
Plate reconstructions as boundary conditions for palinspastic reconstructions and palinspastically reconstructed markers in...
In paleogeography/paleobathymetry/paleotopography
For geodynamic studies
The Indian plate and plume influences on plate motion
Global tectonic reorganizations
Summary
References
Further reading
6 Resolving geological enigmas using plate tectonic reconstructions and mantle flow models
Introduction
The evolution of the plate reconstruction method
Global plate reconstructions
Relative and absolute plate motions
Early plate reconstruction approaches
Linking plate reconstructions with mantle flow
Using global plate reconstructions to better understand the Earth system
India-Eurasia collision
Sundaland and New Guinea
Conclusions
Acknowledgements
References
7 Tectonostratigraphic Megasequences and Chronostratigraphy
References
Further reading
8 Fault classification, fault growth and displacement
Introduction
What is a fault?
Fault anatomy
Fault drag
Fault orientations, stress, strain and kinematics
Relation between faults and stress
Strain and fault orientation patterns
Displacement distributions on faults
Fault initiation
Fault formation from scratch
Faulting by activation of preexisting structures
Fault growth
Fault interaction and linkage
Fault populations
Faults and fluids
Concluding remarks
Acknowledgments
References
9 Thrust systems and contractional tectonics
Historical perspectives – 100 years of thrust belt research
The geometry of thrust systems
Thrusts in three dimension
Balanced cross-sections
Insights from marine seismic imaging
Mechanical context: the critical wedge
Basement and crust
Other structural styles
Thrust sequences and activity
Interpretation – looking ahead
References
Further reading
10 Inverted fault systems and inversion tectonic settings
Introduction
Reactivation of earlier fault systems
Defining a change in stress regime
Recognizing inversion in settings dominated by thin-skinned structures
Recognizing inversion in transpressional and transtensional settings
Role of inversion in facilitating propagation of larger fold-thrust belts
Case Study 1: inversion of extensional faults in a foreland basin, Western Newfoundland, Canada
Case Study 2: inversion of extensional faults in a collapsing compressional orogen, Northeast Thailand
Case Study 3: inversion of intermittent extensional faults in multiple tectonic settings and interaction of thick- and thin...
Case Study 4: variable inversion of extensional faults with different orientations in a failed intracratonic rift, Central ...
Case Study 5: orogen-scale inversion of extensional faults in a rift system that evolved into a back-arc basin and regional...
Case Study 6: inversion of extensional faults in a failed continental rift, the Wessex Basin, southern United Kingdom
Case Study 7: inversion of extensional faults in a rift basin: New Zealand
Case Study 8: inversion of extensional faults in Late-Orogenic molasse basins that evolved to a later continental margin an...
Case Study 9: inversion with significant mechanical contrasts and ductile deformation near a collisional suture on a distal...
Inversion structures and economic implications: petroleum system elements and mineral deposits
Timing of inversion relative to extension
Conclusion
Acknowledgements
References
Further reading
11 Salt- and shale-detached gravity-driven failure of continental margins
Introduction
Gravity-driven failure
Mechanics
Processes
Distribution of detachment layer
Structural styles
Extensional province
Translational province
Contractional province
Strike-slip structures
Diapirism
Extensional salt diapirism
Contractional salt diapirism
Strike-slip salt diapirism
Loading-driven salt evacuation and diapirism
Diapir dissolution
Shale diapirism
Allochthonous salt
Emplacement
Styles
Distribution and impact on gravity-driven failure
Concluding remarks
Acknowledgements
References
12 Carbonate systems
Introduction
The conceptual space in sedimentary geology
Carbonate factories
Food and feeding/follow the food
The proton link
The food web
Boundary layers
Fair-weather and storm wave bases
Preservation potential of sedimentary structures
Light penetration
Pycnoclines
Deep pycnoclines
Shallow pycnoclines
Internal waves
Hummocky cross-stratification
Feed and food: feasting at the pycnoclines
Nutrients and plankton
Mounds and platform margins
Carbonate production modes
The proton play
Biotic carbonate production modes
Biologically induced carbonates
Biologically influenced carbonates
Biologically controlled carbonates
Foraminifers
Coccoliths
Molluscs
Carbonate production systems through Earth’s history
Archean
The great oxygenation event
Banded iron formations, were them the photosynthetic oxygen sinks?
Early Proterozoic carbonates
The Meso-Neoproterozoic carbonate production modes
Prokaryotes, eukaryotes and multicellular forms
The Phanerozoic carbonates
Feed and food again: the eukaryotic phytoplankton
The green plastid lineage
The red plastid lineage
Sclerotization: two episodes
The Phanerozoic performance
Early Palaeozoic
Middle Palaeozoic
Late Palaeozoic
Triassic-Jurassic
Cretaceous
Cenozoic
Platform types
Rimmed platforms
Microbialite rimmed platforms
Carboniferous platforms in Asturias
Triassic Latemar platform
Nonskeletal and skeletal metazoan rimmed platforms
Devonian carbonate platforms from Canning basin
Permian Capitan
Waulsortian-like mud mounds
Skeletal rimmed platforms
The Miocene Llucmajor Platform, Mallorca: sea-level attached rim
Upper Cretaceous Vilanoveta platforms: pycnocline-related attached rim
Palaeocene (Danian) Lizarraga platform
Physical accommodation-predominant platforms: grainy systems
Euphotic shallow-water production
Meso-oligophotic production
The Miocene Migjorn ramp
The Eocene Urbasa-Andia low-angle ramp
The Oligo-Miocene Perla Field, offshore Venezuela
The Eocene Buil nummulitic banks
Permian Upper San Andres Formation
Distally steepened ram versus infralittoral prograding wedge
Oligophotic to aphotic production
The Middle Miocene Lazio-Abruzzo low-angle ramp
Mud dominated–producing biota
Corollary
Mono- versus multifactory platforms
Mono-factory platforms
Multifactory platforms
Interaction among coeval carbonate factories: promotion versus suffocation
Alternation of carbonate factories
Skeletal grain associations: use and abuse
Carbonate platform shedding
Sequence stratigraphy in carbonates: illusion, mirage or hallucination
Acknowledgements
References
13 Lake systems and their economic importance
Introduction to lakes and lake systems
Lakes in time and space; preservation of lakes in the Phanerozoic rock record
Classification of lakes; the different settings for lakes
Conditions needed to create and maintain a tectonic lake with well-developed lake sequences
Megasequences, sequences and cycles in basins containing tectonic lakes
Controls on lake sequences and sequence stratigraphy
Important differences between lake and marine sequence stratigraphy
Principal depositional environments in lake basins
Predicting lake sequences and facies
Major petroleum systems involving lake sequences
Features of potential petroleum source rocks that develop in lakes
Petroleum systems in Early Cretaceous and Tertiary lake basins of South and East Asia
Acknowledgement
References
14 Clastic shorelines and deltas
Introduction
Shoreline and deltaic processes
Waves and associated processes
Tidal processes
River-mouth processes
Sediment gravity flows
Ichnological processes
Universal building blocks of all clastic systems
Deltas: river-fed shorelines
Classification of deltas
Depositional environments of deltas
Facies and architecture of deltas
Case studies of deltas
Wave-dominated, nondeltaic shorelines
Tide-dominated, nondeltaic shorelines
Acknowledgements
References
15 Tidal straits: basic criteria for recognizing ancient systems from the rock record
What are tidal straits?
Why are tidal straits important?
Sedimentary dynamics of modern tidal straits (what we presently know and what we still need to know)
Towards a conceptual model for tidal straits
The hydrodynamics of tidal straits
Definition of ‘flood’ and ‘ebb’ tidal components in a strait
Cross-sectional distribution of the tidal power in a strait
Tidal asymmetry in straits
Main depositional zones in tidal straits
The strait-centre zone
The dune-bedded zone
The strait-end zones
The strait-margin zones
The stratigraphic and sedimentary record of ancient tidal straits with some example
The Cretaceous Western Interior Seaway
The late Miocene multiple straits of the Betic Corridor
The Quaternary straits of Calabria, southern Italy
Criteria for recognizing tidal straits in outcrop or subsurface successions
Stratigraphic criteria
Location of the strait-centre zone
Large-scale stratigraphic architectures of strait-fill dune-bedded complexes
Vertical facies tracts
Ancient tidal dunes in straits and their internal architectures
Sedimentological criteria
Strait-centre facies
Dune-bedded facies
Herringbone cross-stratification
Trough or festoon (three-dimensional) and tabular or planar (two-dimensional) cross-stratification
Simple and compound foreset architectures
Reactivation surfaces
Neap-spring, coarsening-to-fining lamina intervals
Tidal bundles
Strait-end facies
Strait-margin facies
Concluding remarks
References
Further reading
16 Submarine landslides – architecture, controlling factors and environments. A summary
Introduction
Classifications
Types of movement
Frontally confined versus frontally emergent landslides
Attached versus detached landslides
Structural architecture of submarine landslides
Headwall domain
Translational domain
Toe domain
Mechanics of slope failures, preconditioning and triggering factors
Slope steepening
Pore fluid pressure
Earthquakes
Waves
Sediment types
Environments
Fjords
Deltas on continental margins
Submarine canyons
Open continental slopes
Oceanic volcanic islands
Statistics of submarine landslides
Concluding remarks
References
17 Turbidites and turbidity currents
Introduction
A historical perspective
Introduction
The turbidite concept in the 1960s and 1970s: the pioneering works on the field and laboratory experiments
The turbidite concept in the 1980s and early 1980s: the genetic facies tracts by Lowe (1982) and Mutti (1992)
The main insights in the 1990s: Kneller’s model
The latest insights on turbidite facies and processes
Mud-rich transitional flows and slurried ‘hybrid’ facies
Upper flow regimes structures and supercritical fans
Modifications of the facies tract induced by the relationship between flow type and basin morphology
Turbidite systems: relationship between degree of efficiency, degree of tectonic confinement and type of basin
Some concluding remarks
Acknowledgements
References
18 Controls on reservoir distribution, architecture and stratigraphic trapping in slope settings
Introduction
Accommodation
Ponded
Healed slope
Incised submarine valley
Slope
Delivery configuration
Slope profile types
Graded and out-of-grade profiles
Bypass slopes
Toes-of-slope
Above-grade profile
Ponded slopes
Stepped slopes
Discussion
Conclusion
References
19 Geological methods
Introduction
Satellite images and data
Hotspots, rifts, reefs, deltas and cratonic regions – views from space
Geological and tectonic maps
Topography
Surface geology
Tectonic maps
Integrated interpretation – the Appalachians
Acknowledgements
References
Further reading
20 Regional tectonics and basin formation: the role of potential field studies – an application to the Mesozoic West and Ce...
Introduction
Gravity and magnetic coverage over continental areas
Terrestrial gravity
Terrestrial magnetics
Satellite-derived gravity and magnetic coverage over oceanic areas
Satellite gravity
Satellite magnetics
Offshore plate tectonic links to the West and Central Africa Rift System
West and Central Africa Rift System
Evolution of the West and Central Africa Rift System
Stratigraphic unconformities and tectonics
F1 rifting period
F2 rifting period
F3 rifting period
Conclusion and implications
References
Further reading
21 Wide-angle refraction and reflection
Introduction
Wide-angle acquisition
Seismic sources
Receivers
Modelling wide-angle data
Travel-time analysis
Amplitude modelling
Examples of wide-angle seismic interpretations
Oceanic crust
Lithosphere stretching in the North Sea
Volcanic continental margins
Subbasalt imaging
References
22 An introduction to seismic reflection data: acquisition, processing and interpretation
Introduction
The reflection seismic method
Seismic resolution
Acquisition of reflection seismic data
Terrestrial surveys
Marine surveys
Three-dimensional surveys
Broadband seismic
Seismic processing
Data preparation
Demultiplexing
Trace editing
Data correction
Amplitude corrections
Noise attenuation
Static corrections
Velocity analysis (normal move-out and dip move-out)
Migration
Data reduction and enhancements
Common midpoint stacking
Multiple attenuation and deconvolution
Filtering and scaling
Seismic interpretation
Advanced techniques
Depth conversion
Amplitude versus angle analysis
Impedance inversion
Forward modelling
Spectral decomposition
Semiautomated horizon picking
Geographical information system
Summary
Acknowledgements
Glossary of key terms
References
23 Sequence stratigraphy
Introduction
Scope of sequence stratigraphy
Development of sequence stratigraphy
Stratigraphic resolution
Controls on sequence development
Accommodation versus sedimentation
Accommodation
Sedimentation
Concept of ‘base level’
Allogenic versus autogenic controls
Sequence stratigraphic framework
Depositional systems: definition and scales
Systems tracts: definition and scales
Systems tracts in downstream-controlled settings
Falling-stage systems tract
Lowstand systems tract
Transgressive systems tract
Highstand systems tract
Systems tracts in upstream-controlled settings
High-amalgamation systems tract
Low-amalgamation systems tract
Nomenclature of systems tracts
Stratigraphic sequences: definition and scales
Types of stratigraphic sequence
Depositional sequence
Genetic stratigraphic sequence
Transgressive-regressive sequence
Surfaces of sequence stratigraphy
Subaerial unconformity
Basal surface of forced regression
Correlative conformity
Maximum regressive surface
Maximum flooding surface
Transgressive surface of erosion
Regressive surface of marine erosion
Three-dimensional stratigraphic architecture
Hierarchy in sequence stratigraphy
Approaches to stratigraphic classification: absolute versus relative scales
Hierarchy systems: approaches to nomenclature
Hierarchy systems: orderly versus variable patterns
Model-independent hierarchy: basin-specific stratigraphic frameworks
Discussion
Sequence stratigraphy in the context of the ‘modelling revolution’
Workflow of sequence stratigraphy
Standard methodology and nomenclature
Conclusions
Acknowledgements
References
24 Concepts of conventional petroleum systems
Introduction
Clarification of terminology
The essential ingredients
Trap
Spill and leak points
Reservoir
Seal
Hydrocarbon charge
Play-based exploration workflow
Basin evaluation
Rift basins and passive margins
Cenozoic deltas
Deepwater foldbelts
Deep-water passive margins
Fold- and thrustbelts and foreland basins
Play evaluation
Prospect evaluation
References
Further reading
25 The accumulation of organic—matter–rich rocks within an earth system’s framework*
Plate reconstructions
Key aspects of individual plate reconstructions
Cambrian (500Ma)
Ordovician (450Ma)
Late Devonian – Early Carboniferous (Frasnian – Famennian – Tournasian) (375Ma)
Late Permian (250Ma)
Late Jurassic (Callovian – Kimmeridgian) (154Ma)
Late Cretaceous (Cenomanian – Turonian) (89Ma)
Eocene (49Ma)
Present day
Proximate controls on accumulation of organic matter
Source rock settings
Application of concepts to predicting source accumulation
Qualitative analysis of palaeoenvironmental maps for predicting source rock distribution
Applications of the concepts to the Permian
Late Permian
Summary of organic matter–rich rock accumulation in the Late Permian
Applications of the concepts to the Jurassic
Late Jurassic
Summary of organic matter–rich rock accumulation in the Late Jurassic
Quantitative analysis of palaeoenvironmental maps for predicting source rock distribution
Application of concepts to evaluating mudstone/shale reservoir character and distribution
Conclusion
References
26 Modelling fluid flow and petroleum systems in sedimentary basins
Introduction
Principles of fluid flow in sedimentary basins
Fluid flow in porous media
Generation and preservation of abnormal pressure
Overpressure detection and modelling
Measuring pressure
Modelling pressure
Hydrocarbon fluids and charge modelling
Generation and expulsion of hydrocarbon fluids
Migration and entrapment of petroleum fluids
Evolution of properties of petroleum fluids
Modelling migration
Petroleum systems analysis
Source rock presence and quality
Timing of petroleum generation and migration
Trap capacity and trap integrity
Modelling the petroleum system
Petroleum systems of unconventionals
References
Global Maps
27 Tectonic and basin maps of the world
Global geological maps: introduction
Global geological cartography: selected milestones
Stratigraphic nomenclature and the geological time scale
References
Introduction
Global relief models: onshore and submarine morphology and plate tectonic regimes (Plates 27.1 and 27.2A and B)
References
Neotectonics: introduction
Global earthquake distribution (Plates 27.3–27.6)
Well-defined versus diffuse plate boundaries (Plates 27.7–27.10)
Neotectonic plate motions: their relation to a fixed Eurasia and to Cenozoic/Mesozoic fold belts (Plate 27.11)
References
References
Introduction
The continental lithosphere (Plate 27.12)
The continental crust (Plate 27.13)
Crustal layers, rheological models, and conclusions
References
Plates for global topography, neotectonics, the continental lithosphere and crust: segments 27.1–27.3 and 27.5 (For online ...
Introduction to tectonic maps
Recent advances in alpine tectonics: an example of the scope of larger-scale tectonic maps
Simplified tectonic maps of the world (Plates 27.14–27.16A and B)
About Phanerozoic plate tectonic reconstructions
References
Arctic tectonic map (Plate 27.16A)
Antarctic tectonic map (Plate 27.16B)
References
Orogeny versus epeirogeny
Subduction, sutures, and orogens (Plates 27.17 and 27.18)
Active margin fold and thrust belts
Foreland fold and thrust belts
Normal faulting in foreland fold and thrust belts
References
Introduction to basements, that is, the ‘residual’ peneplaned former fold belts
Merging the global tectonic map with a Precambrian basement map
References
Introduction
Large igneous provinces (LIPs) (Plate 27.24)
Giant radiating dike swarms (maps b-6 and b-7)
Is there a ‘canonical progression of tectonic themes’ preceding and/or following the emergence of a plume?
The distribution of active volcanoes (Plate 27.26)
References
Introduction
Subducted oceanic plateaus
Allochthonous accreted oceanic plateaus and intraoceanic island arc terranes
Allochthonous fragments, oceanic, and intraoceanic arc systems, and lower crust and uppermost mantle of hyperextended passi...
Allochthonous, exhumed continental crust–mantle transitions and the Ivrea-Verbano Zone
Conclusion
References
Plates for tectonics, orogenic systems, hot spots, large igneous provinces, volcanoes: segments 27.6–27.10 (for online vers...
Introduction
References
References
References
References
References
References
General reference
Sedimentary basins and rifts: Segment 27.11 (for online version of the plates/figures cited in this chapter, the reader is ...
Index
Back Cover

Citation preview

REGIONAL GEOLOGY AND TECTONICS

REGIONAL GEOLOGY AND TECTONICS Volume 1: Principles of Geologic Analysis SECOND EDITION Edited by

NICOLA SCARSELLI Department of Earth Sciences, Royal Holloway, University of London, Egham, United Kingdom

JU¨RGEN ADAM Department of Earth Sciences, Royal Holloway, University of London, Egham, United Kingdom

DOMENICO CHIARELLA Department of Earth Sciences, Royal Holloway, University of London, Egham, United Kingdom

DAVID G. ROBERTS Department of Earth Sciences, Royal Holloway, University of London, Egham, United Kingdom

ALBERT W. BALLY Department of Earth Science, Rice University, Houston, TX, United States

Elsevier Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom 50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States Copyright © 2020 Elsevier B.V. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress ISBN: 978-0-444-64134-2 For Information on all Elsevier publications visit our website at https://www.elsevier.com/books-and-journals

Publisher: Candice Janco Acquisitions Editor: Amy Shapiro Editorial Project Manager: Ruby Smith Production Project Manager: R. Vijay Bharath Cover Designer: Mark Rogers Typeset by MPS Limited, Chennai, India

Contents

List of Contributors Foreword Introduction

Using plate reconstructions Summary References Further reading

xi xiii xv

1. Regional geology and tectonics of sedimentary basins

6. Resolving geological enigmas using plate tectonic reconstructions and mantle flow models

IAN JAMES STEWART

References

SABIN ZAHIROVIC

5

Introduction The evolution of the plate reconstruction method Global plate reconstructions India-Eurasia collision Sundaland and New Guinea Conclusions Acknowledgements References

2. The Earth: core, mantle and crust C. MARY R. FOWLER

Overview Methods of investigation The lithosphere The mantle The core References Further reading

82 87 88 91

7 9 11 15 17 19 20

93 93 95 101 102 105 108 108

7. Tectonostratigraphic Megasequences and Chronostratigraphy IAN JAMES STEWART

3. Age of the Oceans

References Further reading

LUCI´A PE´REZ DI´AZ

Introduction Dating the oceanic lithosphere Mapping seafloor age Final remarks References Further reading

21 24 30 34 38 40

8. Fault classification, fault growth and displacement HAAKON FOSSEN

Introduction What is a fault? Fault orientations, stress, strain and kinematics Displacement distributions on faults Fault initiation Fault growth Fault interaction and linkage Fault populations Faults and fluids Concluding remarks Acknowledgments References

4. Plate boundaries and driving mechanisms GRAEME EAGLES

Introduction Boundaries Driving mechanisms of plate motion Summary References

41 41 53 57 57

5. Plate kinematic reconstructions

9. Thrust systems and contractional tectonics

GRAEME EAGLES

ROB BUTLER AND CLARE BOND

Introduction Making plate kinematic reconstructions

116 117

Historical perspectives 100 years of thrust belt research The geometry of thrust systems

61 63

v

119 119 124 127 129 134 135 138 141 144 144 144

149 153

vi

CONTENTS

Thrusts in three dimension Balanced cross-sections Insights from marine seismic imaging Mechanical context: the critical wedge Basement and crust Other structural styles Thrust sequences and activity Interpretation looking ahead References Further reading

155 156 158 159 161 163 164 165 166 167

10. Inverted fault systems and inversion tectonic settings MARK COOPER AND MARIAN J. WARREN

Introduction 169 Reactivation of earlier fault systems 173 Defining a change in stress regime 176 Recognizing inversion in settings dominated by thin-skinned structures 177 Recognizing inversion in transpressional and transtensional settings 177 Role of inversion in facilitating propagation of larger fold-thrust belts 179 Case Study 1: inversion of extensional faults in a foreland basin, Western Newfoundland, Canada 179 Case Study 2: inversion of extensional faults in a collapsing compressional orogen, Northeast Thailand 182 Case Study 3: inversion of intermittent extensional faults in multiple tectonic settings and interaction of thick- and thin-skinned compression, British Columbia Foothills, Canada 183 Case Study 4: variable inversion of extensional faults with different orientations in a failed intracratonic rift, Central Africa 185 Case Study 5: orogen-scale inversion of extensional faults in a rift system that evolved into a back-arc basin and regional role of footwall shortcuts, Eastern Cordillera of Colombia 187 Case Study 6: inversion of extensional faults in a failed continental rift, the Wessex Basin, Southern United Kingdom 189 Case Study 7: inversion of extensional faults in a rift basin: New Zealand 190 Case Study 8: inversion of extensional faults in Late-Orogenic molasse basins that evolved to a later continental margin and regional role of footwall shortcuts in an orogenic belt, Pembrokeshire, United Kingdom 191 Case Study 9: inversion with significant mechanical contrasts and ductile deformation near a collisional suture on a distal continental margin and role in orogenic belt development 192 Inversion structures and economic implications: petroleum system elements and mineral deposits 193 Timing of inversion relative to extension 195 Conclusion 198 Acknowledgements 199 References 199 Further reading 204

11. Salt- and shale-detached gravity-driven failure of continental margins MARK G. ROWAN

Introduction Gravity-driven failure Distribution of detachment layer Structural styles Diapirism Allochthonous salt Concluding remarks Acknowledgements References

205 206 211 213 218 226 230 230 230

12. Carbonate systems LUIS POMAR

Introduction Food and feeding/follow the food Boundary layers Feed and food: feasting at the pycnoclines Carbonate production modes Carbonate production systems through Earth’s history Platform types Acknowledgements References

235 239 243 248 252 257 274 300 301

13. Lake systems and their economic importance CHRIS SLADEN AND DOMENICO CHIARELLA

Introduction to lakes and lake systems Lakes in time and space; preservation of lakes in the Phanerozoic rock record Classification of lakes; the different settings for lakes Conditions needed to create and maintain a tectonic lake with well-developed lake sequences Megasequences, sequences and cycles in basins containing tectonic lakes Controls on lake sequences and sequence stratigraphy Important differences between lake and marine sequence stratigraphy Principal depositional environments in lake basins Predicting lake sequences and facies Major petroleum systems involving lake sequences Features of potential petroleum source rocks that develop in lakes Petroleum systems in Early Cretaceous and Tertiary lake basins of South and East Asia Acknowledgement References

313 313 319 323 323 324 324 327 328 328 329 330 339 339

14. Clastic shorelines and deltas M. ROYHAN GANI

Introduction Shoreline and deltaic processes Universal building blocks of all clastic systems

343 343 350

CONTENTS

Deltas: river-fed shorelines Wave-dominated, nondeltaic shorelines Tide-dominated, nondeltaic shorelines Acknowledgements References

351 358 361 362 363

Accommodation Delivery configuration Slope profile types Discussion Conclusion References

15. Tidal straits: basic criteria for recognizing ancient systems from the rock record

19. Geological methods

SERGIO G. LONGHITANO AND DOMENICO CHIARELLA

PATRICIA WOOD DICKERSON AND WILLIAM R. MUEHLBERGER

What are tidal straits? Why are tidal straits important? Sedimentary dynamics of modern tidal straits (what we presently know and what we still need to know) The stratigraphic and sedimentary record of ancient tidal straits with some example Criteria for recognizing tidal straits in outcrop or subsurface successions Concluding remarks References Further reading

365 367 367 378 381 407 409 415

16. Submarine landslides architecture, controlling factors and environments. A summary NICOLA SCARSELLI

Introduction Classifications Structural architecture of submarine landslides Mechanics of slope failures, preconditioning and triggering factors Environments Statistics of submarine landslides Concluding remarks References

417 419 421 426 430 433 435 435

ROBERTO TINTERRI, ANDREA CIVA, MICHELE LAPORTA AND ALBERTO PIAZZA

517 519 524 525 529 535 538 539

20. Regional tectonics and basin formation: the role of potential field studies an application to the Mesozoic West and Central African Rift System Introduction Gravity and magnetic coverage over continental areas Satellite-derived gravity and magnetic coverage over oceanic areas Offshore plate tectonic links to the West and Central Africa Rift System West and Central Africa Rift System Evolution of the West and Central Africa Rift System Conclusion and implications References Further reading

541 542 544 545 547 550 553 555 556

21. Wide-angle refraction and reflection 441 443 453 463 469 472 473 473

ROBERT S. WHITE

Introduction Wide-angle acquisition Modelling wide-angle data Examples of wide-angle seismic interpretations References

557 557 561 562 569

22. An introduction to seismic reflection data: acquisition, processing and interpretation DAVID R. COX, ANDREW M.W. NEWTON AND MADS HUUSE

18. Controls on reservoir distribution, architecture and stratigraphic trapping in slope settings BRADFORD E. PRATHER

Introduction

484 488 488 505 508 508

JAMES DEREK FAIRHEAD

17. Turbidites and turbidity currents

Introduction A historical perspective The latest insights on turbidite facies and processes Modifications of the facies tract induced by the relationship between flow type and basin morphology Turbidite systems: relationship between degree of efficiency, degree of tectonic confinement and type of basin Some concluding remarks Acknowledgements References

Introduction Satellite images and data Hotspots, rifts, reefs, deltas and cratonic regions views from space Geological and tectonic maps Integrated interpretation the Appalachians Acknowledgements References Further reading

vii

481

Introduction The reflection seismic method Acquisition of reflection seismic data Seismic processing

571 571 576 580

viii

CONTENTS

Seismic interpretation Summary Acknowledgements Glossary of key terms References

587 600 601 601 601

OCTAVIAN CATUNEANU

605 612 622 664 671 676 678 678

24. Concepts of conventional petroleum systems JAN DE JAGER

Introduction The essential ingredients Play-based exploration workflow References Further reading

687 689 709 719 720

25. The accumulation of organic—matter rich rocks within an earth system’s framework KEVIN M. BOHACS, IAN O. NORTON, JACK E. NEAL AND DEBBIE GILBERT

Plate reconstructions Key aspects of individual plate reconstructions Proximate controls on accumulation of organic matter Source rock settings Application of concepts to predicting source accumulation Quantitative analysis of palaeoenvironmental maps for predicting source rock distribution Application of concepts to evaluating mudstone/shale reservoir character and distribution Conclusion References

722 724 730 731 734 738 740 741 741

26. Modelling fluid flow and petroleum systems in sedimentary basins JOHANNES WENDEBOURG

Introduction Principles of fluid flow in sedimentary basins Overpressure detection and modelling Hydrocarbon fluids and charge modelling Petroleum systems analysis References

27. Tectonic and basin maps of the world ALBERT W. BALLY, DAVID G. ROBERTS, DALE SAWYER AND ANTON SINKEWICH

23. Sequence stratigraphy Introduction Controls on sequence development Sequence stratigraphic framework Hierarchy in sequence stratigraphy Discussion Conclusions Acknowledgements References

GLOBAL MAPS

745 745 748 750 753 756

27.1 Global topography and platet ectonics

761

Global geological maps: introduction Global geological cartography: selected milestones Stratigraphic nomenclature and the geological time scale References

761 762 764 765

27.2 Global topography and bathymetry: the face of old Earth reworked and modified by present processes

766

Introduction 766 Global relief models: onshore and submarine morphology and plate tectonic regimes 767 References 769 27.3 Neotectonics; earthquakes and conventional (i.e. rigid) versus diffuse plate boundaries

770

Neotectonics: introduction Global earthquake distribution Well-defined versus diffuse plate boundaries Neotectonic plate motions: their relation to a fixed Eurasia and to Cenozoic/Mesozoic fold belts References

770 771 771

27.4 Global stress maps and palaeostress studies

776

References

778

27.5 The continental lithosphere and continental crust

779

772 775

Introduction 779 The continental lithosphere 779 The continental crust 780 Crustal layers, rheological models, and conclusions 782 References 782 Plates for global topography, neotectonics, the continental lithosphere and crust: segments 26.1 26.3 and 26.5 (For online version of the plates/figures cited in this chapter, the reader is referred to http://www.elsevierdirect.com/ companion.jsp?. ISBN: 9780444563576) 783 27.6 Tectonic maps of the world

791

Introduction to tectonic maps Recent advances in alpine tectonics: an example of the scope of larger-scale tectonic maps Simplified tectonic maps of the world About Phanerozoic plate tectonic reconstructions References

791 791 792 793 794

27.7 Polar tectonic maps: introduction Arctic tectonic map Antarctic tectonic map References

796 796 797 798

CONTENTS

27.8 Cenozoic/Mesozoic and Palaeozoic orogenic systems and their fold and thrust belts

800

Orogeny versus epeirogeny Subduction, sutures, and orogens Active margin fold and thrust belts Foreland fold and thrust belts Normal faulting in foreland fold and thrust belts References

800 800 803 806 808 809

27.9 Age of continental basement

811

Introduction to basements, that is, the ‘residual’ peneplaned former fold belts Merging the global tectonic map with a Precambrian basement map References 27.10 Hot spots, linear island chains, large igneous provinces (LIPs), and radiating dike swarms: active volcanoes Introduction Large igneous provinces (LIPs) Giant radiating dike swarms (maps b-6 and b-7) Is there a ‘canonical progression of tectonic themes’ preceding and/or following the emergence of a plume? The distribution of active volcanoes References 27.11 Tectonic settings of mafic/ultramafic oceanic and intraoceanic arc system crust, large igneous provinces, rifted and volcanic passive margins, tectonic setting, and discussion of equivalent allochthonous ‘ophiolitic’ fragments in orogens Introduction Subducted oceanic plateaus Allochthonous accreted oceanic plateaus and intraoceanic island arc terranes Allochthonous fragments, oceanic, and intraoceanic arc systems, and lower crust and uppermost mantle of hyperextended passive margins

811 812 813

815 815 816 818 818 820 820

823 823 823 824

825

ix

Allochthonous, exhumed continental crust mantle transitions and the Ivrea-Verbano Zone Conclusion References Plates for tectonics, orogenic systems, hot spots, large igneous provinces, volcanoes: segments 26.6 26.10 (for online version of the plates/figures cited in this chapter, the reader is referred to http:// www.elsevierdirect.com/companion.jsp?. ISBN: 9780444563576)

830

27.12 Sedimentary basins and rifts (including rifts)

837

Introduction References

837 840

Rift systems on relatively stable/rigid lithosphere References

840 841

Passive margins on relatively stable/rigid lithosphere References

842 845

Cratonic basins on relatively stable/rigid lithosphere References

846 848

Basins on the periphery of orogens: deep-sea trenches and foreland basins References

850 852

Basins located within orogens (episutural basins of Bally and Snelson, 1980, or epi-eugeosynclines and successor basins of earlier authors) References

853 856

Oceanic basins formed by spreading ridges (including oceanic back-arc basins) General reference

857 858

Sedimentary basins and rifts: Segment 26.11 (for online version of the plates/figures cited in this chapter, the reader is referred to http://www.elsevierdirect.com/ companion.jsp?. ISBN: 9780444563576)

858

Index

826 827 828

863

List of Contributors

Albert W. Bally† Department of Earth Science, Rice University, Houston, TX, United States

M. Royhan Gani Department of Geography and Geology, Western Kentucky University, Bowling Green, KY, United States

Kevin M. Bohacs ExxonMobil Upstream Research Company, Spring, TX, United States; KMBohacs GEOconsulting, Houston, TX, United States

Debbie Gilbert ExxonMobil Exploration Company, Spring, TX, United States

Clare Bond Fold-Thrust Research Group, School of Geosciences, University of Aberdeen, Aberdeen, United Kingdom

Mads Huuse Department of Earth and Environmental Sciences, University of Manchester, Manchester, United Kingdom

Rob Butler Fold-Thrust Research Group, School of Geosciences, University of Aberdeen, Aberdeen, United Kingdom

Michele Laporta Earth Sciences Unit, Department of Chemistry, Life Sciences and Environmental Sustainability-Earth Sciences Unit, University of Parma, Parma, Italy

Octavian Catuneanu Department of Earth and Atmospheric Sciences, University of Alberta, Edmonton, AB, Canada

Sergio G. Longhitano Department of Sciences, University of Basilicata, Potenza, Italy William R. Muehlberger University of Texas at Austin, Austin, TX, United States

Domenico Chiarella Clastic Sedimentology Investigation (CSI), Department of Earth Sciences, Royal Holloway, University of London, Egham, United Kingdom

Jack E. Neal ExxonMobil Exploration Company, Spring, TX, United States

Andrea Civa Earth Sciences Unit, Department of Chemistry, Life Sciences and Environmental Sustainability-Earth Sciences Unit, University of Parma, Parma, Italy; Current affiliation: Eni S.p.A., San Donato Milanese, Italy

Andrew M.W. Newton School of Natural and Built Environment, Queen’s University Belfast, Belfast, United Kingdom Ian O. Norton ExxonMobil Upstream Research Company, Spring, TX, United States; Institute for Geophysics, Jackson School of Geosciences, The University of Texas at Austin, Austin, TX, United States

Mark Cooper Sherwood GeoConsulting Inc., Calgary, AB, Canada; School of Geosciences, University of Aberdeen, Aberdeen, United Kingdom David R. Cox Department of Earth and Environmental Sciences, University of Manchester, Manchester, United Kingdom

Alberto Piazza Earth Sciences Unit, Department of Chemistry, Life Sciences and Environmental Sustainability-Earth Sciences Unit, University of Parma, Parma, Italy

Jan de Jager De Jager Geological Consultancy, The Hague Area, Netherlands

Luis Pomar Catedra Guillem Colom, University of the Balearic Islands, Palma de Mallorca, Spain

Lucı´a Pe´rez Dı´az Fault Dynamics Research Group, Royal Holloway University of London, Egham, United Kingdom

Bradford E. Prather CarTerra, LLC, Houston, TX, United States; Department of Geology, University of Kansas, Lawrence, KS, United States

Patricia Wood Dickerson American Geosciences Institute, Alexandria, VA, United States Graeme Eagles Alfred Wegener Institute for Polar and Marine Research, Bremerhaven, Germany

David G. Roberts† Department of Earth Sciences, Royal Holloway, University of London, Egham, Surrey, United Kingdom

James Derek Fairhead School of Earth and Environment, University of Leeds, Leeds, United Kingdom

Mark G. Rowan United States

Haakon Fossen Museum of Natural History/Department of Earth Science, University of Bergen, Bergen, Norway

Dale Sawyer Department of Earth Science, Rice University, Houston, TX, United States

C. Mary R. Fowler Darwin College, Cambridge, United Kingdom

Nicola Scarselli Fault Dynamics Research Group, Department of Earth Sciences, Royal Holloway, University of London, Egham, Surrey, United Kingdom

†

Deceased.

xi

Rowan Consulting, Inc., Boulder, CO,

xii Anton Sinkewich United States

LIST OF CONTRIBUTORS

Anton Sinkewich Studio, Houston, TX,

Chris Sladen Reconnoitre.ltd, Woolley House, Uplyme, Devon, United Kingdom Ian James Stewart Integrated Petroleum Exploration Ltd., Uplands, Pond Road, Woking, Surrey, United Kingdom Roberto Tinterri Earth Sciences Unit, Department of Chemistry, Life Sciences and Environmental Sustainability-Earth Sciences Unit, University of Parma, Parma, Italy

Marian J. Warren Jenner GeoConsulting Inc., Calgary, AB, Canada Johannes Wendebourg

TOTAL Exploration, Paris, France

Robert S. White Bullard Laboratories, Department of Earth Sciences, University of Cambridge, Cambridge, United Kingdom Sabin Zahirovic EarthByte Group, School of Geosciences, The University of Sydney, Camperdown, NSW, Australia

Foreword

I had the great fortune to know very well both Dave Roberts and Albert Bally, the editors of the unrivalled book series Regional Geology and Tectonics. This set of three books was published by Elsevier in 2012. However, as I heard it from Bert a few times, the decisive conversation about the need for such a publication occurred already in 2000. As the story goes, Dave was visiting at Bert’s house in Houston and, over a few drinks, they agreed that there is an obvious gap in the portfolio of geoscience books out there when it comes to ‘how to do regional geology’. They got to work shortly after and, using their enormous networks across the globe, came up with a long list of contributors (146 authors and co-authors, to be exact) whom they considered as top experts either in a particular geo-field and/or in a specific basin/region. The first volume was dedicated to Principles of Geologic Analysis, the second one to Phanerozoic Rift Systems and Sedimentary Basins and the third one to Phanerozoic Passive Margins, Cratonic Basins and Global Tectonic Maps. As the titles suggest, these very ambitious global themes turned out to be a major challenge for both Dave and Bert. For the very long 12 years, making sure that the numerous chapters are finally delivered, the main driving force was Dave. It is very fortunate that both Dave and Bert could still see the impressive outcome of their efforts in print. . . Dave passed away soon after the publication of the books, in 2013, and Bert left us behind in 2019. In the Foreword to the first edition, Dave and Bert mentioned the new generation of ‘Nintendo geoscientists’ whose college and young professional education (either in the industry or academia) was mostly revolving around the specifics of new computer techniques the digital age brought to all of us. Clearly, the emphasis was already increasingly placed on the mastery of various software packages available to analyze geological and geophysical data sets. After almost a decade, the situation did not seem to change much, but with the new wave of technology ruling our everyday life in 2019, let me use the term ‘Smartphone geoscientists’. Whereas it is fantastic to have all kinds of geoscience information literally at our fingertips, this trend is clearly going to the direction of focusing even more on the individual trees in a forest many of our younger colleagues have not even heard about. The appreciation of the big picture of regional geology is clearly eroding away at an alarming rate, both in the academia and in the industry! The readers of this second edition have to appreciate that both Dave and Bert had exceptional professional careers, almost predestined to become the leading advocates of the regional approach in geoscience. Dave started his career at the Institute of Oceanographic Sciences and spent lots of time of various deep-sea drilling projects devoted to the better understanding of continental margins. After this academic experience he joined the industry taking the role of the Head of the Basin Studies Group at British Petroleum (BP). During the 22 years with BP he reached the highest geoscientist position there becoming the Distinguished Advisor in Global Exploration. After BP, Dave kept working as a nonexecutive Director of Premier Oil and also as a Visiting Professor at Royal Holloway. Bert’s career at Shell began in Canada where he first worked as an exploration geologist and later as the Chief Geologist. After 12 years, he became the Manager of Shell Research in Houston, later Chief Geologist, and finally an Exploration Consultant. Whereas Bert’s first 27 years with Shell was an industry career, he already started to teach at the Houston research lab, training and mentoring Shell US geoscientists. After Shell, Bert became the Chair of the Geology Department at Rice University. During his academic years he managed to put together the famous two three-volume sets of seismic atlases that popularized the use of industrial reflection techniques for scientific purposes. Besides this, just like Dave, he managed to teach and supervise lots of graduate students coming from all over the world. Now, imagine the amount of data, quite possibly coming from every basin on this planet, Dave and Bert had the pleasure to look at and digest, one way or the other! No wonder that they were the ones who became the champions for the interpretation of any kind of geological and geophysical data sets with the regional context in mind. Since Dave was working a lot at Royal Holloway, it is very proper and logical that the preparations for the second edition of the Regional Geology books have been made by the faculty there. During his last years, Bert also

xiii

xiv

FOREWORD

supported the efforts in producing an updated version. Nicola Scarselli, Ju¨rgen Adam and Domenico Chiarella, the editors of the second edition, clearly follow the footsteps of Dave and Bert by finding outstanding experts for an updated view on many topics and regions covered in the first edition. Therefore, for the second edition, the new editors mainly went back to the authors who were willing to update their original contributions to the first edition. The first book of the second edition is again devoted to key concepts and methodologies in doing regional geology. Whereas the new second volume will still contain case studies of extensional systems and passive margins, the third volume will provide case studies on contractional and strike-slip systems. I sincerely hope that the readers of the second edition of Regional Geology and Tectonics will enjoy and appreciate the perspective and wisdom Dave and Bert wanted to teach to all of us. The new editors are commended for maintaining the spirit of the first volumes by putting together a superb follow-up compendium on the extremely important regional aspects of geoscience and the Phanerozoic of the world. The second edition will serve as an unparalleled reference book on the subject for many years to come! Gabor Tari Group Chief Geologist, OMV Upstream, Exploration, Vienna, Austria

Introduction

Current United Nations estimates suggest a rise of world population to B9.7 billion people by 2050, with the most significant size increases reserved to booming emerging countries. This worldwide demographic shift raises some global challenges such as the necessity to safeguard growing vulnerable communities from geohazards and the availability of affordable commodities required to secure the prosperity of future generations. This means that 8 years after the publication of the first edition of Regional Geology and Tectonics: Principles of Geologic Analysis, now more than ever there is a need to understand how geological processes work and how these might impact on a wide range of increasingly critical human activities exploration of georesources, water resources management, geohazard mitigation, geological disposal of radioactive waste, subsurface storage of carbon and energy to mention a few. Now more than ever, geology and geologists are at the heart of sustainable global development. Now more than ever, there is a need for ‘Regional Geology and Tectonics’, a resource for academics, students and industry geoscientists to learn ‘how to do regional geology’. Because the validity of any form of geological analysis is underpinned by a sound understanding of the regional background, the so-called ‘bigger picture’. This is essential, as geological processes do not operate in isolation, but are part, and respond to, regional- and ultimately plate-scale events. In devising the new edition of the book, we followed the structure of the original series published in 2012. We approached authors of the first edition for their revised and updated chapters. We also invited a set of key new contributions in the field of plate tectonics, structural geology, sedimentology, and basin dynamics. This has resulted in 27 succeeding chapters arranged in subject groups covering (1) the Earth structure and plate tectonic processes (Chapters 1 6); (2) the modes of deformation of the Earth’s crust by tectonic and gravity-driven processes (Chapters 7 11); (3) the major Earth’s sedimentary systems and associated depositional processes and products, from terrestrial lake systems to deepwater turbidites and submarine landslides (Chapters 12 18); (4) the methods of regional geological analysis and the concepts of subsurface analysis and exploration (Chapters 23 26). The book ends with the renowned ‘Global Maps’ originally drafted by Albert Bally (Chapter 27). We ask for the understanding of the readers for our failure in securing contributions on some critical aspects of regional tectonic analysis and related methodologies, such as seismic tomography, strike-slip systems as well as geodetic methods, and modelling of geological processes. For these, a number of experts were approached, but time constraints and personal commitments of potential contributors clashed. Future editions of this book will have to look again at those critical topics. Despite these shortcomings, we like to think that we kept the spirit and distinction of the 2012 book edited by the late Bert Bally and Dave Roberts, to whom, first and foremost, this new edition is dedicated to. We also like to think that, despite the usual difficulties in planning and editing such an extensive and comprehensive volume, the reader will see the value of a well-illustrated and accessible reference. For this, we thank the authors who have put tremendous effort and much of their precious time in the contributions. A particular thank you is for those diligent authors who submitted their chapters timely and patiently waited for the whole book to come together. It was a real pleasure to work with you! We also thank the Elsevier team and in particular Amy Shapiro, Ruby Smith and Vijay Bharath. They always believed in the success of this project and supported editors and authors throughout the entire process. The editors, Nicola Scarselli, Domenico Chiarella and Ju¨rgen Adam

xv

C H A P T E R

1 Regional geology and tectonics of sedimentary basins Ian James Stewart Integrated Petroleum Exploration Ltd., Uplands, Pond Road, Woking, Surrey, United Kingdom

The most important question every geoscientist should continually ask is ‘why’? Why do I see what I see? Too often it is accepted, for example, that a sedimentary basin ‘is there’, and geoscientists continue to work on surfaces and units within its boundaries, without ever questioning ‘why do I have this hole in the ground’? Why is it here? What governed its geometry and stratigraphic architecture? What controlled the sediment entry points? What dictates the thermal structure? In resource evaluation and field development, the focus is commonly further narrowed, with management only requesting a map on a ‘Green Horizon’ picked on seismic data in a workstation environment covering only a small segment of the total basin area; insights from the broader regional and mega-regional geological understanding are lost. In simple terms, the majority of sedimentary basins develop through contraction and lithospheric loading or crustal extension and subsequent thermal subsidence; both these processes include and can be dominated by oblique- or strike-slip tectonics. Their geometry and internal architecture are commonly strongly influenced by more regional structural fabrics inherited from the broader geological past. The only way we can answer the key questions of ‘why?’, is to build an understanding of the basin’s regional geology. The regional geological understanding forms the bridge from the basin to its province, plate and often multiplatescale context. D.G. Roberts, in his teaching of geology, would always assert the ‘Principle of Least Geological Astonishment’ to basin-scale interpretation: the geological equivalent of Occam’s Razor. The more special geological pleading an interpretation requires, the less likely it is to be correct. We can add another guiding principle: ‘The Principle of Non-Randomness’. Everything is where it is for a reason, and our role in basin interpretation is to understand ‘why’? Regional understanding of the basin’s location and its relationship to the tectonic history of its exposed margins and hinterland is essential. Integrated with the knowledge of far-field tectonic effects on a plate scale, or a palaeo-plate scale commonly provides the most powerful insights in moving towards a valid and viable basin solution. While regional geology began through observations in the field (for a historical review, see Roberts and Bally, 2012), it has been supplemented in recent years by an ever-improving crustal-scale understanding through deep seismic reflection and refraction data, global magnetic and gravity datasets, seismic wave tomography and locally magnetotelluric profiling. The advent of digital elevation modelling and GIS technology allows accurate interpretation of the deeper processes within the mantle and their long-wavelength influence on the dynamic surface topography of the earth (e.g. Flament et al., 2013; Rubey et al., 2017). Mantle plumes are now imaged by tomography below the upper/lower mantle interface (French and Romanowicz, 2015), although their sole contribution to large igneous provinces (LIPs) is still a matter of debate. The important aspects of the regional crustal geometry to the basin-centric geoscientist include the terrane construct, and the thickness of the component crustal layers and the likely heat flow consequences for thermal modelling of the shallower sediment pile. Regional-scale thermo-chronological studies utilizing apatite fission track analysis (AFTA) are now widespread and frequently provide the timing of sediment maturity within the basin as well as changes in basin-margin elevation via structural inversion or thermal processes. Coupled with the broader structural geology and provenance studies, sequence stratigraphy can clarify sediment entry points into the basin.

Regional Geology and Tectonics. DOI: https://doi.org/10.1016/B978-0-444-64134-2.00001-8

1

© 2020 Elsevier B.V. All rights reserved.

2

1. Regional geology and tectonics of sedimentary basins

To the exploration geoscientist at the basin-scale, however, probably the most significant image is that of a deep seismic reflection profile, illustrating a well-defined Moho and a layered crust, with their often relatively insignificant sedimentary basin carried above it at a crustal scale. In the last 20 years, industry has sponsored and supported numerous commercial deep reflection surveys across most of the world’s continental margins. The academic community (often oil company sponsored) had previously demonstrated their value in the 1970s and 1980s with the COCORP, BIRPS, LITHOPROBE, BABEL and ECORS datasets. These surveys focused on elucidating crustal structure, largely over collisional orogens in the northern hemisphere. A thorough review of the early phases of this (still on-going) era of academic-led research and its conclusions is given in Snyder and Hobbs (1999). One of the most profound observations of the original BIRPS programme was the confirmation of frequent reactivation of originally contractional crustal-scale faults, a view long-held by many field geologists. Their detachment in a seismically reflective lower crust or upper mantle, with a shallow near-surface half-graben in their hanging-wall, gave rise to the ‘typical BIRP’ (Matthews and Cheadle, 1986). This observed reactivation of crustal fault zones as extensional structures controlling subsequent basin formation (e.g. Klemperer and Hobbs, 1991), is now widespread. Crustal-scale faults have been frequently shown to be re-used in later contraction during basin inversion as a consequence of tectonic events further afield. Cycles of successive extension and contraction on these faults are common during the geological evolution of a basin. Modern long-record commercial seismic data have increasingly been focused over the ‘new geography’ of increasingly deeper water, as the petroleum industry moves outboard to the limits of exploration and development technology. The revelations from these data on the nature of the thinning of continental crust towards the ocean-continent transition (OCT) and the differential and often depth-dependent behaviour of both upper and lower crust during extension has revolutionized the understanding of passive margins. Coupled with deep-sea drilling programmes, the nature of the OCT has been demonstrated to include exhumed lower crust and locally the mantle beneath low-angle extensional surfaces in largely nonmagmatic margins. Over magma-rich margins, which dominate the majority of continent-ocean transitions, long-record reflection seismic has spectacularly imaged seaward-dipping reflections or SDRs (from largely nonmarine lava flows that approach the base of the seaward tapering ductile crust), magmatic underplating and extensive dyke and sill emplacement associated with interpreted plume-driven activity or adiabatic decompression (Franke et al., 2007; Paton et al., 2017; Quirk et al., 2014). Field geology and observations at a regional scale remain essential to sedimentary basin understanding, as well as informing directly on the crustal processes interpreted from the broader geophysical approaches. Regional geological interpretations have been enhanced in recent years by advances in geochronology, particularly in situ Zircon and Baddeleyite U Pb geochronology, which allows precise dating of both igneous and metamorphic crystallization events (Schoene et al., 2013). In detrital grains, it permits the provenance and maximum depositional age of clastic strata to be determined (Gehrels, 2012). The chronology established from the 35 or so Archaean cratonic nuclei (Bleeker, 2003) and the presence of rifted and collisional margins around their borders reveals a long history of largely lateral crustal accretion since at least Late Archaean times. Each accretionary event leaves an indelible imprint in the lithospheric ‘memory’. The palaeomagnetic and chronological record preserved in extension-related LIPs and their plumbing systems documented as dyke swarms and sills provides LIP ‘barcodes’ (Ernst et al., 2013) for pre-Gondwanan early continents and their progressively accreted margins. This allows not only the correlation in time and space of previous continent and supercontinent configurations, their growth and dispersal, but also the demonstration of the longevity of tectonic processes evident in well-documented younger Phanerozoic terranes of lateral accretion, contraction and subsequent relaxation. Archaean cratonic areas and their progressively accreted Proterozoic margins have been extensively studied by academia and the extractive industries, and few areas are better known than southern Africa. Here, the rock record documents multiple cycles of crustal accretion and dispersal over 3 Ga. In a book focused on Phanerozoic sedimentary basins, the study of Proterozoic tectonic cycles may seem somewhat academic and irrelevant. It is not. The structural template observed in outcrop is a profound reminder of the antiquity of upper crustal inhomogeneity that needs to be studied to understand subsequent basin development. The Zimbabwe and Kaapvaal cratons of southern Africa each consist of numerous Archaean terranes, partitioned by greenstone belts. They are separated by the Limpopo Belt (Fig. 1.1A), a remnant of a Neo-Archaean collisional orogen between them that had stabilized by 2.6 Ga. It is probably one of the most-studied mobile belts in the world. The contractional deformation extended into both craton margins and reactivated the Murchison Thabazimbi zone that delimits an earlier Meso-Archaean suture between the northern Pietersburg terrane and the southern Kaapvaal craton (Fig. 1.1B). Dyke swarms of 2.57 Ga (Great Dyke) and c.2.4 Ga

Regional Geology and Tectonics

Regional geology and tectonics of sedimentary basins

3

FIGURE 1.1 The Limpopo Belt between the Kaapvaal and Zimbabwe cratons. (A) Structural summary showing main crustal domains and Neo-Archaean and Palaeoproterozoic elements of the cratons. Major fault zones and terranes of the Kaapvaal craton labelled. CZ Central Zone (including the Mahalapye (M), Sunnyside (S) and Lechana Shear Zones), HRSZ Hout River Shear Zone, NMZ Northern Marginal Zone, PZSZ Palala-Zoefontein Shear Zone, SMZ Southern Marginal Zone, SSZ Shase Shear Zone, TSZ Triangle Shear Zone, TSSZ Tuli-Sabi Shear Zone, UTZ Umlali Thrust Zone, KT Kimberley Terrane, PT Pietersburg Terrane and WT Witwatersrand Terrane. (B) Palaeoproterozoic age contraction and subsequent relaxation. Magondi Belt c.2.0 Ga with syn- and posttectonic granites. DK Dete Kamatavi Inlier, KI Kubu Island (2.039 Ga: Majaule et al., 2001), HG Hurungwe Granite (1.998 Ga: McCourt et al., 2001), MG Mahalapye Granite (2.040 Ga: Millonig et al., 2010), OK Okwa Block (2.055 Ga; Mapeo et al., 2006), P Palapye rift sediments and volcanics, S Soutpansberg rift sediments and volcanics, BHDS Black Hills Dyke Swarm, M Mashonaland Sills, TDS Tsineng Dyke Swarm and WS Waterberg Basin Sills. (C) Mesoproterozoic. Volcanics and intrusions in Limpopo, Magondi and Ghanzi-Chobe Belts. P Pilanesberg event (c.1.35 Ga: Cawthorn, 2015, Elburg and Cawthorn, 2017), PR Pienaars River Intrusion. Umkondo Province (c.1.11 Ga: Hanson et al., 2004; de Kock et al., 2014). GHF Goha Hills Formation, KF Kwegbe Formation, RD Rakops Dykes, TC Tshane Complex, TG Timbavati Gabbro, TSC Tsetseng Complex, U Umkondo and X Xade Complex. (D) The Late Carboniferous Permian ‘Karoo’ Basins. CBB Cahora Bassa Basin, EB Ellisras Basin, KB Kalahari Basin, MB Mopane Basin, MMB Moatize-Minjova Basin, MZRB Mid-Zambezi Rift Basin, NB Nuanetsi Basin, PB Pafuri Basin, SFB Springbok Flats Basin, TB Tuli Basin and TSB Tshipise Basin.

Regional Geology and Tectonics

4

1. Regional geology and tectonics of sedimentary basins

(Sebanga swarm) are prominent in the Zimbabwe craton but not on the northern margin of the Kaapvaal, and the 2.06 Ga Bushveld-Molopo Farms complexes are not replicated in the Zimbabwe block. This has been interpreted by So¨derlund et al. (2010) to indicate that the two cratons were not in their current configuration by these times. Apparent re-amalgamation and extensive reworking and metamorphic overprinting occurred in the Palaeoproterozoic Magondi orogen at c.2.0 Ga (Eglington et al., 2009 and references therein), although by 1.92-Ga voluminous dyke swarms, sills and basalt flows record at least one phase of extension between the two blocks during deposition of the syn-rift Soutspanburg/Palapye Groups and the marginally older Waterberg Group (Dorland et al., 2006) which carries extensive 1.87-Ga sills (Fig. 1.1B). Massive sill complexes of comparable age (So¨derlund et al., 2010) cover much of Mashonaland on the Zimbabwe craton (Fig. 1.1B), although these intriguingly demonstrate a slightly different palaeomagnetic pole suggesting that the formation of present-day configuration is even more recent (Letts et al., 2011). Much of the subsequent Proterozoic rock record over the Limpopo is lost to erosion; LIP events at 1.3 Ga, and particularly c.1.11 Ga (Umkondo) identify regional thermal events, the latter marking Mesoproterozoic basin formation over previously accreted orogens around the margins of the now broader combined Zimbabwe Kaapvaal craton area (Fig. 1.1C). This period of extensional basin formation immediately postdates the initial contraction during the main phase of the Grenvillian Namaqua-Natal orogeny to the south. The Late Palaeozoic Karoo rift event of much of central and eastern Africa marks the opening of Neo-Tethys to the north and the initial break-up of East and West Gondwana. It is coeval with the early phases of contraction in the Cape fold-belt and the development of its well-studied foreland basin. Catuneanu et al. (2005) ascribe the formation of the southern African basins north of the main Karoo foreland to far-field flexural effects (‘backbulge’), with extension-related basin formation only significant further north in central Africa. Tankard et al. (2012) suggested a model involving lithospheric subsidence due to dynamic mantle flow related to the distal subduction as the drive of the longer wavelength component of early Karoo subsidence. More recent field-based studies, however, coupled with surface, borehole and airborne geophysical data, demonstrate extensional or transtensional half-graben developed over all of the Proterozoic orogens bordering the Zimbabwe craton (Fig. 1.1D) at least as early as Early Permian times (Daszinnies et al., 2009), including the Limpopo fold-belt (e.g. Fourie et al., 2014). The Tuli, Ellisras, Mopane and numerous smaller half-grabens occupy hanging walls to NeoArchaean structures. Their subsequent reactivation demonstrates that the lithospheric ‘memory’ persists. Similarly, the Archaean Murchison Thabazimbi zone in the northern Kaapvaal was reactivated in Permian times and notably again in the Early Jurassic to accommodate the Springbok Flats half-graben (Good and DeWit, 1997). This Jurassic rejuvenation was associated with the LIP of the Karoo igneous province, one of the largest flood basalt provinces on the earth (Fig. 1.2). Re-utilization of the same inherited structural elements as in the Permian

FIGURE 1.2 The ‘Karoo’ igneous event in Southern Africa. TG Dyke Swarm, ODS

Okavango Dyke Swarm, ORS

Tuli Half-Graben, SF Springbok Flats Half-Graben, LDS Olifants River Dyke Swarm, SLDS Save-Limpopo Dyke Swarm.

Regional Geology and Tectonics

Lebombo

References

5

occurred with half-graben relaxation, massive magmatic outpourings, sill emplacement and dyke swarms in Toarcian times. The magma-rich margin marked by the Lebombo escarpment and the equivalent nonmarine SDR sequences proven by wells beneath the Zambezi Delta and seen offshore (Fig. 1.1D) mark the final separation of East and West Gondwana as East Antarctica departed from the eastern Kaapvaal with the opening of the Indian Ocean. The so-called ‘triple junction’ (Burke and Dewey, 1973) of the Okavango, Save-Limpopo and LebomboOlifants River dyke swarms (Fig. 1.2) is now considered to reflect localization by craton margins and the preexisting lithospheric discontinuities of the Limpopo Belt (Jourdan et al., 2006) rather than a plume-related ‘aulacogen’. Comparable dyke studies of South-West Africa similarly conclude that the opening of the modern South Atlantic was controlled by Neoproterozoic Pan-African (or older) structural discontinuities. Here, the inherited structural basement heterogeneity controlled dyke emplacement in Late Jurassic and Early Cretaceous times (Will and Frimmel, 2013). Post-Jurassic modification of the Limpopo Belt is recorded by AFTA in both Early and Late Cretaceous times (Belton and Raab, 2010), with the latter interpreted as inversion coincident with a plate-wide contractional event in Santonian times as the African Plate began to move northwards and obduction of the Oman ophiolites commenced (Guiraud and Bosworth, 1997). The study of the regional tectonic history of the ‘basement’ hinterland and its projection beneath a sedimentary basin is clearly not academic. Geochronologies and long-lived structural influence, such as those established for Southern Africa, are not unique. The so-called basement below any sedimentary basin carries a legacy that invariably has had a profound control on later basin formation through ‘hard-linkage’ (i.e. the direct control on basin geometry, basin margins and often the sediment entry points into the basin) or by ‘soft-linkage’ (i.e. more subtle influence on the architecture of shallower sedimentary sequences and their stacking patterns). If we are to properly answer the question ‘why?’, then perhaps the most relevant question is ‘how regional is regional?’ Clearly, it is no longer acceptable to merely map the ‘Green Horizon’ on a workstation. The widespread recognition of far-field effects, particularly structural inversion of preexisting geometry, or longwavelength mantle-dictated effects demonstrate that a broad, plate-wide knowledge is invariably required to understand present-day basin geometry. Global observations continue to improve as crustal visibility and interpretation technology advances. One of the most frequent observations in basin studies is the role of structural inheritance as a major influence on the basin geometry and its internal architecture throughout its history. This conclusion echoes the observations made by of Drummond Matthews and the BIRPS team in in the North Sea that few, if any, new basin-forming faults can be shown to cut the upper crust during extensional basin development. This crucial observation continues to be reinforced in recent publications (e.g. Fazlikhani et al., 2017) and raises the possibility that basin-forming crustal-scale faults only develop through contractional or oblique-slip processes.

References Belton, D.X., Raab, M.J., 2010. Cretaceous reactivation and intensified erosion in the Archean Proterozoic Limpopo Belt, demonstrated by apatite fission track thermochronology. Tectonophysics 480, 99 108. Bleeker, W., 2003. The Late Archean record: a puzzle in ca. 35 pieces. Lithos 71, 99 134. Burke, K., Dewey, J.F., 1973. Plume-generated triple junctions: key indicators in applying plate tectonics to old rocks. J. Geol. 81 (4), 406 433. Catuneanu, O., Wopfner, H., Eriksson, P.G., Cairncross, B., Rubidge, B.S., Smith, R.M.H., et al., 2005. The Karoo basins of south-central Africa. J. Afr. Earth Sci. 43, 211 253. Cawthorn, G.R., 2015. The geometry and emplacement of the Pilanesberg Complex, South Africa. Geol. Mag. 152 (5), 802 812. Daszinnies, M.C., Jacobs, J., Wartho, J.-A., Grantham, G.H., 2009. Post Pan-African thermo-tectonic evolution of the north Mozambican basement and its implication for the Gondwana rifting. Inferences from 40Ar/39Ar hornblende, biotite and titanite fission-track dating. In: Lisker, F., Ventura, B., Glasmacher, U.A. (Eds.), Thermochronological Methods: From Palaeotemperature Constraints to Landscape Evolution Models, 324. Geological Society, London, pp. 261 286. , Special Publications. De Kock, M.O., Ernst, R., So¨derlund, U., Jourdan, F., Hofmann, A., Le Gall, B., et al., 2014. Dykes of the 1.11 Ga Umkondo LIP, Southern Africa: clues to a complex plumbing system. Precambrian Res. 249, 129 143. Dorland, H.C., Beukes, N.J., Gutzmer, J., Evans, D.A.D., Armstrong, R.A., 2006. Precise SHRIMP U-Pb zircon age constraints on the lower Waterberg and Soutpansberg Groups, South Africa. S. Afr. J. Geol. 109, 139 156. Eglington, B.W., Reddy, S.N., Evans, D.A.D., 2009. The IGCP 509 database system: design and application of a tool to capture and illustrate litho- and chrono-stratigraphic information for Palaeoproterozoic tectonic domains, large igneous provinces and ore deposits; with examples from southern Africa. In: Reddy, S.M., Mazumder, R., Evans, D.A.D., Collins, A.S. (Eds.), Palaeoproterozoic Supercontinents and Global Evolution, 323. Geological Society, London, pp. 27 47. , Special Publications. Elburg, M.A., Cawthorn, G.R., 2017. Source and evolution of the alkaline Pilanesberg Complex, South Africa. Chem. Geol. 455 (2017), 148 165. Ernst, R.E., Bleeker, W., So¨derlund, U., Kerr, A.C., 2013. Large igneous provinces and supercontinents: toward completing the plate tectonic revolution. Lithos 174, 1 14.

Regional Geology and Tectonics

6

1. Regional geology and tectonics of sedimentary basins

Fazlikhani, H., Fossen, H., Gawthorpe, R.L., Faleide, J.I., Bell, R.E., 2017. Basement structure and its influence on the structural configuration of the northern North Sea rift. Tectonics 36, 1 27. Flament, N., Gurnis, M., Muller, R., 2013. A review of observations and models of dynamic topography. Lithosphere 5 (2), 189 210. Fourie, C.J.S., Henry, G., Mare´, L.P., 2014. The structure of the Karoo-age Ellisras Basin in Limpopo Province, South Africa, in the light of new airborne geophysical data. S. Afr. J. Geol. 117 (2), 207 224. Franke, D., Neben, S., Ladage, B., Schreckenberger, K., Hinz, K., 2007. Margin segmentation and volcano-tectonic architecture along the volcanic margin off Argentina/Uruguay, South Atlantic. Mar. Geol. 244 (1 4), 46 67. French, S.W., Romanowicz, B., 2015. Broad plumes rooted at the Earth’s mantle beneath major hotspots. Nature 525, 95 99. Gehrels, G., 2012. Detrital Zircon U-Pb geochronology: current methods and new opportunities. In: Busby, C., Azor, A. (Eds.), Tectonics of Sedimentary Basins: Recent Advances. Wiley-Blackwell, pp. 45 62. , Chapter 2. Good, N., DeWit, M.J., 1997. The Thabazimbi-Murchison lineament of the Kaapvaal Craton, South Africa: 2700 Ma of episodic deformation. J. Geol. Soc. 154 (1), 93 97. Guiraud, R., Bosworth, W., 1997. Senonian basin inversion and rejuvenation of rifting in Africa and Arabia: synthesis and implications to plate-scale tectonics. Tectonophysics 282 (1 4), 39 82. Hanson, R.E., Crowley, J.L., Bowring, S.A., Ramezani, J., Gose, W.A., Dalziel, I.W.D., et al., 2004. Coeval large-scale magmatism in the Kalahari and Laurentian cratons during Rodinia assembly. Science 304 (5674), 1126 1129. Jourdan, F., Fe´raud, G., Bertrand, H., Watkeys, M.K., Kampunzu, A.B., Le Gall, B., 2006. Basement control on dyke distribution in large igneous provinces: case study of the Karoo triple junction. Earth Planet. Sci. Lett. 241, 307 322. Klemperer, S., Hobbs, R., 1991. The BIRPS Atlas. Deep Seismic Reflection Profiles Around the British Isles. Cambridge University Press, 128 pp. Letts, S., Torsvik, T.H., Webb, S.J., Ashwal, L.D., 2011. New Palaeoproterozoic palaeomagnetic data from the Kaapvaal Craton, South Africa. In: Van Hinsbergen, D.J.J., Buiter, S.J.H., Torsvik, T.H., Gaina, C., Webb, S.J. (Eds.), The Formation and Evolution of Africa: A Synopsis of 3.8 Ga of Earth History, 357. Geological Society, London, pp. 9 26. , Special Publications. Majaule, T., Hanson, R.E., Key, R.M., Singletary, S., Martin, M.W., Bowring, S.A., 2001. The Magondi Belt in northeast Botswana: regional relations and new geochronological data from the Sua Pan area. J. Afr. Earth Sci. 32 (2), 257 267. Mapeo, R.B.M., Ramokate, L.V., Corfu, F., Davis, D.W., Kampunzu, A.B., 2006. The Okwa basement complex, Western Botswana: U-Pb zircon geochronology and implications for Eburnean processes in southern Africa. J. Afr. Earth Sci. 46, 253 262. Matthews, D.H., Cheadle, M.J., 1986. Deep reflections from the Caledonides and the Variscides west of Britain and comparisons with the Himalayas. In: Barazangi, M., Brown, L. (Eds.), Reflection Seismology: A Global Perspective, 13. American Geophysical Union, Geodynamics Series, pp. 5 19. McCourt, S., Hilliard, P., Armstrong, R.A., Munyanyiwa, H., 2001. SHRIMP UPb zircon geochronology of the Hurungwe granite northwest Zimbabwe: age constraints on the timing of the Magondi orogeny and implications for the correlation between the Kheis and Magondi Belts. South. Afr. J. Geol. 104 (1), 39 46. Millonig, L., Zeh, A., Gerdes, A., Klemd, R., Barton Jr., J.M., 2010. Decompressional heating of the Mahalapye Complex (Limpopo Belt, Botswana): a response to Palaeoproterozoic magmatic underplating? J. Petrol. 51 (3), 703 729. Paton, D.A., Pindell, J., McDermott, K., Bellingham, P., Horn, B., 2017. Evolution of seaward-dipping reflectors at the onset of oceanic crust formation at volcanic passive margins: insights from the South Atlantic. Geology 45 (5), 439 442. Quirk, D.G., Shakerley, A., Howe, M.J., 2014. A mechanism for construction of volcanic rifted margins during continental breakup. Geology 42 (12), 1079 1082. Roberts, D.G., Bally, W., 2012. Regional geology and tectonics of sedimentary basins: a prologue. In: first ed. Roberts, D.G., Bally, W. (Eds.), Regional Geology and Tectonics: Principles of Geologic Analysis, 2012. Elsevier, pp. 1 15. Rubey, M., Brune, S., Heine, C., Davies, D.R., Williams, S.E., Mu¨ller, R.D., 2017. Global patterns in Earth’s dynamic topography since the Jurassic: the role of subducted slabs. Solid Earth 8, 899 919. Schoene, B., Condon, D.J., Morgan, L., McLean, N., 2013. Precision and accuracy in geochronology. Elements 9, 19 24. Snyder, D.B., Hobbs, R.W., 1999. The BIRPS Atlas II: A Second Decade of Deep Seismic Reflection Profiling. Geological Society, London, CD-ROM. So¨derlund, U., Hofman, A., Klausen, M.B., Olsson, J.R., Ernst, R., Persson, P.-O., 2010. Towards a complete magmatic barcode for the Zimbabwe craton: baddeleyite U-Pb dating of regional dolerite dyke swarms and sill complexes. Precambrian Res. 183, 388 398. Tankard, A., Welsink, H., Aukes, P., Newton, R., Stettler, E., 2012. Geodynamic interpretation of the Cape and Karoo Basins, South Africa. In: first ed. Roberts, D.G., Bally, W. (Eds.), Regional Geology and Tectonics: Principles of Geologic Analysis, 2012. Elsevier, pp. 869 945. Chapter 23. Will, T.M., Frimmel, H.E., 2013. The influence of inherited structures on dike emplacement during Gondwana breakup in southwestern Africa. J. Geol. 121, 455 474.

Regional Geology and Tectonics

C H A P T E R

2 The Earth: core, mantle and crust C. Mary R. Fowler Darwin College, Cambridge, United Kingdom

Overview Though Jules Verne’s adventurers journeyed to the centre of the Earth, we cannot. Our knowledge of the interior is gained from physical information, especially seismology, and also the study of magnetism, gravity and many other physical properties, from geochemistry, especially through the study of lavas and xenoliths and from experimental study of materials making up the interior. The study of the Earth goes back far. The Hebrews and ancient Egyptians thought about it, and Psalm 104 has echoes back to Pharaoh Akhenaton. This hymn of the creation talks of the waters as a mantle round the Earth; nowadays, the same idea of mantle is transferred to the Earth’s interior. The Greeks and Romans thought much about the problem, and the greatest Roman natural historian, Pliny, perished during the eruption of Vesuvius. The Victorian physicists began our modern study of the planet. Both Lord Kelvin and Lord Rayleigh provided the mathematical apparatus, and Rutherford the insight into radioactivity. Several early seismologists made major contributions. Richard Dixon Oldham (1858
1936) discovered the core (Oldham, 1906) and Beno Gutenberg (1889
1960) estimated its depth as 2900 km below the surface (Gutenberg, 1913, 1914). The core
mantle boundary (CMB) is often called the ‘Gutenberg discontinuity’ after him. In 1924, Sir Harold Jeffreys (1891
1989) published his book The Earth, presenting a clear understanding of the structure of the planet. In 1926, Jeffreys’ work on tides led him to establish that the outer core must be fluid (Jeffreys, 1926). In 1936, Inge Lehmann (1888
1993) showed that the core was yet more complex (Lehmann, 1936, 1987). She used seismic energy from a large earthquake that occurred in 1929 near Buller in the South Island of New Zealand and had passed right through the core, to show that there was an inner core within the liquid outer core. In her honour, the boundary between the outer core and the inner core is now called the ‘Lehmann discontinuity’. Shear waves propagating through the inner core were only first clearly identified from the 1996 Flores Sea earthquake (Duess et al., 2000). Ben-Menahem (1995) and Shearer (2009) provide reviews of landmarks in the history of seismology. Understanding the function of the Earth took longer. From the late 1960s, the theory of plate tectonics started development of an integrated understanding of the major surface processes and features. With advances in technology and techniques and with faster numerical analysis, the details of the structure and composition of the deep Earth are steadily being worked out, but there may yet be surprises in store. Broadly speaking, the Earth is made up of a series of concentric shells,1 each shell being physically and chemically distinctive and with fairly sharp transitions between shells (Fig. 2.1). The thinnest and outermost shell is the crust. The base of the crust is marked by the Mohorovicic discontinuity (normally abbreviated to Moho), named 1

In the 1940s, Keith Bullen divided the Earth into a concentric series of spherical shells labelled A
G, using the Jeffreys and Bullen (1940) ‘J-B’ traveltimes as a basis in order to categorize the Earth’s density structure (Bullen, 1947). In this classification, A was the crust, B the mantle down to B400 km, C the mantle between B400 and B1000 km, D the mantle below B1000 km, E and F the outer core and G the inner core. The subdivision into distinct shells was extensively revised and modified through the years, but the alphabetic labels now only survive within the lower mantle: the lowermost 200 km of the mantle is called the Dv (D-double prime) shell and that part of the lower mantle between 670 and 2700 km is sometimes still referred to as the D0 (D-prime) shell.

Regional geology and tectonics. DOI: https://doi.org/10.1016/B978-0-444-64134-2.00002-X

7

© 2020 Elsevier B.V. All rights reserved.

8

2. The Earth: core, mantle and crust

FIGURE 2.1 The Earth is a series of concentric shells having distinct physical and/or chemical properties.

CRUST Lithosphere

Upper

Transition zone

MANTLE Silicate (Mg, Fe)

Lower

solid

D'' layer

OUTER CORE Fe with impurities (Ni, O, S)

liquid

INNER CORE Fe

solid

TABLE 2.1 Volume, mass and density of the Earth. Volume 18

Mass %

(10 kg)

%

Densitya (103 kg/m3)

10

0.9

28

0.5

2.60
2.90

Moho
670

297

27.4

1064

17.8

3.38
3.99

Lower mantle

670
2891

600

55.4

2940

49.2

4.38
5.56

Outer core

2891
5150

169

15.6

1841

30.8

9.90
12.16

Inner core

5150
6371

8

0.7

102

1.7

12.76
13.08

Whole Earth

0
6371

1083

100

5975

100

2.60
13.08

Depth (km)

(10

Crust

0
Moho

Upper mantle

3

m)

21

a

Density values after Dziewonski and Anderson (1981).

after Andrya Mohoroviˇci´c who first reported a crust
mantle boundary (Mohoroviˇci´c, 1910). At the Moho, there is a sudden increase in seismic velocity from typical deep crustal values (B7 km/s) to upper mantle values ( . 8.1 km/s). Successively inward come the upper and lower mantle and then the outer and inner core. The boundary between the mantle and core is 2891 km below the surface (Table 2.1). The distinction between crust, mantle and core was a topic of Jeffreys (1924 and subsequent editions thereof). This distinction is primarily chemical, though it is identified by its physical consequences (e.g. on seismic wave propagation). Since the discovery of plate tectonics, emphasis has shifted to the distinction between the lithosphere, which is the strong conductive outer layer of the Earth, and the asthenosphere below it. The lithosphere, or rock sphere, is relatively cool and strong. Through it, heat is transmitted by conduction. Below this, in the asthenosphere, the rock is hotter and able to creep, so at these depths in the mantle heat is transmitted by convection. Although solid, the mantle acts as a very high-viscosity fluid (rather like a glacier, also solid). The term asthenosphere comes from the Greek word for ‘sphere of weakness or influenza’.

Regional geology and tectonics

9

Methods of investigation

The crust is the uppermost part of the lithosphere. In most places (except along the mid-ocean ridge axes) the uppermost part of the mantle is cool and strong and is also part of the lithosphere. Deeper, the boundary between lithospheric mantle and asthenospheric mantle is transitional, unlike the fairly sharp boundary between crust and mantle.

Methods of investigation The tools of investigation vary according to the problem. The upper crust can be investigated by field mapping where it is exposed by erosion or thrusting and in a few locations by deep drilling. Seismic reflection profiling yields images of fine structure over small areas and is central to oil discovery. Seismic reflection methods are also used for the deeper crust and uppermost mantle, down to at least 50 km. National reflection profiling programmes such as COCORP and USArray in the United States, LITHOPROBE in Canada, DEKORP in Germany, CROP in Italy, ECORS in France and BIRPS in the United Kingdom have been very successful. As seismic reflections arise from contrasts in seismic velocity and/or density, the seismic reflection method is good for identifying contrasts such as sediment/igneous boundaries, but it fails in homogeneous materials, such as large plutons. Wide-angle seismic reflection profiling and seismic refraction methods yield less detail than reflection methods, but they yield seismic velocity
depth structures that can be interpreted in terms of lithology. These seismic methods, used to map the finer detail of the outermost 40
50 km of the Earth, require controlled energy sources such as vibrators or explosives. Gravity, magnetic and electrical surveys are used to model the density, magnetization and resistivity structure and hence are used to additionally constrain properties such as lithology, mineralogy and fluid content. The deeper interior of the Earth cannot be studied by controlled source experiments and is studied by seismologists using earthquakes as the energy source. Networks of seismic recording stations are used to determine the variation of seismic velocities in the interior of the Earth. Initially, only the variation with depth was studied. This is achieved by using travel-time data, as well as the periods of the Earth’s free oscillations, and its mass and moment of inertia. Fig. 2.2 shows the seismic velocity structure for the whole Earth, the culmination of the work first pioneered by Jeffreys and his colleagues Bullen, Lehmann and others (see Bullen and Bolt, 1985). Seismic tomography uses similar techniques to X-ray tomography in medical science but with earthquakes and a global network of seismic recording stations (Romanowicz, 2003, 2008). Development of this methodology has resulted in major advances in understanding of the Earth’s deep interior, allowing lateral variations and inhomogeneities of the mantle to be imaged and studied (Fig. 2.3). In tomographic images of the mantle, variations in seismic FIGURE 2.2

Inner core

Outer core

Mantle 14

α α 15,000

10

ρ 8

β

10,000

6

β

ρ

4

5000

2

0

0

1000

2000

3000

4000

5000

6000

Depth (km)

Regional geology and tectonics

Density (kg/m)

Seismic velocity (km/s)

12

The seismic velocity (α, Pwave; β, S-wave; left-hand axis) and the density (ρ, right
hand axis) structure of the Earth. The Jeffreys and Bullen (1940) ‘J
B’ model for α and β (solid lines) differs only in detail from much more recent determinations such as the preliminary reference Earth model (PREM, broken lines), of Dziewonski and Anderson (1981). The J
B model was on the basis of travel times for body waves, while PREM was determined from a joint inversion of the free oscillation periods of the Earth, its mass and moment of inertia as well as travel-time data. The two regions where the details of the structure have been most improved are in the asthenosphere (lowvelocity zone in the upper mantle) and the inner core
outer core transition zone. Source: After Bullen and Bolt, 1985, An Introduction to the Theory of Seismology, fourth ed. Cambridge University Press, Cambridge, 499 pp.

(A)

(B) (i)

(ii)

(C)

(F)

(D)

(E)

FIGURE 2.3 Variations in seismic velocity through the mantle shown as perturbations from standard Earth velocity models. (A) Longwavelength variations in S-wave velocity shown as perturbations from a standard Earth model at depths of 60, 140, 290, 460, 700, 925, 1225, 1525, 1825, 2125, 2425 and 2770 km. The model was calculated using surface-wave phase-velocity maps, free oscillation data and long-period body-wave travel times. The maximum deviation exceeds 2% at the top of the upper mantle but is 1% in the lower mantle. (B) Large-scale variation through the mantle. Scales indicate % deviation from standard model (note varies with depth). (i) S-wave velocity and P-wave velocity variations shown at depths of 70, 300, 600, 1500 and 2800 km. (ii) Density variations shown at depths of 70, 600, 1500 and 2800 km. (C) Variations in P-wave velocity along a section from the Pacific to the Caribbean. White circles indicate earthquakes (Spakman, personal communication, 2003). (D) Variations in P-wave velocity along a section from China across Japan to the Pacific. White circles indicate earthquakes (Spakman, personal communication, 2003). (E) Variations in P-wave velocity along a section across India and the Himalayas through Nepal to Tibet. White circles indicate earthquakes. (F) Location maps for the tomographic sections shown in (C), (D) and (E) (Spakman, personal communication, 2003). Source: (A) From Masters, G., Laske, G., Bolton, H., Dziewonski, A., 2000. The relative behaviour of shear velocity, bulk sound speed and compressional velocity in the mantle: implications for chemical and thermal structure. In: Karato, S., Forte, A.M., Liebermann, R.C., Masters, G., Sixtrude, L. (Eds.), Earth’s Deep Interior, vol. 117. American Geophysical Union Monograph, pp. 63
87. (B) From Moulik, P., Ekstro¨m, G., 2016. The relationships between large-scale variations in shear velocity, density and compressional velocity in the Earth’s mantle. J. Geophys. Res., 121, 2737
2771. https://doi.org/10.1002/2015JB012679.

The lithosphere

11

velocity are shown as percentage perturbations from a standard Earth velocity model. These perturbations are normally coloured so that those regions having a lower than standard velocity are red and those having a higher than standard velocity are blue, with the gradation in shading indicating the degree of variation in velocity. Such a colour scheme is consistent with anomalies interpreted as arising from thermal variations. Texts such as Stein and Wysession (2013, Chapters 3 and 7) and Fowler (2005, Chapters 4 and 8) provide descriptions of the use of seismic methods in the study of whole Earth structure.

The lithosphere The lithosphere consists of crust and the uppermost mantle. It can be divided into the oceanic lithosphere, which is created at mid-ocean ridges and destroyed in subduction zones, and the continental lithosphere, which has a substantially longer lifetime and is a chimaeric amalgam of many events.

The oceanic crust The oceanic crust is thin, relatively young and uncomplicated compared to the continental crust, and chemically magnesium-rich compared to continental material. The oceanic crust is the product of partial melting of the mantle at the mid-ocean ridges: it is the cooled and crystallized melt fraction. It is typically 7 km thick, though often less along the crest of mid-ocean ridges. Oceanic crust is formed as a result of decompression melting in the mantle at relatively shallow depths below the mid-ocean ridges, as the mantle rises in passive response to plate separation. The oceanic crust is consequently basaltic and relatively uniform in composition with a standard layered structure that is independent of the ridge spreading rate. Oceanic basalts formed at ridges are termed Mid-Ocean Ridge Basalts as distinct from basalts added on top of older oceanic crust by volcanism in volcanoes on ocean islands, which are termed Ocean Island Basalts. All the present oceanic crust is young, not older than Jurassic. Some of the rising magma at mid-ocean ridges erupts on the seabed, typically forming pillow lavas and flows, as well as volcanoclastic debris. The top of the pile has open fractures and hence low seismic velocities. Below the level where cracks close under pressure, seismic velocities increase. Below the extrusive basalts, there is a transition into feeder dykes. The P-wave velocity of this upper volcanic layer increases from c.2.5 to 6.2 km/s. Deeper yet, there are increasing volumes of intrusive and more coarsely crystalline rocks such as gabbros and dunites with P-wave velocities of 6.5
7.2 km/s. The uppermost parts of the oceanic crust have been sampled by drilling, and models of the deeper crust are available from tectonic windows and from ophiolites, for example, in Troodos, Cyprus, or in Oman. The Ocean Drilling Programs (today the International Ocean Discovery Program, IODP and its predecessors, ODP, IPOD and DSDP) have provided much information about the ocean crust and its overlying sediment.

The continental crust In contrast to the oceanic crust, the continental crust is much thicker and of broadly granitoid composition. It is a veneer of silica-rich rocks, which range in age from Hadean to recent, an amalgam of aeons of volcanism, intrusion, erosion, sedimentation, deformation, metamorphism and accident. It is the residue of the relatively incompatible and less refractory fraction from partial melting of the mantle above subduction zones. Its seismic structure is very variable but obeys some general rules as a consequence of its chemical difference from the mantle. It is typically 35
40 km thick, with an average P-wave velocity of 6.5 km/s, varying from very low velocities in surface sediment to velocities in excess of 7.5 km/s in the deep crust. As a rough analogy, the continental crust can be regarded as the light scum that rides on the Earth’s denser mantle. The volume of the continental crust may have increased over time, but this remains controversial. Most models (mainly using isotopic evidence) suggest that the continental crust has grown over time, especially in the late Archaean, but some influential models suggest that the volume has changed little over time though the surface area may have varied (for an overview, see Taylor and McLennan, 1996). The surface of the crust records the most recent sedimentation event. The age of the continental basement is shown in Fig. 2.4. Over 70% of the present global land surface is either older than 450 Ma or underlain by material older than that.

Regional geology and tectonics

12

2. The Earth: core, mantle and crust

FIGURE 2.4 The age of the continents. Source: From Fowler, C.M.R., 2005. The Solid Earth: An Introduction to Global Geophysics, second ed. Cambridge University Press, Cambridge, 684pp., The Solid Earth, Cambridge University Press, on the basis of Sclater et al. (1981).

FIGURE 2.5 (A) The thickness of the continental crust in kilometres (contoured and on the basis of 5 3 5 cells). Model LITHO1.0: (B) Thickness of the crust and (C) thickness of the lithosphere. Source: (A) From Mooney, W.D., Laske, G., Masters, T.G., 1998. CRUST 5.1: a global crustal model at 5 3 5 . J. Geophys. Res., 103, 727
747. (B and C) From Pasyanos, M.E., Masters, T.G., Laske, G., Ma, Z. 2014. LITHO1.0: an updated crust and lithospheric model of the Earth, J. Geophys. Res., 119, 2153
2173, https://doi.org/10.1002/ 2013JB010626.

(A)

(B)

(C)

Regional geology and tectonics

The lithosphere

13

The top 10 km or so of crystalline crust beneath the sediment cover typically has a P-wave velocity of about 6.0
6.3 km/s and is made variably of metasedimentary metamorphic rocks, metavolcanics or intrusions. Beneath this, P-wave velocities are in excess of 6.5 km/s. Under significant regions of continental, the deep crust is broadly granulite or pyroxene granulite with velocities above 7 km/s. In some areas, of complex history and with thrusting events, there may be velocity inversions or low-velocity zones in the crust. Although the continental crust is typically 35
40 km thick, with an average of B38 km (Mooney et al., 1998), there are major regional variations. Fig. 2.5(A) shows the global crustal thickness plotted on a 5-degree grid. More detail and crustal properties are available on 2-degree and 1-degree grids
CRUST2.0, CRUST1.0, LITHO1.0 as shown in Fig. 2.5(B) and (C)
(Bassin et al., 2000; Laske et al., 2013; Pasyanos et al., 2014). Beneath the Andes, Alps, Himalayas and Tibet, the crust is thicker than 50 km. In areas of continental extension, such as the Rhine Graben in Germany and the East African Rift, the crust is locally thinner by 10
15 km than the unextended crust on either side of the rift zone. Much of the surface of the continental crust is covered in sediment. Though most of the sedimentary basins are less than about 550 Ma old, some Proterozoic and even Archaean ( . 2500 Ma old) sedimentary basins survive. In some large areas of shield (e.g. Ontario, parts of Australia, Zimbabwe), Archaean crystalline crust is exposed on the surface, under recent material. There are Archaean cratons in large tracts of Canada, Greenland, Brazil, West and Southern Africa, Finland, Russia, parts of North China, Western Australia and Antarctica. They are the surviving ancient continental nuclei, now surrounded, and in places overlain, by younger rocks. The continual erosion and deposition of sediment by the consequences of plate tectonics mean that a good deal of recycling takes place in crustal rocks. Sediments are accreted onto continents, scraped off subduction zones and collect as low- to medium-grade deformed metamorphic rocks. Others are metamorphosed by igneous intrusion or overlay. There is some ongoing loss of continental crust as sediment is taken down subduction zones, but much of the volatile component of this is recycled when volatiles are brought back via andesitic volcanoes. New crust in the form of andesitic volcanic rock is continually added to the continents. Two deep continental holes have been drilled to mid-crustal levels and are useful for studying rocks under in situ conditions: they are the Kola hole on Russia’s Kola Peninsula near the Norwegian border and the Kontinentales Tiefbohrprogramm de Bundesreplik Deutschland (KTB) hole in southwestern Germany. The Kola SDB-3 Super-Deep Bore was drilled to 12-km depth (Fuchs et al., 1990; Kozlovsky, 1987) and the KTB hole to 9-km depth (Bram and Draxler, 1994; Emmerman et al., 1995). These major projects used numerous geophysical, geotechnical and geochemical methods to investigate the details of the seismic discontinuities, structure and properties of the continental crust.

The subcrustal lithosphere The base of the lithosphere, be it oceanic or continental, is controlled by the temperature and composition of the mantle. As an oceanic plate moves away from the ridge that created it, it progressively cools. This cooling sets the base of the lithosphere, which can be defined in two ways. In thermal terms, the base of the lithosphere lies at the transition between transfer of heat by advective convection in the underlying asthenosphere and transfer of heat by conduction in the overlying lithosphere. In mechanical terms, the base of the lithosphere lies at the transition between the cold, dense, mechanically strong plate, which deforms elastically, and the underlying hotter, less strong asthenosphere, which deforms viscously. A consequence of the conductive cooling of the lithosphere is that the mean oceanic depth increases, and that overall the oceanic lithosphere thickens according to the square root of its age (Fig. 2.6). The oldest continental lithosphere is very thick, but unlike the oceanic situation, there is no straightforward mathematical equation relating continental lithospheric thickness to age. Just as for the oceanic lithosphere, the base of the continental lithosphere is controlled by the temperature of the mantle as heat is transported through the lithosphere by conduction and through the mantle by convection. The rheological boundary between the near-rigid conductive lithosphere and the underlying convecting mantle is not sharp; rather there is a boundary zone within which both conduction and convection operate and which can be 50
80 km in thickness (Sleep, 2005). While the conductive geotherm for the upper continental crust may be as high as 30 C/km, in the mantle at 200-km depth and 1100 C
1200 C it can be as low as c.5
10 C/km. This gradient is low as it is

Regional geology and tectonics

14

2. The Earth: core, mantle and crust

2

Ocean depth (km)

3

4

5

6 0

5 Square root of lithosphere age (Ma1/2)

10

FIGURE 2.6 The mean depth of the oceans (d in km) plotted against the square root of the age (t in Ma) of the underlying oceanic litho-

sphere (grey dots) The solid line d 5 2.604 1 0.344 t1/2 is the best-fitting straight line for ages up to 73 Ma) Source: After Carlson, R.L., Johnson, H.P., 1994. On modeling the thermal evolution of the oceanic upper mantle: an assessment of the cooling plate model. J. Geophys. Res. 99, 3201
3214 on the basis of Stein and Stein (1992).

Differential stress (MPa)

Differential stress (MPa) 0

200

400

600

0

0 0

800 0

Diabase

200

400

600

0 Quartz 200

200

600

OCEANIC LITHOSPHERE

800

Depth (km)

Depth (km) 60

Olivine

20 400

Olivine

600

40

CONTINENTAL LITHOSPHERE 60

Temperature (°C)

400

Temperature (°C)

20

40

800

800

FIGURE 2.7

Yield strength envelopes for the oceanic lithosphere (60 Ma old) and the continental lithosphere in compression. The yield strength envelopes for the oceanic and continental lithosphere in tension have a similar shape to those shown here, but as rocks are not as strong in tension as in compression, the differential stress is reduced by a factor of two to three. Envelopes are shown for a strain rate of 10215/s. Source: From Kohlstedt, D.L., Evans, B., Mackwell, S.J., 1995. Strength of the lithosphere: constraints imposed by laboratory experiments. J. Geophys. Res., 100, 17587
17602.

effectively controlled only by the mantle heat flow, the concentration of radiogenic heat-producing elements in the mantle being very low. In the underlying sublithospheric mantle at c.1400 C, the temperature gradient is close to adiabatic2 and about 0.4 C/km. Fig. 2.7 illustrates how the lithosphere behaves under compression. At shallow depths, the stress at which failure occurs increases linearly with depth. However, as rocks are brittle at low temperatures and become ductile at higher temperatures, reductions in differential stress occur where rocks start to deform by solid-state creep. The quartz mineralogy means that the upper continental crust is strong while the lower continental crust is weak. However, at the Moho, with the change from quartz to olivine mineralogy, the strength of the lithosphere 2

The adiabat is the pressure versus temperature profile within the Earth that would be followed by a body that rises (expands) or contracts (falls) without giving or receiving heat.

Regional geology and tectonics

The mantle

15

increases. In contrast, the yield strength envelope for the oceanic lithosphere is simple—strength increases linearly with depth down to c.35 km and below this the lithosphere deforms by solid-state creep. Just as the oceanic lithosphere thickness is proportional to the square root of the plate age, so is the depth at which the oceanic lithosphere has its maximum strength.

The mantle The uppermost mantle is very heterogeneous and a significant part is within the lithosphere. Beneath the continents, the lithosphere can extend to depths well in excess of 200 km under the old and cold continental cratons (Sleep, 2005). In general, the highest velocities are beneath the old continents, while beneath the mid-ocean ridges and tectonically active regions the velocities are low down to 250 km (Fig. 2.3A). Under the oceans, the lithosphere along the axes of mid-ocean ridges is scarcely thicker than the crust, but the lithosphere of very old oceanic plates thickens to 100
150 km (Fig. 2.3A; depth 60 km shows the lowest velocities along the mid-ocean ridge system). The asthenosphere, that part of the mantle immediately beneath the lithosphere, is characterized by a general low-velocity zone down to about 220
250 km, as established by surface-wave dispersion data and a P-wave shadow zone (Fig. 2.2). Beneath the low-velocity zone, the P- and S-wave velocities increase markedly until 400-km depth. Here, and also at 670 km, there are sharp step-like increases in seismic velocity, linked to phase changes in olivine. At 390
410 km, olivine changes to a b-spinel structure and pyroxene also changes to a garnet structure at these levels. The zone between 400 and 670 km is known as the transition zone. At around 520 km, the b-spinel structure changes again to a g-spinel structure. The boundary between the upper and lower mantle is taken to be at a depth of 670 km. The topography of this boundary can be determined by seismic tomography, which shows that it varies in depth by up to 30 km, with the depressions located in places associated with subduction zones, suggesting that the 670-km boundary can provide a partial impediment to the descent of subducted lithosphere to the lower mantle (Fig. 2.3D). The subducting plates show up on high-resolution tomographic images as high-velocity regions because the lithosphere is colder than the mantle into which it is descending. Along some subduction zones, the descending plate descends through the upper mantle (Japan, Tonga) and some (Farallon, Tethys) clearly cross into the lower mantle (Fig. 2.3C and D), demonstrating that the upper and lower mantle are not entirely separate convecting systems. Part of the seismic heterogeneity in the mantle clearly arises from the subduction of lithosphere (Fig. 2.8) over aeons. The dipping blue (fast) structure in the mantle beneath North America (Fig. 2.3C) is interpreted as the subducted Farallon plate, of which the last remnant is today’s Juan de Fuca plate (Bijwaard and Spakman, 2000). The dipping blue structure in the mantle beneath Korea and China (Bijwaard et al., 1998) is interpreted as the subducting Pacific plate (Fig. 2.3D). A very thick high-velocity region is associated with the India-Eurasia collision zone along with substantial high-velocity regions in the underlying mantle (Fig. 2.3E). These regions in the mid- and lower mantle are interpreted as remnants of subducted oceanic lithosphere dating from the closing of the Neo-Tethys Ocean, while the major fast anomaly at 400
1000-km depth beneath Tibet may be subcontinental lithosphere which has detached from, and has been overridden by, the Indian plate (Van der Voo et al., 1999). The seismic heterogeneity arising from ancient subducted plates with the mantle can be used to study mantle convection and geodynamic plate models (Bunge and Grand, 2000; Bunge et al., 1998, 2003; Lithgow-Bertelloni and Richards, 1998; Schuberth et al., 2009; Stadler et al., 2010; van der Hilst et al., 1997). As the sublithospheric mantle is convecting, the temperature increase with depth is close to adiabatic. This means that the mantle temperature can be expressed in terms of the potential temperature.3 The adiabatic gradient and the potential temperature are very important as they are the basic controls on melting in the mantle. Fig. 2.9 illustrates the difference between the adiabatic gradient and the melting curve. If a mantle rock is rising, its temperature will follow the adiabat. At the depth where the adiabat intersects the melting curve, the rock will begin to melt. When melted, material separates from the solid residue; being less dense, it rises and its temperature follows the melt adiabat (Fig. 2.9). In regions of the mantle where the potential temperature is high, melting will start at greater depths. Below 670 km, the g-spinel form shifts to minerals with postspinel structure, perovskite [(Mg, Fe) SiO3] and magnesiowustite [(Mg, Fe) O]. The lower mantle is a significant part of the Earth’s volume (Table 2.1). The mantle, a ductile solid, is thought to be largely made up of magnesium minerals such as Mg-perovskite. Perovskite 3

The potential temperature of the mantle, Tp, is the temperature at the theoretical intersection of the adiabat with the Earth’s surface.
Tp 5 T exp
αgz=cp , where T(z) is the temperature at depth z, a is the coefficient of thermal expansion and cp is specific heat.

Regional geology and tectonics

16

2. The Earth: core, mantle and crust

FIGURE 2.8 Cold, high-density material is continually added to the mantle along the subduction zones, contributing to mantle heterogeneity. The location of these high-density anomalies has been computed by taking into account the movement of the plates and the viscosity structure of the mantle. Two sets of maps are shown: the present day and 56 Ma ago. Source: From Lithgow-Bertelloni, C., Richards, M.A., 1998. The dynamics of Cenozoic and Mesozoic plate motions. Rev. Geophys., 36 (1), 27
78.

undergoes a phase change to postperovskite towards the base of the mantle. The heterogeneities in the outer part of the lower mantle are of shorter wavelengths than the heterogeneities in the upper mantle (Fig. 2.3). These shorter-wavelength lower mantle anomalies merge into extensive lateral anomalies towards the base of the lower mantle. There are two fast regions (beneath India and the Americas) and two slow regions (beneath Africa and the Pacific) termed large low shear
velocity provinces (LLSVPs). As the slow regions, coloured red, may be hot, and the fast regions, coloured blue, may be cold, many interpretations liken them to hot upwelling zones and cold descending ‘slab graveyards’ that stagnate at the base of the mantle. Broad low-velocity regions have been imaged extending from the base of the mantle towards the upper mantle beneath several major major hotspots (French and Romanowicz, 2015). Models of global convection indicate that plumes often form preferentially near the edges of thermochemical anomalies similar to LLSVPs (Hassan et al., 2014). The lowermost B200 km of the mantle, the Dv shell, is the zone of interaction between the core and mantle. In some ways, this shell is comparable to the lithosphere at the top of the asthenosphere, and may be a thermal as well as a chemical boundary layer, with a temperature contrast across the shell of 1000K 6 500K. The Dv shell is very heterogeneous, with strong lateral variations, up to 3%
4% in seismic velocity, and strong seismic

Regional geology and tectonics

17

The core

Temperature Liquid adiabat

B

Depth

A LIQUID SOLID

ve

adiabat

g cur Meltin

Mantle

FIGURE 2.9 The temperature of rising mantle material follows the adiabat. Melting starts when the rising material reaches the depth where the mantle adiabat and the melting curve intersect (point A). The temperature of the melt will lie on the melting curve until the melt separates from the solid residue (point B). The liquid melt will then rise to the surface with its temperature following a liquid adiabat. Source: After Fowler, C.M.R., 2005. The Solid Earth: An Introduction to Global Geophysics, second ed. Cambridge University Press, Cambridge, p. 684, The Solid Earth, Cambridge University Press.

anisotropies. While some of this variation may be due to local temperature anomalies up to 200K
300K, some may also result from variations in chemical composition (e.g. silicate/oxide ratio), and some may arise from this region at the base of the mantle being in effect a ‘graveyard’ for subducted lithosphere. Laboratory experiments have confirmed that strong chemical reactions take place between the iron and magnesium silicates that make up the mantle and liquid iron and iron alloys. Thus the seismic complexity of the CMB may also reflect the chemical reactions taking place there—it may be the most chemically active part of the planet. The exact form that mantle convection takes is not fully understood. Numerical models taking into account a range of possible compositions, mineralogies and phase changes, viscosity, other physical parameters and assumed temperatures and boundary conditions yield a variety of models (Fig. 2.10). Tomographic images (e.g. Fig. 2.3) show that despite the major increases in density and seismic velocity at c.670 km, descending slabs may be perturbed at this depth, but can penetrate into the lower mantle. Thus while there is geochemical evidence for the existence of a separate repository within the lower mantle, and the flow is substantially affected by the 670-km endothermic phase change for spinel, the mantle is one system. The extent to which mantle structure at the CMB may affect convection in the outer core is unknown.

The core The core accounts for 16% of the volume and 33% of the mass of the Earth. The CMB is marked by the largest physical and chemical contrasts in the interior of the Earth, The major physical and chemical differences between the mantle and the outer and inner core are inferred from the basic seismic wave velocity and density structure (Fig. 2.2). At the CMB, the P-wave velocity falls from 13.71 km/s at the base of the lower mantle to 8.06 km/s, the S-wave velocity falls from 7.26 km/s to zero and the density increases from the lowermost mantle value of 5567 to 9903 kg/m3. Such a change in density rivals the change at the Earth’s surface. Within the outer core, the P-wave velocity increases steadily from 8.06 km/s, reaching 10.36 km/s at the outer core
inner core boundary. At the outer core
inner core boundary, the P-wave velocity increases from 10.36 to 11.03 km/s, the S-wave velocity increases from 0 to 3.50 km/s and the density increases from 12,166 to 12,764 kg/m3. Through the inner core, the P- and S-wave velocities and density increase slightly, reaching maximum values of 11.26 km/s, 3.66 km/s and 13,088 kg/m3, respectively, at the centre of the Earth. A liquid outer core and solid inner core are consistent with all seismological observations as well as studies on tides and the Earth’s rotation which both require a liquid core.

Regional geology and tectonics

18

2. The Earth: core, mantle and crust

(A)

(B)

(C)

(D)

(E)

(F)

(G)

(H)

(I)

FIGURE 2.10 Three-dimensional spherical models of mantle convection illustrate the effects of varying some physical parameters and boundary conditions. The outermost 200-km layer is not shown. Temperatures: red, hot; blue, cold. (A) Mantle is incompressible and has constant viscosity and internal (radioactive) heating only, Rayleigh number for the flow being 4 3 107. (B) As (A) but the viscosity of the lower mantle is 30 times greater than that of the upper mantle. (C) As (B) but showing a constant temperature surface. (D) As (B) but showing the planform of the flow regime. (E) Mantle is compressible and has constant viscosity, with Rayleigh number 108. (F) As (E) but with 38% of the heat coming from the core. (G) As (E) but with an endothermic (requiring heat) phase change at 670-km depth. (H) As (E) but the viscosity of the lower mantle is 30 times greater than that of the upper mantle. (I) As (H) but with 38% of the heat coming from the core (Bunge et al., personal communication, 2003). Source: (I) After Bunge, H.P., Richards, M.A., Baumgardner, J.R., 1997. A sensitivity study of 3-D spherical mantle convection at 108 Rayleigh number: effects of depth dependent viscosity, heating mode and an endothermic phase change. J. Geophys. Res. 102, 11991
12007.

The composition of the core is very hard to verify as we do not have samples to study. Rather, we have to rely on ingenuity and analogue. The abundance of elements in the Sun and in meteorites suggests that in order to balance the silicate mantle, the core should be predominantly iron, and in bulk BFe2O, with some nickel and a small proportion of light elements. Other constraints on the composition of the core come from its seismic velocity and density structure (Fig. 2.2), mineral physics calculations and laboratory experiments conducted at high pressure and temperatures on potential materials. The outer core is less dense than would be expected for pure iron or an iron
nickel alloy at core conditions and it is thought to contain B10% of lighter elements. Iron, nickel,

Regional geology and tectonics

References

19

silicon or oxygen and sulphur, together with small amounts of chromium, phosphorus, carbon and hydrogen, are the probable constituents (McDonough, 2003). The solid inner core is a very small part of the Earth: making up less than 2% of the mass and less than 1% of the volume of the whole Earth (Table 2.1) or about 5% of the mass but slightly less than 5% of the volume of the core. Its density is slightly less, perhaps 4%
5% less, than that expected for pure iron at core conditions (e.g. Hemley and Mao, 2001) and it may also contain a small proportion of light elements, possibly including oxygen. The Earth’s liquid outer core plays an important role as the source of the magnetic field. It effectively acts as a giant spherical self-exciting dynamo, in which less dense, rising and spiralling convection currents of liquid iron also carry electric currents (Buffett, 2000). The interaction of these electric currents with the Earth’s magnetic field then results in an enhancement of that magnetic field. This phenomenon occurs because the outer core is liquid, convects strongly and can conduct electricity.

References Bassin, C., Laske, G., Masters, G., 2000. The current limits of resolution for surface wave tomography in North America, EOS Trans AGU, 81, F897. ,http://igppweb.ucsd.edu/Bgabi/rem.html.. Ben-Menahem, A., 1995. A concise history of mainstream seismology: origins, legacy, and perspectives. Bull. Seismol. Soc. Am. 85 (4), 1202
1225. Bijwaard, H., Spakman, W., 2000. Non-linear global P-wave tomography by iterated linearized inversion. Geophys. J. Int. 141, 71
82. Bijwaard, H., Spakman, W., Engdahl, E.R., 1998. Closing the gap between regional and global travel time tomography. J. Geophys. Res. 103, 30055
30078. Bram, K., Draxler, J.K., 1994. Basic Research and Borehole Geophysics (Final Report). KTB Research 94-1, ISBN 3-928559-11-7, p. 460. Buffett, B.A., 2000. Earth’s core and the geodynamo. Science 288, 2007
2012. Bullen, K.E., 1947. Introduction to the Theory of Seismology. Cambridge University Press, Cambridge, p. 276pp. Bullen, K.E., Bolt, B.A., 1985. An Introduction to the Theory of Seismology, fourth ed. Cambridge University Press, Cambridge, p. 499pp. Bunge, H.-P., Grand, S.P., 2000. Mesozoic plate-motion history below the northeast Pacific Ocean from seismic images of the subducted Farallon slab. Nature 405, 337
340. Bunge, H.P., Richards, M.A., Baumgardner, J.R., 1997. A sensitivity study of 3-D spherical mantle convection at 108 Rayleigh number: effects of depth dependent viscosity, heating mode and an endothermic phase change. J. Geophys. Res. 102, 11991
12007. Bunge, H.P., Richards, M.A., Lithgow-Bertelloni, C., Baumgartner, J.R., Romanowicz, B.A., 1998. Time scales and heterogeneous structure in geodynamic Earth models. Science 280, 91
95. Bunge, H.P., Hagelberg, C.R., Travis, B.J., 2003. Mantle circulation models with variational data assimilation: inferring past mantle flow and structure from plate motion histories and seismic tomography. Geophys. J. Int. 152, 280
301. Available from: https://doi.org/10.1046/ j.1365-246X.2003.01823.x. Carlson, R.L., Johnson, H.P., 1994. On modeling the thermal evolution of the oceanic upper mantle: an assessment of the cooling plate model. J. Geophys. Res. 99, 3201
3214. Duess, A., Woodhouse, J.H., Paulssen, H., Trampert, J., 2000. The observation of inner core shear waves. Geophys. J. Int. 142, 67
72. Dziewonski, A.M., Anderson, D.L., 1981. Preliminary reference Earth model. Phys. Earth Planet. Inter 25, 297
356. Emmerman, R., Althaus, E., Giese, P., Sto¨ckhert, B., 1995. KTB Hauptbohrung
Results of a Geoscientific Investigation in the KTB Field Laboratory. Final Report 0-9101m, KTB Report 95-2, ISBN 3-928559-15-X. Fowler, C.M.R., 2005. The Solid Earth: An Introduction to Global Geophysics, second ed. Cambridge University Press, Cambridge, p. 684pp. French, S.W., Romanowicz, B.A., 2015. Broad plumes rooted at the base of the Earth’s mantle beneath major hotspots. Nature 525, 95
99. Fuchs, K., Kozlovsky, E.A., Krivtsov, A.I., Zobak, M.D. (Eds.), 1990. Super-Deep Continental Drilling and Deep Geophysical Sounding. Springer Verlag, Berlin. Gutenberg, B., 1913. Uber die Konstitution der Erdinnern, erschlossen aus Erdbebenbeobach-tungen. Phys. Z 14, 1217
1218. ¨ ber Erdbebenwellen, VII. A Beobachtungen an Registerierungen von Fernbeben in Go¨ttingen und Folgerungen u¨ber die Gutenberg, B., 1914. U Konstitution des Erdko¨rpers, Nachrichten von der Ko¨niglichen Gesellschaft der Wissenschaften zu Go¨ttingen, Mathematisch-physikalische Klasse. pp. 125
176. Hassan, R., Flament, N., Gurnis, M., Bower, D.J., Mu¨ller, D., 2014. Provenance of plumes in global convection models. Geochem. Geophys. Geosyst. 16, 1465
1489. Hemley, R.J., Mao, H.K., 2001. In situ studies of iron under pressure: new windows on the Earth’s core. Int. Geol. Rev. 43, 1
30. Jeffreys, H., 1924. first ed. The Earth, Cambridge University Press, Cambridge, p. 278pp., 1976, sixth ed. 574 pp). Jeffreys, H., 1926. The rigidity of the earth’s central core. Mon. Not. R. Astron. Soc. Geophys. (Suppl. 1), 371
383. Jeffreys, H., Bullen, K.E., 1940. Seismological Tables. British Association Seismological Committee, London, p. 50pp. Kohlstedt, D.L., Evans, B., Mackwell, S.J., 1995. Strength of the lithosphere: constraints imposed by laboratory experiments. J. Geophys. Res. 100, 17587
17602. Kozlovsky, Y., 1987. The Superdeep Well of the Kola Peninsula. Springer Verlag, Berlin, p. 558pp. Laske, G., Masters, G., Ma, Z., Pasyanos, M., 2013. Update on CRUST1.0
A 1-degree global model of earth’s crust, Geophys. Res. Abstracts, 15, EGU2013-2658. ,https://igppweb.ucsd.edu/Bgabi/crust1.html.. Lehmann, I., 1936. P’, Publications du Bureau Central Se´ismologique International, Se´rie A, Travaux Scientifiques 14, 87
115. Lehmann, I., 1987. Seismology in the days of old. EOS 68 (3), 33
35. Lithgow-Bertelloni, C., Richards, M.A., 1998. The dynamics of Cenozoic and Mesozoic plate motions. Rev. Geophys. 36 (1), 27
78.

Regional geology and tectonics

20

2. The Earth: core, mantle and crust

Masters, G., Laske, G., Bolton, H., Dziewonski, A., 2000. The relative behaviour of shear velocity, bulk sound speed and compressional velocity in the mantle: implications for chemical and thermal structure. In: Karato, S., Forte, A.M., Liebermann, R.C., Masters, G., Sixtrude, L. (Eds.), Earth’s Deep Interior, vol. 117. American Geophysical Union Monograph, pp. 63
87. McDonough, W.F., 2003. Compositional model for the earth’s core. In: Carlson, R.W. (Ed.), The Mantle and Core, vol. 2. Elsevier-Pergamon, Oxford, pp. 547
568. , Holland, H.D., Turkian, K.K., (Eds.), Treatise on Geochemistry. Mohoroviˇci´c, A., 1910. Das Beben vom 8.X.1909, Jahrbuch des Meteorologischen Observatoriums in Zagreb (Agram) fuer das Jahr 1909, 9 (part 4, Section 1), pp. 1
63. Mooney, W.D., Laske, G., Masters, T.G., 1998. CRUST 5.1: a global crustal model at 5 3 5 . J. Geophys. Res. 103, 727
747. Moulik, P., Ekstro¨m, G., 2016. The relationships between large-scale variations in shear velocity, density and compressional velocity in the Earth’s mantle. J. Geophys. Res. 121, 2737
2771. Available from: https://doi.org/10.1002/2015JB012679. Oldham, R.D., 1906. The constitution of the earth as revealed by earthquakes. Q. J. Geol. Soc 62, 456
475. Pasyanos, M.E., Masters, T.G., Laske, G., Ma, Z., 2014. LITHO1.0: an updated crust and lithospheric model of the Earth. J. Geophys. Res. 119, 2153
2173. Available from: https://doi.org/10.1002/2013JB010626. Romanowicz, B., 2003. Global mantle tomography: progress status in the past 10 years. Annu. Rev. Earth Planet. Sci. 31, 303
328. Romanowicz, B., 2008. Using seismic waves to image Earth’s internal structure. Nature 451, 266
268. Available from: https://doi.org/ 10.1038/nature06583. Schuberth, B.S.A., Bunge, H.-P., Ritsema, J., 2009. Tomographic filtering of high-resolution mantle circulation models: can seismic heterogeneity be explained by temperature alone? Geochem. Geophys. Geosyst 10, . Available from: https://doi.org/10.1029/2009GC002401Q05W03. Sclater, J.R., Parsons, B., Jaupart, C., 1981. Oceans and continents: similarities and differences in the mechanisms of heat loss. J. Geophys. Res. 86, 11535
11552. Shearer, P., 2019. Introduction to Seismology, third ed. Cambridge University Press, ISBN: 9781316635742. Sleep, N.H., 2005. Evolution of the continental lithosphere. Annu. Rev. Earth Planet. Sci. 33, 369
393. Available from: https://doi.org/ 10.1146/annurev.earth.33.092203.122643. Stadler, G., Gurnis, M., Burstedde, C., Wilcox, L.C., Alisic, L., Ghattas, O., 2010. The dynamics of plate tectonics and mantle flow: from local to global scales. Science 329, 1033
1038. Available from: https://doi.org/10.1126/science.1191223. Stein, C.A., Stein, S., 1992. A model for the global variation in oceanic depth and heat flow with lithospheric age. Nature 359, 123
129. Stein, S., Wysession, M., 2013. An Introduction to Seismology, Earthquakes and Earth Structure. Wiley, ISBN: 978-1-118-68745-1p. 512. Taylor, S.R., McLennan, S.M., 1996. The evolution of continental crust. Sci. Am. 76
81. January. van der Hilst, R.D., Widiyantoro, S., Engdahl, E.R., 1997. Evidence for deep mantle circulation from global tomography. Nature 386, 578
584. Available from: https://doi.org/10.1038/386578a0. Van der Voo, R., Spakman, W., Bijwaard, H., 1999. Tethyan subducted slabs under India. Earth Planet. Sci. Lett. 171, 7
20.

Further reading Dziewonski, A.M., Romanowicz, B., 2015. Deep earth seismology: an introduction and overview. In: Schubert, G. (Ed.), Treatise on Geophysics, second ed. Elsevier, Oxford, pp. 1
28. Grand, S., van der Hilst, R.D., Widiyantoro, S., 1997. Global seismic tomography: a snapshot of convection in the Earth. GSA Today 7 (4), 1
7.

Regional geology and tectonics

C H A P T E R

3 Age of the oceans Lucı´a Pe´rez Dı´az Fault Dynamics Research Group, Royal Holloway University of London, Egham, United Kingdom

Introduction Since the first maps of the American continents were published in the early 16th century, the oceans between them and the European and African continents intrigued geographers and naturalists. ‘The vestiges of the rupture [between the Americas, Africa and Europe] reveal themselves, if someone brings forward a map of the world and considers carefully the coasts of the three’. This statement by Dutch map maker Abraham Ortelius (1596) put forward the speculation that continents might have drifted apart through time until reaching their present configuration. The similarity between continental coastlines (and especially those of South America and Africa) was noted by many between the 16th and 19th centuries (e.g. Bacon, 1620; Lilienthal, 1756; Von Humboldt, 1801, 1845), as was the similitude in the in-land geology of distant landmasses. These early notions and maps may have contributed, later on, to the development of the theory of plate tectonics. However, at the time, the mechanism leading to the separation of South America and Africa had little to do with continental drift. According to Placet (1658), South America and Africa, once joined by the lost continent of Atlantis, were separated when the lost continent disappeared below water after the great biblical floods. Similarly, Alexander von Humboldt (1801) argued that the Atlantic Ocean represents a large and ancient river bed, flooded by the biblical catastrophe (Fig. 3.1). Although the great flood idea had become obsolete by the early 19th century, ‘fixist’ notions remained very popular until the establishment and broad acceptance of plate tectonic theory in the 1960s. This was partly due to the absence of a convincing mechanism that is able to produce the forces needed to move entire continents apart as depicted by Antonio Snider-Pellegrini as early as 1858 (Fig. 3.2). According to fixism, the formation of ocean basins resulted from the subsidence of continental masses, without the need for substantial extension but simply the transformation of continental into oceanic (thinner) lithosphere. In 1890, a letter written in the Proceedings of the Geological Society by Dr Wallace read ‘. . .while the general permanence of ocean basins and continental areas cannot be said to be established on firm proof, the general evidence in favour of this view is very strong’. With the onset of extensive ocean basin research in the first half of the 20th century, it became clear that, on the geological timescale, the seafloor is far from permanent. The present ocean basins are being created by seafloor spreading and recycled by subduction on a timescale just short of 200 million years (Fig. 3.3). Unlike continental lithosphere, which over time may be deformed, fragmented, reassembled and eroded, oceanic lithosphere is relatively simple and thus the information recorded on it is preserved through time. For this reason and despite the fact that oceanic lithosphere only provides information about the most recent 4% of Earth’s history, the study of ocean basins provided the basis for the development of plate tectonics.

Ocean basin research

historical perspective

Ocean exploration with the modern scientific method started in the 19th century. The most significant of early oceanic expeditions was that of research ship HMS Challenger between 1872 and 1876. Funded by the British Regional Geology and Tectonics. DOI: https://doi.org/10.1016/B978-0-444-64134-2.00003-1

21

© 2020 Elsevier B.V. All rights reserved.

22

3. Age of the oceans

FIGURE 3.1 Original illustration from Thomas Burnet’s ‘The Sacred Theory of the Earth’ 1684 showing how the earth of the creation is transformed by waters unleashed from below the crust. FIGURE 3.2 Continent configuration before (left) and after (right) the separation of Africa and South America, as interpreted by Snider-Pellegrini (1858).

FIGURE 3.3 Schematic diagram illustrating the creation and recycling of lithosphere according to the theory of global tectonics developed in the 1960s. Source: Redrawn from Isacks, B., Oliver, J., 1968. Seismology and the new global tectonics. J. Geophys. Res. 71(18), with permission. Permission obtained via Rightslink.

Royal Society, this expedition collected information about seafloor depth and water temperatures, as well as samples of ocean bottom rocks and deep-sea marine life. The abundant and diverse fossil content of dredged samples contrasted with the simplicity in sediment types found, when compared to terrestrial strata. One of the most surprising observations arising from this campaign was that the ocean floor is not deepest in the middle of the oceans, hinting at the existence of the vast mid-ocean ridge system whose full recognition would later be instrumental in the establishment of the concept of seafloor spreading. The modern era of deep-ocean exploration began between the first and second world wars with the establishment of the Scripps Institution of Oceanography and Woods Hole Oceanographic Institution. However, it was not until after World War II (1939 45) that the forms and structures of all the major oceanic basins were clearly mapped. Electronic navigation systems developed for precision bombing were employed by the United States

Regional Geology and Tectonics

Introduction

23

FIGURE 3.4 The top line shows a ship-track magnetic anomaly profile across the Pacific-Antarctic Ridge centred at 110W/45S and projected onto N90E. The bottom line is a synthetic magnetic anomaly profile arising from the series of previously known magnetic epochs indicated by the black (normally magnetized) and white (reversely magnetized) boxes beneath, generated from the timescale of Gradstein et al. (2012). Red and green dots labelled ‘y’ and ‘o’ mark the position of the young and old edges of the Gauss magnetic chron.

Coast and Geodetic Survey to conduct hydrographic surveys, and magnetometers developed to hunt submarines became a valuable method of acquiring information about the ocean floor. Research carried out during the war led to the development of many other tools and techniques for ocean exploration, including deep-ocean camera systems, side-scan sonar instruments and early technology for guiding remotely operated vehicles. Careful mapping of the ocean basins during the 1950s and early 1960s revealed the seafloor’s rough topography and the existence of mid-ocean ridges. Understanding their significance began with the realization that they are not isolated features, but are connected to one another forming part of a seismically active system encircling the entire planet (Ewing et al., 1956). This was not immediately understood in terms of seafloor spreading. Some researchers, including Tuzo Wilson, initially surmized that the Earth was expanding and therefore ‘cracking’ along this system of globally encircling mid-ocean ridges. Magnetic field measurements taken by oceanographic ships revealed the existence of stripes of alternating normal and reversed polarity in seafloor magnetization running parallel to mid-ocean ridges. These findings were used as confirmatory evidence for the theory of seafloor spreading by Dietz (1961) and Hess (1962). According to this theory, new crust is created at mid-ocean ridges, carried away by mantle convection cells (Griggs, 1939; Holmes, 1926; Vening Meinesz, 1933) and recycled into the interior of deep oceanic trenches. Publication of the Eltanin-19 magnetic anomaly profile across the South Pacific ridge greatly contributed to the broad acceptance of the theory of seafloor spreading (Pitman and Heirtzler, 1966) (Fig. 3.4). The nearly perfect symmetry either side of the mid-ocean ridge was noted by Walter Pitman, who was a student at Lamont Observatory at the time. Lamont scientists had been mapping and studying the seafloor for many years, since the establishment of the institution in 1949. The symmetry of this profile provided breakthrough evidence for the theory of seafloor spreading, illustrating the process of lithosphere formation along mid-ocean ridges and subsequent spreading outwards on both sides over time. By February 1966, seafloor spreading had been embraced by many at Lamont and elsewhere. Further evidence in favour of this hypothesis would come from rock coring campaigns such as that of the research vessel Glomar Challenger. During a year-long campaign starting in 1968 as part of the Deep-Sea Drilling Project (DSDP), rock samples between South America and Africa were collected and later dated by paleontological and isotopic methods, revealing how they become progressively older with distance from mid-ocean ridges. The first attempts at mapping the age of the world’s oceans date from the early 1980s and were done in sketch form (e.g. Sclater et al., 1981). With the progressive improvement in the quality and availability of magnetic anomaly and satellite altimetry data and increase in computer capabilities, seafloor age maps proliferated. Printed maps of magnetic anomaly isochrons drawn using published and unpublished magnetic and bathymetric information (e.g. Larson, 1985; Sclater et al., 1981) gave way to computer-generated products (e.g. Mu¨ller et al., 2016; Pe´rez-Dı´az and Eagles, 2017a).

Regional Geology and Tectonics

24

3. Age of the oceans

Dating the oceanic lithosphere Dated magnetic anomaly isochrons together with seafloor fabric features such as fracture zones, transform faults, extinct and active spreading centres help define the spreading history of ocean basins and form the basis for plate tectonic models. These, in turn, can be used to produce high-resolution gridded seafloor age maps, which can be further tested for reliability by using age determinations obtained from other sources such as drilled rock samples.

Magnetic anomaly data Earth’s magnetic field can be modelled to a good approximation as a simple dipole field, with an axis slightly tilted at approximately 10 degrees to the rotational axis of the planet. In the 1920s, studies of ancient lava flows had shown that some older igneous rocks had magnetic properties that could be explained by a reversely oriented magnetic field. This was reinforced by the study of cores taken from the ocean floor (Cox et al., 1963) where it was discovered that the orientation of mineral magnetization in the core varied with depth. Later, the striped patterns of alternating normal and reversed polarity found in the ocean floor were explained by Vine and Matthews (1963) as further evidence of a series of reversals of the Earth’s dipole field, recorded next to each other in strips of crust formed simultaneously all along the length of mid-ocean ridges (later confirmed by Cox et al., 1963; Dickson et al., 1968; Vine, 1966) (Fig. 3.5). At any point on Earth, the total magnetic field can be defined as the vector sum of the fields globally by convection in the iron-rich core and locally by induced and remanent rock magnetization. The induced component is dependent on, and aligned parallel to, the present-day geomagnetic field developed in the convecting outer core. Remanent magnetization may arise by various means, but for rocks formed by the cooling of magma near midocean ridges, its most important form is thermoremanent magnetization (TRM). TRM, by being acquired geologically instantaneously as basaltic magma cools to rock over the range 800 C 400 C, provides valuable information about the Earth’s magnetic field at an instant in the past. Its origin lies in the tendency of magnetic domains within minerals condensing out of a cooling magma to align themselves with the local magnetic field at the time they cool below the so-called Curie temperature. This magnetization becomes permanent when rocks cool to a lower temperature (sometimes referred to as blocking temperature), below which the orientation of the domains will not change unless the rock is reheated or subjected to further chemical or tectonic processes. Unlike the induced field, the remanent component does not need an external field to persist. TRM is not acquired by the whole of the oceanic crust, but layer 2A and the upper part of layer 2B (e.g. Fox and Opdyke, 1973). These layers (Fig. 3.6), formed essentially by pillow lavas, sheeted dykes and containing relatively high proportions of magnetic minerals, experience very rapid cooling in contact with sea water. As a result, they acquire a stable TRM. The interaction between the magnetic fields raised by normally and reversely magnetized blocks and the Earth’s present-day dipole field results in marine magnetic anomalies with predictable shapes. At mid- to high latitudes, strips of crust magnetized in the direction of the present-day magnetic field will give rise to positive anomalies. Conversely, strips magnetized in opposition to the present field will result in negative magnetic anomalies. These anomalies, running parallel to (and either side of) mid-ocean ridges, form the zebra-like pattern known as magnetic striping and when calibrated with numerical ages provide the basis for mapping the age of the seafloor. Marine magnetic anomaly profiles are generally measured by towing a total field magnetometer on the ocean surface behind a ship at sufficient distance so that the magnetic effects of the ship are below the noise level of the magnetometer. They can also be measured by flying an aircraft equipped with a suitable magnetometer at a known height above the ocean surface (Gubbins and Herrero-Bervera, 2007). Isochrons can be picked along these profiles by eye or by numerically modelling the locations of the magnetic contrasts. The orientations and spacings of magnetic isochrons either side of spreading centres provide qualitative information about spreading rates and azimuths, past and present, which can be further quantified if ages are assigned to isochrons on the basis of geomagnetic timescales. Geomagnetic timescales Identifications of particular patterns of magnetic anomaly lineations, to which ages can be assigned using geomagnetic timescales, can be used to divide the seafloor into age provinces (i.e. Cande et al., 1989; Scotese et al., 1988).

Regional Geology and Tectonics

FIGURE 3.5 Magnetic anomaly map of the total magnetic field out off the western coast of North America, revealing a linear striped pattern. Coloured bands indicate normal polarity. Source: Redrawn from Raff, A., Mason, R., 1961. Magnetic survey off the west coast of North America, 40 N latitude to 52 N latitude. Geol. Soc. Of. Am. Bull. 72, 1267 1270. https://doi.org/ 10.1130/0016-7606(1961)72. Reproduced with permission of the Geological Society of America. Permission obtained via Rightslink/CCC.

FIGURE 3.6 Schematic ‘Penrose-type’ oceanic lithosphere model, proposed in the 1970s based on observations made from ophiolite studies and geophysical imaging of the seafloor. The uppermost layer is not always present, and it consists of semiconsolidated or unconsolidated sediments deposited on the seafloor. Layer 2, divided into two parts, consists of extrusive lava flows and sheeted dykes. The interaction of this layer’s thermoremanent magnetization with the Earth’s present-day magnetic field gives rise to marine magnetic anomalies. A sharp contrast in seismic velocities marks the transition between layer 2 (slow velocities, steep velocity gradients) and layer 3 (fast velocities, low gradients), composed of lower crust massive and layered gabbros and peridotites.

26

3. Age of the oceans

Early geomagnetic timescales, such as that put forward by Cox et al. (1963), were largely based on K/Ar age determinations from young lava flows and measurements of magnetic polarity from widely spaced land locations worldwide. Cox et al. (1963) broke up the polarity sequence into ‘epochs’ (now also referred to as ‘chrons’) of dominantly normal or reversed polarity and also recognized the existence of shorter intervals or ‘events’ (now ‘sub-chrons’). Epochs were named after geoscientists that had made a significant contribution to geomagnetism (Brunhes, Matuyama, Gauss and Gilbert). Because volcanic activity is intermittent and dating errors with K/Ar become too large to delineate polarity reversals beyond 5 Ma, other methods of age determination were necessary to extend timescales past that time. The first long-term geomagnetic timescale extended back in time to nearly 80 Ma. Heirtzler et al. (1968), using the radiometric dates obtained from the younger reversals seen on land as a starting point, matched a series of synthetic profiles computed under the assumption of constant seafloor spreading with the observed anomaly pattern in the South Atlantic. This timescale would be later confirmed by independent seafloor age determinations obtained from continuously deposited sediments sampled in oceanic drilling campaigns as part of the DSDP, which started in 1968. Prominent anomalies were assigned to periods of normal polarity (referred to as magnetic isochrons or simply chrons), which were numbered from 1 to 32 with increasing time. By the early 1970s, marine magnetic anomalies had been assigned to a sequence of chrons dating back to 83 Ma (Chron 34). These anomalies terminate at vast regions of normal polarity seafloor, referred to as the Cretaceous quiet zone (CQZ), which is interpreted as the signal of a long-lived Cretaceous Normal Superchron (CNS). The discovery of older magnetic anomalies beyond the old edge of the CQZ (B121 Ma) in the western Pacific (Larson and Pitman, 1972) allowed the Heirtzler et al. (1968) timescale to be extended back to 160 Ma (M0 M29 chrons, ‘M’ for Mesozoic). Several geomagnetic polarity timescales have subsequently been published, using Heirtzler et al. (1968) as a starting point and refining it by drawing on new magnetostratigraphic studies of sedimentary sequences on land and at sea, radiometric dating of layer 2 basalts and layer 1 sediments returned by the DSDP and Oceanic Drilling Program Campaigns (1968 1983), and correlations of biostratigraphic and magnetostratigraphic information (Berggren et al., 1995; Cande and Kent, 1992; Cande and Kent, 1995; Channell et al., 1995; Gradstein et al., 2004; Gradstein et al., 2012; Harland et al., 1989, 1982; Malinverno et al., 2012). Despite the frequent revisions to the geomagnetic polarity timescale, the inferred pattern and timing of geomagnetic reversals have changed relatively little since the pioneering work of Heirtzler et al. (1968). Two of the most widely used modern timescales are those of Gradstein et al. (2004, 2012). These timescales result from the review of GTS1989 (Harland et al., 1989), which was preceded by GTS1982 (Harland et al., 1982) on the light of advances and improvements in data coverage and timescale research methods developed since 1989 (described in detail by Gradstein et al., 2004). Within the framework of the International Stratigraphical Chart, construction of these geomagnetic timescales was done using a number of methods dependent upon the quality and availability of data for different intervals. For the Cenozoic and late Mesozoic, these included studies of radiogenic and stable isotopes, studies of seafloor spreading records and orbital tuning based on astronomical cycles. The development of geomagnetic timescales (Gee and Kent, 2007) provides the necessary information to visualize a plate’s relative motions in time, based on the identification of specific anomalies within that plate. Once the ages of magnetic isochrons are assigned to specific magnetic anomalies, a map showing the positions of those anomalies is equivalent to a chronological map of the ocean basins’ formation (Fig. 3.7). Limitations and uncertainty in magnetic anomaly data The minimum uncertainty one might expect on seafloor ages derived from magnetic anomaly data is the greater of (1) that related to the uncertainty in location of the magnetic anomaly pick that is given an isochron age (navigational and process-related uncertainty in location) or (2) the uncertainty in age assigned to the isochron from a chosen timescale. Shipborne and airborne magnetic anomaly data have been collected for more than half a century, providing global coverage of variable density and quality. Due to the changing main field from the Earth’s core, differences in quality and coverage and lack of magnetic data in oceanic regions in the vicinity of the equator and within the CNS, combining these data to a consistent global magnetic grid is challenging. Efforts to collect and compile aeromagnetic and marine data worldwide have led to the development of the World Digital Magnetic Anomaly Map (Lesur et al., 2016) and NOAA’s Earth Magnetic Anomaly Grid (EMAG and EMAG2, Meyer et al., 2017), who provide open-access downloadable datasets of gridded magnetic measurements across the globe.

Regional Geology and Tectonics

Dating the oceanic lithosphere

27

FIGURE 3.7 Ages of magnetic anomaly picks (Seton et al., 2014) based on the timescale of Gee and Kent (2007). Plate boundaries are those of Coffin et al. (1998) and are symbolized by double black lines (constructive), thick black lines (destructive) or white dashed lines (conservative). This figure was produced using the Generic Mapping Tools (GMT, Wessel and Luis, 2017). Robinson projection.

Navigational inaccuracies in the collection of magnetic data are related to the date at which data were collected, with undoubtedly greater uncertainties existing in datasets acquired before the use of GPS became standard (pre-1985). However, taken on the whole, uncertainties in navigational techniques tend to be smaller than those one might expect to arise from geological processes (Eagles, 2004). One example is errors arising from the crustal accretion process itself (such as pillow basalt lava flows spreading laterally across the floor of a median valley), which can be up to 30 km wide (Parmentier and Forsyth, 1985; Smith et al., 1999), rather than forming a thin ribbon along its axis. Navigational and process-related uncertainties are likely to dominate in areas where seafloor age changes significantly over short distances (e.g. across fracture zone traces), or where interpretations of magnetic anomaly isochrons are made from particularly old magnetic data. The process of magnetic reversal is thought to complete over the course of a few thousand years and can be considered geologically as instantaneous. Ages are assigned to these instants on the basis of radiometric ages determined from volcanic rocks in continuous sequence displaying reversals and are usually quoted with numerical uncertainties. Estimated errors range between 6 0.5 and 6 5 My for times between Late Jurassic and Mid-Cretaceous and decrease to a few hundreds of thousands of years for younger ages. Because of this, these uncertainties are likely to dominate towards the continental margins, where the ocean floor is of mid-Cretaceous ages. Further errors may arise from the picking procedure itself. Most often, either the ‘old’ or ‘young’ ends of magnetic chrons are picked by researchers (Fig. 3.4), and ages can be assigned to them directly. In other cases, the ‘centre’ or ‘middle’ may be chosen instead, for example where chron edges are uncertain, and an age needs to be interpolated by taking into account the distance of the chosen pick and the young/old dated ends. Researchers’ choice of picking location is normally indicated after the chron name, by appending ‘y’, ‘o’ or ‘c’. The absence of this information can lead to large errors.

Drill-core data Advances in piston coring in the 1940s and advances in dynamic positioning in the 1960s allowed research ships to begin studying Earth’s subseafloor geology by recovering sediment and rock sections from drill holes. Drilling and coring campaigns across all of the world’s major oceans began in the 1960s and continue to be a key part of marine exploration today, collecting sediment, rock, fluids and faunal samples from the seafloor and below. Ocean drilling over the past 50 years has been done as part of the DSDP (1966 1983), Ocean Drilling Program (ODP, 1983 2003), Integrated Ocean Drilling Program (IODP, 2003 2013) and, most recently, the International Ocean Discovery Program (IODP, 2013 present). The locations of all holes drilled during these campaigns are shown in Fig. 3.8. In these localities, age information can be obtained by direct dating of igneous basement rocks or extrapolations of dates determined by analysis of the sediments above them.

Regional Geology and Tectonics

28

3. Age of the oceans

FIGURE 3.8 World map showing the location of drill-core sites from DSDP, ODP and IODP campaigns between 1969 and 2017. This figure was produced using the Generic Mapping Tools (GMT, Wessel and Luis, 2017). Mercator projection.

Limitations and uncertainty in dated samples Ages obtained from dated oceanic rock and sediment samples collected as part of drilling campaigns are of limited use to map seafloor age over large areas. Their sparse distribution makes them most suitable to test the reliability of a given age grid or the validity of thermal models (e.g. Johnson and Carlson, 1992) and not as the sole basis on which to map seafloor age. The most reliable source of seafloor age in recovered deep-sea cores is that given by radiometric dating of unaltered igneous basement. This technique is based on the premise that radioactive elements such as 238U (Uranium) or 40K (potassium) within rocks decay at a predictable rate. By measuring the quantity of unstable atoms left in a rock and comparing it to the quality of stable daughter atoms in the rock, the time that has passed since that rock formed can be estimated (Boltwood, 1907; Holmes, 1911; Nier, 1939; Parrish, 1990; Parrish and Krough, 1987). The probable age of the sample should be taken into consideration when choosing a dating method. Ideally, the decay scheme used should have a half-life in the same order of magnitude as the expected age of the sample. Another important factor is the expected concentrations of parent and daughter elements in the rock, and whether the rock was a closed system (it has not gained or lost the parent nuclide as a result of processes other than radioactive decay during the time since its formation). For most rocks, this is a false assumption. Igneous rocks formed by the solidification of magma along a mid-ocean ridge are no exception. Hydrothermal water transports radioactively derived lead, which is deposited together with uranium as these rocks cool. In these cases, U Pb dating would result in an overestimation of age due to the extra amount of daughter lead present (Chelle-Michou and Schaltegger, 2018). Another widely used isotope system is K Ar, because potassium is a widespread element that can be found in most rocks. Because argon is not preserved in minerals until they cool below their closure temperatures (Dodson, 1973), K Ar dating of a rock does not always reflect the rock’s age of formation, but the age at which the rock last cooled below the closure temperature. When attempting to date oceanic lithosphere, the choice and application of radiometric techniques to rock samples require knowledge of a number of factors concerning not just the methods but also the samples. Additional errors, several orders of magnitude greater than those related to the dating method used, can be introduced where pre- or postbreakup volcanic additions are erroneously attributed to true seafloor basement.

Regional Geology and Tectonics

Dating the oceanic lithosphere

29

True oceanic basement has only been reached in a handful of sites from those cored by oceanic drilling campaigns over the past 60 years. Where drilling is not able to penetrate deep enough to reach basement, the oldest sediments cored are often used to estimate the age of the seafloor below them. Doing this relies on the correct identification and dating of fossil assemblages and requires making assumptions about the depth below the oldest sediment at which basement might be located, as well as the rates of sediment deposition. In these instances, extrapolations of basement ages made over great distances are unlikely to draw confident conclusions.

Age proxies The age of the oceanic lithosphere can also be extrapolated, albeit more loosely, from the present-day depth of the seafloor, the ages of large igneous provinces (LIPs) and aseismic ridges and features within the extended continental margins either side of an ocean. The numerical ages delivered using these proxies are best regarded in a qualitative sense, making them perhaps most suitable for understanding relative ages and so to test the reliability of an existing age grid but they rarely form a useful basis on which to build one. Present-day oceanic lithosphere One of the most frequently revisited problems in geodynamics regards the thermal evolution of oceanic lithosphere through time. The basic principle is simple: the oceanic lithosphere forms by conductive cooling of the upper mantle, which is exposed by the divergence of plates at mid-ocean ridges. Consequently, oceanic lithosphere is hot when it is created at mid-ocean ridges and cools down as it moves away from them (e.g. Menard and Smith, 1966; Vogt and Ostenso, 1967). As it cools, it becomes denser and sinks into the underlying asthenosphere in order to maintain isostatic equilibrium (Pratt, 1859). Although horizontal conduction, hydrothermal convection and advection also contribute to the loss of heat in oceanic lithosphere, they are generally neglected in calculations of its thermal evolution. This is due to the expectation or assumption that, with the exception of areas in the vicinity of mid-ocean ridges, their magnitude is much smaller than that of vertical heat conduction. Thermal models describing the way in which the oceanic lithosphere cools, thickens, and subsides as it spreads away from mid-ocean ridges can be used, in principle, to estimate the age of oceanic lithosphere from its present-day depth, heat flow and thickness. Observations of the decrease in heat flow and increase in depth with seafloor age have prompted two main groups of models. In ‘Half-space’ cooling models, the lithosphere behaves as the cold upper boundary layer of a cooling half-space (Davis and Lister, 1974; Parker and Oldenburg, 1973; Turcotte and Oxburgh, 1967). Conversely, ‘Plate-cooling’ models treat oceanic lithosphere as a cooling plate with an isothermal lower boundary (Crough, 1975; Doin and Fleitout, 2002; Langseth et al., 1966; McKenzie, 1967; Stein and Stein, 1992). Because plate-cooling models predict the observed seafloor flattening in old ocean basins, they have generally been preferred over half-space models for describing thermal subsidence. A comparison of age-depth and age-heat flow curves as modelled by both half-space and plate cooling is shown in Fig. 3.9, illustrating the significant differences between thermal models. As well as the choice of thermal model, estimates of age based on present-day depth and heat flow are complicated by the fact that thermal subsidence is not the only process controlling the evolution of seafloor depth through time. Processes such as sedimentation and postbreakup volcanism, variations in crustal thickness and dynamic topography need to be accounted for. Furthermore, determinations of oceanic age from observed seafloor depth done in this way involve a certain degree of circularity due to the fact that depth data are often used as an input to define agedepth curves in the first place. Volcanic additions and extended continental margin features LIPs, aseismic ridges and volcanic seamount chains emplaced on oceanic lithosphere constrain the minimum age of the seafloor beneath them (Fig. 3.10). Similarly, LIPs found on land and interpreted as immediate precursors of break-up can be used as proxies to estimate the age of the oldest oceanic lithosphere in a particular basin. Although of little value for mapping the age of oceans due to their scattered distribution and potentially large interpretational errors, they may be used as independent datasets to test existing seafloor age maps. Other features, such as continent-ocean boundaries or so-called ‘breakup unconformities’ found within extended continental margins may be used as a proxy for the age of break-up and formation of the first oceanic crust. These features, as well as age information derived from them, can, in theory, provide constraints on the extent of oceanic lithosphere and time of break-up. In practice, interpretations of so-called break-up markers are subject to large observational, interpretational and dating uncertainties (Fig. 3.11) (Eagles et al., 2015).

Regional Geology and Tectonics

30

3. Age of the oceans

FIGURE 3.9 Graphic comparison of six model curves describing the relationship between age and depth of oceanic lithosphere (XBY: Crosby et al., 2006; HS: Davis and Lister, 1974; CH: Doin and Fleitout, 2002; HW: Hillier and Watts, 2005; PSM: Parsons and Sclater, 1977; GDH1: Stein and Stein, 1992).

FIGURE 3.10 Radiometric ages of rock samples along the Walvis Ridge displayed over satellite-derived free-air gravity dataset (Sandwell et al., 2014). Data are obtained from O’Connor and Duncan (1990). This figure was produced using the Generic Mapping Tools (GMT, Wessel and Luis, 2017). Mercator projection.

Mapping seafloor age Analogue maps of the age of the world’s oceans date from the early 1980s and were compiled from magnetic anomaly isochron data (Larson, 1985; Pitman, 1971; Sclater et al., 1981). With the improvement in magnetic anomaly data quality and increase in quantity, the availability of more sophisticated kinematic models and increases in computing power and software capabilities, printed maps of magnetic anomaly isochrons

Regional Geology and Tectonics

Mapping seafloor age

31 FIGURE 3.11 Alternative interpretations of the location of the continent-ocean boundary along the extended continental margins of South America and Africa as compiled by Eagles et al. (2015), displayed over gravity anomaly data (Sandwell et al., 2014). This figure was produced using the Generic Mapping Tools (GMT, Wessel and Luis, 2017). Oblique Mercator projection.

FIGURE 3.12 Gridded global map of seafloor age grid (Mu¨ller et al., 2008). This figure was produced using the Generic Mapping Tools (GMT, Wessel and Luis, 2017). Plate boundaries are those of Coffin et al. (1998) and are symbolized by double black lines (constructive), thick black lines (destructive) or white dashed lines (conservative). Robinson projection.

have given way to computer-generated products accompanied by calculations of the uncertainties in them (Mu¨ller et al., 1997; Mu¨ller et al., 2008; Mu¨ ller et al., 2016; Pe´rez-Dı´az and Eagles, 2017a). A modern global gridded map of seafloor age is shown in Fig. 3.12. Remnants of ancient oceans (older than B200 million years) are mostly limited to on-land ophiolite assemblages (e.g. Aitchison et al., 1994; Skrzypek et al., 2012; Arenas et al., 2018). Although disputed (Cowie and Kusznir, 2012; de Voogd et al., 1992; Domeier and Torsvik, 2014; Garfunkel, 1998; Stampfli and Borel, 2002), the eastern Mediterranean may be an exception. Granot (2016) recently identified oceanic lithosphere as old as 340 Ma and interpreted it as belonging to the remnants of the Neotethys ocean, which formed as the supercontinent Pangea was breaking apart and is currently in the process of disappearing between Africa and Europe. Although anomalously old for oceanic lithosphere, this maximum age contrasts with that of the oldest rocks found in continents, which are up to 3.6 Giga years old. Larger domains of old oceanic lithosphere can be found in the North Atlantic (along the coasts of North America and the western part of Africa) as well as in the northwestern the Pacific and along the eastern coast of Africa.

Regional Geology and Tectonics

32

3. Age of the oceans

Gridding methods Constructing complete and accurate age grids for the world’s oceans that describe the spreading process in detail requires the identification and integration of magnetic anomaly, gravity anomaly and satellite altimetry data and plate kinematic model predictions, as well as knowledge of present-day plate boundary geometries. Creating a smooth grid of seafloor ages requires a number of steps, the first of which is to compile a collection of densely interpolated isochrons covering the area of interest and based on the locations and ages of available magnetic anomaly data (Fig. 3.13A). Interpolation between magnetic isochron picks requires a number of assumptions about spreading rate and spreading direction and should be informed by seafloor gravity fabric and the need for isochrons to unite at a single paleo-ridge crest. Additionally, a set of fracture zone traces and other tectonic boundaries (mid-ocean ridges, extinct ridge crests and triple junction traces, Fig. 3.13B) digitized from satellite altimetry data are required to inform the gridding procedure about the location of lines across which the age function can be thought of as discontinuous. Fracture zone traces may be picked and interpolated along strike by hand or automatically (Wessel et al., 2015). Picking differences between these two approaches are comparable to the absolute reliability in the location of a fracture zone axis picked from satellite alternative gravity data (Mu¨ller et al., 1991) and comparably less than the total width of a typical fracture zone-related gravity anomaly. The choice of fracture zone mapping technique is therefore unlikely to be of significance for confident interpretation of the grid. Seafloor age can then be interpolated using the modeller’s choice of gridding algorithm (Lam, 1983; Maslyn, 1987) (Fig. 3.14A). Recent global and regional age grids use minimum curvature rules implemented through contouring and surface modelling software packages. Minimum curvature methods generate smooth surfaces that pass through or are close to the input data points in such a way that the amount of bending on the surface is reduced to a minimum. The fracture zone and tectonic barrier dataset can be used to impose explicit breaks to the interpolation. In most cases, no interpolation should be calculated for portions of the grid with insufficient data coverage. In areas of low density of age data points, the results of minimum curvature algorithm approximate a linear interpolation.

Uncertainty quantification At any given age grid cell, the minimum uncertainty that should be expected depends on (1) the errors related to the input datasets, (2) the proximity of the cell to a point of constrained age and (3) the proximity of the cell to a point of potential discontinuity-related error (fracture zone traces and other tectonic boundaries). Age grid accuracy will be best for cells coincident with magnetic anomaly identifications, albeit susceptible to errors related to navigation limitations or wrongly identified magnetic anomalies. For age grid cells not coincident with magnetic or plate kinematic model-derived data points, uncertainty will increase with interpolation distance from the nearest of such points. The largest of these uncertainties exist for regions with sparse data coverage (such as parts of the Southern Ocean) and regions for which there are no magnetic isochrons. These include seafloor that formed during periods of time without changes in the polarity of the Earth’s magnetic field (such as the CNS) and within areas of east west spreading at low latitudes (such as the equatorial Atlantic) (Fig. 3.7). Uncertainty in age estimates is also greater around the highly fractured crust in the vicinity of fracture zone traces because of the strong lateral contrast in seafloor age from one side of the fracture zone to the other. Wrongly located magnetic anomalies near fracture zone, be this the result of navigational errors or poorly digitized fracture zone traces, will give rise to errors. Total age uncertainty can be calculated by taking into account, for each grid cell, its proximity to a point of constrained age, the age discrepancy (if any) between magnetic anomaly picks, kinematic model isochrons and age grid predictions, and the distance to fracture zone traces and other age discontinuities (Pe´rez-Dı´az and Eagles, 2017a). Figs 3.15 3.18 show seafloor age according to Mu¨ller et al. (2008). For these grids, age was mapped on the basis of global magnetic isochron data and associated plate reconstruction rotation schemes, as well as geological and geophysical data along the extended continental margins. Uncertainty in these grids is dominated by the choice of underlying plate kinematic models and their limitations. At times, potential age grid inaccuracies can be identified by eye where age contours follow unnatural trends. An example of this is the smooth contours at 100E 30S in the eastern part of the Indian Ocean. However, in most cases, uncertainty only becomes apparent when age grids produced from alternative datasets or differing interpretations of data are compared side by side.

Regional Geology and Tectonics

33

Mapping seafloor age

FIGURE 3.13 (A) Magnetic anomaly isochron picks (coloured squares) and ‘kinematic’ isochron picks (black circles) as interpolated by Pe´rez-Diaz and Eagles (2014). (B) Fracture zone traces (yellow dashed lines), active and extinct mid-ocean ridges (black and blue double lines, respectively) and other tectonic boundaries (blue solid lines) (Pe´rez-Diaz and Eagles, 2014) displayed over satellite-derived gravity anomaly data (Sandwell et al., 2014). This figure was produced using the Generic Mapping Tools (GMT, Wessel and Luis, 2017). Oblique Mercator projection.

In the South Atlantic, stark differences can be seen in the regions of the Vema Channel (20S) and around the fossil boundaries of the Malvinas Plate between the age grids of Mu¨ller et al. (2008) and Pe´rez-Dı´az and Eagles (2017a) (Figs 3.16A and 3.17 respectively). They are the result of ridge jumps in these two regions interpreted only by Pe´rez-Diaz and Eagles (2014). Additional differences exist in the central part of the Atlantic, where these two age grids differ by up to 18 million years. In order to assess age grid fidelity, their predicted ages can be compared to age determinations at point locations corresponding to magnetic anomaly picks and drill-core samples. Doing this for the two South Atlantic grids discussed above favours that of Pe´rez-Dı´az and Eagles (2017a).

Regional Geology and Tectonics

34

3. Age of the oceans

FIGURE 3.14 (A) Seafloor age gridded on the basis of the datasets in Fig. 3.13, using minimum curvature rules (Pe´rez-Dı´az and Eagles, 2017a). (B) Spreading rates (in mm/year) derived from gradients in the age grid in (A) (Pe´rez-Dı´az and Eagles, 2017a). This figure was produced using the Generic Mapping Tools (GMT, Wessel and Luis, 2017). Oblique Mercator projection.

Final remarks Constructing well-constrained digital age grids requires the integration of magnetic anomaly isochron data and seafloor fabric features and benefits from a good understanding of past plate motions and the location of present and past plate boundaries. Global age grids such as that shown in Fig. 3.12 provide a wealth of information about the evolution of ocean basins, including spreading rates, directions and asymmetry, as well as the distribution of oceanic age. However, in order to draw confident conclusions regarding these aspects of oceanic evolution, the uncertainties in gridded age estimates must be carefully considered. These are the combined result

Regional Geology and Tectonics

Final remarks

35

FIGURE 3.15 Seafloor age map of the Pacific Ocean (Mu¨ller et al. 2008). This figure was produced using the Generic Mapping Tools (GMT, Wessel and Luis, 2017). Plate boundaries are those of Coffin et al. (1998) and are symbolized by double black lines (constructive), thick black lines (destructive) or white dashed lines (conservative). Robinson projection.

FIGURE 3.16 Seafloor age map of the Indian Ocean (Mu¨ller et al. 2008). This figure was produced using the Generic Mapping Tools (GMT, Wessel and Luis, 2017). Plate boundaries are those of Coffin et al. (1998) and are symbolized by double black lines (constructive), thick black lines (destructive) or white dashed lines (conservative). Robinson projection. The colour bar is the same as in Fig. 3.15.

Regional Geology and Tectonics

FIGURE 3.17 Seafloor age map of the Atlantic Ocean (Mu¨ller et al. 2008). This figure was produced using the Generic Mapping Tools (GMT, Wessel and Luis, 2017). Plate boundaries are those of Coffin et al. (1998) and are symbolized by double black lines (constructive), thick black lines (destructive) or white dashed lines (conservative). Oblique Mercator projection. The colour bar is the same as in Fig. 3.15.

FIGURE 3.18 Seafloor age map of the Southern Oceans (Mu¨ller et al. 2008). This figure was produced using the Generic Mapping Tools (GMT, Wessel and Luis, 2017). Plate boundaries are those of Coffin et al. (1998) and are symbolized by double black lines (constructive), thick black lines (destructive) or white dashed lines (conservative). Stereographic projection. The colour bar is the same as in Fig. 3.15.

FIGURE 3.19 For 40 Ma, (A) paleoage grid, (B) seafloor depths as predicted by plate-cooling model GDH1 (Stein and Stein, 1992) and (C) modelled paleobathymetry (Pe´rez-Dı´az and Eagles, 2018). Plate positions at 40 Ma are those of Pe´rez-Diaz and Eagles (2014). This figure was produced using the Generic Mapping Tools (GMT, Wessel and Luis, 2017). Oblique Mercator projection.

38

3. Age of the oceans

of uncertainties inherent to the input datasets (both observational and interpretational) and those associated with the gridding method itself. Grids of basement age with quantified uncertainties are an extremely powerful tool for modern geodynamic and paleoceanographic studies. The age of the ocean floor plays an essential role in studies of plate kinematics, plate driving forces and paleobathymetry, amongst others. One example is the calculation of seafloor spreading azimuths and rates from the two components of gradient vectors in an age grid (Fig. 3.14B). At the first order, the age of oceanic lithosphere also provides the keys to modelling the evolution of seafloor depth through time (Pe´rez-Dı´az and Eagles, 2017b, 2018) (Fig. 3.19B and C) and, in turn, carry out studies of paleocirculation, water mass formation and paleoclimate. Additionally, knowledge of the age of oceans can provide constraints for seismic tomography and mantle convection models (Mu¨ller et al., 2019).

References Aitchison, J., Blake Jr., M., Flood, P., Jayko, A., 1994. Paleozoic ophiolitic assemblages within the southern New England Orogen of eastern Australia: implications for growth of the Gondwana margin. Tectonics 13 (5), 1135 1149. Arenas, R., Ferna´ndez-sua´rez, J., Montero, P., Dı´ez, R., Andonaegui, P., Sa´nchez, S., et al., 2018. The Calzadilla Ophiolite (SW Iberia) and the Ediacaran fore-arc evolution of the African margin of Gondwana. Gondwana Res. 58, 71 86. Available from: https://doi.org/10.1016/j. gr.2018.01.015. Bacon, F., 1620. Novum Organum Scientiarum (textbook). Berggren, W.A., Kent, D.V., Swisher, C.C., Aubry, M.-P., 1995. A revised Cenozoic geochronology and chronostratigraphy, Geochronology Time Scales and Global Stratigraphic Correlation, vol. 54. SEPM Special Publication. Boltwood, B., 1907. On the ultimate disintegration products of the radioactive elements. Philos. Mag. ix, 613. Cande, S.C., Kent, D.V., 1992. A new geomagnetic polarity time scale for the late Cretaceous and Cenozoic. J. Geophys. Res. 97 (B10). Cande, S., Kent, D., 1995. Revised calibration of the geomagnetic polarity time scale for the Late Cretaceous and Cenozoic. J. Geophys. Res. 100, 6093 6095. Cande, S.C., La Brecque, J., Larson, R., Pitman III, W.C., Golovchenko, X., 1989. Magnetic lineations of the world’s ocean basins. Am. Assoc. Pet. Geol. 13. Channell, J., Erba, E., Nakanishi, M., Tamaki, K., 1995, January 1. Late jurassic-early cretaceous time scales and oceanic magnetic anomaly block models. In: Berggren, W.A., Kent, D.V., Aubry, M.-P., Hardenbol, J. (Eds.), Geochronology, Time Scales and Global Stratigraphic Correlation, vol. 54, p. 0. ,https://doi.org/10.2110/pec.95.04.0051.. Chelle-Michou, C., Schaltegger, U., 2018. U Pb dating of mineral deposits: from age constraints to oreforming processes. In: Huston, D., et al. (Ed.), Isotopes in Economic Geology, Metallogenesis and Exploration, SGA Special Volume. ,https://doi.org/10.31223/osf.io/b7pw9.. Coffin, M.F., Gahagan, L.M., Lawver, L.A., 1998. Present-day plate boundary digital data compilation. University of Texas Institute of. Geophysics Technical Report 174. Cowie, L., Kusznir, N., 2012. Gravity inversion mapping of crustal thickness and lithosphere thinning for the eastern Mediterranean. Lead Edge 31 (7), 810 814. Cox, A., Doell, R., Dalrymple, G.B., 1963. Geomagnetic polarity epochs and Pleistocene geochronometry. Nature (4885), 1049 1051. Crosby, A.G., McKenzie, D., Sclater, J.G., 2006. The relationship between depth, age and gravity in the oceans. Geophys. J. Int. 166 (2), 553 573. Available from: https://doi.org/10.1111/j.1365-246X.2006.03015.x. Crough, S., 1975. Thermal model of oceanic lithosphere. Nature 256, 388 390. Davis, E.E., Lister, C.R.B., 1974. Fundamentals of ridge crest topography. Earth Planet. Sci. Lett. 21 (4), 405 413. Available from: https://doi. org/10.1016/0012-821X(74)90180-0. de Voogd, B., Truffert, C., Chamot-Rooke, N., Huchon, P., Lallemant, S., Le Pichon, X., 1992. Two-ship deep seismic soundings in the basins of the Eastern Mediterranean Sea (Pasiphae Cruise). Geophys. J. Int. 109, 536 552. Dickson, G.O., Pitman III, W.C., Heirtzler, J.R., 1968. Magnetic anomalies in the South Atlantic and ocean floor spreading. J. Geophys. Res. 73 (6). Available from: https://doi.org/10.1029/JB073i006p02087. Dietz, R.S., 1961. Continent and ocean basin evolution by spreading of the sea floor. Nature 190, 854 857. Available from: https://doi.org/ 10.1038/190854a0. Dodson, M., 1973. Closure temperature in cooling geochronological and petrological systems. Contr. Miner. Pet. 40, 259 274. Doin, M.P., Fleitout, L., 2002. Thermal evolution of the oceanic lithosphere: an alternative view. Earth Planet. Sci. Lett. 142 (1 2), 121 136. Available from: https://doi.org/10.1016/0012-821x(96)00082-9. Domeier, M., Torsvik, T.H., 2014. Plate tectonics in the late Paleozoic. Geosci. Front. 5 (3), 303 350. Available from: https://doi.org/10.1016/j. gsf.2014.01.002. Eagles, G., 2004. Tectonic evolution of the Antarctic-Phoenix plate system since 15 Ma. Earth Planet. Sci. Lett. 217 (1 2), 97 109. Available from: https://doi.org/10.1016/S0012-821X(03)00584-3. Eagles, G., Pe´rez-Dı´az, L., Scarselli, N., 2015. Getting over continent ocean boundaries. Earth-Sci. Rev. 151. Available from: https://doi.org/ 10.1016/j.earscirev.2015.10.009. Ewing, M., et al., 1956. Some problems of Antarctic submarine geology. Geophys. Monogr. Ser. 1, 75 81. Fox, P.J., Opdyke, N.D., 1973. Geology of the oceanic crust: magnetic properties of oceanic rocks. J. Geophys. Res. 78 (23). Garfunkel, Z., 1998. Constraints on the origin and history of the eastern Mediterranean basin. Tectonophysics 298, 5 37. Gee, S., Kent, D., 2007. Source of oceanic magnetic anomalies and the geomagnetic polarity time scale. In: Kono, M. (Ed.), Geomagnetism. Elsevier. Gradstein, F., Ogg, J., Smith, A. (Eds.), 2004. A Geologic Time Scale. ,https://doi.org/10.1017/CBO9780511536045..

Regional Geology and Tectonics

References

39

Gradstein, F.M., Ogg, J., Schmitz, M., Ogg, G., 2012. The geologic time scale. In: Gradstein F.M., Ogg, J.G., Schmitz M.D., Ogg, G. (Eds.), The Geologic Time Scale 2012. ,https://doi.org/10.1016/B978-0-444-59425-9.02001-1.. Granot, R., 2016. Palaeozoic oceanic crust preserved beneath the eastern Mediterranean. Nat. Geosci. 9, 701 705. Griggs, D., 1939. A theory of mountain building. Am. J. Sci. 237. Available from: https://doi.org/10.1111/j.1365-246X.1961.tb00435.x. Gubbins, D., Herrero-Bervera, E., 2007. Encyclopedia of Geomagnetism and Paleomagnetism. Springer. Harland, W., Cox, A., Llewellyn, P., Pickton, C., Smith, A., Walters, R., 1982. A Geologic Time Scale. Cambridge University Press, Cambridge. Harland, W., Armstrong, R., Cox, A., Craig, L., Smith, A., Smith, D., 1989. A Geologic Time Scale. Cambridge University Press, Cambridge. Heirtzler, J.R., Dickson, G.O., Herron, E.M., Pitman, W.C., Le Pichon, X., 1968. Marine magnetic anomalies, geomagnetic field reversals, and motions of the ocean floor and continents. J. Geophys. Res. 73 (6), 2119 2136. Available from: https://doi.org/10.1029/jb073i006p02119. Hess, H., 1962. History of ocean basins. In: Petrologic Studies: A Volume to Honor A. F. Buddington. Princeton University Press, Princeton, NJ, pp. 599 620. Hillier, J.K., Watts, A.B., 2005. Relationship between depth and age in the North Pacific Ocean. J. Geophys. Res. Solid Earth 110 (2), 1 22. Available from: https://doi.org/10.1029/2004JB003406. Holmes, A., 1911. The association of lead with uranium in rock minerals, and its application to the measurement of geological time. Proc. R. Soc. A Math. Phys. Eng. Sci. 85, 248. Holmes, A., 1926. Contributions to the theory of magmatic cycles. Geol. Mag. LXII, 306 329. Isacks, B., Oliver, J., 1968. Seismology and the new global tectonics. J. Geophys. Res. 71 (18). Johnson, H.P., Carlson, R.L., 1992. Variation of sea floor depth with age: A test of models based on drilling results. Geophysic. Res. Lett 19 (19), 1971 1974. Lam, N., 1983. Spatial interpolation methods: a review. Am. Cartogr. 10 (2), 129 150. Langseth Jr, M., Le Pichon, X., Ewing, M., 1966. Crustal structure of the mid-ocean ridges: 5. Heat flow through the Atlantic Ocean floor and convection currents. J. Geophys. Res. 71 (22), 5321 5355. Larson, R.L., 1985. Bedrock Geology of the World. W.H. Freeman, New York. Larson, R., Pitman III, W.C., 1972. World-wide correlation of mesozoic magnetic anomalies and its implications. Geol. Soc. Am. Bull. 83 (12), 3645 3662. Available from: https://doi.org/10.1130/0016-7606(1975)86 , 270:wcomma . 2.0.co;2. Lesur, V., Hamoudi, M., Dyment, J., The´baut, E., 2016. Building the second version of the World Digital Magnetic Anomaly Map (WDMAM). Earth Planets Space 68 (27). Available from: https://doi.org/10.1186/s40623-016-0404-6. Lilienthal, T. C., 1756. Die gute Sache der go¨ttlichen Offenbarung, 8 vols, Ko¨nigsberg, Hartung. Malinverno, A., Hildebrandt, J., Tominaga, M., Channell, J.E.T., 2012. M-sequence geomagnetic polarity time scale (MHTC12) that steadies global spreading rates and incorporates astrochronology constraints. J. Geophys. Res. Solid Earth 117 (6), 1 17. Available from: https:// doi.org/10.1029/2012JB009260. Maslyn, R.M., 1987. Gridding advisor: an expert system for selecting gridding algorithms. Geobyte 2 (4), 42 43. McKenzie, D., 1967. Some remarks on heat flow and gravity anomalies. J. Geophys. Res. 72 (24), 6261 6273. Menard, H.W., Smith, S.M., 1966. Hypsometry of ocean basin provinces. J. Geophys. Res. 71 (18), 4305 4325. Available from: https://doi.org/ 10.1029/jz071i018p04305. Meyer, B., Saltus, R., Chulliat, A., 2017. EMAG2: Earth Magnetic Anomaly Grid (2-arc-minute resolution) Version 3. ,https://doi.org/ 10.7289/V5H70CVX.. Mu¨ller, R.D., Sandwell, D.T., Tucholke, B.E., Sclater, J.G., Shaw, P.R., 1991. Depth to basement and geoid expression of the Kane Fracture Zone: a comparison. Mar. Geophys. Res. 13 (2), 105 129. Available from: https://doi.org/10.1007/BF00286284. Mu¨ller, R.D., Roest, W.R., Royer, J.-Y., Gahagan, L.M., Sclater, J.G., 1997. Digital isochrons of the world’s ocean floor. J. Geophys. Res Solid Earth 102, 3211 3214. Available from: https://doi.org/10.1029/96jb01781. Mu¨ller, R.D., Sdrolias, M., Gaina, C., Roest, W.R., 2008. Age, spreading rates, and spreading asymmetry of the world’s ocean crust. Geochem. Geophys. Geosyst. 9 (4), 1 19. Available from: https://doi.org/10.1029/2007GC001743. Mu¨ller, R.D., Seton, M., Zahirovic, S., Williams, S.E., Matthews, K.J., Wright, N.M., et al., 2016. Ocean basin evolution and global-scale plate reorganization events since Pangea breakup. Annu. Rev. Earth Planet. Sci. 44 (1), 107 138. Available from: https://doi.org/10.1146/annurev-earth-060115-012211. Mu¨ller, R., Zahirovic, S., Williams, S., Cannon, J., Seton, M., Bower, D., et al., 2019. A global plate model including lithospheric deformation along major rifts and orogens since the Triassic. Tectonics 38. Available from: https://doi.org/10.1029/2018TC005462. Nier, A., 1939. The isotopic constitution of uranium and the half-lives of the uranium isotopes. Phys. Rev. 55, 150 153. O’Connor, J.M., Duncan, R.A., 1990. Evolution of the Walvis Ridge-Rio Grande rise hot spot system: implications for African and South American Plate motions over plumes. J. Geophys. Res. 95 (B11), 17475 17502. Available from: https://doi.org/10.1029/jb095ib11p17475. Parker, R., Oldenburg, D., 1973. Thermal model of ocean ridges. Nature 242, 137 139. Parmentier, E., Forsyth, D., 1985. Three dimensional flow beneath a slow spreading ridge axis: a dynamic origin of the deepening of the median valley toward fracture zones. J. Geophys. Res. 90, 678 684. Parrish, R., 1990. U-Pb dating of monazite and its application to geological problems. Can. J. Earth Sci. 27, 1431 1450. Parrish, R., Krough, T., 1987. Synthesis and purification of 205Pb for U-Pb geochronology. Chem. Geol. Isot. Geosci. Sect. 66, 103 110. Parsons, B., Sclater, J.G., 1977. An analysis of the variation of ocean floor bathymetry and heat flow with age. J. Geophys. Res. 82 (5), 803 827. Available from: https://doi.org/10.1029/jb082i005p00803. Pe´rez-Diaz, L., Eagles, G., 2014. Constraining South Atlantic growth with seafloor spreading data. Tectonics 33 (9), 1848 1873. Available from: https://doi.org/10.1002/2014TC003644. Pe´rez-Dı´az, L., Eagles, G., 2017a. A new high-resolution seafloor age grid for the South Atlantic. Geochem. Geophys. Geosyst. 18 (1), 457 470. Available from: https://doi.org/10.1002/2016GC006750. Pe´rez-Dı´az, L., Eagles, G., 2017b. South Atlantic paleobathymetry since early Cretaceous. Sci. Rep. 7 (1). Available from: https://doi.org/ 10.1038/s41598-017-11959-7.

Regional Geology and Tectonics

40

3. Age of the oceans

Pe´rez-Dı´az, L., Eagles, G., 2018. Estimating palaeobathymetry with quantified uncertainties: a workflow illustrated with South Atlantic data. In: Geological Society, London, Special Publications, SP476.1. ,https://doi.org/10.1144/sp476.1.. Pitman III, W.C., 1971. Seafloor spreading and plate tectonics. Eos Trans. Am. Gephys. Union 52 (5). Available from: https://doi.org/10.1029/ EO052i005pIU130. Pitman III, W., Heirtzler, J.R., 1966. Magnetic anomalies over the Pacific-Antarctic Ridge. Science 154 (3753), 1164 1171. Available from: https://doi.org/10.1126/science.154.3753.1164. Placet, T., 1658. The break up of large and small world’s, as being demonstrated that America was connected before the flood with the other parts of the world (booklet). Pratt, J., 1859. On the deflection of the plumb-line in India, caused by the attraction of the Himalaya Mountains and of the elevated regions beyond. Philos. Trans. R. Soc. Lond. 149. Raff, A., Mason, R., 1961. Magnetic survey off the west coast of North America, 40 N latitude to 52 N latitude. Geol. Soc. Of. Am. Bull. 72, 1267 1270. Available from: https://doi.org/10.1130/0016-7606(1961)72. Sandwell, D.T., Mu¨ller, R.D., Smith, W.H.F., Garcia, E., Francis, R., 2014. New global marine gravity model from CryoSat-2 and Jason-1 reveals buried tectonic structure. Science 346 (6205), 65 67. Available from: https://doi.org/10.1126/science.1258213. Sclater, J., Parsons, B., Jaupart, C., 1981. Oceans and continents similarities and differences in the mechanisms of heat-loss. J. Geophys. Res. 86, 8. Scotese, R.I., Gahagan, L.M., Larson, L., Deuelopment, S., 1988. Plate tectonic reconstructions of the Cretaceous and Cenozoic ocean basins. Tectonophysics 155, 27 48. Seton, M., Whittaker, J.M., Wessel, P., Mu¨ller, R.D., DeMets, C., Merkouriev, S., et al., 2014. Community infrastructure and repository for marine magnetic identifications. Geochem. Geophys. Geosyst. 15 (4), 1629 1641. Available from: https://doi.org/10.1002/2013GC005176. Skrzypek, E., Tabaud, A., Edel, J., Schulmann, K., Cocherie, A., Guerrot, C., et al., 2012. The significance of Late Devonian ophiolites in the Variscan orogen: a record from the Cosges Klippen Belt. Int. J. Earth Sci. 101 (4), 1 22. Smith, D., Tivey, M., Schouten, H., Cann, J., 1999. Locating the spreading axis along 80 km of the Mid-Atlantic Ridge south of the Atlantis Transform. J. Geophys. Res. Solid Earth 104 (84), 7599 7612. Available from: https://doi.org/10.1029/1998JB900064. Stampfli, G., Borel, G., 2002. A plate tectonic model for the Palaeozoic and Mesozoic constrained by dynamic plate boundaries and restored synthetic oceanic isochrones. Earth Planet. Sci. Lett. 169, 17 33. Stein, C.A., Stein, S., 1992. A model for the global variation in oceanic depth and heat flow with lithospheric age. Nature 359 (10), 123 129. Turcotte, D., Oxburgh, E., 1967. Finite amplitude convection cells and continental drift. J. Fluid Mech. 28, 29 42. Vening Meinesz, F., 1933. The mechanism of mountain formation in geosynclinal belts. Koninkl. Ned. Adad. Wetenschop. Proc. . Vine, F.J., 1966. Spreading of the ocean floor: new evidence. Science 154 (3755), 1405 1415. Vine, B.F.J., Matthews, D.H., 1963. Magnetic anomalies over oceanic ridges. Nature 4897, 947 949. Vogt, P.R., Ostenso, N.A., 1967. Steady state crustal spreading. Nature 215 (5103), 810 817. Available from: https://doi.org/10.1038/ 215810b0. von Humboldt, A., 1801. Alexander von Humboldt’s neue physikalische Beobachtungen im spanischen Amerika. Annalen der Physik 7, 329 347. von Humboldt, A., 1845. Kosmos: Entwurf einer physischen Weltbeschreibung, vol. 1, 493 pp. Stuttgart and Tu¨bingen, J. G. Cotta’scher Verlag. Wessel, P., Luis, J., 2017. The GMT/MATLAB toolbox. Geochem. Geophys. Geosyst. 18, 811 823. Wessel, P., Matthews, K.J., Mu¨ller, R.D., Mazzoni, A., Whittaker, J.M., Myhill, R., et al., 2015. Semiautomatic fracture zone tracking. Geochem. Geophys. Geosyst. 16 (7), 2462 2472. Available from: https://doi.org/10.1002/2015GC005853.

Further reading Le Pichon, X., Heirtzler, J.R., 1968. Magnetic anomalies in the Indian Ocean and sea-floor spreading. J. Geophys. Res. 73 (6).

Regional Geology and Tectonics

C H A P T E R

4 Plate boundaries and driving mechanisms Graeme Eagles Alfred Wegener Institute for Polar and Marine Research, Bremerhaven, Germany

Introduction Beneath Earth’s crust, which rarely exceeds a thickness of 40 50 km, lies a 2900 km thick layer of silicate rocks known as the mantle (Fowler, 2004). Throughout most of this thickness, the high pressure and temperature maintain rock viscosities that enable the mantle to flow over timescales longer than those of earthquakes. This capacity to flow enables convection, which is the principal means by which the planet loses its heat. The higher viscosities at shallow depths (shallower than 100 200 km), in contrast, do not permit convection by viscous deformation on the same timescales as in the mantle. Instead, they enable rock behaviour that closely approximates ideal elasticity over a wide range of timescales (Turcotte and Schubert, 2014). Forces raised as consequences of interior convection are consequently transferred over long distances without leading to pervasive deformation at the surface. Most near-surface deformation is instead concentrated into narrow zones (Heidbach et al., 2016; Fig. 4.1). Based on observations of the distribution of these zones, the theory of plate tectonics states that the planet’s surface shell is broken up into a finite number of mobile interlocking fragments, which are rigid internally but undergo deformation at the margins with their neighbours. This rigid outer shell is referred to as the lithosphere, and its individual fragments are termed plates. Studies using GPS data, very long baseline interferometry, earthquake distributions and first motions, the global distribution of active volcanism and the spacings of recent conjugate magnetic anomalies in the oceans all show the theory is a good approximation on observable to short geological timescales (e.g. DeMets et al., 2010). The narrow zones of tectonic interaction between neighbouring plates are referred to as plate boundaries. Plate boundaries can be classified kinematically, that is in terms of the relative motions of the plates they separate and the change in area affected by their action (Fowler, 2004; Turcotte and Schubert, 2014). The contrasts in relative plate motion at the different plate boundary types give rise to distinct tectonic and volcanic processes that produce distinct landforms and geological environments, which interact with rocks, climate and biology in distinctive ways. This makes it possible to interpret past plate motion from its geological products as preserved in rocks at multiple levels in the crust. Interpreted plate boundary settings, in turn, offer an essential context within which to ensure a truly self-consistent and mutually consistent understanding of the full range of features participating in a regional geological analysis. Ultimately, however, plate motion is a consequence of mantle convection. This chapter concludes with a brief consideration of how the two might be related.

Boundaries Divergent/constructive Plate boundaries from which the bounding plates move away are referred to as divergent. These zones are also referred to as constructive plate boundaries because new area is generated by the plates’ divergence.

Regional Geology and Tectonics. DOI: https://doi.org/10.1016/B978-0-444-64134-2.00004-3

41

© 2020 Elsevier B.V. All rights reserved.

42

4. Plate boundaries and driving mechanisms

FIGURE 4.1 Global distribution of large earthquakes (ISC data, Storchak et al., 2013; coloured discs, colour scale indicates hypocentre depths) over the 2004 16 period. Pink lines delineate plate boundaries according to Bird (2003). The great majority of large earthquakes occur at narrow plate boundaries. Intraplate seismicity shows that the plates do, however, deform internally.

In oceanic lithosphere At nearly all divergent plate boundaries in the oceans, the oceanic lithosphere is a primary product of plate divergence, rather than a preexisting body undergoing modification by it. The increasing volume created by the plates’ divergence is filled by convective ascent of warm mantle rocks from beneath the lithosphere. The rate of ascent is fast compared to the rate of cooling, such that the rocks decompress whilst still hot (e.g. McKenzie and O’nions, 1991; White et al., 1992). Partial melting is the result, leading to the formation of mafic magma, rich in silicates of iron and magnesium. The storage, transport and eruption of this magma produce a layered oceanic crust of basalt and gabbro rocks (Fowler, 2004; White et al., 1992). Beneath the crust, the lithosphere forms by conductive cooling of the top layer of the mantle. As time continues, ongoing cooling means that the lithosphere thickens and densifies. The lithosphere’s increasing density causes it to sink deeper and deeper into the underlying mantle, leading to deepening of the overlying ocean. By about 80 90 million years after its formation, the lithosphere reaches a thickness of around 100 km, and the ocean overlying it is over 5.5 km deep. Divergent plate boundaries in the oceanic lithosphere are generally marked by a mid-ocean ridge. Globally, the crests of these ridges lie at an average depth of around 2.6 km (Smith and Sandwell, 1997). The ridge crests overlook surfaces that slope downwards to extensive abyssal plains of the surrounding older oceanic lithosphere. They are arranged in a series of segments, whose lengths vary between tens to over a thousand kilometres. The segment ends may be defined by simple bends or deviations, which can propagate to form so-called pseudofaults, or by discrete stationary offsets known as transform faults, which accommodate strike-slip motion (Macdonald, 1982; Macdonald et al., 1988). Variable melt supply along the ridge segments is a response to the local pattern of mantle upwelling beneath them, cooling and alteration by sea water near the segment ends and along-segment transport. Melt oversupply can lead to a situation in which neighbouring ridge crests overlap and in extreme cases to the development of small plates in the overlap zone (e.g. Larson et al., 1992). Mid-ocean ridges are subdivisible observationally with reference to the crestal morphology, segmentation, basement morphology and crustal thickness (e.g. Macdonald, 1982; Fig. 4.2). At some ridges, broad crestal rises with relatively shallow (200 m) axial summit calderas associate with a corrugated basement surface off-axis that is textured by long ridges of volcanic origin; the long segments of this association are separated by either long transforms or overlapping ridge ends. In a second association, the ridge crest is the site of a deep ( . 1500 m), rifted median valley with steep walls and a broad (50 km) floor populated by volcanic centres; here, the ridge segments tend to be shorter and offset by more numerous transform faults, which also tend to be shorter (50 300 km). The upper basement surface topography is rough, textured by closely spaced normal faults and central volcanoes. In both of these associations, the crustal thickness is remarkably uniform at close to 7 km (e.g. White et al., 1992). A third association is with often-thinner and very much more variable crustal thickness, or even the exposure of mantle peridotites where a crust is absent (Dick et al., 2003). Here, ridge segments tend to host median valleys and may be dominated by normal faulting or volcanic centres that tend not to correlate with the presence of well-developed transform faults. These ridges are frequently oriented oblique to the direction of plate divergence.

Regional Geology and Tectonics

Boundaries

43 FIGURE 4.2 Diversity of mid-ocean ridges revealed in vertical gradient of gravity anomalies. Left: The East Pacific Rise between the Nazca and Pacific plates, with assemblage of features typical of fast spreading, at which normal mantle decompresses at a rate that ensures near-constant melt supply. Centre: The central Mid-Atlantic Ridge between the North American and Nubian plates, typical of slow spreading rates at which normal mantle decompresses to produce episodic melt supply. Right (top): Mohns and Knipovitch Ridges between the North American and Eurasian plates, sites of ultraslow spreading, at which rates normal mantle decompresses at a rate slow enough for conductive cooling to depress melting to the extent that melt supply is weak and strongly interrupted. Right (bottom): Kolbeinsey Ridge between the same plate pair, a slow spreading ridge over a region of warmer mantle, whose decompression produces an abundant and near-constant supply of melt. The ridge characteristics are similar to those of fast spreading rates. In the extreme, melt is so abundant that the igneous crust it builds is thick enough for the plate boundary to operate subaerially (Iceland). Yellow arrows indicate orientations of presentday relative plate motion, based on the MORVEL model (DeMets et al., 2010). Source: Data from Sandwell, D.T., Mu¨ller, R.D., Smith, W.H., Garcia, E., Francis, R., 2014. New global marine gravity model from CryoSat-2 and Jason-1 reveals buried tectonic structure. Science, 346 (6205), 65 67.

The three associations are often classified according to a spreading rate terminology, owing to their correlation with plate divergence rates (Macdonald, 1982; Dick et al., 2003). A key agent seems however to be the pattern of melt supply. Whilst the mantle in most locations seems well-mixed and of a uniform-enough temperature for plate divergence at most rates to lead to the construction of a 7 km thick oceanic crust, its delivery from the mantle at slower divergence rates is more episodic. This leads to the occurrence of periods in which plate divergence is accommodated by the development and rotation of large normal faults and low-angle detachments. Slow divergence also tends to associate with a higher along-axis frequency of transform faulting, perhaps relating to some combination of cooling of the lower crust (Bell and Buck, 1992) and the pattern of upwelling in the shallowest mantle (Morgan and Parmentier, 1995). It is only where plate divergence is extremely slow (slower than B12 mm/year) that the rate of mantle ascent is slow enough for conductive cooling to repress decompression melting, leading to a three-dimensional distribution of thin oceanic crust (Dick et al., 2003). In continental lithosphere Continental lithosphere is not produced in large quantities at divergent plate boundaries. Instead, mechanical and magmatic processes at these boundaries modify the preexisting continental lithosphere by thinning and/or

Regional Geology and Tectonics

44

4. Plate boundaries and driving mechanisms

replacing it. If allowed to continue, the eventual product of this is a plate boundary zone in which the continental crust has been entirely removed or replaced. At this point, plate divergence involves processes affecting the lithospheric mantle as described in the preceding section. Before this occurs, plate divergence in continents leads to the formation of rift zones. The detailed morphology of rift zones is a response to their volcanic and tectonic evolution as steered by a number of factors. This is because the new area created in the upper crust by plate divergence can be accommodated by a combination of newly generated volcanic rocks and the rotation of blocks of preexisting rocks between normal faults. The occurrence and distribution of these processes are determined by a variety of factors and their evolution through time. Old, thick, cold crust of the so-called continental cratons rarely undergoes rifting. It may be simply too strong, in comparison to its surroundings, to do so. Where regional heat flow is high and the crust is thicker than the mantle lithosphere, small basins form but in doing so strengthen the lithosphere, by cooling and/or by the introduction of compressive gravitational stresses (Buck, 1991; Buck et al., 1999). This results in the accommodation of plate divergence by rifting over a wide (800 km) region leading to a broad rift zone of multiple small faultbounded basins and localized central volcanoes that transport strongly fractionated melt to the surface. One example of such a broad rift zone is to be found in the Basin and Range region of the western United States (e.g. Thatcher et al., 1999; Fig. 4.3). Away from areas of high heat flow and thick crust, rifting does not result in local strengthening and plate divergence is accommodated by the action of a narrower (,100 km) and more focused rift zone with closely spaced normal faults associated with more extensive basaltic volcanism. The East African Rift zone is an example of this kind of setting (e.g. Ebinger, 1989; Fig. 4.3). With ongoing plate divergence, the ability to accommodate newly created area by ongoing fault rotations reduces as fault angles flatten, causing the frictional resistance to further motion on them to rise. With lithospheric thinning focusing on a smaller, and therefore narrower, selection of active faults, mantle upwelling and partial melting are locally enhanced and volcanism may consequently come to dominate over the mechanical extension of preexisting rocks. Examples of magma-dominated rifts include the Main Ethiopian Rift and Afar Triangle (e.g. Hayward and Ebinger, 1996; Fig. 4.3). Studies of extended continental margins however show that it is also possible for large amounts of divergence to be accommodated without an abundance of melting. Melting in these circumstances may be suppressed by low mantle temperatures, slow ascent and decompression or unfavourable mantle compositions. In these circumstances, it seems that near-surface mechanical processes transition from fault rotation to broad-scale horizontal displacement on low-angle or near-horizontal detachment surfaces along which the frictional resistance to sliding is reduced by the presence of serpentinite, a product of the alteration of mantle rocks by sea water (Pe´rez-Gussinye´ and Reston, 2001). In extreme cases, the action of such detachments can lead to uncovering or exhumation not only of the middle and lower crust but also of the underlying mantle lithosphere. The best-known example of such a situation is studied at the paired Iberian and Newfoundland conjugate margins in the north Atlantic (Whitmarsh et al., 2001; Tucholke et al., 2004). Beyond these general principles, the evolution of rift zones is more specifically influenced by a range of local factors including the history of divergence rates (and therefore strain rate), lithospheric temperature and heat production, mantle chemistry, crustal rheology, climate and erosion. The accompanying literature is vast and cannot be summarized here.

Convergent/destructive Plate boundaries towards which the bounding plate pairs move are termed convergent. Based on the loss of area that this movement causes, they may also be referred to as destructive. Involving subduction The formation of oceanic lithosphere takes place by cooling of mantle rocks and by volcanism of rocks generated from the mantle at mid-ocean ridges. This results in a buoyancy contrast between the cold dense lithosphere and the warmer and therefore less dense area of the mantle beneath it. Through much of the oceans, this contrast is not great enough to overcome the flexural strength of the lithosphere and allow sinking to occur. At destructive plate boundaries, however, this strength is much reduced by the presence of a fault zone cutting through the lithosphere, across which the vertical motion of sinking is accommodated. This process is known as subduction; it serves to accommodate the loss of area between converging plate pairs. The conditions that permit the initiation of subduction faults are the subject of ongoing research (Stern and Gerya, 2018). It has been repeatedly shown that they may develop when a change in relative plate motion requires a preexisting active normal or

Regional Geology and Tectonics

Boundaries

45

FIGURE 4.3 Two active rift zones showing variable modes of continental extension. Top: Basin and Range province, North America, a broad zone of rifting on thick continental crust. Multiple narrow surface basins separated by N-trending mountain ranges comprising normal fault footwall blocks. The Basin and Range is characterized by distributed seismicity and modest volcanism. Bottom: The East African Rift, a focused rift zone comprising one (in the north) or two (south) linear rift branches with more intense volcanism and mechanical extension focused on large faults bounding discrete large deep basins. The northern end of the rift zone is dominated by magmatism. Topography from Smith and Sandwell (1997). Volcanoes in red triangles (Global Volcanism Program, 2013). Earthquakes, all since 1990, located with black rings (ISC; Storchak et al., 2013). Black arrows indicate orientation of present-day relative plate motion (Nubian and Somalian plates), based on the MORVEL model (DeMets et al., 2010).

strike-slip fault zone to accommodate plate convergence (e.g. Guilmette et al., 2018). With mobile plates, therefore, once initiated subduction can be considered as a self-sustaining phenomenon. More controversy surrounds the possibility of so-called spontaneous subduction initiation, which is envisaged to take place in plate interiors rather than at an existing nondestructive plate boundary. This requires buoyancy contrasts to become large enough to overcome the strength of intact lithosphere, leading to the development or reactivation of a translithospheric fault. It has been suggested that spontaneous subduction initiation might occur near old fracture zones, or near the already fault-bounded lateral density contrasts between old, cold, oceanic lithosphere and lighter continental crust near mature extended continental margin. Such contrasts are suggested only to have been possible on Earth since Neoproterozoic times, prior to which the planet may have been too warm to allow subduction zones to exist (Sleep, 2000). There are no easily interpretable records of spontaneous subduction initiation in

Regional Geology and Tectonics

46

4. Plate boundaries and driving mechanisms

FIGURE 4.4 Convergent plate boundary between the oceanic Pacific (in the east) and Philippines (in the west) plates. Inset: seismic tomography model PRI-P05 [Montelli et al., 2006; imaged using Submachine (Hosseini et al., 2018)] with deep fast anomalies interpretable in terms of slab of oceanic lithosphere passing through the transition zone (dashed lines) into the lower mantle. Topography and bathymetry from Smith and Sandwell (1997). Volcanoes in red triangles (Global Volcanism Program, 2013). Earthquakes, all since 1990 (ISC; Storchak et al., 2013). Yellow arrow indicates orientation of present-day relative plate motion, based on the MORVEL model (DeMets et al., 2010).

rocks this old. Available interpretations of more recent spontaneous subduction initiation tend to be closely associated with preexisting convergent plate margins (e.g. Arculus et al., 2015) and may instead be interpretable in terms of relatively short-distance propagation of strain from an existing active fault onto a favourably aligned preexisting feature nearby. The oceanic lithosphere must bend in order to travel downwards into the mantle at a subduction zone. The bent upper surface of the lithosphere forms a deep (6 11 km) trench, which may or may not be filled by sedimentation (e.g. Figs 4.4 and 4.5). Previously deposited sediments and sedimentary rocks, if present, may be scraped off the subducting plate as it rubs against the subduction interface, stacking up on the overriding plate as thick, broad foreland wedges that eventually become gravitationally and tectonically unstable. Sedimentary basins may develop in various settings on the wedge in response to this instability (e.g. Smith et al., 2012). On its descent, the oceanic lithosphere is heated by conduction from the surrounding mantle rocks and is subjected to intense pressure. Starting at about 100 km depth, the changing pressure and temperature conditions lead to the occurrence of phase changes in the subducted slab’s hydrous minerals (Fowler, 2004; Turcotte and Schubert, 2014). These phase changes see water and other volatile materials expelled from the minerals. After migrating into the wedge of hotter mantle overlying the dipping slab, these volatiles promote decompression melting of the otherwise-dry mantle rocks. The presence of volatiles means the melt produced in the wedge is chemically distinct from the basaltic melts generated by decompression melting at mid-ocean ridges. They tend to be poorer in iron but richer in alkali metal elements and oxides of magnesium and calcium (Green and Ringwood, 1968). This

Regional Geology and Tectonics

Boundaries

47 FIGURE 4.5 Convergent boundary between the oceanic Nazca and Antarctic plates (in the northwest and southwest) and the South American plate (in the east). The arc-related topography is wider, and with a broader distribution of seismicity, than that in Fig. 4.4. Note the lack of a deep bathymetric trench south of 35 S, where mid-latitude westerly winds deliver abundant moisture to the Andes, resulting in strong rainfall and erosion that supply sediments in quantities great enough to fill the trench (e.g. Strecker et al., 2007). Topography and bathymetry from Smith and Sandwell (1997). Volcanoes in red triangles (Global Volcanism Program, 2013); note the nonvolcanic ‘Pampean Gap’ centred on 30 S, a consequence of very low-angle subduction. Earthquakes, all since 1990 (ISC; Storchak et al., 2013). Yellow arrows indicate orientations of present-day relative plate motion, based on the MORVEL model (DeMets et al., 2010).

calc-alkaline melt is introduced to the lithosphere, where it gives rise to intrusive magmatism and volcanism, at any given time forming an arc of magmatically-altered crust topped by active andesitic volcanoes, which lies parallel to the subduction zone. Despite the overall setting of plate convergence, many subduction zones show evidence for localized extension or even plate divergence within or behind the volcanic arc. This may transport part of the arc away from its deeper source in the devolatilizing slab (Karig, 1972). Subduction interfaces can and do migrate with respect to their overriding plates, for example by trench rollback or as a consequence of ‘subduction erosion’ (Stern, 2011), the result of processes by which parts of the overriding plate detach and are transferred to the subducting plate or lost into the mantle. A convergent plate boundary may thus migrate across an ocean until it reaches a continental margin. Subduction initiation, either forced or spontaneous, can occur at a continental margin. In either case, the resulting convergent plate boundary is one where oceanic lithosphere undergoes subduction beneath a continent (e.g. Fig. 4.5). In these situations, most or all of the region landwards of the trench consists of features developed in or on continental crust. Despite this, the overall structure of these settings is similar to that in intraoceanic subduction. The trench in oceanic/continental subduction zones is more likely to be sediment-filled owing to the neighbouring continental sediment source; it may not form a strong bathymetric feature. The volcanic arc is more chemically diverse and

Regional Geology and Tectonics

48

4. Plate boundaries and driving mechanisms

distinct owing to contamination of the melt by its interactions with the diverse lithologies of the continental crust. The type example of this kind of setting is the Andean margin of South America (Fig. 4.5). The detailed variability of subduction zones is attributable to the influences of a range of factors on the processes responsible for their formation. These include slab angle, which determines the width of the so-called arctrench gap and even the very occurrence of volcanism (e.g. Ramos et al., 2002; Von Huene et al., 1997); slab width, which may correlate with the presence or absence of back-arc basins (e.g. Dvorkin et al., 1993; Schellart et al., 2007); the speed of plate convergence (e.g. Huang and Lundstrom, 2007; Eagles and Scott, 2014); the age and thermal structure of the subducting slab, which influences volcanism via volatile release (e.g. Green and Harry, 1999), and even the roughness of the incoming crust, which influences the seismicity of the subduction interface (e.g. Bu¨rgmann et al., 2005). In continental lithosphere Subduction in the oceans can lead to complete transport of a basin’s oceanic lithosphere to the mantle interior. Instances like this culminate in the juxtaposition of continents across an active convergent plate boundary. In such situations, near the surface continental crust is forced into collision with continental crust. The distinct bulk chemistry of the continental crust is such that its density remains less than that of the mantle under a wide range of temperature and pressure conditions. The lithospheric mantle beneath cratonic continental crust is also chemically distinct as a consequence of its depletion during formation of the overlying crust, potentially leaving it unusually strong (Pearson et al., 1995). As such, although the continental lithosphere can be deeply buried or incorporated into the mantle lithosphere by folding and thrust faulting, it cannot achieve a large-scale negative buoyancy contrast that leads to long-lived or widespread subduction. Because of this, it is not possible for destructive plate boundaries in continental crustal settings to accommodate hundreds or thousands of kilometres of convergent plate motion as vertical motion on a single fault zone. What happens instead is the subject of debate and geodynamic modelling, informed by competing interpretations of geological, geodetic, seismic and geophysical datasets. Suggested mechanisms, which need not be mutually exclusive, include long-distance underthrusting of one continent beneath the other, distributed shortening and crustal thickening by stress transmission far into the interior of one or both of the colliding continents, the redistribution of crustal volume by viscous flow, removal of mantle lithosphere by ‘dripping’ or delamination, subduction of subcontinental mantle by small stepwise amounts at migrating subduction fronts and lateral extrusion of blocks of the lithosphere (e.g. Argand, 1922; England and McKenzie, 1982; Dewey and Burke, 1973; Molnar and Tapponnier, 1975, Tapponnier et al., 2001; Royden et al., 1997; Molnar, 1988). The type example of a continent continent collision zone at a convergent plate boundary is the Himalayan Tibetan plateau system (Fig. 4.6). Its defining feature is the broad (B1000 km across strike) region of very high ( . 5 km) elevation known as the Tibetan Plateau. Folding and thrust faulting express the convergent plate boundary setting most clearly in the B150 300 km wide Himalaya, at the southern edge of the Tibetan Plateau. Consistent with the convergent setting, elevations are the highest on Earth and the lithosphere is extraordinarily thick. The loading effect produces a broad flexural basin in northern India. The Tibetan Plateau, in contrast, is dominated by extensional and strike-slip faulting in thick crust over a very thin lithosphere, which can be understood in terms of gravitational collapse of thick crust over lithosphere that has been thermally eroded and/or detached from the overlying crust (Chen et al., 2017). The lack of volcanism in the Himalaya may reflect lack of volatiles in subcontinental lithosphere undergoing compression and/or the great thickness of continental crust leading to intracrustal storage. Tibetan volcanism is more widespread, a distribution that can be seen as consistent with the recent detachment of the lower part of a thick lithosphere allowing for decompression melting in the mantle that welled up to take its place.

Conservative/strike-slip Two plates that move parallel to the strike of their mutual boundary see no change in area; such boundaries are termed conservative. They are also referred to with the term strike-slip. A further description, as ‘transform’ plate boundaries, refers to the observation that these features often accommodate a transformation in the sense of motion between combinations of divergent and convergent plate boundary segments. As all of the relative motion is in the plane of the lithosphere, these plate boundaries are not necessarily associated with strong topography. A newly formed strike-slip fault zone in homogeneous crust may be a finite-width zone of pure shear at depth, but towards the free upper surface of the crust it develops by the amalgamation of two en echelon sets of small shears, known as Riedel shears, that form as conjugate sets at 30 degrees to the short Regional Geology and Tectonics

Boundaries

49

FIGURE 4.6 Convergent boundary between continental parts of the Indian and Eurasian plates. Tibetan Plateau earthquake mechanisms are a mixture of normal on north south normal faults and strike-slip on east west regional faults. Topography and bathymetry from Smith and Sandwell (1997). Volcanoes in red triangles (Global Volcanism Program, 2013). Earthquakes, all since 1990 (ISC; Storchak et al., 2013). Black arrow indicates orientation of present-day relative plate motion, based on the MORVEL model (DeMets et al., 2010). Right: PRI-P05 tomography model [Montelli et al., 2006; imaged using Submachine (Hosseini et al., 2018)]. The image can be interpreted to show a doublethickness lithosphere under India as a result of underthrusting without subduction and a thin or absent lithosphere under the Tibetan Plateau as a result of thermal erosion.

axis of the instantaneous strain ellipse associated with the plate boundary that is at B15 and B75 degrees to the main fault zone. These shears coalesce and link with ongoing strain to form a fault zone of 10 30 km width that is marked by a modest surface depression that forms in response to oblique extensional reactivation of the Riedel shears during ongoing strain. In many cases, however, preexisting fault zones are sufficiently weaker than the rest of the crust that a new strike-slip fault does not form. Instead, the preexisting features undergo reactivation despite potentially lying oblique to the axes of the strain ellipse. Transpression is the term applied to a situation in which transform motion is not perfectly achieved along upper crustal features, because a small-to-moderate component of plate convergence is accommodated across the plate boundary zone. Transpression can lead to the development of folded and thrust-bounded ridges separated by strike-slip faults, or situated along lengths of them, on a variety of length scales. Plate boundaries that act in transtension, in contrast, accommodate a component of plate divergence within a dominantly strike-slip setting. The normal faults that develop in these settings potentially transform very long horizontal displacements into vertical motions. As a result, they can bound extremely deep basins whose floors may even experience the formation of oceanic crust or exhumation of mantle rocks. In oceanic lithosphere The section about divergent plate boundaries in oceanic lithosphere described the distribution of transform faults along mid-ocean ridges as part of the response to plate divergence. It was suggested that these faults may form as a response to the divergent plate boundary’s need to accommodate itself to the detailed pattern of viscous flow in the lower oceanic crust or underlying upwelling mantle. These ridge-offset transform faults may be a few kilometres to a few hundred kilometres long (Fig. 4.7). As they affect newly formed oceanic lithosphere, they frequently form as simple strike-slip fault zones 5 30 km wide, which are marked by a surface trough. Where the seafloor age contrast across the transform fault is large, the trough may be superimposed on a longerwavelength topographic step. Changing plate motion orientations, nonetheless, do lead to situations in which the stress fields around active preexisting fault zones rotate. Short- and medium-length transform plate boundary segments are sometimes able to accommodate such rotations by modification of tectonic processes within their deformation zones that achieve a change in those zones’ overall strikes. As transform faults lengthen, it becomes less and less possible to accommodate all of the necessary change within their deformation zones. Instead, even slight rotations can lead to the Regional Geology and Tectonics

50

4. Plate boundaries and driving mechanisms

FIGURE 4.7 Conservative plate boundaries in the oceans. Top: Transform faults on the Chile Rise, between the Nazca and Antarctica plates, in SRTM15 1 data (Olson et al., 2016). Middle: Macquarie Ridge complex between the Australian and Pacific plates, with history of transpression that has proceeded to the point of incipient subduction at its southern and northern ends (note the deep trenches on the western edge of the ridge complex). Bottom: Cayman Trough between the North American and Caribbean plates, with history of transtension, giving rise to a deep basin with an active mid-ocean ridge segment near 81 W, 17.5 N. Bottom and middle topographic and bathymetric data from Smith and Sandwell (1997). Volcanoes in red triangles (Global Volcanism Program, 2013). Large earthquakes, all since 1990 (ISC; Storchak et al., 2013). Yellow arrows indicate orientations of present-day relative plate motion, based on the MORVEL model (DeMets et al., 2010).

localized onset of transpression or transtension in which parts of these fault zones or their surroundings are reactivated in new senses. Reactivation may lead to the development of more complex morphology such as deep basins, possibly even featuring small mid-ocean ridge segments, along so-called releasing bends/basins (e.g. Cayman Trough; Fig. 4.7) or thrust-bound ridges at so-called pop-ups (e.g. St Peter and St Paul archipelago, Maia et al., 2016; Macquarie Ridge, Fig. 4.7). In continental lithosphere Preexisting structures are far more common and available for reactivation in continental crust and lithosphere, by virtue of its long history of involvement in multiple plate tectonic events. As a result, any single conservative

Regional Geology and Tectonics

Boundaries

51

plate boundary might express both transpression and transtension at different locations along its length where it responds to ongoing rotation of plate motion vectors and/or encounters and reactivates differently oriented preexisting structures. The active NNE-trending left-lateral Dead Sea Transform in the Middle East presents a case in point (Fig. 4.8). In the south, four segments of the plate boundary host deep ( . 10 km) sediment-filled depressions. Three of these are submarine features located within the Gulf of Aqaba, the fourth is at the Dead Sea. The deep basins are separated from one another by more subdued linear troughs. Two of the depressions have been suggested to mark locations where the plate boundary zone intersects preexisting (B100 Ma) crustal-scale dense magmatic bodies, reactivating their margins as normal faults bounding deep offset basins (Ben-Avraham et al., 2010). Further north along the same plate boundary zone, regional structures rotate to trend NE in the 3 km high Palmyride mountain belt. These structures appear to be of Cretaceous age and to have formed by continental extension during a past period of plate divergence. Despite their origin in extension, today’s earthquake focal mechanisms reveal the NE-trending faults to fail with thrust and oblique mechanisms, consistent with their

FIGURE 4.8 Conservative plate boundary between continental parts of the Arabian (in the east) and Nubian (in the west) plates: the Dead Sea Transform. Dead Sea and Gulf of Aqaba releasing offsets near 29 N and 31 N. Palmyride mountain belt restraining bend on reactivated Jurassic Cretaceous rift faults near 34 N. Simple valley segments between. Topography from SRTM15 1 (Olson et al., 2016). Volcanoes in red triangles (Global Volcanism Program, 2013). Large earthquakes, all since 1990 (ISC; Storchak et al., 2013). Black arrows indicate orientations of present-day relative plate motion, based on the MORVEL model (DeMets et al., 2010).

Regional Geology and Tectonics

52

4. Plate boundaries and driving mechanisms

adoption to accommodate NNE-trending sinistral transform motion on the Arabia Nubia plate boundary (Brew et al., 2003). Between oceanic and continental lithosphere Conservative plate boundaries can also form at the transition between continental and oceanic lithosphere. Short active segments of such a setting can be seen along the margins of the Gulf of California, illustrating the most likely cause of their formation to be in the early stages of plate divergence as continental extension gives way to the formation of oceanic lithosphere. Conservative plate boundary segments like these go on to produce very steep continental margins with relatively sharp continent-ocean transition zones. The equatorial conjugate margins of West Africa and South America host good examples.

Oblique, partitioned-strain, and diffuse plate boundaries The introduction to this chapter described how the theory of plate tectonics envisages the lithosphere to consist of a patchwork of rigid plates that deform only at their margins. Fig. 4.1 and many of the subsequent figures have also mapped intraplate seismicity that clearly shows how this description is not accurate. Based on the pattern of this seismicity and various ancillary observations, it can be shown that plate boundary zones at the surface can and do become as wide as or wider than whole plates in various locations worldwide. One example of this was described in the section on destructive plate boundaries in continental lithosphere; the Himalaya Tibet collision zone exceeds 1000 km in width, ultimately as a consequence of the fact that stable subduction zones cannot be sustained in continental crust. Elsewhere, the continental lithosphere at oceanic-continental convergence zones can also deform over a broad plate boundary zone if the convergence involves a strong component of obliquity. In these instances, instead of the trench-parallel component of relative plate motion being accommodated by oblique thrusting along the subduction thrust fault, the trench-normal and trench-parallel components may be partitioned onto parallel but geographically separate fault zones. A well-known example of a pair of such faults is the Great Sumatran Fault, which accommodates strike-slip motion, and the thrust-dominated Sumatran subduction zone (Fitch, 1972; McCaffrey et al., 2000). The Great Sumatran Fault is located 300 km further into the overriding plate than the trench and is a focus for intense volcanic activity. The distribution of earthquakes between the subduction trench and Great Sumatran Fault (Fig. 4.9) suggests the intervening region is one of continuous complex deformation and not an independently moving rigid small plate. Broad lithospheric-scale plate boundaries are also to be found confined to regions of oceanic lithosphere. Their presence is betrayed not only by intraplate seismicity but by evidence for active basement folding and faulting in seismic profiles, bathymetric and potential field data, and by small deviations from the expected rate of seafloor spreading measured using paired magnetic anomaly isochrons along lengths of mid-ocean ridges near them. The best-known example of this kind of boundary is situated in the eastern part of the Indian Ocean between India and FIGURE 4.9 Strain partitioning in the Sumatra trench region between the Indian (bottom) and Sundaland (top) plates. Grey discs: earthquakes with undifferentiated thrust and normal mechanisms. Yellow discs: earthquakes with strike-slip mechanisms. Red triangles: active volcanoes, which follow the trace of the right-lateral Great Sumatran Fault. Topography and bathymetry from Smith and Sandwell (1997). Volcanoes in red triangles (Global Volcanism Program, 2013). Large earthquakes, all since 1990 (ISC; Storchak et al., 2013). Yellow arrow indicates orientation of present-day relative plate motion, based on the MORVEL model (DeMets et al., 2010).

Regional Geology and Tectonics

Driving mechanisms of plate motion

53

FIGURE 4.10 (A) Detailed bathymetry data in part of the Wharton Basin, west of Indonesia, showing pervasive deformation in response to NNW-directed regional convergent stress between the Indian and Australian component plates. The crust in the basin responds by (B and C) transtensional reactivation of N S trending fracture zones involving the development of NNE- and WNW-oriented neotectonic Riedel shear zones and, locally, thrusting along reactivated NE-trending volcanic features. Figure reproduced from Hananto et al., 2018; other figures referenced are to be found in that publication and not this chapter.

Australia (Fig. 4.10; Royer and Gordon, 1997; DeMets et al., 2005; Hananto et al., 2018). The imbalance in spreading rates and the pattern of seismic energy release in the region are both consistent with it experiencing ongoing very slow (,5 mm/year; DeMets et al., 2010) north south compression, which possibly started around 15 20 million years ago. It has been suggested that such rates of deformation are too slow, and/or that the lithosphere in such settings is too strong, to permit a localization of deformation onto a narrow boundary (Zatman et al., 2001).

Driving mechanisms of plate motion The plate kinematic record shows clearly not only that the plates are in motion, but also that their motions undergo slow but constant change (e.g. Seton et al., 2012). This requires a slowly and constantly changing balance of torques (turning forces) to act on the plates. This section examines the torque balance. At any location within a lithospheric plate, the torque balance responsible for the plate’s motion contributes to stress, a measurable descriptor of the internal force per unit area acting within a material. Stress measurement techniques vary, for example by study of the energy released during earthquakes or by measurements of changes

Regional Geology and Tectonics

54

4. Plate boundaries and driving mechanisms

FIGURE 4.11 2016 World Stress Map showing orientation of maximum horizontal intraplate stress in the upper 40 km of the crust based on multiple methods (Heidbach et al., 2016). Stress regimes are colour-coded in consistency with the occurrence of normal, strike-slip and thrust faulting. Topography ETOPO1 (NGDC). The plate boundary data set plotted in the background figure is that of Bird (2003).

in the shapes of boreholes. For these reasons, measurements tend to be concentrated in the upper part of the crust. The pattern of available stress measurements collated in the World Stress Map (Fig. 4.11) is not constant over large areas within the plate interiors. This allows the conclusion that the torques participating in the lithospheric balance come from a variety of sources and locations at plate, regional and local scale (e.g. Heidbach et al., 2016). This heterogeneity of stress orientations in the plate interiors is consistent with the idea that the torques involved in the balance responsible for plate motion are the consequences of (1) buoyancy contrasts and (2) viscous dissipation or drag, both being generated within and between the plates and the mantle beneath them (Fig. 4.12). Concentrating on convection in the mantle as the ultimate source of both, it can be shown that the pattern of plate tectonics is not a simple response to the pattern of mantle convection in which mid-ocean ridges mark the sites of warm rising material and subduction zones the sites of cold material sinking. Tomographic imaging shows, instead, that the pattern of upwelling appears unrelated to the distribution of ridges. It is likely instead to be influenced by some combination of internal radioactive heating and heat conduction across the core mantle boundary. It should be noted that these factors in turn may respond to plate motion over very long timescales, via the distribution of subduction zones that determine the composition and thermal conductivity distribution of mantle rocks at the core mantle boundary (e.g. McNamara, 2019). Despite all this, in exploring whether mantle convection may play a dominant role in determining plate motion despite this complexity, some two-dimensional and three-dimensional convection models, fed with boundary conditions derived from interpretations of present-day mantle tomography images, have been able to reproduce near-surface material transport fields that are plausibly reminiscent of observed plate motion patterns (Vigny et al., 1991; Forte and Peltier, 1994; Becker and O’Connell, 2001). The accuracy and applicability of these models can, however, be questioned on the basis of the specific parameterizations they require for modelling convection, in particular those for conductive heating across the core mantle boundary and for the plates themselves.

The torque balance An often-followed alternative approach to examining the reasons for plate motion does not explicitly consider convection as a whole-mantle kinematic process. Instead, it is based on observation of plate motion and forward

Regional Geology and Tectonics

Driving mechanisms of plate motion

Transform resistance Buoyancy contrasts Collision resistance Sl

ab

pu

ll

rm

al

ab

Sl no

ab

Downwelling

Upwelling

Sl

n io ct su

Basal shear

Slab resistance

55 FIGURE 4.12 Sketch of a torque balance maintaining the motion of a green plate. The balance can be modelled as a set of torques assumed to be experienced by the lithosphere (forces shown by black arrows). Ultimately, all of these are a consequence of convection in the mantle (large white arrows). Downwelling as part of this convection is understood to be represented by the subduction of oceanic lithosphere. The sites of upwelling, in contrast, are not obliged to be located beneath mid-ocean ridges; instead they are determined deeper within the mantle. Longwavelength mantle flow between upwellings and downwellings is therefore not obliged to be parallel to the direction of plate motion. The basal shear force may therefore act on either side of the force balance equation.

modelling of the torque balance itself. A weakness of this is that it needs to assume in advance that all of the torques, and their roles in the balance, are known. A strength is that, by considering the torques as acting on or within the plates, many of the necessary parameters and observations for modelling them are more plausibly constrained by observations than those required for modelling mantle convection. In some of this class of studies, plate speed and direction have been statistically compared with variables such as slab length, plate area or continental area (Forsyth and Uyeda, 1975). Others have undertaken forward modelling of parameterized torque balances for individual plates (e.g. Warners-Ruckstuhl et al., 2012). Some of the forces that can be considered to contribute to stresses acting within the lithosphere are raised by buoyancy contrasts and physical processes at their boundaries. These forces can be considered to be transmitted into the plate interiors as a consequence of the plates’ very high viscosity. The local budget of torques at subduction zones is fairly complex and due to local as well as whole-mantle convection. Some considerations see the balance as almost locally complete, leaving gravitational forces within the lithosphere and possibly deeper mantle to dominate plate motion. Negative buoyancy of slabs (slab pull and slab-normal forces) Lithospheric slabs are the result of conductive cooling of mantle rocks. They can sink as a consequence of the negative buoyancy that results. Sinking is vertical but slab pull acts along the slab. The accompanying slabnormal component of the vertically oriented negative buoyancy can be viewed as the reason for trench rollback, upper-plate extension and back-arc basin development. A whole row of studies that parameterize slab pull has concluded it is to be the largest magnitude torque in the plate-driving balance and therefore the most important influence on plate motion (e.g. Forsyth and Uyeda, 1975; Conrad and Lithgow-Bertelloni, 2002). This conclusion is epitomized in plots of plate speed against slab length, in which the plates with the longest lengths of margin attached to slabs in the upper mantle tend also to be those that move fastest over the mantle at the present day. It should be recalled, however, that correlation is not evidence of causation. That is to say, is it possible instead that the fastest-moving plates are those most likely to require the longest lengths of slab to accommodate their fast motion with respect to their neighbours? Slab resistance and collision resistance Subducting slabs will experience viscous resistance to their passage through the upper mantle. At crustal and lithospheric levels, frictional and viscous resistance at various depths along the subduction interface and thrust faults will have a similar effect. These undoubtedly resisting torques are thought to locally balance and be cancelled by slab pull, which allows to question the supposed dominance of slab pull in the torque balance.

Regional Geology and Tectonics

56

4. Plate boundaries and driving mechanisms

Slab suction Viscous shear forces are raised between the moving slabs of subducted lithosphere and the surrounding mantle. This may set up local convection cells in the mantle whose upper parts are directed towards the subduction zone, attracting nearby plates towards it (Conrad and Lithgow-Bertelloni, 2002). Slab suction forces have been suggested as the reason for trench rollback in instances where the slab is not strongly attached to the surface part of the plate and, on overriding plates, back-arc extension. The localized convection has, however, proved difficult to image tomographically. Transform shear Davies (1978) suggests that the pattern of stress drop from earthquakes at large transform faults implies a resistive shear force so strong as to be potentially able to balance the sum of all suggested driving torques on the plates. If so, this might mean that slab and collision resistance forces should be regarded, as above, as localized phenomena. Gravitational body forces/rifts and ‘ridge push’ A large component of the stress field in terrestrial rocks results is simply gravitational: the rocks’ weight. This gravitational stress depends on mass and thus varies along with lithospheric thickness and density. It is the reason for overall compressive stress at the edges of continents and extensional stresses within orogenic plateaux. Similarly, the World Stress Map compilation (Fig. 4.11) reveals a tendency for old oceanic lithosphere to exist largely in a state of compressive stress, which is consistent with gravitational driving forces playing a strong role in intraplate stress. The best-known instance of such forces is the gravitational force that can be expected as a response to the cooling-related topography and thickness change of the oceanic lithosphere. In numerous publications, this has been referred to as a distinct ‘ridge push’ force. Plots of the lengths of mid-ocean ridges at plate margins against their speed over the mantle do not reveal a strong correlation between the two, however, suggesting that ridge push at the very least is not the only or dominant driving torque related to plate thickness and density variations. Global mantle seismic tomographic imaging suggests that upwellings in the mantle are not always or even usually located underneath plate edges, ruling out the simplest understanding of plate motions as a consequence of mantle convection (slabs as direct return flow of plumes). Where upwellings do interact with plate edges, however, it can be envisaged that the plate moves down the potential slope at the top of the convecting mantle. Like ‘ridge push’, this contributes to the overall gravitational body force acting on a plate by wholesale tilting. Basal shear force Basal shear is a term used to refer to the viscous force raised owing to the contrast in the horizontal components of plate motion and motion of the mantle rocks beneath the plates. In much of the world, determinations of lattice-preferred orientations in upper mantle minerals, based on seismic shear-wave splitting calculations, suggest that the upper mantle dominantly undergoes strain with the same orientation as the motion of the plates above it (e.g. Kreemer, 2009). This can be interpreted to mean either that basal drag drives the plates, or alternatively that plate motion viscously deforms the asthenosphere. The latter possibility is consistent with the observation of widespread compressional stress in old flat oceanic lithosphere (Fig. 4.11; Richardson, 1992). This in turn has been taken as the basis for assuming a passive, thermally homogeneous, mantle in constructing forward models of the torque balance (e.g. Forsyth and Uyeda, 1975). Despite this, local mismatches do occur between plate motion and lattice-preferred orientation, as well as between plate motion and mantle flow based on tomographic images interpreted as temperature-dependent. These mismatches suggest that the upper mantle does flow, at least locally, on different azimuths to plate motion and can be interpreted in terms of the effects of deep mantle buoyancy contrasts on the upper mantle flow field. This invites an interpretation of basal shear as a locally complex product of two independently moving fluids at an interface and of the likelihood of there being instances where basal shear contributes on either side of the torque balance (Forte et al., 2009). More knowledge about upper mantle rheology and mechanics of the lithosphere asthenosphere boundary is needed to understand basal shear more clearly.

Regional Geology and Tectonics

References

57

Summary Plate motion is maintained due to the detailed balance of (1) buoyancy forces generated by their thickness variations, by the distribution and size of their subducted parts and by upwelling in the mantle beneath them, with (2) resistance at their various interfaces with other plates and the underlying mantle. The pattern of plate motion may dictate or be dictated by the pattern of mantle convection. The torque balance maintains, and is maintained by, steady plate motion and mantle convection over long periods. Relative to their shared margins, movements between pairs of plates may be divergent, forming continental rifts and mid-ocean ridges, convergent, giving rise to subduction and collision zones, or be parallel, causing the action of transform faults. The detailed structure and development of all these features depend on a wide variety of specific parameters that vary strongly according to the speed of relative motion, crustal and lithospheric rheology, preexisting structure, and composition, and the composition and structure of the shallow convecting mantle.

References Arculus, R.J., Ishizuka, O., Bogus, K.A., Gurnis, M., Hickey-Vargas, R., Aljahdali, M.H., et al., 2015. A record of spontaneous subduction initiation in the Izu Bonin Mariana arc. Nat. Geosci. 8 (9), 728. Argand, E., 1922, June. La tectonique de l’Asie. Confe´rence faite a´ Bruxelles, le 10 aouˆt 1922. In Congre`s ge´ologique international (XIIIe session)-Belgique 1922, pp. 171 372. Becker, T.W., O’Connell, R.J., 2001. Predicting plate velocities with mantle circulation models. Geochem. Geophy. Geosyst. 2 (12). Bell, R.E., Buck, W.R., 1992. Crustal control of ridge segmentation inferred from observations of the Reykjanes Ridge. Nature 357 (6379), 583. Ben-Avraham, Z., Lyakhovsky, V., Schubert, G., 2010. Drop-down formation of deep basins along the Dead Sea and other strike-slip fault systems. Geophys. J. Int. 181 (1), 185 197. Bird, P., 2003. An updated digital model of plate boundaries. Geochem. Geophys. Geosyst. 4 (3). Brew, G., Best, J., Barazangi, M., Sawaf, T., 2003. Tectonic evolution of the NE Palmyride mountain belt, Syria: the Bishri crustal block. J. Geol. Soc. 160 (5), 677 685. Buck, W.R., 1991. Modes of continental lithospheric extension. J. Geophys. Res. Solid Earth 96 (B12), 20161 20178. Buck, W.R., Lavier, L.L., Poliakov, A.N., 1999. How to make a rift wide. Philos. Trans. R. Soc. Lond. A Math. Phys. Eng. Sci. 357 (1753), 671 693. Bu¨rgmann, R., Kogan, M.G., Steblov, G.M., Hilley, G., Levin, V.E., Apel, E., 2005. Interseismic coupling and asperity distribution along the Kamchatka subduction zone. J. Geophys. Res. Solid Earth 110 (B7). Chen, M., Niu, F., Tromp, J., Lenardic, A., Lee, C.T.A., Cao, W., et al., 2017. Lithospheric foundering and underthrusting imaged beneath Tibet. Nat. Commun. 8, 15659. Conrad, C.P., Lithgow-Bertelloni, C., 2002. How mantle slabs drive plate tectonics. Science 298 (5591), 207 209. Davies, G.F., 1978. The roles of boundary friction, basal shear stress and deep mantle convection in plate tectonics. Geophys. Res. Lett. 5 (3), 161 164. DeMets, C., Gordon, R.G., Royer, J.Y., 2005. Motion between the Indian, Capricorn and Somalian plates since 20 Ma: implications for the timing and magnitude of distributed lithospheric deformation in the equatorial Indian Ocean. Geophys. J. Int. 161 (2), 445 468. DeMets, C., Gordon, R.G., Argus, D.F., 2010. Geologically current plate motions. Geophys. J. Int. 181 (1), 1 80. Dewey, J.F., Burke, K.C., 1973. Tibetan, Variscan, and Precambrian basement reactivation: products of continental collision. J. Geol. 81 (6), 683 692. Dick, H.J., Lin, J., Schouten, H., 2003. An ultraslow-spreading class of ocean ridge. Nature 426 (6965), 405. Dvorkin, J., Nur, A., Mavko, G., Ben-Avraham, Z., 1993. Narrow subducting slabs and the origin of backarc basins. Tectonophysics 227 (1 4), 63 79. Eagles, G., Scott, B.G., 2014. Plate convergence west of Patagonia and the Antarctic Peninsula since 61 Ma. Glob. Planet. Change 123, 189 198. Ebinger, C.J., 1989. Tectonic development of the western branch of the East African rift system. Geol. Soc. Am. Bull. 101 (7), 885 903. England, P., McKenzie, D., 1982. A thin viscous sheet model for continental deformation. Geophys. J. Int. 70 (2), 295 321. Fitch, T.J., 1972. Plate convergence, transcurrent faults, and internal deformation adjacent to southeast Asia and the western Pacific. J. Geophys. Res. 77 (23), 4432 4460. Forsyth, D., Uyeda, S., 1975. On the relative importance of the driving forces of plate motion. Geophys. J. Int. 43 (1), 163 200. Forte, A.M., Peltier, W.R., 1994. The kinematics and dynamics of poloidal toroidal coupling in mantle flow: the importance of surface plates and lateral viscosity variations, Advances in Geophysics, vol. 36. Elsevier, pp. 1 119. Forte, A.M., Moucha, R., Rowley, D.B., Que´re´, S., Mitrovica, J.X., Simmons, N.A., et al., 2009. Recent tectonic plate decelerations driven by mantle convection. Geophys. Res. Lett. 36 (23). Fowler, C.M.R., 2004. The Solid Earth: An Introduction to Global Geophysics., second ed. Cambridge University Press. Global Volcanism Program, 2013. Volcanoes of the World, v. 4.8.0. In: Venzke, E. (Ed.), Smithsonian Institution. Downloaded 29 May 2019. ,https://doi.org/10.5479/si.GVP.VOTW4-2013.. Green, N.L., Harry, D.L., 1999. On the relationship between subducted slab age and arc basalt petrogenesis, Cascadia subduction system, North America. Earth Planet. Sci. Lett. 171 (3), 367 381. Green, T.H., Ringwood, A.E., 1968. Genesis of the calc-alkaline igneous rock suite. Contrib. Miner. Petrol. 18 (2), 105 162. Guilmette, C., Smit, M.A., van Hinsbergen, D.J., Gu¨rer, D., Corfu, F., Charette, B., et al., 2018. Forced subduction initiation recorded in the sole and crust of the Semail Ophiolite of Oman. Nat. Geosci. 11 (9), 688.

Regional Geology and Tectonics

58

4. Plate boundaries and driving mechanisms

Hananto, N., Boudarine, A., Carton, H., Singh, S.C., Avianto, P., Dyment, J., et al., 2018. Evidence of pervasive trans-tensional deformation in the northwestern Wharton Basin in the 2012 earthquakes rupture area. Earth Planet. Sci. Lett. 502, 174 186. Hayward, N.J., Ebinger, C.J., 1996. Variations in the along-axis segmentation of the Afar Rift system. Tectonics 15 (2), 244 257. Heidbach, O., Rajabi, M., Reiter, K., Ziegler, M., the WSM Team, 2016. World Stress Map database release 2016. GFZ Data Services. ,https:// doi.org/10.5880/WSM.2016.001.. Hosseini, K., Matthews, K.J., Sigloch, K., Shephard, G.E., Domeier, M., Tsekhmistrenko, M., 2018. SubMachine: web-based tools for exploring seismic tomography and other models of Earth’s deep interior. Geochem. Geophys. Geosyst. 19 (5), 1464 1483. Huang, F., Lundstrom, C.C., 2007. 231Pa excesses in arc volcanic rocks: constraint on melting rates at convergent margins. Geology 35 (11), 1007 1010. Karig, D.E., 1972. Remnant arcs. Geol. Soc. Am. Bull. 83 (4), 1057 1068. Kreemer, C., 2009. Absolute plate motions constrained by shear wave splitting orientations with implications for hot spot motions and mantle flow. J. Geophys. Res 114, B10405. Larson, R.L., Searle, R.C., Kleinrock, M.C., Schouten, H., Bird, R.T., Naar, D.F., et al., 1992. Roller-bearing tectonic evolution of the Juan Fernandez microplate. Nature 356 (6370), 571. Macdonald, K.C., 1982. Mid-ocean ridges: fine scale tectonic, volcanic and hydrothermal processes within the plate boundary zone. Annu. Rev. Earth Planet. Sci. 10 (1), 155 190. Macdonald, K.C., Fox, P.J., Perram, L.J., Eisen, M.F., Haymon, R.M., Miller, S.P., et al., 1988. A new view of the mid-ocean ridge from the behaviour of ridge-axis discontinuities. Nature 335 (6187), 217. Maia, M., Sichel, S., Briais, A., Brunelli, D., Ligi, M., Ferreira, N., et al., 2016. Extreme mantle uplift and exhumation along a transpressive transform fault. Nat. Geosci. 9 (8), 619. McCaffrey, R., Zwick, P.C., Bock, Y., Prawirodirdjo, L., Genrich, J.F., Stevens, C.W., et al., 2000. Strain partitioning during oblique plate convergence in northern Sumatra: geodetic and seismologic constraints and numerical modeling. J. Geophys. Res. Solid Earth 105 (B12), 28363 28376. McKenzie, D., O’nions, R.K., 1991. Partial melt distributions from inversion of rare earth element concentrations. J. Pet. 32 (5), 1021 1091. McNamara, A.K., 2019. A review of large low shear velocity provinces and ultra low velocity zones. Tectonophysics 760, 199 220. Molnar, P., 1988. A review of geophysical constraints on the deep structure of the Tibetan Plateau, the Himalaya and the Karakoram, and their tectonic implications. Philos. T. R. Soc. A 326, 33 88. Molnar, P., Tapponnier, P., 1975. Cenozoic tectonics of Asia: effects of a continental collision. Science 189 (4201), 419 426. Montelli, R., Nolet, G., Dahlen, F.A., Masters, G., 2006. A catalogue of deep mantle plumes: new results from finite-frequency tomography. Geochem. Geophys. Geosyst. 7 (11). Morgan, J.P., Parmentier, E.M., 1995. Crenulated seafloor: evidence for spreading-rate dependent structure of mantle upwelling and melting beneath a mid-oceanic spreading center. Earth Planet. Sci. Lett. 129 (1 4), 73 84. Olson, C.J., Becker, J.J., Sandwell, D.T., 2016. SRTM15_PLUS: Data fusion of Shuttle Radar Topography Mission (SRTM) land topography with measured and estimated seafloor topography (NCEI Accession 0150537). Version 1.1. NOAA National Centers for Environmental Information. Dataset. Pearson, D.G., Carlson, R.W., Shirey, S.B., Boyd, F.R., Nixon, P.H., 1995. Stabilisation of Archaean lithospheric mantle: a ReOs isotope study of peridotite xenoliths from the Kaapvaal craton. Earth Planet. Sci. Lett. 134 (3-4), 341 357. Pe´rez-Gussinye´, M., Reston, T.J., 2001. Rheological evolution during extension at nonvolcanic rifted margins: onset of serpentinization and development of detachments leading to continental breakup. J. Geophys. Res. 106, 3961 3975. Ramos, V.A., Cristallini, E.O., Pe´rez, D.J., 2002. The Pampean flat-slab of the Central Andes. J. South Am. Earth Sci. 15 (1), 59 78. Richardson, R.M., 1992. Ridge forces, absolute plate motions, and the intraplate stress field. J. Geophys. Res. Solid Earth 97 (B8), 11739 11748. Royden, L.H., Burchfiel, B.C., King, R.W., Wang, E., Chen, Z., Shen, F., et al., 1997. Surface deformation and lower crustal flow in eastern Tibet. Science 276 (5313), 788 790. Royer, J.Y., Gordon, R.G., 1997. The motion and boundary between the Capricorn and Australian plates. Science 277 (5330), 1268 1274. Sandwell, D.T., Mu¨ller, R.D., Smith, W.H., Garcia, E., Francis, R., 2014. New global marine gravity model from CryoSat-2 and Jason-1 reveals buried tectonic structure. Science 346 (6205), 65 67. Schellart, W.P., Freeman, J., Stegman, D.R., Moresi, L., May, D., 2007. Evolution and diversity of subduction zones controlled by slab width. Nature 446 (7133), 308. Seton, M., Mu¨ller, R.D., Zahirovic, S., Gaina, C., Torsvik, T., Shephard, G., et al., 2012. Global continental and ocean basin reconstructions since 200 Ma. Earth-Sci. Rev. 113 (3 4), 212 270. Sleep, N., 2000. Evolution of the mode of convection within terrestrial planets. J. Geophys. Res. 105 (E7), 17563 17578. Smith, W.H., Sandwell, D.T., 1997. Global sea floor topography from satellite altimetry and ship depth soundings. Science 277 (5334), 1956 1962. Smith, G., McNeill, L., Henstock, T.J., Bull, J., 2012. The structure and fault activity of the Makran accretionary prism. J. Geophys. Res. Solid Earth 117 (B7). Stern, C.R., 2011. Subduction erosion: rates, mechanisms, and its role in arc magmatism and the evolution of the continental crust and mantle. Gondwana Res. 20 (2-3), 284 308. Stern, R.J., Gerya, T., 2018. Subduction initiation in nature and models: a review. Tectonophysics 746, 173 198. Storchak, D.A., Di Giacomo, D., Bonda´r, I., Engdahl, E.R., Harris, J., Lee, W.H., et al., 2013. Public release of the ISC GEM global instrumental earthquake catalogue (1900 2009). Seismol. Res. Lett. 84 (5), 810 815. Strecker, M.R., Alonso, R.N., Bookhagen, B., Carrapa, B., Hilley, G.E., Sobel, E.R., et al., 2007. Tectonics and climate of the southern central Andes. Annu. Rev. Earth Planet. Sci. 35, 747 787. Tapponnier, P., Zhiqin, X., Roger, F., Meyer, B., Arnaud, N., Wittlinger, G., et al., 2001. Oblique stepwise rise and growth of the Tibet Plateau. Science 294 (5547), 1671 1677.

Regional Geology and Tectonics

References

59

Thatcher, W., Foulger, G.R., Julian, B.R., Svarc, J., Quilty, E., Bawden, G.W., 1999. Present-day deformation across the Basin and Range province, western United States. Science 283 (5408), 1714 1718. Tucholke, B.E., Sibuet, J.C., Klaus, A., 2004. Leg 210 summary. In: Proceedings of the Ocean Drilling Program. Turcotte, D., Schubert, G., 2014. Geodynamics. Cambridge University Press. Vigny, C., Ricard, Y., Froidevaux, C., 1991. The driving mechanism of plate tectonics. Tectonophysics 187 (4), 345 360. Von Huene, R., Corvala´n, J., Flueh, E.R., Hinz, K., Korstgard, J., Ranero, C.R., et al., 1997. Tectonic control of the subducting Juan Ferna´ndez Ridge on the Andean margin near Valparaiso, Chile. Tectonics 16 (3), 474 488. Warners-Ruckstuhl, K.N., Govers, R., Wortel, R., 2012. Lithosphere mantle coupling and the dynamics of the Eurasian Plate. Geophys. J. Int. 189 (3), 1253 1276. White, R.S., McKenzie, D., O’Nions, R.K., 1992. Oceanic crustal thickness from seismic measurements and rare earth element inversions. J. Geophys. Res. Solid Earth 97 (B13), 19683 19715. Whitmarsh, R.B., Manatschal, G., Minshull, T.A., 2001. Evolution of magma-poor continental margins from rifting to seafloor spreading. Nature 413 (6852), 150. Zatman, S., Gordon, R.G., Richards, M.A., 2001. Analytic models for the dynamics of diffuse oceanic plate boundaries. Geophys. J. Int. 145 (1), 145 156.

Regional Geology and Tectonics

C H A P T E R

5 Plate kinematic reconstructions Graeme Eagles Alfred Wegener Institute for Polar and Marine Research, Bremerhaven, Germany

Introduction Reconstruction maps showing repositioned continental outlines have been made since early in the history of postclassical cartography. These early reconstructions served simply to illustrate the idea of continental mobility, which was originally considered as no more than a quirk. Accompanying the realization that geological understanding can be applied in Earth system science, the range of uses for such maps broadened as the range of implications of mobile continents came to be appreciated. Using arbitrarily-constructed maps showing the modern continents reconstructed to hypothetical all-polar and all-equatorial configurations, Charles Lyell speculated in his Principles of Geology (1830 33) on the possible paleoclimatological consequences of contrasting latitudinal distributions of land and sea (Fig. 5.1). Using more data-driven supercontinent reconstructions, Alfred Wegener, as part of work on his hypothesis of continental drift, commented on the distribution of glacial and desert paleoenvironments indicated by large-scale sedimentary facies (Fig. 5.2). The reliability and range of uses of reconstructions increased with the interpretational context supplied by the success of the seafloor spreading hypothesis and, onwards, the geometrical discipline of the plate tectonic paradigm through the 1960s and early 1970s. Advances in the understanding and precision of paleomagnetic measurements and in radiometric dating were also hugely influential in dating the rates of change that Lyell had speculated on. The recognition of the comparative geological simplicity of oceanic crust following the GSA Penrose Field Conference of 1972 paved the way for reconstructions that did not rely on patchworks of disputed geological interpretations and correlations, but instead offered the ability to more objectively test and refine them. Together with all this, ever-increasing computing power led to the ability to treat reconstruction data and reconstructions with mathematical rigour and so to present and understand models as approximations to reality with estimable uncertainties. These reconstructions did not all agree with the earlier geological correlation-based reconstructions and in many cases led to revisions of geological understanding and interpretation (compare, e.g. the Indian sector reconstructions of Gondwana by Smith and Hallam, 1970 and Crawford, 1970). Today, high-resolution studies (i.e. featuring multiple closely spaced and/or evenly spaced static reconstructions; Fig. 5.3) give the opportunity to determine plate motions and their changes through time. They can be presented amenably as animations for laypeople and Earth scientists whose expertise lies elsewhere. These high-resolution reconstructions form the basis of seafloor age grids and, based on them in turn, paleobathymetric models that embody important boundary conditions in paleoceanographic modelling studies. Other modern studies see global plate kinematic patterns explicitly embedded as boundary conditions in geodynamic models, allowing to illustrate the plates’ roles in convection of the mantle on a cooling planet.

Regional Geology and Tectonics. DOI: https://doi.org/10.1016/B978-0-444-64134-2.00005-5

61

© 2020 Elsevier B.V. All rights reserved.

62

5. Plate kinematic reconstructions

FIGURE 5.1 ‘. . . what rigors might we not anticipate in a winter generated by the transfer of the mountains of India to our arctic circle!’. Charles Lyell illustrated his question with maps showing the present-day continents in all-polar and all-equatorial distributions. True to his principle of uniformitarianism, Lyell justified the possibility or likelihood of such distributions having occurred at some time in the past as the result of unspecified very slow changes acting over extremely long timescales. His speculations on the climatic consequences were based on his appreciation of the differences in absorption of solar heat by ocean, ice and land surfaces, and the subsequent redistribution of that heat by ocean and atmospheric currents.

FIGURE 5.2 Wegener’s (1929) reconstruction of the North Atlantic region. The dashed lines outline the edges of Quaternary ice sheets based on glacial landscapes and deposits. For this reconstruction, Wegener assumed the presence of a single continuous ice sheet prior to the opening of the North Atlantic ocean. Only later, with the development and improvement of paleomagnetic and radiometric dating techniques, would it become possible to discount the Quaternary ice sheets as a valid constraint on the reconstruction of the much older (Paleogene) North Atlantic.

Regional Geology and Tectonics

Making plate kinematic reconstructions

63

FIGURE 5.3 A set of modern global plate kinematic reconstructions for the past 200 100 Ma period (Matthews et al., 2016). Green fill: Present-day onshore parts of continents. Brown fill: Submerged or subsequently-shortened limits of continental crust. Blue lines: Divergent and transform plate boundary segments. Purple lines: Plate boundary segments with subduction/collision zones.

Making plate kinematic reconstructions The history and workspace of plate kinematic reconstructions The first reconstructions of displaced continents were drawn by hand. Wegener’s global-scale reconstructions of the supercontinents Pangaea, Laurasia and Gondwana, and the oceans Panthalassa and Tethys are prominent examples (e.g. Fig. 5.2). Regional-scale reconstructions were built by moving mapped features across projected maps without laboriously reprojecting them in the process (e.g. Du Toit, 1937) or by more ingenious work with physical globes and transparent or removable overlays (Choubert, 1935; Carey, 1955). With a careful choice of map projection, some of the distortions associated with relocating parts of the planet’s curved surface on flat maps can be avoided completely, but in many circumstances, the distortions involved in moving parts of flat maps with respect to each other were admitted on the assumption of them being smaller than the bounding geological or observational uncertainties (e.g. Barker and Griffiths, 1972). With the first microcomputers, it became possible to rapidly and reliably rotate and reproject mapped features on the surface of a model sphere. First to do this for publication were Bullard et al. (1965), who produced an iconic reconstruction based on least-squares fitting of the conjugate continental shelves of the Atlantic. The following years saw the use of computers for reconstruction work confined to specialist working groups using a wider variety of geological constraints, whose models were made available in printed form as atlases of global and regional reconstructions (e.g. Smith et al., 1981; Ziegler, 1988; Fig. 5.4).

Regional Geology and Tectonics

64

5. Plate kinematic reconstructions

FIGURE 5.4 Regional north Atlantic facies zone reconstruction from the compilation atlas by Ziegler (1988). The reconstructions were drafted manually by transfer of information from regional maps onto computer-generated palinspastic reconstruction basemaps. Palinspastic reconstruction involves the rotation of supposed-rigid (cratonic) regions according to constraints generated from seafloor-spreading data and further correction for the shapes of their edges based on estimates or models of the changes in area and shape that occur as a result of regional deformation (e.g. in extended continental margins or fold-and-thrust belts).

By the mid-1990s, with ongoing improvements in the power and availability of personal computers, interactive computer programs for exploring and developing plate reconstructions started to become available to individual researchers and smaller research groups interested in using them as part of their workflows. The majority of these programs were sponsored or funded by the resource exploration industry, and some were available commercially. Three examples are Cambridge Paleomap Systems’ ATLAS (e.g. Hall, 2002), Torsvik’s GMAP (http:// www.earthdynamics.org/Bugs/GMAP_2012.htm) and the University of Texas’ PLATES (https://ig.utexas.edu/ marine-and-tectonics/plates-project/). The open-source era of the 21st century saw academic and academic/industry-led development of freely available and powerful reconstruction tools and services like those included in GMT (https://github.com/ GenericMappingTools/gmt), ODSN (http://www.odsn.de/odsn/services/paleomap/paleomap.html) and GPlates (https://www.gplates.org/). This availability, along with improvements in speed, ease of use and the range of applications, ensure that cartographically precise plate kinematic reconstructions now play leading and supporting roles in a huge range of industry and academic settings ranging from undergraduate projects to multinational and multidisciplinary studies in Earth system science or resource exploration.

Rotations As noted above, distortions can occur when manipulating features on projected maps. Working to avoid these distortions requires explicit account to be taken of the fact that the manipulations are intended to depict motions on the surface of a sphere. Doing this is possible computationally by making use of the Swiss mathematician Leonhard Euler’s 1776 proof about motions on a sphere. The proof states that there is always one diameter on a sphere whose orientation does not change when the sphere is rotated about its centre point. On Earth, the equator is an example of such a diameter. For plate tectonics, the theorem is restated with reference to an axis, a line normal to an Euler diameter that also passes through the centre of the sphere and exits the surface at two poles. By way of example, the axis associated with the equator is a straight line that passes through both the north and south poles. Motions across the surfaces of spheres can thus be described by defining an appropriate Euler axis in terms of the location of either of its poles and an angle of rotation about that axis. In plate kinematic studies, by convention, positive rotation angles are considered to be right-handed (anticlockwise). Euler’s theorem can be

Regional Geology and Tectonics

Making plate kinematic reconstructions

65 FIGURE 5.5 Relative displacements of a pink and a yellow plate separated by a mid-ocean ridge (double black line), as described by different kinds of rotations. F1 is the displacement described by a finite rotation, in this case designed to unite the continental shelves (thick coloured lines) of the two plates whilst holding the pink plate fixed in its present-day position. The pink plate contributes to the reference frame for rotation Sa as well, but in this case the reference frame is dated at time t2, when the pink plate would have been differently located than at the present day. Sb describes the displacement of the yellow plate in the same period as Sa, but its reference frame is the yellow plate. In this instance, the displacements described by Sa and Sb are intended to be close approximations to the motion of each of the two plates in the period t3 2 t2.

proved and made use of by building matrices populated with trigonometric terms of the axis’ location. The construction and manipulation of these matrices, whilst relatively simple, are repetitive and time-consuming tasks that are best done with computer assistance. Numerical plate kinematic studies that define Euler rotations tend to use geological records of plate divergence as their input data. They distinguish between two main types of rotations (Fig. 5.5), depending on the study aims. Finite rotations reproduce a past state without explicit regard for any intervening states; they depict the consequences of some past motion or motions, but not the motions themselves. Often the starting state is the present day, and the past state is that prior to any plate divergence. Reconstructions made in this way are useful as base maps in geological and paleo-Earth system studies. When the need is to describe the motions of plates, rather than their relative locations, stage rotations are used instead. Plate motions are governed by a balance of continuously evolving torques raised within and around the plates themselves. As the torques evolve, the plate motion they cause also undergoes continuous change. It is unlikely therefore that the instantaneous rotations describing the relative motions of a pair of plates at two different moments in time will occur about the same pole. Because of this, studies that use stage rotations require a simplifying assumption that the system of torques, and with it the stage pole, remained fixed at some level of accuracy appropriate to the aims of the study. In the present-day geographical reference frame, stage rotations for past periods that end before the present occur about differing poles for different plates in a pair. These two poles are separated from one another by a finite rotation for the plate pair that dates from the end of the stage pole period. This occurs because of the need to adopt the two plates’ interiors as the paleo-reference frames for their stage rotations. Given the short time periods necessary to justify the assumption of constant motion, stage rotations are generally by much smaller angles than finite rotations. As the shorter areal displacements achieved by applying such rotations to tectonic markers approach the same length scales as those of observational uncertainty in the displaced markers’ locations, stage rotations tend to be estimated with large statistical uncertainty.

Requirements for reliable reconstructions Generating plate kinematic reconstructions involves characterizing geological, geophysical or morphological markers in their present-day locations and then moving them into coincidence with some estimate or marker of their location prior to their displacement by plate motion. Geometrically, the markers used can be represented as points or contours on maps (Fig. 5.6). Examples of point markers include hotspots, apparent paleomagnetic poles and piercing-point markers where pairs of linear features intersect. Examples of contour markers include magnetic isochrons, shelf-edge depths and continent ocean boundaries (COBs). Doing so requires the assumption that all of the displacement occurred as part of some rigid expression of a plate. As shown in the ‘Choices of markers’ section, this assumption is never fully achieved. Geological markers come of necessity from the plate’s paleo-margins, which by definition are those parts that deform during plate

Regional Geology and Tectonics

66

5. Plate kinematic reconstructions

FIGURE 5.6 Use of points and contours in geometrical fitting for plate reconstructions. (A) Two plates separated by a mid-ocean ridge, with two piercing points of predivergence-aged shear zones with the COBs, Py and Pp, to be used for reconstructing to prespreading times, and two sets of conjugate contours from magnetic isochrons for reconstruction to times t2 and t3. (B) The reconstruction of a pair of points can be very precise, but its accuracy may be very poor each of the five reconstructed shear zones reunites the piercing points within their locational error (purple disc) but only one of them can be correct. (C and D) Fitting of contours consisting of chains of points is unlikely to be possible as precisely for all points within their attendant errors, but its accuracy is good because of the requirement for the rotation to reconstruct both location and orientation. This accuracy improves along with contour length as the orientation (and shape) become increasingly prescriptive.

motion. Virtual geomagnetic poles (VGPs), whilst not required to be located at paleo-plate margins, are pointapproximations of a feature taken from a continuously moving sample set. Hotspots are the surface expressions of mantle plumes that are understood to move laterally within the mantle beneath the moving plates. Detailed understanding of these difficulties leading to careful choices and preparation of reconstruction markers can minimize the effects of nonrigidity, but should be borne in mind when building, comparing, using and interpreting reconstructions. Point-to-point fitting is 100% precise beyond the assumption that the two points were previously coincident. As well as the effects of nonrigidity listed above, the real precision of point-to-point fitting is reduced slightly depending on how precisely the point marker and its target are located. The accuracy of a reconstruction built with point-to-point fitting is, however, unavoidably low by virtue of the fact that a single pair of points can be reunited in one location by an enormous range of rotations (Fig. 5.6B). Locational precision is also important for the accuracy of fitting contours (i.e. chains of points at arbitrary crossings of linear features) together. Added to this, however, the fitting task must also take into consideration the orientation, or shape, of the contour. As contours lengthen, the complexity of their orientation increases, decreasing the range of rotations that might be deemed to fit them together within the precision prescribed by locational errors. A further consideration with linear features on spheres is their curvature. Lines can be presented as segments of great circles, whose centre points are also the centre point of the sphere, or as small circles, whose centre points do not coincide with the centre point of the sphere. Great circles all have the same curvature but sets of small circles display a range of curvatures that vary according to distance from the pole. The choice of circle type to which a linear feature is likened in modelling can thus have a strong influence on the model’s governing rotation poles. This choice can be constrained given relatively straightforward knowledge of the tectonic environment of a linear feature’s formation. Segments of purely divergent plate boundary lie on great circles that all pass through the Euler pole describing the motion the boundary accommodates. Segments of purely conservative plate boundaries lie on small circles. Mid-ocean ridges consist of alternations of these instances in the form of spreading segments and transform faults and thus present a range of unique and complementary constraints on rotation pole locations. A final consideration is the precision and accuracy with which features to be reunited are dated. For this, ‘piercing-point’ features are attractive at face value, because it should be enough merely to demonstrate that they are definitively older than the plate motion in question. In reality, this is difficult to do because of the likelihood of their reactivation or deformation within the paleo-plate boundary zone, to which they must be mapped in order to be useful in fitting (e.g. Gibson et al., 2013). Contour or other point features from within the paleo-plate boundary for a given time slice must be dateable to the time of that boundary’s activity to be useful.

Regional Geology and Tectonics

67

Making plate kinematic reconstructions

Choices of markers Apparent paleomagnetic poles Apparent paleomagnetic poles are features that can undergo test rotations but that are not physically part of the paleo-plate or any of its boundaries. Instead, they are indications of the past location of either the north or south geomagnetic pole as seen by rocks on the paleo-plate. Indications like these are calculated based on measurements of the paleo-inclination and declination of the magnetic field that is preserved in components of the so-called remanent magnetization of rock samples. At the present day, the inclination and declination vary predictably under the assumption that the field they describe is that of a dipole source aligned with the spin axis and with its origin at the centre of the planet. Magnetic poles calculated according to this assumption are known as VGPs. VGPs can be shown to migrate over large distances away from the spin axis through geological time; the migration is referred to as apparent polar wander and attributable to plate motion that has occurred since the date of the VGP. To build plate reconstructions, displaced VGPs, so-called apparent paleomagnetic poles, are rotated back into coincidence with the present-day spin axis pole and/or with a contemporaneous apparent paleomagnetic pole estimate for any or more other plate or plates (Fig. 5.7). Earth’s magnetic field is a consequence of slow thermal convection in the iron-rich outer core, whose movement within the field induces a system of electrical currents that, in turn, maintains the magnetic field. This socalled self-exciting dynamo is ultimately maintained by the size and primordial heat of the planet, which keep the temperature and pressure in the outer core high enough to sustain convection. Owing to the evolving and detailed pattern of convection, the true core field is neither a perfect dipole field, nor is it fixed in time. A measurable consequence of this at the present day is that the poles of the spin axis are not geographically coincident with the geomagnetic poles. Direct observations and determinations based on magnetic observations recorded in historical ship’s logbooks show that this secular variation has seen the geomagnetic pole drift by around 2000 km over the last 500 years. The scatter of VGPs collected for individual plates over comparably short periods in the FIGURE 5.7

Use of apparent paleomagnetic poles for reconstructions of North American (blue) and Eurasian (red) plates. Blue line and dots: apparent polar wander path for North American plate. Red: Eurasian plate. 95% confidence ellipses are shown around the apparent paleomagnetic poles along the paths. F1 is a finite rotation that could be used to reconstruct the Eurasian plate with respect to the North American plate in the period 309 365 Ma. F2 would reconstruct Eurasia with respect to the terrestrial spin axis at 216 232 Ma. Coloured symbols show secular variation of north magnetic pole in the CE 1590 2020 period (data from US Geological Survey). Source: Data from Gordon, R. G., Van der Voo, R., 1995. Mean paleomagnetic poles for the major continents and the Pacific plate. Global Earth Physics: A Handbook of Physical Constants (AGU Reference Shelf), 1, 225 239.

Regional Geology and Tectonics

68

5. Plate kinematic reconstructions

geological past (Butler, 2004) suggests that this kind of drift is a long-lived feature of the field. The effects of secular variation on individual VGPs needs to be accounted for by calculating long-term averages of multiple VGPs in sampling windows of 10,000 1,000,000 years duration. Statistical reliability increases with the number of VGPs, but increasing the duration of the averaging window risks including errors in the VGP locations due to real plate motion occurring whilst the window was open. VGPs, as indirect determinations based on microscopic rock properties, can be affected further by smallerscale processes than plate motion alone. These include structural rotations occurring during or after acquisition of the magnetization, for example by regional- and smaller-scale tectonics and sedimentary compaction, and a diversity of chemical and thermal processes that affect rock magnetization. Structural rotations can be corrected for by numerical adjustments to the measured declination and inclination, based on field-based measurements of geological indicators of paleo-horizontal and, if available, macro- and microstructural strain indicators. Smallerscale chemical and thermal effects can be mitigated by careful use of demagnetization techniques, which aim to remove the magnetization components expressed by the rock’s various populations of mineral carriers on a stepby-step basis until the desired initial magnetization state, or ‘characteristic remanent magnetization’, is isolated for the determination of a VGP. Given the range of factors that can affect VGP determinations, it is essential that a number of criteria are used to assess their goodness. A reliable VGP should be determined from measurements repeated in a large number of individual samples, and from a large number of sites, dated to within an acceptable time range, in order to generate a meaningful long-term average pole; to be useful, the statistical confidence region for this pole should not be too large. The exact criteria for establishing confidence vary from laboratory to laboratory. Besse and Courtillot (2002), for example, describe criteria for sampling (at least six sites per pole and at least six samples per site), for statistical confidence (95% confidence interval smaller than 15 degrees), for laboratory procedures and tests (evidence for successful alternating-field and/or thermal demagnetization), and for dating (maximum age uncertainties of 15 Myr). In this latter case, it is worth noting that a 15 Myr dating error would translate to a plate location error of 150 1200 km of north south plate motion assuming spreading rates in the range typical of the major plates during the last 200 million years. When determined from small regions, plate kinematic interpretations of VGPs can be ambiguous and therefore contentious. The reason for this is that rotations due to structural deformation can affect rocks within paleo-plate boundary zones. These zones can be of a similar width to some small plates, or at least the continental components of those plates, leaving open the possibility to interpret their VGPs either in terms of lithospheric plate rotation or local vertical-axis rock rotations. Prominent examples of contested or enigmatic rotations in such regions include those of the Falkland Islands (Mitchell et al., 1986; Richards et al., 1996) and the Patagonian orocline (Poblete et al., 2016; Eagles, 2016). Intraplate volcanic chains/hotspots Thermal and/or chemical contrasts between mantle rocks are thought to play a role in convection of parts or all of the mantle below the lithosphere. Melt may be delivered to the lithosphere as a consequence of decompression where these plumes are positively buoyant and rise. A proportion of this melt may be evident at the surface as volcanic rocks. Assuming the rising plume has little or no horizontal motion with respect to the mantle it ascends through, a plate moving over it may gain an age-progressive sequence of such volcanic rocks (Fig. 5.8). Dating the volcanic trace with radiometric techniques allows for the possibility to rotate different-aged points along the traces to one another, thus giving an idea of the plate’s progress over the mantle below. In its simplest form, this technique involves point-to-point fitting, with its attendant nonuniqueness (cf. Fig. 5.6B). It works best for plates that have been affected by several plumes over a long time period, increasing the number of time steps for which multiple points must undergo simultaneous rotation and fitting. The best example of this is the African plate, over which numerous age-progressive tracks can be identified in the period back to B130 Ma (O’Neill et al., 2005). The precision of this technique is modulated by the observation that the diameters of intraplate volcanoes and seamounts may be several tens of kilometres (e.g. the size of the Big Island of Hawaii, beneath which the Hawaiian hotspot is assumed to lie, is B90 km). This leads to ambiguity in the location of the plume at any given stage in its volcanic history. Added to this, plume diameters and structures themselves are not known well from seismic studies, opening the question of whether a single plume may give rise to multiple discrete contemporaneous volcanic centres and making it difficult to determine the locations of some present-day hotspots. The duration of plume-related volcanism is another complicating factor. With dated products in many locations ranging over tens of millions of years it is possible that less-well sampled sites may not be attributed an accurate age of

Regional Geology and Tectonics

Making plate kinematic reconstructions

69

FIGURE 5.8 Age progression of seamounts and islands of the Hawaii-Emperor seamount chain. The Emperor chain is the N S segment between Detroit and Koko seamounts. The progression from old to young magmatism ages is interpretable in terms of the Pacific plate’s motion over a fixed melting cause in the mantle below. Interpreted in this way, the Pacific plate seems to have moved rapidly northwards over the mantle between 85 and 43 Ma and, after an abrupt change to NWdirected motion, rather more slowly.

volcanic and plume-influence onset. Another question is that of whether the assumption that plumes do not move laterally within the mantle is tenable. A row of studies has worked on the possibility that it is not (e.g. Tarduno et al., 2003; Steinberger and Antretter, 2006). The plume-fitting technique is the basis for constructing a plate motion reference frame with respect to the centre of the planet and not to any point on its surface (‘Hotspot reference frame’ section). Magnetic reversal isochrons Fig. 5.9 shows an excerpt of a global magnetic anomaly compilation map for the North Atlantic. The grid is based on computer-generated interpolations between magnetic anomaly calculations made using measurements taken along thousands of magnetic anomaly data profiles collected by ships and aircraft; an example of one such profile is shown at the bottom of the figure. The map and the profiles reveal a pattern of magnetic anomaly peaks and troughs that are symmetrically arranged east and west of the central and northern Mid-Atlantic Ridge. The source of the anomalies is largely but not completely confined to a 0.5 1.0 km thick layer of pillow lavas near the top of the basaltic crust that erupts all along the linear mid-ocean ridge segments (Banerjee, 1984). The lavas cool rapidly and preserve the direction and strength of the geomagnetic field at the time of cooling in their ironrich minerals. With ongoing volcanism and plate motion, this information is gradually transported out into the plate interiors. The field information sampled by magnetometers is strongly influenced by previous occurrences of geomagnetic reversals. In these geologically rapid events, which are complex and apparently random properties of the field generated by convection in the outer core, the polarity of the core field reverses. Each reversal leads to a near-180-degree change in the magnetic field recorded at the mid-ocean ridge. The sequence and timing of these polarity reversals are known at high resolution from radiometric studies of onshore volcanic sequences and can be further interpolated using the assumption of orbital control on cyclicity in sedimentary sequences with good paleomagnetic records of their own. In the oceanic crust, this known and dated reversal sequence can be combined with knowledge of the present-day field and seafloor spreading rates to model the expected shapes of magnetic anomalies formed at any given separation from the mid-ocean ridge. At many mid-ocean ridges, the erupted lavas flow relatively short distances before solidifying and may be confined in doing so by the walls of a median valley or axial summit caldera. The eruption frequency is in many locations high enough to maintain a constant thickness of as much as 1 km of solidified lava. These circumstances

Regional Geology and Tectonics

70

5. Plate kinematic reconstructions

FIGURE 5.9 Magnetic reversal isochrons in the central and North Atlantic. EMAG2.3 grid data (https:// doi.org/10.7289/V5H70CVX). The grid is generated from hundreds of magnetic anomaly profiles collected by ships and aircraft and at much longer wavelength by satellites in low orbits. Bottom panel shows a ship’s profile (Ocean Drilling Program leg 150, https://doi.org/ 10.2973/odp.proc.ir.150.1994) and two picks of the conjugate 34y (B84 Ma) reversal anomaly.

combine to ensure that the transitions from normally to reverse-magnetized basaltic crust occur over a distance of a few kilometres, a value which can be assumed to contribute to the locational accuracy of the paleo-plate edge. Anomaly shape is also important, changing markedly with ridge orientation, latitude and paleolatitude, basalt chemistry, and possibly in response to the thermal effects of thick sediments (Granot and Dyment, 2019). A point along a magnetic profile where the anomaly shape marks an underlying field reversal to be preserved can be assigned a date, known or modelled for the reversal based on magnetization studies of onshore or drilled rocks as part of a magnetic reversal anomaly timescale, and is referred to as an isochron pick. Alignments of these picks can be referred to as magnetic reversal isochrons. Thousands of anomaly picks and isochrons have been made throughout the oceans and used and presented in hundreds of publications (e.g. Seton et al., 2014). Contrasting interpretations of anomaly sequences are not uncommon and mean that any available set of profiles might deliver multiple sets of conflicting picks (e.g. in the region south of Australia: Weissel and Hayes, 1978; Cande and Mutter, 1982; Eagles, 2019). The Global Seafloor Fabric and Magnetic Lineation Data Base Project (GSFML; http://www.soest.hawaii.edu/PT/GSFML/) curates many of these picks. Fracture zones Fig. 5.10 shows examples of fracture zones in bathymetric and gravity data sets from the region south of Australia and New Zealand. Fracture zones are the scars left in plate interiors by the action of transform faults that offset mid-ocean ridge segments. This transform faulting occurs at the ends of spreading segments where melt supply may be either reduced or enhanced, leading to crustal thickness contrasts across transform faults. Tectonically, however, the transforms themselves are relatively short-lived zones of pure strike-slip faulting developed in newly formed and thus relatively homogeneous oceanic crust. This comparative lack of complexity means that fracture zones tend to appear as simple topographic troughs, and in some circumstances steps, populated by strike-slip fault systems. Active transform faults at the present-day trend to within less than 3 degrees of the orientations expected of plate motion as determined by earthquake slip vectors. The variation is, however, systematic, with left-lateral transforms trending clockwise of slip vectors and right lateral transforms anticlockwise. The reasons for this may lie in the detailed structure of the mantle near transforms and how it affects the rupture process during earthquakes. These biases appear however not to strongly affect determinations of present-day plate motions, which make extensive use of earthquake data (DeMets, 1993). At any instant, transform faults along an active mid-ocean ridge crest align along small circle segments common to the instantaneous pole of rotation. Carefully chosen finite lengths of fracture zones or sets of fracture

Regional Geology and Tectonics

Making plate kinematic reconstructions

71 FIGURE 5.10 Fracture zones. On left, as present in global predicted bathymetry data (Smith and Sandwell, 1997). On right, more prominent in vertical gradient of gravity (Sandwell et al., 2014). (1) Anomaly is asymmetrical at long-offset features, (2) symmetrical VGG anomaly at short offset, (3) symmetrical with amplified marginal peaks where the fracture zone has experienced cross-axis strain and (4) Sediment-filled fracture zone troughs with no bathymetric signature are still evident in gravity owing to density contrasts between sediments and basement.

zones, therefore, can be useful in determining the locations of stage rotation poles for plate kinematic reconstructions. Changes in the locations of these poles are recorded by changes in transform orientation, which become manifest in the plate interiors as bends and kinks in fracture zone traces. The longer a transform offset, the more complicated the local tectonic changes that achieve the change in its orientation need to be. This can lead to the development of transpressional ridges or transtensional basins as previously strike-slip fault segments along the transform begin to accommodate components of strain across their axes, corrupting the signal of plate motion orientation change (e.g. Bonatti et al., 2003). The observational accuracy in the locations of fracture zone troughs determined from satellite altimeter-based gravity data is around 5 km, based on careful comparison with high-resolution bathymetry data (Mu¨ller et al., 1991). Another source of error is interpretational: a range of tectonic processes in the formation and evolution of oceanic crust can lead to linear features that resemble fracture zones. Fig. 5.11 shows some examples. These features can often be distinguished from fracture zones on the basis of their strike, which is unlikely to be parallel to the set of fracture zones formed in neighbouring parts of the same plates. Continent ocean boundaries The complementary shapes of the opposing edges of pairs of continents are observations that have most often been cited in support of the idea of crustal mobility on Earth. Combined with this, the recognition of chemical and physical divisions of the crust into buoyant continental crust rich in aluminium silicates and less buoyant, magnesium silicate-rich, oceanic crust led to the expectation that it should be possible to map the separated edges of continents with some accuracy. These so-called COBs are frequently used as linear markers to generate predivergence aged finite rotations. Geochemically based mapping of COBs is, however, not a realistic prospect, given that basement rocks at continental margins are widely covered by thick layers of sediments shed by rivers and glaciers from the continental interiors. Instead, therefore, a long history of studies has chosen to map COBs using second-order interpretations based on contrasting physical properties, themselves interpreted from geophysical observations (Fig. 5.12). Perhaps the simplest of such interpretations is based on the expectation for the buoyancy and thickness contrasts between continental and oceanic crust to give rise to bathymetric variability, with COBs being mapped from bathymetric slopes. In another approach, strong flat seismic reflections from the Moho discontinuity at the base of the crust have been assumed to be indications of the presence of oceanic crust, based on the expectation of its

Regional Geology and Tectonics

72

5. Plate kinematic reconstructions

FIGURE 5.11 Tectonic features that might be mistaken for fracture zones, in vertical gradient of gravity data (Sandwell et al., 2014). Yellow: A set of fracture zones formed at transform faults along the Pacific-Antarctic Ridge (red) in the southern Pacific ocean. Other linear features that did not form in this way are highlighted by blue lines, (1) A pseudofault, the scar left by a migrating second-order offset of the ridge crest, (2) an abandoned segment of ridge crest (Cande et al., 1982), (3) a linear ridge formed by interactions between a mid-ocean ridge segment and excess melt delivered from a nearby off-axis source (Okal and Langenhorst, 2000) and (4) a triple junction propagation trace (Larter et al., 2002).

FIGURE 5.12 A global compilation of COB location estimates at extended continental margins. The mean widths across the ensembles of estimates are shown next to each margin. The most closely clustered set of estimates is also the smallest (eight members), suggesting a tendency for a large component of uncertainty in COB locations to be attributable to contrasting interpretations of multidisciplinary data sets. Source: Data from Eagles, G., Pe´rez-Dı´az, L., Scarselli, N., 2015. Getting over continent ocean boundaries. Earth-Sci. Rev. 151, 244 265.

formation by magmatism involving the presence of a long-lived steady-state magma chamber at whose base gabbro cumulates form in seismically-reflective horizontal sheets. Other geophysical approaches have mapped COBs based on contrasting ‘continental’ and ‘oceanic’ seismic velocity stratigraphies modelled in refraction data or rock density and magnetization contrasts modelled in potential field data. Features like this might be expected to be locatable with relatively high accuracy in geophysical data, and yet where groups of COB interpretations made using multiple variable data sets can be compared, they tend to disagree on the location within a zone that is on average 165 km wide, and in many cases much wider (Eagles et al., 2015). The ages of COBs, a crucial attribute if they are to be used for plate kinematic studies, are also determined by second-order interpretation by extrapolation from the last identified magnetic reversal anomaly or by reference to the ages of ‘breakup unconformities’. These unconformities are likely however formed in response to rheological changes in the evolving plate boundary zone that may not be straightforwardly relatable to changes in crustal lithology or formation processes at divergent plate boundaries. This indicates overall that COBs are unlikely to be a meaningfully measurable starting point for high-resolution plate kinematic studies. Piercing points from intracontinental structures Pairs of ‘piercing points’, the locations where prekinematic lineaments are truncated at continental margins, have been long and widely used for generating or validating plate reconstructions (Fig. 5.6; Choubert, 1935; White et al., 2013; Aitken et al., 2014). To be considered geometrically useful the lineaments producing piercing points should be defined by sharp, near-vertical geological contrasts. The sharpness of the truncated lineaments

Regional Geology and Tectonics

Making plate kinematic reconstructions

73

FIGURE 5.13 Partial Gondwana reconstruction map from De Wit et al. (2008) showing the locations of ten pairs of piercing points at the coastlines of the South Atlantic. Note the long cross-shelf distances that some are interpolated over.

is necessary to ensure that points determined along them are indeed laterally equivalent, and the criterion that structures should be vertical ensures they cannot have been relocated with respect to the continental interiors, or each other, by syn- or postkinematic erosion to different structural levels. In practice, these criteria must be validated using offshore geophysical data where prekinematic lineaments are truncated by the COB. The indistinctness of the COB concept makes this task difficult to complete with much precision. In addition to this, for confidence in the piercing points, De Wit et al. (2008) noted the requirement that geological interpretations of any given pair of prekinematic lineaments should be both congruent and subjects of consensus. He discussed one example of the kind of difficulty this might lead to (annotated ‘10’ in Fig. 5.13). In this example, the African correlative to vertical shear zones at the southern edge of the Dom Feliciano Belt in Uruguay has been suggested to run either along the southern edge of the Damara Belt or possibly NE edge of the Saldania Belt, several hundred kilometres further south. White et al. (2013) describe similar controversies from the conjugate margins of the SE Indian Ocean. Based on these considerations, piercing points are unlikely to offer a strong starting point for highresolution plate kinematic reconstruction work. Where possible, and as with COB segments, their use should instead be restricted as coarse end-member constraints on the very earliest phases of models based on seafloor spreading data (e.g. Pe´rez-Dı´az and Eagles, 2014). Diffuse and regional markers As well as continental tectonic markers, a variety of other constraints have been brought to bear on plate reconstructions. Early on, rudimentary biogeographic zones in the form of the ranges of fossil Mesosaurs (e.g. Wegener, 1929; du Toit, 1937) were taken as evidence of former continental connectivity. The Mesosaurs’ ranges included parts of South Africa and South America in Permian times, but as limited swimmers confined to coastal habitats their fossil distribution was inconsistent with the present-day separation on each side of the South Atlantic. Later work using fossil trilobite assemblages each side of the North Atlantic related the rate and amount of evolutionary divergence to the growing width of that ocean’s Paleozoic predecessor, Iapetus (e.g. Cocks and Fortey, 1990). Climate zones have been used as markers, as revealed by climatically sensitive lithologies, mineralogies or assemblages, such as moraines, glendonites, cap dolomites and other glacial-deglacial assemblages (e.g. Wegener, 1929; Adie, 1952). Continental flood basalts and their related igneous complexes and dyke swarms represent a further class of markers that have been widely used because of their distinctive and precisely dateable

Regional Geology and Tectonics

74

5. Plate kinematic reconstructions

FIGURE 5.14 Differing reconstruction locations for Sri Lanka (SL) in Gondwana under differing constraints. (A) Yoshida et al. (1992), who correlated metamorphic grades and P-T paths between Proterozoic rocks of SL and Antarctica. The reconstruction also aligns Permian basins in the Mahanadi and Lambert grabens (MG and LG) and is widely used in the literature (see Gibbons et al., 2013, for an influential recent example). (B) Crawford (1970), who correlated cratons and mobile belts (which he termed ‘blocks’) determined using geochronology on rocks from India, Australia and SL, as well as a set of younger basins including in the Damovar Valley (DV) and Fitzroy Trough (FT). This arrangement is closely comparable to that of Wegener. (C) Eagles and Ko¨nig (2008), who used fits of the shelf edges in India and SL to those of the remainder of East Gondwana after it had been reconstructed using joint inversion of fracture zone and magnetic data.

lithologies that are considered to be intimately related in time to the continental breakup process (e.g. Torsvik et al., 2009 for the Etendeka and Parana´ flood basalts at the opposing margins of the South Atlantic or Coffin et al. (2002) for the Bunbury and Rajmahal Traps basalts of SW Australia and NE India). Elsewhere, distinctive or rare metamorphic zones and metamorphism dates have been taken as leading or supporting constraints on prebreakup relative continental positions (e.g. Yoshida et al., 1992 for Sri Lanka and Lu¨tzow-Holm Bay in Antarctica). These constraints tend to be regional and geographically nonprecise, either because of incomplete sampling or by their very nature (climate zones are broadly constrained by planetary-scale atmospheric circulation patterns in belts tens of degrees of latitude wide; the widths of metamorphic zones may be determined by the locations of thick and/or hot crust in collisional plate boundaries integrated over the several tens of millions of years of their action). Despite this, constraints like these still play leading, and sometimes controversial, roles in numerous regional reconstructions. Fig. 5.14 uses the location of Sri Lanka in Gondwana as an example. Early geochronological work on Precambrian metamorphic ages suggested a location off the western coast of Australia (Crawford, 1970), but more detailed comparisons came to suggest correlations with the Lu¨tzow-Holm Bay region in Antarctica (Yoshida et al., 1992). This correlation can be supported by VGPs determined on a small number of sites and samples in Sri Lanka (Funaki et al., 1990). Studies aiming to match COBs can be taken to support this location (e.g. Gibbons et al., 2013) or, by changing their emphasis, in turn to suggest a third possible location to the east of Antarctica’s Napier Peninsula (Eagles and Ko¨nig, 2008).

Regional Geology and Tectonics

Making plate kinematic reconstructions

Mantle constraints

75

subducted slabs in tomography

Another source of constraint, from within the mantle, has become possible to use in recent years with improvements in the availability and reliability of travel-time tomography models from teleseismic measurements (Thurber et al., 2007). These models are built using various approaches to recognizing and mapping the causes of small deviations in the arrival times of earthquake energy transmitted along multiple paths between thousands of earthquake-receiver pairs. These deviations are typically attributed to the distribution of thermal and chemical heterogeneities in the mantle, which in turn are interpretable in terms of the distribution of slabs of subducted oceanic lithosphere and of buoyant plumes involved in mantle convection. Analyses of the lengths and locations of slabs have been taken as constraints on the widths of partially or completely subducted oceans and marginal seas. Taken together with assumptions about or forward models of the rate of subduction and geological evidence for the past locations of trenches, these analyses can place quantitative limits on past plate motions. This approach has proved most valuable in regions affected by relatively recent subduction, where the interpretations of slab remnants in the upper mantle are more readily justifiable than those at greater depth (e.g. Handy et al., 2010; van der Meer et al, 2018). Handy et al. (2010) suggested that a 30% shortfall between the area of slabs interpreted in tomography when compared to the area of lithosphere expected to have been lost to subduction on the basis of a reconstruction model should be acceptable on the basis of limitations in interpreting the tomographic images. Disagreements and differences between tomographic models are widespread and can be related to the specific uncertainties arising from the data sets and approaches used to generate them. It is crucial to understand these uncertainties, as they can lead to the production of model artefacts that resemble slabs. The lower mantle is especially difficult to interpret with great confidence in the face of these problems, casting doubt on the usefulness of slab-based constraints for very long timescales. Attempting to overcome this obstacle, Shephard et al. (2017) consider an ensemble of models in the search for consistently modelled slabs in the lower mantle. They find that the most-consistently resolved slabs are those related to fast recent or long-lived and ongoing subduction west of the Americas and in the former Tethyan realm at depths of 1000 1400 km. In addition to this, they concluded slabs in depths of 1400 2000 km may be routinely less well imaged over the range of models they investigated, or alternatively that changes in the amount of subduction with time have given rise to a heterogeneous global vertical distribution of slabs.

Model production and assessment Once a set of constraints has been decided on and gathered for a reconstruction project, any of a variety of modelling procedures and workflows can be followed. The previous section “The history and workspace of plate kinematic reconstructions” described how, classically, this involved working with carefully chosen flat maps but how, with advances in computer power, it became possible to work quantitatively in the face of the difficulties of spherical geometry. By now, two broad patterns of approach can be followed and, ideally, combined in such a way that their advantages mutually reinforce for the highest confidence in model interpretation. Interactive visual forward modelling The first of these approaches is an interactive forward modelling approach. The modeller’s chosen set of constraints are loaded into a Geographical Information System-like computer modelling package such as GPlates. Rotation parameters are then applied from a database to reconstruct those constraints at the time of interest. The reconstruction is then altered, either by changing the rotations in the database or by a click-and-drag procedure on individual mapped features, in order to visually improve the fits of the constraints to one another or to some notion of the geological setting that the map is expected to portray. This procedure can be repeated or expanded for large numbers of constraints, ideally ones that each reduce uncertainty by mutually excluding alternative models suggested by the others, until the modeller is satisfied with the product. Fig. 5.15 shows an example of this procedure in action using the GPlates software (https://www.gplates.org). At the time of writing, GPlates is the leading product for forward modelling of plate kinematics and generation of reconstruction maps, by virtue of its open-source licence that makes it free to download and use and its very wide range of functions and tools for modelling and depicting plate kinematics. Other well-known and widely used examples of large forward models are those produced for the fragmentation of Gondwana by Colin Reeves (e.g. Reeves et al., 2016) and for the complex plate tectonic development of SE Asia by Robert Hall (e.g. Hall, 2002), both using the Cambridge Paleomap Services ATLAS software.

Regional Geology and Tectonics

76

5. Plate kinematic reconstructions

FIGURE 5.15 A GPlates screenshot working with its included dynamic topography data and a selection of plate-outline polygons for plates and continental blocks at the margins of the south, central and north Atlantic. The polygons, here representing constraints for an Atlantic-centred regional plate reconstruction, can be altered by digitization within the application.

Statistical modelling Here, the best-fitting set of Euler rotation parameters for uniting a given set of constraints is sampled or searched for by a computer. A grid search approach may test all available combinations of rotation parameters for how closely they unite features, or a best-fitting set of parameters might be approached by adjusting the rotation parameters incrementally in ways to improve the overall misfit of the constraints to some modelled estimate of their true shape or location. First to do this were Bullard et al. (1965), who united shelf-edge contours in the Atlantic at a variety of contoured depths, noting the fit at 500 fathoms to be most convincing. Their aim was merely to quantify to what degree the continental margin shapes on opposing margins of the Atlantic were conjugate shapes, which at the time was still a matter of debate. Although the field has moved on from the hypothesis of conjugacy, the numerical approach retains a strong advantage in that it allows for an objective statistical assessment of the model’s reliability. Subjective choices still play a role in the model design stage, when a feature set is chosen to model and the ages and locational errors of the features are assigned. In general, it can be expected that the more predictable or reproducible (that is to say, simpler) the shapes of the features chosen to fit are, the more confidence there can be that an appropriate interpretation is being applied for the modelling. In view of this, the statistical modelling techniques are widely applied to magnetic isochron picks, with their predictable and near-cylindrical (i.e. stackable) waveforms, and fracture zone crossings that appear as simple long linear troughs or steps in gravity or bathymetry data. The best known of these techniques is that of Hellinger (1981) (Fig. 5.16A). It has been widely studied and used and is embedded in GPlates. Thanks to numerous extensions built on the work of specialists in the statistics of spherical geometry, it is a very powerful technique (e.g. Chang et al., 1990). It is, however, limited to the use of data in conjugate sets, and to the use of incomplete samples of fracture zones. Engebretson et al. (1984) and Shaw and Cande (1990) overcame the fracture zone-related limitations by modelling entire fracture zone shapes using the notion of fixed points on ridge crests moving relative to the interiors of the two plates flanking the ridge. Nankivell (1997; Fig. 5.16B) extended this methodology with a provision for the rotation of isochron picks to nonconjugate targets. This approach provides the ability to use data from the

Regional Geology and Tectonics

Making plate kinematic reconstructions

77

FIGURE 5.16 Two approaches to determining finite rotations from seafloor spreading data on a pair of plates, a and b. Hellinger’s approach relies on the availability of a conjugate pair of pick sets (C1) from either side of a divergent plate boundary (double line). Picks are taken in magnetic anomaly profiles and, along fracture zones (thin grey lines), in bathymetric, magnetic or gravity data. The two sets are rotated towards coincidence with one another by a finite rotation (Fab1), and their misfits to a set of great circle segments are calculated for iterative least-squares minimization. Nankivell’s technique can also make use of data without conjugates by calculating a set of stage rotations for each plate (e.g. Sa01, Sa12, Sb01, Sb12) that can be summed for finite rotations. This approach allows the use of data from along the full lengths of fracture zones.

complete traces of fracture zones and leads to considerably smoother models than those built with Hellinger’s criteria, which can only use parts of the fracture zone traces. This smoothness has been suggested as a locally undesirable consequence of the tectonic alteration of transform faults occurring around the times of rapid plate kinematic changes (Mu¨ller et al., 1999). Since then, the smoothness of rotation pole sequences generated with extensive fracture zone fitting has been shown to be comparable to that generated by the application of sophisticated Bayesian inference-based noise reduction techniques on rotation pole sequences generated from Hellingertype models (Iaffaldano et al., 2014).

Regional/global models with multiple plates Most of the techniques described above are used to derive rotation parameters for pairs of plates. Multipleplate statistical reconstruction schemes also exist and can make use of the powerful constraint that the sum of relative motions across a chain of plate boundaries that begins and ends with the same plate should be zero (e.g. Cande and Patriat, 2015; Nankivell, 1997). Building models such as this is not a simple task, in no small part because of the difficulty of harmonizing data sets in such a way that all of the modelled plate pairs offer constraints for all of the modelled times. Forward models for multiple plates, in contrast, are easier to build and extend. A database depicting a hierarchy or tree of rotations describing relative motions (Fig. 5.17) is used. From its outermost branches to its trunk, the database prescribes the order in which rotations for pairs of plates are summed in order to generate relative motions between nondiverging or nonneighbouring pairs. Fig. 5.17 shows that it is possible to build these hierarchies in alternative ways, for example for South American-East Antarctic motions via the Weddell Sea, instead of the South Atlantic and SW Indian Oceans, for Australian-East Antarctic motions via the Wharton, NW and central Indian Oceans instead of the SE Indian Ocean or for motions between east and west Antarctica via the SE Indian Ocean, Tasman Sea and South Pacific Ocean instead of the West Antarctic Rift. Exercises in all three show that the resulting sets of motions are not identical, despite the expectation from considering plate rigidity that they should be (Steinberger et al., 2004; Eagles, 2016, 2019). In some cases, the reason for this may be misinterpretation of geophysical data on one or more legs of the circuit, leading to falsely located isochron pick sets, or to their attribution or nonattribution to inappropriate plates. In others, the differences may merely express some combination of the uncertainties inherent in building reconstructions in general and their amplification when applied through a series of rotations. Ideally, forward models should be tested for internal consistency by

Regional Geology and Tectonics

78

5. Plate kinematic reconstructions

FIGURE 5.17

Part of Mu¨ller et al.’s (2016) hierarchy of plate pairs for some of their post-Jurassic global plate reconstructions, indicated by plate names in boxes and, in blue, the locations or names of divergent plate boundaries from which their relative motions are calculated. Within that study, the hierarchy changes according to reconstruction age and the availability of constraints. Outside of that study, alternative hierarchies have been used. Three examples are illustrated in magenta.

balancing multiple possible pathways through the appropriate hierarchy in such a way that inacceptable fits or motion attributes are eliminated or minimized to smaller than observational or model resolution. For the most part this is undertaken by visual means only, by examining the fits of features along alternate paths as they appear in reconstruction maps built using the main or preferred path (e.g. Gibbons et al., 2013). The top level of the hierarchy in Fig. 5.17 describes motions not between any pair of plates, but between one plate (Nubia) and Earth’s spin axis. Motions like this can be derived in a variety of ways, including by the hotspot and paleomagnetic reconstruction techniques described in the ‘Intraplate volcanic chains/hotspots’ and ‘Apparent paleomagnetic poles’ sections. With increasing geological age, undeformed constraints on the relative motions of pairs of plates, such as can be derived from basins, continental margins or oceans, become less numerous and confined to ever-smaller geographical ranges so that their mapping, interpretation, and dating may become more contentious. Because of this, hierarchies for more ancient times tend to appear flatter than in Fig. 5.17, with more and more elements reconstructed solely with respect to the spin axis via VGPs, whose determination requires only outcrop-sized extents of rock to have survived to the present day. The resolution of VGPbased reconstructions is inherently lower than that of two-plate reconstructions based on seafloor spreading data and not necessarily of any higher resolution than chain-additions of models of divergent pairs. Hence, their main value is for pre-Jurassic times for which large areas of oceanic crust with isochrons and fracture zones have been lost to subduction. Absolute plate motion and reference frames for reconstructions Without its top row, the hierarchy in Fig. 5.17 is rooted in its Nubian plate, meaning that any reconstruction built using it for any age would show the Nubian plate in its present-day position, for instance with Cape Town near 34 S, 18.5 E and Cairo near 31 N, 31 E. A map produced in this way for late Carboniferous and Permian times would challenge its users to explain how and why a continental-scale ice sheet, whose presence at that time is betrayed by striated glacial rocks and tillites in outcrops along the western Cape, could have formed and covered Cape Town at such a low latitude. A Carboniferous-Permian reconstruction would be easier to understand and use if it could also take account of the fact that the Nubian plate has not only moved with respect to the other plates, but also with respect to the spin axis, taking it away from a near-polar location in Carboniferous times. Evidently, some way of fixing the hierarchy within a latitudinal framework, another reference frame, is needed. Hotspot reference frame

One widely used reference frame is that of the vertically ascending plumes in the convecting mantle (‘Intraplate volcanic chains/hotspots’ section; Fig. 5.18). Although it consists of silicate rocks at very high temperatures, the great pressure at depth in the mantle means that these rocks do not exist in a molten state. Their viscosity under the prevailing pressure and temperature regime is nonetheless low enough that they flow like a fluid on geological timescales. This capacity for flow furthermore means that the mantle rids itself of the heat of

Regional Geology and Tectonics

Making plate kinematic reconstructions

79

FIGURE 5.18 Top: Sketch map illustrating the basis of a mantle-based reference frame for plate motions. A source for the generation of excess quantities of volcanic rocks (at the surface, a so-called hot spot) is situated in the mantle beneath plate a. Continuous partial melting at the hot spot has produced a chain of seamounts (triangles) in plate a. At time 1 before present (orange seamount), the hot spot was situated near the centre of plate a. By time 0 (red seamount), the hot spot was situated at its SW corner. Assuming the hotspot exists over an ascending plume with no lateral motion through the mantle, it can be stated that plate a followed a NE-directed path over the mantle in the period between times 1 and 0. Over the same period, plate b has been moving westwards with respect to plate a, as shown by seafloor spreading data on the flanks of their shared mid-ocean ridge (double red line at time 0). Plate b has no hot spot or seamount chain of its own, but its overall NW-directed velocity over the mantle between times 1 and 0 can be resolved by addition of its velocity with respect to plate a to the velocity of plate a with respect to its hot spot. Moderate changes in plate a’s motion over the mantle prior to time 1 resolve to a complex displacement history between plate b and the mantle involving a period of near still-stand at times 1 2 between two NW-directed phases of faster motion. This history of interaction with the underlying mantle may have left products in plate b’s geological record that do not correlate with any observable change in its motion relative to plate a.

radioactive decay generated within the silicate mantle, as well as that of conduction from the outer core, by convection. Fluid dynamic principles suggest that the upward component of this convection should take the form of cylindrical plumes of rock with relatively lower viscosity than that in the surrounding mantle. However, it is not clear whether this upwelling dominates the convection process or whether it is more strongly modulated by plate tectonics, which constrains the pattern of downwelling to the sheets of cold material that enter the mantle at subduction zones. Added to this, the mantle is internally layered, for example with a prominent mineral phase boundary close to 670 km depth that alters the buoyancy contrasts between cooler and warmer rocks, hindering the sinking of slabs. The precise style of mantle convection is the subject of ongoing research in seismology, geochemistry and numerical modelling (e.g. Langemeyer et al., 2018). Regardless of the exact pattern of their ascent, the upwelling warm rocks undergo decompression, which promotes partial melting and can eventually give rise to surface volcanism. The volcanic rocks are dateable by geochronological techniques. The dates thus constrain the times at which an upwelling plume might be expected to have been present in the mantle beneath the site from which the volcanic rock sample was taken. Assuming that the motion of these upwelling plumes was dominantly vertical within the mantle (i.e. radial as seen from the centre of the planet), then they can be thought of as fixed with respect to the spin axis and fixed with respect to each other. In this way, plume-related volcanism can be used to place plate motion in an independent observational reference frame. This reference frame is largely limited to use in the period since 100 Ma, as the record of hotspot volcanism from older times is too fragmentary. Early work on the hotspot reference frame was undertaken using low-resolution plate kinematic models and relatively few and relatively unreliable dates on intraplate volcanic rocks (Morgan, 1971). With increasing precision in kinematic models and more and better intraplate volcanic dates, it became possible to discern a misfit between those global reference frame solutions developed from the starting point of intraplate volcanism and relative plate motions in the Indo-Atlantic network of plates and those developed from the Pacific network. O’Neill et al. (2005) quantified this misfit and found the Indo-Atlantic and Pacific reference frame solutions not to be unifiable within observational uncertainties related to available volcano locations and dates. A variety of explanations for this misfit have been suggested. One is as a consequence of unidentified or poorly quantified relative

Regional Geology and Tectonics

80

5. Plate kinematic reconstructions

motions, particularly between east and west Antarctica, the only divergent plate boundary across which the Pacific and Indo-Atlantic plate circuits can be linked (Fig. 5.17; Cande et al., 2000). An alternative class of explanations suggests that the problem is a consequence of the real (as opposed to assumed purely vertical) nature of upwelling within the mantle. Some in this class strongly question the assumption of plume upwelling from the very base of the lower mantle (e.g. Foulger, 2005). Others consider the possibility of true relative motion between plumes rising beneath the Pacific and Indo-Atlantic realms, describing it in terms of plumes bending like saplings in the ‘mantle wind’ of material flowing laterally between trenches and ridges. Geodynamic model studies have been set to the task of quantifying this nonradial component of mantle flow, after assuming the surface plate motions to be reliable at a level of uncertainty smaller than the expected relative plume motions (e.g. Steinberger, 2002). An unrelated source of uncertainty is likely to affect the hotspot reference frame over long timescales. This socalled ‘true polar wander’ involves rotation of Earth’s entire silicate shell, the crust and mantle together, with respect to the spin axis. It is an understandable consequence of the redistribution of mass (e.g. continents, by plate tectonics) in order to maintain optimal alignments of the planet’s moments of inertia with respect to the spin axis. True polar wander on Earth has been demonstrated using hotspot (Steinberger and Torsvik, 2008) and magnetic anomaly studies (Horner-Johnson and Gordon, 2010) and estimated to occur at rates of not more than 1 /Myr. With the distribution of volcanic samples of hotspot presence not uniform on all of the plates, it may be necessary to correct the hotspot reference frame for bias arising from uneven sampling of true polar wander signals. Paleomagnetic reference frame

As noted in the ‘Apparent paleomagnetic poles’ section, the magnetic north and south poles wander constantly and apparently randomly over distances of thousands of kilometres across the polar regions. This secular variation can nonetheless be averaged over long time periods to show that the mean locations of the magnetic poles lie very close to the poles of the spin axis. This is thought to be a consequence of the fact that the planet’s rotation influences the pattern of convection in the outer core so that it stays relatively constant with respect to the spin axis. If so, then large numbers of paleopoles preserved as remanent magnetizations in rocks of similar ages should be useful to deliver a statistically meaningful estimate of the paleo-positions of the poles of the spin axis. Reconstructions in the paleomagnetic reference frame are established by aligning these estimated poles with the present-day spin axis (Torsvik et al., 2008a,b). Like reconstructions built using the paleomagnetic technique, the accuracy of the paleomagnetic reference frame is limited by the resolution of observations of VGPs. The ‘Apparent paleomagnetic poles’ section shows that this is unlikely to exceed locational accuracy of more than 100 200 km and in most cases less accurate. Misfits between the paleomagnetic and hotspot reference frames can be attributed to true polar wander. For any given time slice, the paleomagnetic reference frame is, in essence, a point-to-point fitting exercise, promising constraint in only one (latitudinal) dimension (cf. Fig. 5.6). With only the relative latitudes of continents constrainable, the paleomagnetic reference frame must be combined with comparative geological constraints in order to build working plate or continental reconstructions. This can lead to considerable controversy (e.g. Muttoni et al., 2003). Lower mantle slabs reference frame

Teleseismic global tomography models, as described above in the ‘Diffuse and regional markers’ section, allow interpretations of subducted slabs of mantle lithosphere to be made. After assuming that slabs sink vertically within the mantle, Van Der Meer et al. (2010) adjusted global reconstruction models so that 28 selected lower mantle slab signals were located vertically beneath the surface orogens that they assumed to mark the corresponding sites of plate convergence. This exercise forms the basis of a lower mantle slab reference frame that allows the reconstruction of the plates’ paleolongitudes for periods as far back as Permian times, older than the B100 Ma that is possible with the hotspot reference frame. It has been criticized on the basis of the difficulty of interpreting lower mantle velocity variations consistently and unequivocally in terms of slabs of subducted oceanic lithosphere. Large Low-shear-velocity provinces

Advances in global seismology have revealed the possible presence of two large ( . 3000 km wide) areas of low shear-velocity material (large low-shear-velocity provinces; LLSVPs) coinciding with parts of the D’’ layer of the lowermost mantle. Whilst D’’ is typically around 200 km thick, the LLSVPs may rise up to 1000 km above the

Regional Geology and Tectonics

Making plate kinematic reconstructions

81

core mantle boundary (McNamara, 2018). The exact composition of the material in which sound waves travel so slowly here is speculated on, but it has been interpreted as a product of long-term subduction of oceanic lithosphere into the mantle. A series of publications has noted that the locations of Phanerozoic mantle plumes and kimberlites, when projected downwards from the sites of their eruptions towards D’’, appear to coincide with the edges of the LLSVPs (Fig. 5.19). Based on this, it is speculated that plumes incubate at thermochemical contrasts between the LLSVPs and surrounding rocks of the deepest mantle. Only one or two of the modelled hotspots and kimberlites seem not to project downwards to one of the LLSVP edges. This consistency has led to the conclusion that the LLSVPs are fixed on long timescales, perhaps because they are products of long-lived subduction zones that persisted around the margins of the late Precambrian supercontinent Rodinia. Assuming this fixity, and given that the LLSVP edges are lines, as opposed to points, they can be used for locating plates, via their hotspot traces, in paleolongitude as well as paleolatitude (Torsvik et al., 2008a,b). The asserted longevity and applicability of the LLSVP reference frame have been questioned. Among the reasons for this are observations in global surface plate motion models that suggest the LLSVPs are likely to have been deformed by slabs descending from subduction zones situated above them during times prior to the amalgamation of Pangea, and that reconstructions built with strict adherence to the LLSVP reference frame may imply continental or slab motions over the mantle that were implausibly fast in comparison to what is known from the independently and well-constrained last 200 Myr (e.g. Zhong and Rudolph, 2015; Flament et al., 2017).

FIGURE 5.19 (A) A cross-sectional schematic of Earth’s two LLSVPs from a South Pole vantage point, illustrating their role in defining plume generation zones (PGZs) at their edges. (B) Vote map outlining numbers of velocity tomography studies featuring a low-velocity anomaly close to the core mantle boundary. A mean 1% slowness contour at 2800 km depth, based on comparison of a set of tomographic studies, is also shown. The vote map and contour outline the presence of two main LLSVPs, one beneath southern and western Africa and another beneath much of the central and western Pacific. Squares indicate the locations of hotspots and reconstructed locations of plumes from which large igneous provinces of the last 300 Myr would have formed. Source: From Torsvik, T.H., van der Voo, R., Doubrovine, P.V., Burke, K., Steinberger, B., Ashwal, L.D., et al., 2014. Deep mantle structure as a reference frame for movements in and on the Earth. Proc. Natl. Acad. Sci. U.S.A. 111, 8735 8740.

Regional Geology and Tectonics

82

5. Plate kinematic reconstructions

Using plate reconstructions As context for regional geological and tectonic studies Tectonics is the study of rock deformation. A central tenet of plate tectonics is that the lithospheric plates only deform at their margins with other plates. Plate reconstructions can thus be used as powerful boundary conditions for regional tectonic studies, by enforcing the expectation that examples of paleostrain in rocks should be attributable to and located on a similarly aged plate boundary whose kinematics match those associated with the observed strain. The Alps as a previously extended continental margin Extensive similarities between the geology of the western Alps and facies associations of deep-water basins have long been known or suspected (e.g. Steinmann, 1905; Bernoulli and Jenkyns, 1974). Deep-sea drilling and geophysical surveying confirmed and extended the comparisons (e.g. Lemoine et al., 1987; Wilson et al., 2001). It is concluded that the western Alps consist of remnants of the extended continental margins of an ocean that existed between Gondwana and Laurasia. This ocean, the Alpine Tethys, could not be reconstructed using fits of magnetic anomalies or fracture zones, as by now it has almost completely been destroyed by subduction beneath Eurasia. Responding to this challenge, Ricou (1994) set about generating a set of reconstructions of the Alpine Tethys based on identifications of the extended margin segments and correlations of their geological histories to one another. The reconstructions show that the western Alps could indeed have experienced an earlier stage in their geological history as parts of the extended continental margins of Gondwana and Laurasia. This work was extended and elaborated on by Gerard Stampfli and coworkers at the University of Lausanne through the 1990s and early 21st century as part of their much larger-scale reconstructions of Neotethys and its neighbours in space and time (e.g. Stampfli and Kozur, 2006; Fig. 5.20). More recent reconstructions have concentrated on improving the use of available constraints on understanding the development and subsequent destruction of the western Tethys. With variable emphasis placed on the various available techniques, they have reached contrasting conclusions. Capitanio and Goes (2006) concluded that the region’s development involved relative motions of only the set of relatively large plates that can currently be concluded on from seafloor spreading evidence in the north Atlantic (Africa, Eurasia and Iberia). Other studies have concluded on the supplementary activity of a number of small plates whose oceanic parts have been largely or completely lost to subduction. The existence and actions of these plates are largely determined from geological interpretations and models and are suggested to have acted independently of one another either during oblique Pangea breakup (e.g. Handy et al.’s (2010) Alkapecia, Adria, Alcapia and Tiszia) and/or later as parts of a broad African-Eurasian collision zone (e.g. Hosseinpour et al.’s (2016) Adria, Carpathian-Tauride, Kir¸sehir and others). The collision of India with Eurasia and shortening of the Indian continent Geological similarities between the upper plate of the Himalaya, the southwards vergence of near-surface structures in the orogen and the presence in seismic images of extensive north-dipping surfaces at depth make it hard to escape the conclusion that a salient of Indian crust and lithosphere lies beneath part of the Tibetan Plateau, north of the Himalaya (Zhao et al., 1993). Apart from its narrow outcrop in the Himalaya, this salient, known as Greater India, is not accessible anywhere at outcrop. Its extent prior to the collision of India with Eurasia can only be indirectly determined. Estimates of this extent, and its variations along strike, are important boundary conditions for understanding and modelling the extent to which it may be possible to subduct continental crust and lithosphere (Ingalls et al., 2016; van Hinsbergen et al., 2017), for understanding the crustal shortening budget of the largest continental-scale orogens and for estimating the importance and processes of lateral escape along regional strike-slip faults (e.g. Tapponnier et al., 1982, 1986). Uncertainty in the size and shape of Greater India also translates directly into uncertainty in our understanding of the timing and style of the collision. Late geological estimates of the collision timing, based on the progression from flysch- to molasse-type associations, can be attributed to collision of a relatively narrow Greater India, whereas earlier estimates have led to much wider candidates for Greater India and/or precursor collisions involving Tethyan island arc terranes separated by one or more ‘Greater India basins’ (as discussed by van Hinsbergen et al., 2012, 2017). Approaches to determining the size of Greater India with kinematic reconstructions vary. In one, Greater India is estimated by tuning its northward extent so that it arrives into the collision at an assumed-correct geological estimate of the collision’s age. These reconstructions are of course strongly dependent on the time chosen as representative of the collision. Early collisions require very large Greater India estimates (Matte et al., 1997) or that

Regional Geology and Tectonics

Using plate reconstructions

83

FIGURE 5.20

Reconstruction of African-Eurasian plate convergence and loss of the Neotethys Ocean and diverse oceanic basins to subduction (Stampfli and Kozur, 2006).

island arc bodies preceded Greater India into the collision zone (Khan et al., 2009). In another, the salient is fitted according to assumptions about the shape of Gondwana’s NE continental margin, with particular attention paid to the supposed continuation of a sheared continental margin along Greater India’s northern edge eastwards along Australia’s Wallaby Zenith Plateau (Ali and Aitchison, 2005; Fig. 5.21). For these, the time of first collision is then estimated by running reconstructions featuring the salient forwards in time. The accuracy of this approach strongly depends on the precision and the accuracy of the Gondwana reconstruction. Plate reconstructions as boundary conditions for palinspastic reconstructions and palinspastically reconstructed markers in plate reconstructions Palinspastic restoration of extensional and compressional plate boundaries in continental crust were classically carried out along profiles, either measured using seismic techniques or calculated using geological field measurements. Their grounding assumption is that the area of crust sampled by the profile should balance in their preand postdeformational depictions (Chamberlin, 1910). Dunbar and Sawyer (1987) and Williams et al. (2011) extended this process to regional maps of extended continental margins using gridded representations of crustal thickness. They generated profiles of crustal cross-sectional area for balancing by sampling the crustal thickness

Regional Geology and Tectonics

84

5. Plate kinematic reconstructions

FIGURE 5.21 Ali and Aitchison’s (2005) Greater India (in black) depending on the fit of extant continental India in Gondwana. Zenith and Wallaby plateaus (ZP, WP) are key corroborative constraints, based on an interpretation that they were excised from Greater India by transport along a sheared continental margin segment.

grid along lines drawn across the extended margin along model flow lines derived from plate kinematic rotations. After balancing the cross-sectional area by shortening the profile, points along the COB were displaced towards the plate interior along the model flow line by the shortening distance, producing a ‘restored COB’. Eagles et al. (2015) questioned the technique’s use as a reconnaissance- or plate-scale tool by noting that its sensitivity to the typical range of uncertainty in COB locations is similar in magnitude to the statistical uncertainty of a plate kinematic model derived using the continental slope as a passive prerift marker.

In paleogeography/paleobathymetry/paleotopography Understanding how the Earth system works at the present day requires knowledge of the distribution of its geographical, biological and geological features. An understanding of past states of the Earth system is equally dependent on accurate and precise models of these features during past time slices. Fundamentally, models such as this permit users to understand something about the Earth system in the past, whether for commercial or academic purposes, and are most accessibly portrayed in map form (e.g. Markwick, 2019). The design of paleomaps is ultimately dependent on their anticipated use, but a fundamental component they all share is the need to portray the areal distribution of continental and oceanic crust in the past. This can be determined from the presentday distribution of continents along with assumptions that plate interiors are rigid, that continental crust is not dense enough to be completely subducted in significant quantities and that new continental crust has not been created in large quantities since at least Archaean times. With further information from geological and geophysical data and interpretations on plate boundary types and their locations, the information on crustal distribution can be supplemented with the topography and relief of the ocean floor and land surface. Less commonly, paleomaps are adorned with geological and paleobiological information. The full modern diversity of this kind of work owes a lot to Peter Ziegler and a large cast of coworkers starting at the University of Chicago in the 1960s. Much of it has been made available as parts of the PALEOMAP PaleoAtlas project, originally for Esri’s ArcMap software (e.g. Scotese, 2014), and more recently for GPlates (https://www.earthbyte.org/paleomap-paleoatlas-for-gplates/). Fundamentally, the approach is to generate maps that are led by and illustrative of (e.g. https://deeptimemaps.com/), or modelled from and consistent with (Markwick, 2019; Pe´rez-Dı´az and Eagles, 2017) geological and geophysical observations (Figs 5.22 and 5.23). That is, plate-scale landforms can be depicted in reconstructions in two main ways. The first of these is by direct

Regional Geology and Tectonics

85

Using plate reconstructions

illustration of observations or interpretations of topography made using the rock record onto a reconstruction basemap (e.g. Figs 5.4 and 5.22). In the second, topography and relief are modelled in situ according to an understanding of the tectonic and climatic processes that lead to their formation. Fig. 5.22 shows an illustrative reconstruction for Permian times from the Paleomap project (Scotese, 2014). The Deep Time Maps project of Colorado Plateau Geosystems, Inc. produces similar maps using a similar workflow (Blakey and Ranney, 2018). These striking reconstructions are offered both academically and commercially and are widely used in situations where global and regional paleogeography play a supporting role. The relative continental positions are based on tectonic models chosen from the literature. Topographic and climatic indications, also based on a literature review for the period, are illustrated using colours intended to be reminiscent of photographic images of the planet taken by orbiting satellites. Shades of blue from light to dark indicate increasing ocean depths, determined from the distribution of mid-ocean ridges in the tectonic model and paleo-seafloor age. Browns and greens are used to denote the interplay of relief and vegetation cover on land, determined mostly from sedimentation patterns as interpreted from regional geological and paleoenvironmental studies, and white shows the distribution of ice sheets. The colouring is applied using a commercial computer drawing and illustration package. Fig. 5.23 shows a process-based model of paleobathymetry in the South Atlantic (Pe´rez-Dı´az and Eagles, 2017). This model is built using an empirically derived quantitative relationship between seafloor age and ocean depth, describing the process of thermal contraction of the cooling oceanic lithosphere. This is applied to a custom-built South Atlantic tectonic model and age grid, describing the progress of the oceanic lithosphere’s formation. The thermal subsidence surface is adjusted numerically for the isostatic effects of further processes including sedimentation over time, crustal formation with variable thickness at mid-ocean ridges, intraplate crustal thickening by hotspot volcanism and vertical displacement of the lithosphere by viscous stresses transmitted from the

FIGURE 5.22 An illustrative paleogeography for Late Permian times (Scotese, 2014). Blue shades are intended to represent relative oceanic water depths (lighter shades shallower). Brown and green shades indicate topography and vegetation cover, effectively on the basis of the known Permian rock record. White indicates perennial ice sheets, again based on Permian rocks.

FIGURE

5.23 Process-based paleobathymetry and uncertainty range for the South Atlantic at 60 Ma (Pe´rez-Dı´az and Eagles, 2017).

Regional Geology and Tectonics

86

5. Plate kinematic reconstructions

convecting mantle (so-called dynamic topography). The two side panels show the likely uncertainties in the quantification, based on uncertainties in the governing observations and equations, as maximum (deepest) and minimum (shallowest) paleobathymetric surfaces for the same period.

For geodynamic studies Plate driving forces are difficult to interpret or measure from paleostress indicators, which may be imprecisely dated, strongly influenced by local lithological variations or simply too sparsely distributed, to form the basis for a reliable analysis. Balancing forward-modelled plate boundary and gravitational forces with paleostress indications is intensely challenging (e.g. Warners-Ruckstuhl et al., 2013). One of the key tasks in doing this is identifying the start and end points of steady-state epochs in which the evolution of plate boundary forces can be thought to have changed little, indicating a balanced set of plate boundary torques. A proxy observation for the plate boundary forces is the motion it causes, and this in turn can be determined from seafloor spreading data and/or models. These, in their turn, can be correlated with the distribution and extent of plate boundaries, at which at least some of the forces driving plate motion are generated. The Indian plate and plume influences on plate motion One of the most remarkable signals of changing plate driving forces in the seafloor spreading record is the rapid northwards motion of the Indian plate in late Cretaceous times. A long period of very fast spreading rates is suggested to have started around 90 million years ago, around 20 million years later reaching rates close to the fastest rates currently observed on Earth (100 120 mm/yr from the East Pacific Rise) and appears to have been punctuated by a hyperfast ( . 180 mm/yr) event of short duration around 66 Ma. Numerous suggestions have been made to explain the rapid northwards motion of India, involving unusually strong driving forces on the one hand or unusually weak braking forces on the other (Kumar et al., 2007; van Hinsbergen et al., 2011; Eagles and Wibisono, 2013). Mantle plumes play a role in several of these suggestions. van Hinsbergen et al. (2011) suggested that the arrival of the Marion plume at the southern edge of the Indian plate around 90 Ma led to its northwards acceleration by introducing a northwards-sloping upper mantle surface down which the plate slid. More detailed modelling of the seafloor spreading rate, from the Mascarene Basin that started opening upon the plume’s arrival, showed no unusual acceleration beyond what might be expected as a gravitational response to the development and thickening of the oceanic lithosphere within that basin (Eagles and Wibisono, 2013). The shorter-lived period of hyperfast spreading centred around 66 Ma is, however, unusual when viewed against the backdrop of steadily increasing spreading rates. Eagles and Wibisono (2013) suggested that the unusually sharp onset and cessation of the seafloor spreading pulse at 67 64 Ma are not consistent with the regional gravitational effects of this second plume’s arrival beneath the southern edge of the plate. After assuming that the pulse is not an artefact of errors in the magnetic reversal timescale, they suggested that the plume may have acted via the agency of an unusually tall mid-ocean ridge in the short-lived Laxmi Basin. In contrast, Cande and Stegman (2011) noted how, simultaneously with India’s acceleration at this time, the northwards progress of the African plate over the mantle (measured in spreading rates in the South Atlantic and Southwest Indian Ocean) appears to have slowed. They suggested that the arrival of the Re´union plume beneath the two plates’ shared margin tilted the base of the Indian plate’s lithosphere northwards, adding a northwards ‘downhill’ component to the forces driving it northwards, and simultaneously tilted the base of the African plate southwards, retarding that plate’s northwards progress. Like van Hinsbergen et al.’s (2011) hypothesis for a Marion Plume-related downhill force, this conclusion is not strongly supported by higher-resolution models of spreading rates in those settings (Fig. 5.24). Global tectonic reorganizations The main torques that drive the plates to move at the surface of the planet are generated at their margins. The negative buoyancy of subducting slabs pulls the surface parts of the plates towards the trenches where they undergo subduction. The base of the lithosphere slopes away from the mid-ocean ridges as a consequence of its gradual cooling, giving rise to a gravitational body force known as ridge push. Additional topography at the top surface of the mantle, a result among other influences of its convection, may also contribute to the gravitational driving forces. These forces are balanced by drag between the base of the lithosphere and the underlying mantle, and by friction at transform faults. For much of the time, the balance of these forces evolves only slowly because the plates have great inertia, ensuring they move along constant paths, which in turn ensures that their boundaries, and the forces generated at them tend to maintain their orientations and senses of motion with respect to the neighbouring plates.

Regional Geology and Tectonics

Summary

87 FIGURE 5.24 Changing spreading rates and azimuths (in degrees, increasing clockwise) over the 90 20 Ma period. Indian-African plate divergence rates (INDAFR) from Eagles and Hoang (2013), South American-African plate divergence rates (SAMAFR) from Pe´rez-Dı´az and Eagles (2014), African-Antarctic plate divergence rates (AFRANT) from TuckMartin et al. (2018) and Indian-Antarctic plate divergence rates (INDANT) from Eagles (2019). Grey bars show the magnetic reversal anomaly timescale of Gradstein et al. (2004). Yellow bars indicate the timing of eruptions related to selected plume arrivals. Deccan plume arrival signals appear to be restricted to a short-lived spreading rate increase in relative motions involving the Indian plate and a change in azimuth of African plate motion away from the Antarctic plate. It is not clear why this azimuth change was subsequently (at 53 Ma) reversed.

However, because the surface area of the planet is finite and unchanging but the plates’ areas and perimeters are free to increase or decrease over time, plate boundaries must interact with one another. This interaction can be gradual, for example the slow lengthening of mid-ocean ridges at ridge-ridge-ridge triple junctions like the Rodriguez triple junction in the central Indian Ocean, or catastrophic, for example by the collision and subduction of a mid-ocean ridge at a destructive plate boundary such as between the Chile Rise and Andean trench. Catastrophic changes can remove or introduce one of the forces contributing to the balance of drivers on a plate in a geological instant. The plate’s motion is impelled to change as a result. This, in turn, may lead to the requirement for changes in the relative motions at the plate’s other boundaries, resulting in the introduction of changed or new driving forces not only there, but also at the boundaries of the neighbouring plates. This, again, leads to changes in the force balances on the neighbouring plates and to attendant changes in all their relative motions. As a consequence of this kind of chain reaction, a whole circuit of plates and their relative motions can be reorganized over the course of a few million years until a new equilibrium state is established. Proposed examples of reorganizations include a set of events occurring around 100 Ma, and well expressed in the Indian Ocean, and at 70 and 50 6 2 Ma (Mu¨ller et al., 2016). Those authors summarize a selection of regional relative plate-motion models that describe how, by the end of the youngest event the orientation of North Atlantic spreading had rotated clockwise, South Atlantic spreading had accelerated and rotated clockwise, Southwest Indian ocean spreading had accelerated and rotated clockwise, Indian-Antarctic spreading decelerated, Pacific-Antarctic had spreading decelerated and rotated slightly anticlockwise and Pacific-Farallon spreading had decelerated and rotated slightly clockwise. They suggest that the end of subduction of oceanic lithosphere with the entry of continental India into the convergent plate boundary may have seen an increase in collisional resistive forces that led to the 50 Ma global reorganization. Some, but not all of these observations reappear in Fig. 5.24, suggesting that model choice is an important factor in the identification and interpretation of global and regional tectonic reorganization events. To illustrate this further, Whittaker et al. (2007) presented a different compilation of regional models, centred on a reinterpretation of the history of seafloor spreading between Australia and East Antarctica, which dated the youngest reorganization to a slightly earlier time (50 53 Ma) and attributed it to ridge-trench collision in the northwestern Pacific at 60 55 Ma.

Summary Models of past plate motions are of fundamental significance as boundary conditions for regional tectonic and geological analyses, and paleogeographical and geodynamic studies. These models are built by quantitative or graphical application of the plate tectonic paradigm to geological and geographical data. Whilst regional

Regional Geology and Tectonics

88

5. Plate kinematic reconstructions

geological data can be confidently and quantitatively interpreted in terms of plate reconstruction models, plate tectonics, as a geometrical discipline, is best modelled on the basis of larger-scale data sets for which a priori geological interpretation can be kept to a minimum. In practice, this means the best constraints are those that can be taken from the geologically-short lived oceanic lithosphere, for large areas of which the a priori interpretation of formation during plate divergence can be legitimately upheld. Moving backwards in time, the available extent of oceanic lithosphere for high-resolution plate kinematic work becomes ever smaller as a result of subduction of the oldest oceanic lithosphere. The diminishing oceanic area from which high-quality constraints can be taken leads to a general decrease in statistical reliability with increasing reconstruction age. For pre-Jurassic times, no oceanic constraints are available, and the reconstruction task relies ever more strongly on geometrically mediumto low-precision paleomagnetic reconstructions and a priori interpretations of kinematic setting from eversmaller and frequently reworked continental terranes and outcrops. Consequently, kinematic reconstructions based on these constraints become progressively more uncertain and progressively less reliable as quantitative or with great age even qualitative, constraints or checks on regional geological understanding.

References Adie, R.J., 1952. The position of the Falkland Islands in a reconstruction of Gondwanaland. Geol. Mag. 89, 401 410. Aitken, A., Young, D.A., Ferraccioli, F., Betts, P.G., Greenbaum, J.S., Richter, T.G., et al., 2014. The subglacial geology of Wilkes Land, East Antarctica. Geophys. Res. Lett. 41, 2390 2400. Ali, J., Aitchison, J.C., 2005. Greater India. Earth-Sci. Rev. 72, 169 188. Banerjee, S.K., 1984. The magnetic layer of the ocean crust—How thick is it? Tectonophysics 105, 15 27. Barker, P.F., Griffiths, D.H., 1972. A Discussion on volcanism and the structure of the Earth-The evolution of the Scotia Ridge and Scotia Sea. Philos. Trans. R. Soc. London. A Math. Phys. Sci. 271, 151 183. Bernoulli, D., Jenkyns, H.C., 1974. Alpine Mediterranean and Central Atlantic Mesozoic Facies in Relation to the Early Evolution of the Tethys. Special Publication, SEPM. Besse, J., Courtillot, V., 2002. Apparent and true polar wander and the geometry of the geomagnetic field over the last 200 Myr. J. Geophys. Res. Solid Earth 107, EPM 6-1 EPM 6-31. Blakey, R.C., Ranney, W.D., 2018. Flat-Slab Subduction, the Laramide Orogeny, Uplift of the Colorado Plateau and Rocky Mountains: Paleocene and Eocene: Ca. 65 35 Ma. Ancient Landscapes of Western North America. Springer, Cham, pp. 131 148. Bonatti, E., Ligi, M., Brunelli, D., Cipriani, A., Fabretti, P., Ferrante, V., et al., 2003. Mantle thermal pulses below the Mid-Atlantic Ridge and temporal variations in the formation of oceanic lithosphere. Nature 42, 499. Bullard, E., Everett, J.E., Gilbert Smith, A., 1965. The fit of the continents around the Atlantic. Philos. Trans. R. Soc. Lond. A Math. Phys. Sci. 258, 41 51. Butler, R.F., 2004. Paleomagnetism: Magnetic Domains to Geologic Terranes, Electronic edition. ,https://www.geo.arizona.edu/ Paleomag/.. Cande, S.C., Mutter, J.C., 1982. A revised identification of the oldest sea-floor spreading anomalies between Australia and Antarctica. Earth Planet. Sci. Lett. 58, 151 160. Cande, S.C., Patriat, P., 2015. The anticorrelated velocities of Africa and India in the Late Cretaceous and early Cenozoic. Geophys. J. Int. 200, 227 243. Cande, S.C., Stegman, D.R., 2011. Indian and African plate motions driven by the push force of the Reunion plume head. Nature 475, 47. Cande, S.C., Herron, E.M., Hall, B.R., 1982. The early Cenozoic tectonic history of the southeast Pacific. Earth Planet. Sci. Lett. 57, 63 74. Cande, S.C., Stock, J.M., Mu¨ller, R.D., Ishihara, T., 2000. Cenozoic motion between east and west Antarctica. Nature 404, 145. Capitanio, F.A., Goes, S., 2006. Mesozoic spreading kinematics: consequences for Cenozoic Central and Western Mediterranean subduction. Geophys. J. Int. 165, 804 816. Carey, S.W., 1955. Wegener’s South America Africa assembly, fit or misfit? Geol. Mag. 92, 196 200. Chamberlin, R.T., 1910. The Appalachian folds of central Pennsylvania. J. Geol. 18, 228 251. Chang, T., Stock, J., Molnar, P., 1990. The rotation group in plate tectonics and the representation of uncertainties of plate reconstructions. Geophys. J. Int. 101, 649 661. Choubert, B., 1935. Recherches sur la gene`se des chaıˆnes pale´ozoı¨ques et ante´cambriennes. Rev. Ge´ogr phys. Ge´ol dyn. VIII 1, 5 50. Cocks, L.R., Fortey, R.A., 1990. Biogeography of Ordovician and Silurian faunas. Geol.l Soc. Lond. Mem. 12, 97 104. Coffin, M.F., Pringle, M.S., Duncan, R.A., Gladczenko, T.P., Storey, M., Mu¨ller, R.D., et al., 2002. Kerguelen hotspot magma output since 130 Ma. J. Pet. 43, 1121 1137. Crawford, A.R., 1970. The Precambrian geochronology of Rajasthan and Bundelkhand, northern India. Can. J. Earth Sci. 7, 91 110. DeMets, C., 1993. Earthquake slip vectors and estimates of present-day plate motions. J. Geophys. Res. Solid Earth 98, 6703 6714. De Wit, M.J., Stankiewicz, J., Reeves, C., 2008. Restoring Pan-African-Brasiliano connections: more Gondwana control, less trans-Atlantic corruption, Geological Society, London, Special Publications., 294. pp. 399 412. Dunbar, J.A., Sawyer, D.S., 1987. Implications of continental crust extension for plate reconstruction: an example from the Gulf of Mexico. Tectonics 6, 739 755. Du Toit, A.L., 1937. Our Wandering Continents: An Hypothesis of Continental Drifting. Oliver and Boyd. Eagles, G., 2016. Plate kinematics of the Rocas Verdes Basin and Patagonian orocline. Gondwana Res. 37, 98 109. Eagles, G., 2019. A little spin in the Indian Ocean plate circuit. Tectonophysics 754, 80 100. Eagles, G., Hoang, H.H., 2013. Cretaceous to present kinematics of the Indian, African and Seychelles plates. Geophys. J. Int. 196, 1 14.

Regional Geology and Tectonics

References

89

Eagles, G., Ko¨nig, M., 2008. A model of plate kinematics in Gondwana breakup. Geophys. J. Int. 173, 703 717. Eagles, G., Wibisono, A.D., 2013. Ridge push, mantle plumes and the speed of the Indian plate. Geophys. J. Int. 194, 670 677. Eagles, G., Pe´rez-Dı´az, L., Scarselli, N., 2015. Getting over continent ocean boundaries. Earth-Sci. Rev. 151, 244 265. Engebretson, D.C., Cox, A., Gordon, R.G., 1984. Relative motions between oceanic plates of the Pacific basin. J. Geophys. Res. Solid Earth 89, 10291 10310. Flament, N., Williams, S., Mu¨ller, R.D., Gurnis, M., Bower, D.J., 2017. Origin and evolution of the deep thermochemical structure beneath Eurasia. Nat. Commun. 8, 14164. Foulger, G.R. (Ed.), 2005. Plates, Plumes, and Paradigms, vol. 388. Geological Society of America. Funaki, M., Yoshida, M., Vitanage, P.W., 1990. Natural remanent magnetization of some rocks from southern Sri Lanka. Proc. NIPR Symp. Antarct. Geosci., 4, 231 240. Gibbons, A.D., Whittaker, J.M., Mu¨ller, R.D., 2013. The breakup of East Gondwana: assimilating constraints from Cretaceous ocean basins around India into a best-fit tectonic model. J. Geophys. Res. Solid Earth 118, 808 822. Gibson, G.M., Totterdell, J.M., White, L.T., Mitchell, C.H., Stacey, A.R., Morse, M.P., et al., 2013. Pre-existing basement structure and its influence on continental rifting and fracture zone development along Australia’s southern rifted margin. J. Geol. Soc. 170, 365 377. Gordon, R.G., Van der Voo, R., 1995. Mean paleomagnetic poles for the major continents and the Pacific plate, Global Earth Physics: A Handbook of Physical Constants (AGU Reference Shelf), 1. pp. 225 239. Gradstein, F.M., Ogg, J.G., Smith, A.G., 2004. A Geologic Timescale. Cambridge University Press. Granot, R., Dyment, J., 2019. The influence of post-accretion sedimentation on marine magnetic anomalies. Geophys. Res. Lett. 46, 4645 4652. Hall, R., 2002. Cenozoic geological and plate tectonic evolution of SE Asia and the SW Pacific: computer-based reconstructions, model and animations. J. Asian Earth Sci. 20, 353 431. Handy, M.R., Schmid, S.M., Bousquet, R., Kissling, E., Bernoulli, D., 2010. Reconciling plate-tectonic reconstructions of Alpine Tethys with the geological geophysical record of spreading and subduction in the Alps. Earth-Sci. Rev. 102, 121 158. Hellinger, S.J., 1981. The uncertainties of finite rotations in plate tectonics. J. Geophys. Res. Solid Earth 86, 9312 9318. Horner-Johnson, B.C., Gordon, R.G., 2010. True polar wander since 32 Ma BP: a paleomagnetic investigation of the skewness of magnetic anomaly 12r on the Pacific plate. J. Geophys. Res. Solid Earth 115. Hosseinpour, M., Williams, S., Seton, M., Barnett-Moore, N., Mu¨ller, R.D., 2016. Tectonic evolution of Western Tethys from Jurassic to present day: coupling geological and geophysical data with seismic tomography models. Int. Geol. Rev. 58, 1616 1645. Iaffaldano, G., Hawkins, R., Bodin, T., Sambridge, M., 2014. REDBACK: open-source software for efficient noise-reduction in plate kinematic reconstructions. Geochem. Geophys. Geosyst. 15, 1663 1670. Ingalls, M., Rowley, D.B., Currie, B., Colman, A.S., 2016. Large-scale subduction of continental crust implied by India-Asia mass-balance calculation. Nat. Geosci. 9, 848 853. Khan, S.D., Walker, D.J., Hall, S.A., Burke, K.C., Shah, M.T., Stockli, L., 2009. Did the Kohistan-Ladakh island arc collide first with India? Geol. Soc. Am. Bull. 121, 366 384. Kumar, P., Yuan, X., Kumar, M.R., Kind, R., Li, X., Chadha, R.K., 2007. The rapid drift of the Indian tectonic plate. Nature 449, 894. Langemeyer, S.M., Lowman, J.P., Tackley, P.J., 2018. The Sensitivity of Core Heat Flux to the Modeling of Plate-Like Surface Motion. Geochem. Geophys. Geosyst. 19, 1282 1308. Larter, R.D., Cunningham, A.P., Barker, P.F., Gohl, K., Nitsche, F.O., 2002. Tectonic evolution of the Pacific margin of Antarctica 1. Late Cretaceous tectonic reconstructions. J. Geophys. Res. Solid Earth 107, EPM-5. Lemoine, M., Tricart, P., Boillot, G., 1987. Ultramafic and gabbroic ocean floor of the Ligurian Tethys (Alps, Corsica, Apennines): in search of a genetic imodel. Geology 15, 622 625. Lyell, C., 1830. Principles of Geology, Being an Attempt to Explain the Former Changes of the Earth’s Surface, by Reference to Causes Now in Operation. John Murray, London. Matte, P., Mattauer, M., Olivet, J.M., Griot, D.A., 1997. Continental subductions beneath Tibet and the Himalayan orogeny: a review. Terra Nova 9, 264 270. Matthews, K.J., Maloney, K.T., Zahirovic, S., Williams, S.E., Seton, M., Mueller, R.D., 2016. Global plate boundary evolution and kinematics since the late Paleozoic. Glob. Planet. Change 146, 226 250. Markwick, P.J., 2019. Palaeogeography in exploration. Geol. Mag. 156, 366 407. McNamara, A.K., 2018. A review of large low shear velocity provinces and ultra low velocity zones. Tectonophysics . Mitchell, C.E., Taylor, G.K., Cox, K.G., Shaw, J., 1986. Are the Falkland Islands a rotated microplate? Nature 319, 131. Morgan, W.J., 1971. Convection plumes in the lower mantle. Nature 230, 42. Mu¨ller, R.D., Sandwell, D.T., Tucholke, B.E., Sclater, J.G., Shaw, P.R., 1991. Depth to basement and geoid expression of the Kane Fracture Zone: a comparison. Mar. Geophys. Res. 13, 105 129. Mu¨ller, R.D., Royer, J.Y., Cande, S.C., Roest, W.R., Maschenkov, S., 1999. New constraints on the Late Cretaceous/Tertiary plate tectonic evolution of the Caribbean, Sedimentary Basins of the World, vol. 4. Elsevier, pp. 33 59. Mu¨ller, R.D., Seton, M., Zahirovic, S., Williams, S.E., Matthews, K.J., Wright, N.M., et al., 2016. Ocean basin evolution and global-scale plate reorganization events since Pangea breakup. Annu. Rev. Earth Planet. Sci. 44, 107 138. Muttoni, G., Kent, D.V., Garzanti, E., Brack, P., Abrahamsen, N., Gaetani, M., 2003. Early Permian Pangea ‘B’to Late Permian Pangea ‘A’. Earth Planet. Sci. Lett. 215, 379 394. Nankivell, A.P., 1997. Tectonic evolution of the Southern Ocean between Antarctica, South America and Africa over the past 84 Ma. Doctoral dissertation, University of Oxford. Okal, E.A., Langenhorst, A.R., 2000. Seismic properties of the Eltanin transform system, South Pacific. Phys. Earth Planet. Inter. 119, 185 208. O’Neill, C., Mu¨ller, D., Steinberger, B., 2005. On the uncertainties in hot spot reconstructions and the significance of moving hot spot reference frames. Geochem. Geophys. Geosyst. 6. Pe´rez-Dı´az, L., Eagles, G., 2014. Constraining South Atlantic growth with seafloor spreading data. Tectonics 33, 1848 1873.

Regional Geology and Tectonics

90

5. Plate kinematic reconstructions

Pe´rez-Dı´az, L., Eagles, G., 2017. South Atlantic paleobathymetry since early Cretaceous. Sci. Rep. 7, 11819. Poblete, F., Roperch, P., Arriagada, C., Ruffet, G., de Arellano, C.R., Herve´, F., et al., 2016. Late Cretaceous early Eocene counterclockwise rotation of the Fueguian Andes and evolution of the Patagonia Antarctic Peninsula system. Tectonophysics 668, 15 34. Reeves, C.V., Teasdale, J.P., Mahanjane, E.S., 2016. Insight into the Eastern Margin of Africa from a new tectonic model of the Indian Ocean, Geological Society, London, Special Publications, 431. pp. 299 322. Richards, P.C., Gatliff, R.W., Quinn, M.F., Fannin, N.G.T., Williamson, J.P., 1996. The geological evolution of the Falkland Islands continental shelf, Geological Society, London, Special Publications, 108. pp. 105 128. Ricou, L.E., 1994. Tethys reconstructed: plates, continental fragments and their Boundaries since 260 Ma from Central America to Southeastern Asia. Geodin. Acta 7, 169 218. Sandwell, D.T., Mu¨ller, R.D., Smith, W.H., Garcia, E., Francis, R., 2014. New global marine gravity model from CryoSat-2 and Jason-1 reveals buried tectonic structure. Science 346, 65 67. Scotese, C., 2014. Atlas of Permo-Carboniferous Paleogeographic Maps (Mollweide Projection), Maps 53 64, Volumes 4, The Late Paleozoic, PALEOMAP Atlas for ArcGIS, PALEOMAP Project, Evanston, IL. Seton, M., Whittaker, J.M., Wessel, P., Mu¨ller, R.D., DeMets, C., Merkouriev, S., et al., 2014. Community infrastructure and repository for marine magnetic identifications. Geochem. Geophys. Geosyst. 15, 1629 1641. Shaw, P.R., Cande, S.C., 1990. High-resolution inversion for South Atlantic plate kinematics using joint altimeter and magnetic anomaly data. J. Geophys. Res. Solid Earth 95, 2625 2644. Shephard, G.E., Matthews, K.J., Hosseini, K., Domeier, M., 2017. On the consistency of seismically imaged lower mantle slabs. Sci. Rep. 7, 10976. Smith, A.G., Briden, J.C., Hurley, A.M., 1981. Phanerozoic Paleocontinental Maps. Cambridge University Press. Smith, A.G., Hallam, A., 1970. The fit of the southern continents. Nature 225, 139. Smith, W.H., Sandwell, D.T., 1997. Global sea floor topography from satellite altimetry and ship depth soundings. Science 277, 1956 1962. Stampfli, G.M., Kozur, H.W., 2006. Europe from the Variscan to the Alpine cycles. Geol. Soc. Lond. Mem. 32, 57. Steinberger, B., 2002. Motion of the Easter hot spot relative to Hawaii and Louisville hot spots. Geochem. Geophys. Geosyst. 3 (11), 1 27. Steinberger, B., Antretter, M., 2006. Conduit diameter and buoyant rising speed of mantle plumes: implications for the motion of hot spots and shape of plume conduits. Geochem. Geophys. Geosyst. 7. Steinberger, B., Torsvik, T.H., 2008. Absolute plate motions and true polar wander in the absence of hotspot tracks. Nature 452, 620. Steinberger, B., Sutherland, R., O’connell, R.J., 2004. Prediction of Emperor-Hawaii seamount locations from a revised model of global plate motion and mantle flow. Nature 430, 167. Steinmann, G., 1905. Geologische Beobachtungen in den Alpen: die Schardt’sche Ueberfaltungstheorie und die geologische Bedeutung der Tiefseeabsa¨tze und der ophiolitischen Massengesteine. Ber. Natf. Ges. Freiburg. 16, 18 67. Tapponnier, P., Peltzer, G., Le Dain, A.Y., Armijo, R., Cobbold, P., 1982. Propagating extrusion tectonics in Asia: New insights from simple experiments with plasticine. Geology 10, 611 616. Tapponnier, P., Peltzer, G., Armijo, R., 1986. On the mechanics of the collision between India and Asia, Geological Society, London, Special Publications, 19. pp. 113 157. Tarduno, J.A., Duncan, R.A., Scholl, D.W., Cottrell, R.D., Steinberger, B., Thordarson, T., et al., 2003. The Emperor Seamounts: southward motion of the Hawaiian hotspot plume in Earth’s mantle. Science 301, 1064 1069. Thurber, C., Ritsema, J., Dziewonski, A., Romaonwicz, B., 2007. Seismic tomography and inverse methods, Treatise on Geophysics, 1. Elsevier, Oxford, pp. 1323 1360. Torsvik, T.H., Mu¨ller, R.D., Van der Voo, R., Steinberger, B., Gaina, C., 2008a. Global plate motion frames: toward a unified model. Rev. Geophys. 46. Torsvik, T.H., Steinberger, B., Cocks, L.R.M., Burke, K., 2008b. Longitude: linking Earth’s ancient surface to its deep interior. Earth Planet. Sci. Lett. 276, 273 282. Torsvik, T.H., Rousse, S., Labails, C., Smethurst, M.A., 2009. A new scheme for the opening of the South Atlantic Ocean and the dissection of an Aptian salt basin. Geophys. J. Int. 177, 1315 1333. Torsvik, T.H., van der Voo, R., Doubrovine, P.V., Burke, K., Steinberger, B., Ashwal, L.D., et al., 2014. Deep mantle structure as a reference frame for movements in and on the Earth. Proc. Natl. Acad. Sci. U.S.A. 111, 8735 8740. Tuck-Martin, A., Adam, J., Eagles, G., 2018. New plate kinematic model and tectono-stratigraphic history of the East African and West Madagascan Margins. Basin Res. 30, 1118 1140. Van Der Meer, D.G., Spakman, W., Van Hinsbergen, D.J., Amaru, M.L., Torsvik, T.H., 2010. Towards absolute plate motions constrained by lower-mantle slab remnants. Nat. Geosci. 3, 36. van der Meer, D.G., van Hinsbergen, D.J., Spakman, W., 2018. Atlas of the underworld: slab remnants in the mantle, their sinking history, and a new outlook on lower mantle viscosity. Tectonophysics 723, 309 448. van Hinsbergen, D.J., Steinberger, B., Doubrovine, P.V., Gassmo¨ller, R., 2011. Acceleration and deceleration of India-Asia convergence since the Cretaceous: roles of mantle plumes and continental collision. J. Geophys. Res. Solid Earth 116. van Hinsbergen, D.J., Lippert, P.C., Dupont-Nivet, G., McQuarrie, N., Doubrovine, P.V., Spakman, W., et al., 2012. Greater India Basin hypothesis and a two-stage Cenozoic collision between India and Asia. Proc. Natl. Acad. Sci. USA 109, 7659 7664. van Hinsbergen, D.J., Lippert, P.C., Huang, W., 2017. Unfeasible subduction? Nat. Geosci. 10, 878 879. Warners-Ruckstuhl, K.N., Govers, R., Wortel, R., 2013. Tethyan collision forces and the stress field of the Eurasian Plate. Geophys. J. Int. 195, 1 15. Wegener, A., 1929. Die Entstehung der Kontinente und Ozeane, fourth ed. Friedrich Vieweg & Sohn Akt.-Ges, Braunschweig. Weissel, J.K., Hayes, D.E., 1978. Magnetic anomalies in the southeast Indian Ocean. Antarctica Oceanology II: Australian New Zealand Sector 19, 165 196. White, L.T., Gibson, G.M., Lister, G.S., 2013. A reassessment of paleogeographic reconstructions of eastern Gondwana: bringing geology back into the equation. Gondwana Res. 24, 984 998.

Regional Geology and Tectonics

Further reading

91

Whittaker, J.M., Mu¨ller, R.D., Leitchenkov, G., Stagg, H., Sdrolias, M., Gaina, C., et al., 2007. Major Australian-Antarctic plate reorganization at Hawaiian-Emperor bend time. Science 318, 83 86. Williams, S.E., Whittaker, J.M., Mu¨ller, R.D., 2011. Full-fit, palinspastic reconstruction of the conjugate Australian-Antarctic margins. Tectonics 30 (6). Wilson, R.C.L., Manatschal, G., Wise, S., 2001. Rifting along non-volcanic passive margins: stratigraphic and seismic evidence from the Mesozoic successions of the Alps and western Iberia, Geological Society, London, Special Publications, 187. pp. 429 452. Yoshida, M., Funaki, M., Vitanage, P.W., 1992. Proterozoic to Mesozoic East Gondwana: the juxtaposition of India, Sri Lanka, and Antarctica, Tectonics, 11. pp. 381 391. Zhao, W., Nelson, K.D., Che, J., Quo, J., Lu, D., Wu, C., et al., 1993. Deep seismic reflection evidence for continental underthrusting beneath southern Tibet. Nature 366, 557. Zhong, S., Rudolph, M.L., 2015. On the temporal evolution of long-wavelength mantle structure of the Earth since the early Paleozoic. Geochem. Geophys. Geosyst. 16, 1599 1615. Ziegler, P.A., 1988. Evolution of the Arctic-North Atlantic and the Western Tethys: a visual presentation of a series of paleogeographicpaleotectonic maps. AAPG Mem. 43, 164 196.

Further reading Smith, A.G., Smith, D.G., Funnell, B.M., 2004. Atlas of Mesozoic and Cenozoic Coastlines. Cambridge University Press.

Regional Geology and Tectonics

C H A P T E R

6 Resolving geological enigmas using plate tectonic reconstructions and mantle flow models Sabin Zahirovic EarthByte Group, School of Geosciences, The University of Sydney, Camperdown, NSW, Australia

Introduction The unifying Theory of Plate Tectonics has revolutionized Earth sciences and our ability to contextualize geological processes. A number of leaps were required to adapt Alfred Wegener’s ideas of continental drift (Wegener, 1915) into a substantiated, self-consistent and quantifiable understanding of Earth’s plate-mantle system. The sketches of continental arrangements in a Pangea supercontinent have evolved into modern geographic information systems that can model the tectonic evolution of the planet over more than 1 billion years of Earth history. These models synthesize a wide range of geological and geophysical constraints into digital plate motion models that are integral to better understanding the formation and evolution of economic geological resources, as well as changes in the Earth’s biogeographic evolutionary trajectory (Rolland et al., 2015; Whitmore, 1982), climate (Raymo and Ruddiman, 1992; Hay, 1996; Arthur et al., 1985), oceanic circulation (Heine et al., 2004; Exon et al., 2002) and sea level (Mu¨ller et al., 2008) as a result of changing arrangements of oceans and continents.

The evolution of the plate reconstruction method Earth’s surface topography and geology have been interpreted for hundreds of years, but the mechanism for the formation of geomorphological and topographic features has evolved over time. Prior to plate tectonics, continental ‘permanence’ was argued primarily through (variants of) the geosynclinal theory in the 19th century. The formation of mountain belts and sedimentary basins was generally explained by ‘fixist’ ideas arguing for purely vertical motions of continents. One notable mechanism to explain relative vertical motions was through Earth contraction, resulting from ongoing planetary cooling, as argued by Eduard Suess (Austrian geologist). Similarly, James Dwight Dana and James Hall (both North American geologists) argued for vertical continental motions, with sedimentary loading typically invoked to explain subsidence through ‘downwarping’, leading to the propagation of ‘lateral pressure’ to induce folding of continental crust and thus the uplift of mountain ranges (Knopf, 1948). To explain the cooccurrence of the same fossils on either side of ocean basins, many of the fixist theories required the presence of ephemeral land bridges in the ocean basins, thus allowing the exchange of plants and animals between continents (Romm, 1994; Frisch et al., 2011). The earliest implied suggestion of horizontal continental mobility came from Abraham Ortelius (Flemish cartographer) in 1596, who invoked catastrophic flooding and earthquakes in tearing the Americas from Africa and Eurasia (Frisch et al., 2011). Similarly, Antonio Snider-Pellegrini (French geographer) in his 1858 book ‘The Creation and its Mysteries Revealed’ proposed that the similarity of the circum-Atlantic coastlines may be explained by a prior geographic affinity. His initial sketches of this continental arrangement largely constrained Regional Geology and Tectonics. DOI: https://doi.org/10.1016/B978-0-444-64134-2.00006-7

93

© 2020 Elsevier B.V. All rights reserved.

94

6. Resolving geological enigmas using plate tectonic reconstructions and mantle flow models

by coastline morphology have a surprisingly similar arrangement as the Pangea supercontinent that was proposed by Wegener (1915). However, it was only Alfred Wegener (German meteorologist, polar researcher and geophysicist) that compiled multiple independent pieces of geological evidence to infer the prior geographic proximity and affinity of the continents that made up the Pangea supercontinent, which required a ‘mobilist’ interpretation of continental geology. However, Wegener’s ‘continental drift’ remained controversial (and largely rejected) for decades as it implied that the weaker continents must plough through the stronger oceanic crust, leading to a lack of a convincing mechanism for the horizontal motion of continents (Oreskes, 1988). Perhaps the only positive outcome of World War I/II and the Cold War in the 20th century was the development of technologies that could better map the ocean basins. For example, more accurate depth soundings were possible through SONAR (SOund Navigation And Ranging), while magnetic anomalies could be detected with higher-precision magnetometers. Both of these technologies became more useful with better navigational accuracies through improvements in radio-based navigational systems, later to evolve into satellite-based Global Positioning Systems (Oreskes, 2003; Doel et al., 2006). The morphology of the seafloor was only revealed through painstaking plotting of sonar measurements by Marie Tharp, who, with her colleague Bruce Heezen, produced the first global bathymetric map in the 1950s (Barton, 2002). This work revealed that the oceans were not featureless basins, but instead dominated by a network of submarine mountain ranges that were later identified as seafloor spreading centres (Oreskes, 2003). As this mapping project preceded the formalized ideas around plate tectonics, Bruce Heezen and Marie Tharp both initially believed that the mid-oceanic ridge valleys best represented the proposed ‘stretch marks’ argued by the ‘expanding Earth’ hypothesis of Samuel W. Carey (Australian geologist) (Heezen and Tharp, 1965). The idea that new oceanic crust was generated at divergent mid-oceanic ridges (Hess, 1954; Hess, 1962) only gained momentum, once it was established that the Earth’s magnetic field underwent reversals over geological time, with symmetrical magnetic anomalies discovered on either flank of the spreading centre (Vine and Matthews, 1963; Vine, 1966) that were calibrated to a magnetic polarity reversal timescale (Cox et al., 1968). These discoveries partly motivated Edward Bullard (British geophysicist) to construct the first computational fit of the circum-Atlantic continents, where the conjugate bathymetric contours of the continental shelves were rotated to their predrift positions (Bullard et al., 1965). The method applied by Bullard et al. (1965) also minimized the misfits of the reconstructed bathymetric contours and inferred that the 500 fathom (B1000 m) bathymetric contour was the most reliable reference. Although later reconstructions would formalize the use of continent ocean boundaries rather than bathymetric contours, the regional reconstruction using computational methods by Bullard et al. (1965) was an important methodological advance at the dawn of plate tectonic theory. Complementing the creation of seafloor at the mid-oceanic spreading centres, were dipping planes of compressional earthquakes descending from the deep-sea trenches and into the mantle, which were first described by Kiyoo Wadati (Japanese seismologist) and Hugo Benioff (American seismologist) (Wadati, 1928; Benioff, 1949). To accommodate seafloor spreading and the required consumption of seafloor at the newly described subduction zones, transform boundaries were the last component required to formalize plate tectonics. Transform boundaries were first described by John Tuzo Wilson (Canadian geophysicist) (Wilson, 1965b) and were first invoked to be necessary for the motion of rigid plates on a sphere by McKenzie and Parker (1967). This foundational work led to the first global plate tectonic reconstructions were presented by Le Pichon (1968) who used fracture zones for directional constraints and limited seafloor magnetic anomaly data as age constraints to derive Euler rotations describing rigid lithospheric motions on a sphere. One important stride in confirming the new ideas of moving lithospheric plates came from the analysis of seamounts, notably the Hawaiian seamount chain, where the idea of stationary mantle hotspots burning a trail of intraplate volcanoes was formalized by Wilson (1965a). The model of mantle plumes, as the source of the thermal perturbations linked to hotspots, was another step towards better understanding the link between moving lithospheric plates and the convecting mantle beneath (Wilson, 1965a; Hess, 1962; Morgan, 1972). The motion of lithospheric plates was then attributed to principal plate driving forces (Forsyth and Uyeda, 1975), with slab pull as the dominant driver and ridge push as a secondary component, and additional forces that enhance plate motions (e.g. slab suction, etc.) or resist them (e.g. mantle drag, etc.). These new ideas together, formed over decades in different parts of the world, encapsulate the Theory of Plate Tectonics, leading to decades of research into reconstructing the Earth’s changing arrangement of continents, ocean basins and plate boundaries. In particular, the resulting models have shed light on important aspects of the plate-mantle system on a global scale, but have been vital in piecing together the chronology and physical mechanisms of major regional geological events.

Regional Geology and Tectonics

Global plate reconstructions

95

Global plate reconstructions Models of global plate motions require a comprehensive synthesis of relative plate kinematics, usually constrained by seafloor spreading histories, which are then embedded into a plate motion (‘absolute’) reference frame, typically constrained by a collection of hotspot tracks. However, as oceanic crust is lost to subduction over time, reconstructions of the Mesozoic and earlier often partly or entirely rely on paleolatitudinal constraints from paleomagnetic data, as well as using the contiguity of major geological features (sutures, orogens, etc.) and inferences on paleogeographic affinities using fossils and stratigraphic clues, much like the first-principles approach applied by Alfred Wegener.

Relative and absolute plate motions The most reliable indicator of relative plate motions are magnetic polarity reversals that are recorded in basaltic oceanic crust (Fig. 6.1). As oceanic crust is generated, the magnetic grains in the basaltic magma and lava orient with the contemporary magnetic field. When the temperature of the cooling basalt drops below the Curie point at B540 C (Zablocki and Tilling, 1976), the magnetic mineral orientations are trapped through subsequent magnetic reversals. Ocean-going expeditions that collect magnetic field measurements are planned to maximize coverage of unmapped regions of the oceanic crust (Fig. 6.1A), with a preference to cover seafloor perpendicular to the strike of the mid-oceanic ridge (i.e. roughly parallel to the spreading direction). As ocean-going expeditions capture the precise measurement of the strength of Earth’s magnetic field typically using a towed magnetometer (to avoid interference from the metal hull of the ship), extracting the magnetic anomaly requires subtracting the regional average magnetic field strength (defined by the International Geomagnetic Reference Field) from the measured magnetic field. The measured magnetic anomalies arise because the oceanic crust interacts with and modulates the magnetic field, where oceanic crust that formed during a ‘reversed’ polarity tends to dampen (i.e. reduce amplitude) the strength of the magnetic signal, while oceanic crust formed during ‘normal’ polarity tends to amplify the magnetic field strength. Magnetic anomalies can then be identified by correlating the recovered pattern of reversals to the pattern in the global magnetic polarity timescale (Fig. 6.1D). This process is usually manual, requiring user decisions and interpretations (Mendel et al., 2005). Some constraints on the absolute age of the oceanic crust, such as radiometric ages of dredged basalt and/or drilled seafloor, can help the user increase the confidence in their interpretations. However, the identification process is hindered by remagnetizations that may occur as a result of subsequent volcanic overprinting (such as intraplate hotspot volcanism), and the pattern itself can be obscured by ridge jumps and a thick cover of sediments (Mendel et al., 2005; Hellinger, 1981). In particular, there are no magnetic reversals in the period between B120 and 83 Ma during the Cretaceous Normal Superchron (CNS, Fig. 6.1). In addition, seafloor generated near the geographic equator, where the magnetic field lines are generally horizontal with the Earth’s surface, leads to ambiguous preservations of the magnetic inclination and hence usually cannot be used to infer magnetic polarity reversals (Vine and Matthews, 1963). This ambiguity is most obvious in the equatorial Atlantic and parts of the Pacific. The last complicating factor in magnetic anomaly identifications is that seafloor that formed as a result of exhumed mantle rocks (e.g. peridotite), usually in ultraslow spreading centres, has a more complex magnetic signature than seafloor underlain by basaltic rocks (e.g. Bronner et al., 2014). The resulting magnetic anomaly identifications, often referred to as ‘magnetic picks’, can be compiled as a collection of points on the seafloor along the ship-tracks where they were interpreted. The magnetic picks can be compiled for a region of interest, or globally (Fig. 6.1B) (Seton et al., 2014), and used to interpolate the age of the oceanic crust (Fig. 6.1C) and to compute relative plate motions. The magnetic anomalies of the same age on either flank of the mid-oceanic ridge are rotated in a spherical coordinate system, with fracture zones that are typically derived from satellite gravity data used as directional constraints (e.g. Matthews et al., 2011b). The misfits are minimized during the rotation computation, usually using the Hellinger (1981) method, which computes the best-fit Euler rotations that can be used in plate tectonic reconstructions. Each Euler rotation is defined by a pole of rotation (geographic coordinate), as well as an angle of rotation. The magnitude of the angle of rotation is an indicator of the distance that is covered, and the sign of the angle provides a direction of motion about the pole of rotation (where positive is clockwise motion). Where seafloor has been lost to subduction but one flank of the seafloor spreading system has been preserved, an assumption of spreading symmetry can be applied to resurrect subducted oceanic crust.

Regional Geology and Tectonics

96

6. Resolving geological enigmas using plate tectonic reconstructions and mantle flow models

FIGURE 6.1 (A) Magnetic anomaly measurements from ship track and satellite data (Maus et al., 2009), (B) magnetic anomaly identifications (coloured points) (Seton et al., 2014) and fracture zones (grey lines) (Matthews et al., 2011b), (C) interpolated seafloor age-grid (Mu¨ller et al., 2016), plate velocities (arrows) and hotspots (white triangles) (Whittaker et al., 2015) and (D) idealized magnetic polarity reversals pattern for the oceanic crust using the magnetic polarity timescale of Gee and Kent (2007), compared with the seafloor age, and the global timescale of Ogg et al. (2016).

Characterizing plate boundaries and their evolution within continents is markedly more difficult. Quantifying relative motions across continental rifts prior to seafloor spreading, such as in the East African Rift (Chorowicz, 2005), requires robust age and kinematic constraints. The onset of rifting is typically estimated by dating the base of syn-rift sediments, where they may be exposed or drilled and in the ideal situation complemented by the chronology of rift-related volcanism. However, these ages are often stratigraphic or biostratigraphic ages with broader ranges than radiometrically defined ages. The magnitude of extension across a rift can also be obscured by asymmetric rifts controlled by lithospheric-scale detachment faults (Wernicke, 1981), rather than through pure shear uniform symmetrical stretching (McKenzie, 1978). In addition, continental plate boundaries tend to be diffuse

Regional Geology and Tectonics

Global plate reconstructions

97 FIGURE 6.2 Generalized plate motion hierarchy (dashed blue lines) during Pangea breakup highlights that the trunk of absolute plate motions is tied to South Africa (red and yellow circle). Relative plate motions are preferably linked through rifts or midoceanic ridges, which have far less uncertainty than links through subduction zones.

and thus distribute deformation over a larger region than typical oceanic boundaries, although even intra-oceanic diffuse boundaries like those surrounding the present-day Capricorn Plate exist (Kreemer et al., 2003; Bird, 2003). As a result, incorporating deformation into plate reconstructions has been a long-standing challenge, which has only recently been tackled for the post-Pangea timeframe (Mu¨ller et al., 2019). As the seafloor spreading record only provides constraints on relative plate motions, other constraints are used to infer how these relative plate motions move in an absolute sense with respect to a particular reference frame. The most reliable indicator of absolute plate motions, with respect to the base of the mantle, can be derived from hotspot tracks, which were first described in a plate tectonic context by John Tuzo Wilson (Canadian geophysicists) in the 1960s (Wilson, 1965a; Wilson, 1963). Multiple hotspot tracks across different plate systems can be used to derive mantle reference frames (e.g. Mu¨ller et al., 1993; O’Neill et al., 2005; Torsvik et al., 2008), but the reliable hotspot trails themselves extend only back to approximately 70 100 Ma due to the fact that much of the seafloor from earlier times has been consumed by subduction. For earlier times, a synthesis of (paleomagnetic) apparent polar wander paths across multiple continental blocks can be used to generate a paleomagnetic absolute reference frame, where plate motions are tied back to Earth’s spin axis (Torsvik et al., 2008). However, paleomagnetic data can also contain true polar wander, which represents the wholesale rotation of the mantle and lithosphere with respect to the Earth’s core. In order to isolate the plate-mantle system, especially for models of mantle convection, the true polar wander component of motion is removed (Steinberger and Torsvik, 2008) in constructing a paleomagnetic-based reference frame for the plate-mantle system. It is the combination of relative plate motions embedded in an appropriate reference frame that absolute plate motions, whether with respect to the spin axis or the base of the mantle, can be calculated (Fig. 6.2).

Early plate reconstruction approaches The relative plate motions derived from seafloor magnetic anomalies can undo seafloor spreading to constrain the post-Pangea tectonic evolution over the last B200 Myr. Early plate reconstruction models relied on the limited data that were available, with the first global reconstruction made by Le Pichon (1968) who modelled the motion of six large rigid blocks at several snapshots for the Cenozoic, and tentatively back to 120 Ma in the Cretaceous. These reconstructions are complemented by the rigid fit restoration of the Atlantic blocks in their

Regional Geology and Tectonics

98

6. Resolving geological enigmas using plate tectonic reconstructions and mantle flow models

FIGURE 6.3 Present-day plate boundaries of the Australasian region from Bird (2003) and topography from Amante et al. (2009) and seafloor age (Mu¨ller et al., 2016; Zahirovic et al., 2016b) (colour scale in Fig. 6.1). Volcanic provinces, largely related to mantle plume eruptions, from Johansson et al. (2018), are plotted as grey polygons, with presentday hotspots from Whittaker et al. (2015) plotted as white triangles (those with question marks indicate somewhat uncertain hotspot designations). The blue star is used in Fig. 6.4 to extract full seafloor spreading rates and azimuths for the Southeast Indian Ridge (SEIR), and the green star is used to compute the convergence rates and azimuths for India and Eurasia using the Zahirovic et al. (2016b) plate reconstructions. The orange outline of Greater India represents the small end-member option for the precollisional size of India (e.g. Ali and Aitchison, 2005; Gibbons et al., 2012), and purple represents a larger estimated size (e.g. Lee and Lawver, 1995). Yellow stars indicate approximate locations where dynamic topography is extracted from the numerical models of mantle flow (see Fig. 6.7). Red polygons with white outlines are ophiolites that represent major suture zones where ocean basins have been lost to subduction. The sutures (blue) indicate consumed ocean basins relevant to this chapter, highlighted by ophiolite belts (Hutchison, 1975; Zahirovic et al., 2016b; Zahirovic et al., 2014), with the (1) IndusTsangpo, (2) Woyla, (3) Luk-Ulo, (4) Meratus and (5) Sepik suture zones.

Pangea configuration made earlier by Bullard et al. (1965) and thus provided the foundation for future refinement to plate reconstruction approaches. One of the earliest plate reconstructions that used newly identified magnetic anomalies of the eastern Pacific also considered the evolution of regional plate boundaries, namely the subduction and demise of the Farallon Plate and the establishment of the San Andreas Fault system on the western North American margin in the Cenozoic (Atwater, 1970). Following significant data collection from the ocean basins, new global reconstructions that had higher spatial and temporal resolutions and also extended further back in time to cover the entire Cretaceous were presented by Scotese et al. (1988). Crucially, the Euler rotations that made up the plate motion model were made available, allowing others to build on these global reconstructions. A similar approach was applied to the circum-Indian Ocean region by Besse and Courtillot (1988), which allowed the authors to extract additional information from the models in order to address regional geological problems. For example, the velocity of India’s northward motion indicated a slowdown at B50 Ma, which was interpreted by Besse and Courtillot (1988) as the onset of the India-Eurasia collision (and a similar analysis using the latest plate reconstructions is presented in Fig. 6.3). The geometrical requirement in plate tectonics implies that lithospheric plates are in motion relative to one another, thus requiring a network of plate boundaries to accommodate the relative motions. Jurdy and Van der Voo (1975) presented some of the first work that modelled both the restored positions of lithospheric blocks and a continuous network of plate boundaries for three snapshots since the Cretaceous. Using these plate topologies, particularly the snapshot at 55 Ma in the Eocene, Solomon et al. (1977) highlighted how global plate reconstructions can be used to decipher global tectonic enigmas, while also providing an insight into specific regional problems. The authors compared the computed plate velocity field at 55 Ma and the present to understand the range

Regional Geology and Tectonics

Global plate reconstructions

99

FIGURE 6.4 Global plate reconstructions can be used to estimate relative and absolute plate kinematics, in this case derived from

Zahirovic et al. (2016b). In this example, the convergence rate between India and Eurasia, using a present-day coordinate (30 N, 80 E), is represented by the thick green line. The full seafloor spreading rate along the Southeast Indian Ridge (see blue star in Fig. 6.3) is represented by the thick blue line. The thin coloured lines indicate directional information for the kinematics of the two points.

of plate velocities at these times, with a focus on whether viscous drag beneath continental keels led to a differential velocity of the oceanic and continental lithosphere. Notably, the 55 Ma snapshot highlighted the rapid motion of India prior to collision with Eurasia, which was similar to the speed of oceanic plates, and the authors argued that continents (and their keels) likely do not slow down plate velocities (Solomon et al., 1977). This early work highlighted the power of using global plate reconstructions to better understand the mechanisms influencing plate velocities, while the benefit of hindsight and decades of additional data may instead suggest that the 55 Ma precollision rapid motion of India may not be representative of typical speeds attainable by continental lithosphere. For example, the precollision anomalous speed of India (Fig. 6.4) may be the result of a combined effect of the Reunion plume head lubrication of the asthenosphere (leading to lower coupling with the underlying mantle) from B65 Ma as well as slab pull and slab suction from two contemporaneous subduction zones in the NeoTethys north of India (Jagoutz et al., 2015; Cande and Stegman, 2011). Gordon and Jurdy (1986) used a global plate reconstruction that included plate boundaries at six intervals in the Cenozoic, which allowed them to extract the velocity field across each plate and thus compute the role of continents in modulating plate velocities. Gordon and Jurdy (1986) argued that plates with high continentality, namely high portion of plate area as continental lithosphere, tended to move more slowly. The opposite argument was essentially made by Stoddard and Abbott (1996), although plate motion models with higher temporal resolution were needed to guide further progress in this debate.

Linking plate reconstructions with mantle flow By tracking the evolution of plate boundaries, it became easier to also couple the motion of plates on the surface with the convective evolution of the mantle beneath. Richards and Engebretson (1992) and LithgowBertelloni and Richards (1998) presented plate reconstructions with plate boundaries back to 180 Ma over a number of stages and crucially tied the modelled subduction history to a model of mantle convection. Such a global approach allowed the plate reconstructions to be tested against seismic tomography at the present-day, and the predicted geoid could be compared with the observed values. In addition, the semianalytical mantle convection model of Lithgow-Bertelloni and Richards (1998) predicted the time-evolving contribution to Earth’s surface topography since 180 Ma as a result of mantle flow, called ‘dynamic topography’, driven by the imposed subduction history. This evolving and transient topographic signal resulting from mantle flow, typically of several hundred metres in amplitude and hundreds to thousands of kilometres in wavelength, has been long overlooked in both industry and academia. However, dynamic topography has led to large-scale regional subsidence and emergence of continents that could not be previously explained by other mechanisms such as tectonic topography (i.e. uplift from collisions, subsidence from rifting) or eustasy (i.e. global sea-level change). A community open-source and cross-platform software tool in the form of GPlates (www.gplates.org; Boyden et al., 2011; Mu¨ller et al., 2018), first released in 2008, enabled users to interactively create and edit global and

Regional Geology and Tectonics

100

6. Resolving geological enigmas using plate tectonic reconstructions and mantle flow models

regional plate reconstructions. Notably, GPlates functionality was extended significantly by 2012 to model evolving plate boundaries in 1 Myr intervals, with the workflow described and an accompanying global reconstruction back to 140 Ma provided in Gurnis et al. (2012). This was followed by global reconstructions extending back to 200 Ma (Seton et al., 2012), 230 Ma (Mu¨ller et al., 2016) and 410 Ma (Domeier and Torsvik, 2014; Matthews et al., 2016; Young et al., 2019), with work underway to extend full-plate reconstruction to 1 Ga and beyond (Merdith et al., 2017). Several approaches have been developed to link global plate motions with numerical models of mantle flow, including the application of traction and plate velocities on the surface of mantle convection codes. The plate velocities drive mantle flow, which also interacts with heating from the core mantle boundary, leading to large-scale global convection in a spherical shell representing the mantle. These coupled models have been used to demonstrate the Cenozoic north-eastward tilt of the Australian continent resulting from the dynamic topography subsidence signal as Australia overrode subducting slabs from the West Pacific and Southeast Asia (DiCaprio et al., 2009; DiCaprio et al., 2011). Similarly, the subsidence history interpreted from wells on the Australian continent could be used to constrain the distance of the enigmatic subduction zone east of Australia in the Cretaceous (Matthews et al., 2011a). A similar approach was applied to study the history of inundation of North America during the Cretaceous, as a combination of higher global sea levels and dynamic subsidence from the sinking Farallon slab led to the formation of the Western Interior Seaway (Spasojevic et al., 2009). The assimilation of plate reconstructions into global mantle flow models was improved by Bower et al. (2015) who not only incorporated the plate velocities on the surface, but also the thermal (and compositional) structure of the lithosphere through time. Although this approach is semianalytical as it imposes the slab structure, it results in more realistic single-sided and uninterrupted subduction that can be problematic where the slab structure is not assimilated in the uppermost mantle. This technique builds on the work of Lithgow-Bertelloni and Richards (1998) and has been widely used to estimate the time-evolving dynamic topography (Flament et al., 2013) and mantle structure (Hassan et al., 2016) that provide vital insights into addressing outstanding global and regional geological enigmas.

Using global plate reconstructions to better understand the Earth system Plate tectonics has strong influences on many components of the Earth system, especially on geological timescales. Global plate tectonic reconstructions have been used to better understand the tectonic component of global sea-level change, for example, faster spreading rates typically lead to shallower ocean basins and hence higher sea levels, as occurred in the Cretaceous (Mu¨ller et al., 2008; Conrad, 2013). Ocean chemistry, and namely the oscillation between aragonite and calcite sea, can also be inferred using global plate tectonic reconstructions, where the length of mid-oceanic ridge systems and the ridge flank areas have a strong influence over the resulting Mg/Ca ratios of seawater (Mu¨ller et al., 2013). The changing arrangement of continents and ocean basins over time has a fundamental control on the distribution of oceanic gateways and related ocean currents. Plate reconstructions have linked the closure of the equatorial Tethyan gateway as a result of the collisions of Africa, Arabia and India with Eurasia since the Eocene to changes in the equator-to-pole heat transport efficiency (Berggren, 1982; Sijp et al., 2014). Similarly, the closure of the Indonesian gateway during Australia’s northward journey towards Asia has been implicated in the aridification of Africa in the last 4 Myr (Cane and Molnar, 2001). As the Indonesian gateway closed, the Southern Ocean gateway opened and led to the establishment of the Antarctic Circumpolar Current (Exon et al., 2002). This rearrangement of oceanic circulation has also been implicated in contributing to the thermal isolation and glaciation of Antarctica, which led to long-term sea-level falls since B 35 Ma (Barker and Thomas, 2004). Plate reconstructions are also crucial in tracking the exchange of carbon between the mantle and Earth’s surface as part of the deep carbon cycle. For example, the eruption of large igneous provinces resulting from the arrival of mantle plumes on the surface has significantly perturbed atmospheric CO2 concentrations, with perhaps the best example being the Siberian Traps volcanism (and related decarbonation of the continental crust) at the Permo-Triassic boundary (B252 Ma) and the associated most significant mass extinction event in the geological record (Wignall, 2001). However, the motion of subaerial large igneous provinces through the near-equatorial humid belt leads to longer-term sequestration of atmospheric CO2 through chemical weathering (Johansson et al., 2018), highlighting the importance of paleogeographic reconstructions in accounting for fluxes and perturbations to the deep carbon cycle. As the more recent plate reconstructions capture evolving plate boundaries at higher temporal resolutions (up to 1 Myr) and extend further back in time than previous models, it is also possible to extract more kinematic information that is embedded in the history of plate motions. In particular, previous research had either

Regional Geology and Tectonics

India-Eurasia collision

101

supported the idea that continents tend to slow down plate motions (Gordon and Jurdy, 1986), while other work argued for the opposite (Stoddard and Abbott, 1996; Solomon et al., 1977). An analysis of the Seton et al. (2012) plate reconstructions that extend to 200 Ma (in 1 Myr intervals) strongly indicates that a high portion of continental plate area (referred to as ‘high continentality’), especially in terms of Archean cratons, leads to generally much slower plate velocities than largely oceanic lithospheric plates (Zahirovic et al., 2015). The likely mechanism for the slowdown of plates with thick continental keels is the increased traction between the lithosphere and upper mantle, considering that the lower-viscosity asthenosphere is more constricted by the deep continental roots. More broadly, plate motion models provide an important framework for better understanding major regional events, especially in terms of their mechanisms and chronology. Although ongoing debate surrounds the nature and chronology of the India-Eurasia collision, which has uplifted the world’s most significant mountain belt and modified regional (and likely global) climate, relative and absolute plate motions shed important clues on this debate that may be more difficult to interpret from the complex suture zone in the Himalayas.

India-Eurasia collision The India-Eurasia collision, which occurred sometime in the Eocene, consumed the youngest Tethyan ocean basin and uplifted the world’s tallest mountain range. However, the nature and chronology of the collision remain unresolved. One significant attempt at resolving this issue was tackled by Lee and Lawver (1995), who presented a regional and entirely reproducible model with published Euler rotations, which is often not provided in publications for the region. These reconstructions allowed Lee and Lawver (1995) to compute the convergence rate between India and Eurasia, who noted a two-step decrease in convergence velocities. The precollisional speed of India relative to Eurasia was approximately 17 cm/yr, and dropped to B10 cm/yr at B60 Ma, and dropped again significantly to 6 cm/yr at B45 Ma. These kinematically constrained results indicated a two-stage slowdown in India-Eurasia convergence, leading the authors to interpret a ‘soft’ continent-continent collision at B60 Ma, followed by a ‘hard’ collision at B45 Ma due to the arrival of stronger Indian cratonic lithosphere near the collision zone. However, the biggest uncertainty in this approach is actually the size of the colliding continent, namely Greater India. Much of this continental promontory is underthrust beneath Eurasia, crumpled up into the Tethyan Himalayas and/or eroded into the major regional sedimentary depocentres (namely, the Indus and Bengal fans). To accommodate the B60 55 Ma initial continent-continent collision, Lee and Lawver (1995) infer a larger end-member that extends B2000 km north of the present-day suture zone (Fig. 6.3). In contrast, other studies prefer a much smaller Greater India, extending no further than B1000 km north of the present-day suture zone. In the eastern portion of Greater India, this smaller estimate is consistent with the required geometric fit with Australia and the areal limit imposed by the Wallaby-Zenith Fracture Zone system (Gibbons et al., 2012; Ali and Aitchison, 2005) during the Gondwana fit of India with Australia. However, the western part of Greater India is largely unconstrained. When reconstructed, a smaller Greater India is nowhere near the Eurasian margin at B60 55 Ma, which requires an alternative explanation for the dramatic slowdown in the India-Eurasia convergence that is observed. Instead of a continent-continent collision at this time, many authors argue for an initial arc-continent collision, based on geological evidence from the suture zone for intraoceanic subduction south of the Eurasian margin prior to the India-Eurasia collision (Aitchison et al., 2007; Aitchison et al., 2000). This idea was first raised decades ago by Bard (1983) who suggested that Greater India may have first collided with the intra-oceanic Kohistan Arc, but has only recently gained more traction (Van der Voo et al., 1999; Zahirovic et al., 2012; Zahirovic et al., 2014). An intra-oceanic arc would lead to two collisions, first between Greater India and the island arc, followed by a continent-continent collision with a two-step decrease in plate velocities and without requiring an unnecessarily large Greater India. Independent of the India-Eurasia convergence rates, which require a complex plate circuit where errors can propagate, the changes in seafloor spreading rate and direction in the Indian Ocean provides additional clues (Fig. 6.4). A comprehensive analysis of the seafloor spreading history in the Indian Ocean by Cande et al. (2010) revealed a significant slowdown in spreading rates between 51.9 and 45.3 Ma, followed by a significant change in the seafloor spreading direction between B43.0 and 41.5 Ma. These kinematic constraints, along with evidence for intra-oceanic subduction preceding the India-Eurasia collision, suggest a progressive arc-continent collision between B60 and 52 Ma, followed by a continent-continent collision at B43 Ma (Zahirovic et al., 2016b; Zahirovic et al., 2012; Zahirovic et al., 2014; Gibbons et al., 2015). The collision timings are crucial in understanding the uplift of the Tibetan Plateau, and the resulting modification to ocean circulation, and the changing intensity of the Asian monsoon.

Regional Geology and Tectonics

102

6. Resolving geological enigmas using plate tectonic reconstructions and mantle flow models

Sundaland and New Guinea Further east of the India-Eurasia collision zone, the eastern Tethyan tectonic evolution is defined by a more complex interaction of converging Indo-Australian, Eurasian and Pacific plates following Pangea breakup. The global plate reconstructions, incorporating this complex Tethyan-Pacific linking region, have been used to better understand both regional and global processes. For example, the closure of the Indonesian gateway following the collision of the Southeast Asian continental promontory, Sundaland, and the New Guinea margin has significantly modified the oceanic exchange between the Pacific and Indian oceans (Heine et al., 2004; Gaina and Mu¨ller, 2007). In addition, the interaction between the Earth’s surface and underlying mantle in this region has controlled regional flooding patterns that are out of sync with global sea level, which highlights the need to consider the plate-mantle system in interpreting stratigraphic and other geological constraints. The first tectonic synthesis of Southeast Asia was presented by Hamilton (1970) and Katili (1971), approaches that evolved into schematic reconstructions of the region (Katili, 1975; Pigram and Panggabean, 1984; Metcalfe, 1988; Audley-Charles, 1988; Rangin et al., 1990). Further incorporation of seafloor spreading constraints led to more detailed plate reconstructions that also provided reproducible Euler rotation parameters (Besse and Courtillot, 1988; Scotese et al., 1988; Lee and Lawver, 1995). More detailed reconstructions of Southeast Asia have since been presented (Hall, 1996; Hall, 2002; Hall, 2012), but these have not provided any Euler rotation parameters. Plate tectonic reconstructions of the Southeast Asia region that also consider plate boundary evolution, embedded in global models, have been generated using GPlates, where the Euler rotations and digital model geometry files have been linked to numerical models of mantle flow (Zahirovic et al., 2016b; Zahirovic et al., 2014). By applying plate reconstructions as surface boundary conditions to global mantle flow models (Bower et al., 2015), the resulting mantle evolution can be used to track the interaction of subducted slabs from the complex Indo-Australia, Eurasia and Pacific tectonic convergence zone (Figs 6.5 and 6.6). The predicted present-day mantle structure from the numerical models can be compared with P- and S-wave seismic tomography constraints (Fig. 6.7), which help identify more plausible end-member tectonic scenarios where the geological constraints may be vague or incomplete. Where the present-day mantle structure is relatively well constrained, the timeevolving mantle flow models can be interrogated to provide insight into regional geological enigmas. Despite falling long-term global sea levels since B35 Ma, largely driven by the growth of significant and permanent continental ice sheets on Antarctica (DeConto and Pollard, 2003), the long-term paleogeographic evolution of Sundaland and New Guinea is one of progressive flooding rather than emergence (Figs 6.8 and 6.9). In the case of Sundaland, the accretion of the Woyla Arc onto the Sumatra margin in the Late Cretaceous (B80 Ma) (Zahirovic et al., 2016a) led to a B15 Myr interruption in arc volcanism (McCourt et al., 1996) and subduction, leading to broad uplift of the region (Fig. 6.10). The combination of high global sea levels and strong negative dynamic topography, resulting from sinking slabs beneath Sundaland, led to broad flooding of the region. However, following the interruption of subduction at B80 Ma, the entire region experienced up to several hundred metres of dynamic uplift, enough to trigger emergence and widespread erosional and nondepositional environments on Sundaland (Fig. 6.11). This process manifested as a regional unconformity across much of Southeast Asia spanning the Late Cretaceous to Eocene times (Clements et al., 2011; Clements and Hall, 2011). Renewed subduction by B60 Ma led to progressive dynamic subsidence of Sundaland, which was strongest since B30 Ma (Yang et al., 2016; Yang et al., 2018; Zahirovic et al., 2016a). Similarly, the broad topography of the New Guinea margin was also modified by mantle convection (Fig. 6.10), with progressive dynamic uplift between B155 and 60 Ma, resulting from the opening of a back-arc basin and the oceanward shift in subduction (Fig. 6.11). However, as Australia advanced northward (Fig. 6.10), the back-arc basin was consumed as the Sepik Arc collided with the New Guinea margin in the Eocene. As Australia overrode the sinking slabs, the southwestward subduction of the (proto) Solomon Sea Plate initiated by B20 Ma to produce the Maramuni Arc volcanics. The interaction of these sinking slabs and the continental lithosphere led to broad subsidence since B60 Ma (Figs 6.5, 6.6, and 6.10). The paleogeographic maps indicate that the subsidence dominated the effect of falling sea levels from B40 Ma, leading to widespread long-term flooding of the New Guinea margin (Harrington et al., 2017). The example of long-term flooding of Sundaland and New Guinea since B40 Ma that is out of sync with global sea level highlights the need to consider the interaction between plate motions on the surface and the convective mantle beneath. It also highlights that the role of changing trends in dynamic topography from changes in subduction has a significant influence over regional patterns of erosion and sedimentation. This broad regional signal is ephemeral due to the nature of the convecting mantle where slabs in the lower mantle contribute less to

Regional Geology and Tectonics

Sundaland and New Guinea

103

FIGURE 6.5 The mantle flow models, driven by plate motions (Bower et al., 2015), can be visualized as three-dimensional volumes, as shown here from GPlates, where sinking slabs are shown as blue volumes and hotter upwelling material from the core mantle boundary is shown as red. These four-dimensional models enable the user to track the interaction of sinking slabs from multiple proximal subduction zones in a global whole-mantle context. The numerical mantle flow model presented in this chapter follow the setup of Case 5 in Zahirovic et al. (2016b).

Regional Geology and Tectonics

104

6. Resolving geological enigmas using plate tectonic reconstructions and mantle flow models

FIGURE 6.6 More detailed interrogations of the models can be achieved by extracting two-dimensional vertical slices of the mantle temperature and in this case with the profiles attached to the moving overriding plate for Sundaland and New Guinea. The vertical slices highlight the evolving mantle structure that is driven by the plate motions on the surface, namely the evolution of subduction and sinking slabs from a number of regional subduction zones. CAR, Caroline Plate; MT, Meso-Tethys; PSCS, Proto South China Sea; PSP, Philippine Sea Plate; PSS, Proto Solomon Slab; S, Sundaland; SBA, Sepik back-arc basin; SPK, Sepik terrane; W, Woyla terrane.

Regional Geology and Tectonics

Sundaland and New Guinea

105

FIGURE 6.7 The present-day predicted mantle structure from the mantle flow models can be compared to seismic tomographic constraints. In this example for the Sundaland and New Guinea, the predicted slab contours (black) are superimposed on P-wave seismic tomography models from MIT-P08 (Li et al., 2008) and UU-P07 (van der Meer et al., 2018; Amaru, 2007). These qualitative comparisons highlight an overall consistency between the numerical mantle flow models and the present-day mantle structure, but also highlight aspects of the models that require further refinement.

dynamic subsidence than slabs closer to the surface, while the lithosphere is also in constant motion and hence is exposed to different upwelling or downwelling mantle domains. These approaches can then be used in conjunction with surface process models to better model the erosion and deposition of sediments (Salles and Hardiman, 2016) and thus better capture the complex interactions between deep Earth and surface processes.

Conclusions The emergence of the unifying Theory of Plate Tectonics in the geosciences has revolutionized our understanding of Earth processes. It has been applied to better understand the changing distribution of land and sea as part of the supercontinent cycle, which itself modulates global sea level, ocean circulation, climate, biogeographic dispersal and the formation of mineral and energy resources. Plate tectonic reconstructions have evolved over the decades to include more data from marine and continental settings, leading to an overall reduction in uncertainties. However, gaps in data remain, often due to the destructive processes of subduction and continental collisions. Deformation of continental lithosphere presents a challenge for robust restorations of past continental configurations (Mu¨ller et al., 2019), which will require better age constraints and estimates of shortening and extension. In preserved oceanic crust, magmatic overprinting, complex series of ridge jumps and large areas of oceanic crust without magnetic reversals formed during the CNS remain as challenges highlighted for future work. Global and regional plate reconstructions, often coupled with numerical models of mantle flow, have been used to explore end-member scenarios and help better understand some key regional geological enigmas. Dynamic topography, as the vertical motion of the surface resulting from mantle flow, has been a longoverlooked phenomenon in understanding regional paleogeographic evolution. In particular, the examples of long-term subsidence and flooding of Sundaland and New Guinea in Southeast Asia since B40 Ma that is entirely out of sync with the global trends in sea level can only be reconciled when dynamic topography is considered. More broadly, these examples highlight the need for community digital plate tectonic reconstructions that capture evolving plate boundaries and hence enable the linking between surface plate motions and

Regional Geology and Tectonics

106

6. Resolving geological enigmas using plate tectonic reconstructions and mantle flow models

FIGURE 6.8 Paleogeographic reconstructions of Sundaland (left, A D) and New Guinea (right, E H) indicate progressive inundation of continental crust (blue) since B30 Ma, which is out of sync with falling long-term global sea levels (Fig. 6.10). For Sundaland, the palaeogeography also highlights the long-term emergence and erosional environments (orange) from the Late Cretaceous to Oligocene times, resulting in a widely discussed regional unconformity (Clements et al., 2011). Sundaland palaeogeography is synthesized by Zahirovic et al. (2016a) from Golonka et al. (2006) and New Guinea palaeogeography is synthesized by Harrington et al. (2017) from Norvick (2003).

Regional Geology and Tectonics

FIGURE 6.9 (A) The long-term global sea level (Haq et al., 1987) has been falling since the onset of Antarctic glaciation at B34 Ma, while (B) both New Guinea and Sundaland experienced progressive long-term inundation based on the paleogeographic reconstructions. This counterintuitive relationship requires a mechanism other than eustasy to be driving the paleogeographic evolution of these regions.

FIGURE 6.10 The global plate reconstructions since the latest Jurassic (left), with refinements for the eastern Tethys (Zahirovic et al., 2016b), are used to drive global mantle flow models and estimate the dynamic topography (right) acting on the lithosphere. Dynamic topography is calculated following Flament et al. (2014) and Zahirovic et al. (2016a). The subduction of the Tethyan and (proto-) Pacific plates drives dynamic subsidence and those sinking slabs (as well as the complementary seafloor spreading elsewhere) help drive the mantle return flow and the broad regions of dynamic uplift. Continents and terranes experience an evolving dynamic topography signal as they move over different convective domains or experience the onset or cessation of a major subduction zone. SUN, Sundaland; NG, New Guinea.

108

6. Resolving geological enigmas using plate tectonic reconstructions and mantle flow models

FIGURE 6.11 The dynamic topography predicted by the numerical mantle flow models is spatially and temporally transient due to the evolving mantle, but also due to the moving lithosphere on the surface. The yellow stars from Fig. 6.10 are used to sample the dynamic topography acting on New Guinea and Sundaland since the latest Jurassic. For New Guinea, a prolonged dynamic uplift during the Cretaceous is predicted due to slab roll-back to open the Sepik back-arc basin. The increasing distance between the New Guinea margin and the sinking slabs causes this relative dynamic uplift, with dynamic subsidence starting in the latest Cretaceous and early Cenozoic due to the arrival of the Sepik terrane (and associated subduction system). Australia’s rapid northward motion from B45 Ma results in the New Guinea margin overriding more subducted slab volumes, leading to a general dynamic subsidence trend. The New Guinea paleogeographic maps indicate flooding at sea-level high-stands since B25 Ma, which is in contradiction to the long-term falling sea levels since B35 Ma. For Sundaland, Woyla accretion at B80 Ma interrupts subduction and leads to relative dynamic uplift during the latest Cretaceous to latest Eocene times. Renewed subduction from B60 Ma along the Sundaland margin causes a progressive increase in subducted slab volumes in the upper mantle, leading to strong dynamic subsidence from B30 Ma predicted by the models, generally contemporaneous with the regional flooding pattern, but out of sync with long-term global sea-level trends.

underlying mantle convective processes. These approaches form a pathway towards also modelling the interaction between deep Earth and surface processes resulting from mantle flow, isostatic and flexural topography, eustatic sea level and atmospheric processes.

Acknowledgements The author was supported by the Australian Research Council grant IH130200012 and the Deep Carbon Observatory through Alfred P. Sloan Foundation grants G-2017-9997 and G-2018-11296. The numerical mantle flow models presented were computed using resources from the National Computational Infrastructure (NCI), which is supported by the Australian Government. GPlates development was supported by the AuScope National Collaborative Research Infrastructure System (NCRIS), and CitcomS development was supported by the Computational Infrastructure for Geodynamics (CIG).

References Aitchison, J., Davis, A., Liu, J., Luo, H., Malpas, J., Mcdermid, I., et al., 2000. Remnants of a Cretaceous intra-oceanic subduction system within the Yarlung-Zangbo suture (southern Tibet). Earth Planet. Sci. Lett. 183, 231 244. Aitchison, J., Ali, J., Davis, A., 2007. When and where did India and Asia collide? J. Geophys. Res. 112, 1 19. Ali, J., Aitchison, J., 2005. Greater India. Earth-Sci. Rev. 72, 169 188. Amante, C., Eakins, B., Boulder, C., 2009. ETOPO1 1 Arc-Minute Global Relief Model: Procedures, Data Sources and Analysis. NOAA Technical Memorandum. Amaru, M., 2007. Global Travel Time Tomography With 3-D Reference Models. Utrecht University. Arthur, M.A., Dean, W., Schlanger, S., 1985. Variations in the global carbon cycle during the Cretaceous related to climate, volcanism, and changes in atmospheric CO2. In: The Carbon Cycle and Atmosphere CO2: Nature Variations Archean Present, vol. 32, pp. 504 529. Atwater, T., 1970. Implications of plate tectonics for the Cenozoic tectonic evolution of western North America. Geol. Soc. Am. Bull. 81, 3513 3536. Audley-Charles, M.G., 1988. Evolution of the southern margin of Tethys (North Australian region) from early Permian to late Cretaceous, Geological Society, London, Special Publications, 37. pp. 79 100. Bard, J.P., 1983. Metamorphism of an obducted island arc: example of the Kohistan sequence (Pakistan) in the Himalayan collided range. Earth Planet. Sci. Lett. 65, 133 144. Barker, P., Thomas, E., 2004. Origin, signature and palaeoclimatic influence of the Antarctic Circumpolar Current. Earth-Sci. Rev. 66, 143 162.

Regional Geology and Tectonics

References

109

Barton, C., 2002. Marie Tharp, oceanographic cartographer, and her contributions to the revolution in the Earth sciences, Geological Society, London, Special Publications, 192. pp. 215 228. Benioff, H., 1949. Seismic evidence for the fault origin of oceanic deeps. Geol. Soc. Am. Bull. 60, 1837 1856. Berggren, W.A., 1982. Role of ocean gateways in climatic change, Climate in Earth History, 20. pp. 118 125. Besse, J., Courtillot, V., 1988. Paleogeographic maps of the continents bordering the Indian Ocean since the Early Jurassic. J. Geophys. Res. Solid Earth 93, 11791 11808. Bird, P., 2003. An updated digital model of plate boundaries. Geochem. Geophys. Geosyst. 4, 1027. Bower, D.J., Gurnis, M., Flament, N., 2015. Assimilating lithosphere and slab history in 4-D Earth models. Phys. Earth Planet. Inter. 238, 8 22. Boyden, J., Mu¨ller, R., Gurnis, M., Torsvik, T., Clark, J., Turner, M., et al., 2011. Next-generation plate-tectonic reconstructions using GPlates. In: Keller, G., Baru, C. (Eds.), Geoinformatics: Cyberinfrastructure for the Solid Earth Sciences. Cambridge University Press, Cambridge. Bronner, A., Sauter, D., Munschy, M., Carlut, J., Searle, R., Cannat, M., et al., 2014. Magnetic signature of large exhumed mantle domains of the Southwest Indian Ridge results from a deep-tow geophysical survey over 0 to 11 Ma old seafloor. Solid Earth 5, 339 354. Bullard, E., Everett, J.E., Smith, A.G., 1965. The fit of the continents around the Atlantic. Philos. Trans. R. Soc. Lond. A Math. Phys. Sci. 258, 41 51. Cande, S.C., Patriat, P., Dyment, J., 2010. Motion between the Indian, Antarctic and African plates in the early Cenozoic. Geophys. J. Int. 127 149. Cande, S.C., Stegman, D.R., 2011. Indian and African plate motions driven by the push force of the Reunion plume head. Nature 475, 47 52. Cane, M.A., Molnar, P., 2001. Closing of the Indonesian seaway as a precursor to east African aridification around 3 4 million years ago. Nature 411, 157. Chorowicz, J., 2005. The east African rift system. J. Afr. Earth Sci. 43, 379 410. Clements, B., Hall, R., 2011. A record of continental collision and regional sediment flux for the Cretaceous and Palaeogene core of SE Asia: implications for early Cenozoic palaeogeography. J. Geol. Soc. 168, 1187 1200. Clements, B., Burgess, P., Hall, R., Cottam, M., 2011. Subsidence and uplift by slab-related mantle dynamics: a driving mechanism for the Late Cretaceous and Cenozoic evolution of continental SE Asia? Geological Society, London, Special Publications, 355. pp. 37 51. Conrad, C.P., 2013. The solid Earth’s influence on sea level. Bulletin 125, 1027 1052. Cox, A., Doell, R., Dalrymple, G., 1968. Radiometric time-scale for geomagnetic reversals. Q. J. Geol. Soc. 124, 53 66. Deconto, R.M., Pollard, D., 2003. Rapid Cenozoic glaciation of Antarctica induced by declining atmospheric CO2. Nature 421, 245 249. DiCaprio, L., Gurnis, M., Mu¨ller, R.D., 2009. Long-wavelength tilting of the Australian continent since the Late Cretaceous. Earth Planet. Sci. Lett. 278, 175 185. DiCaprio, L., Gurnis, M., Mu¨ller, R.D., Tan, E., 2011. Mantle dynamics of continentwide Cenozoic subsidence and tilting of Australia. Lithosphere L140, 1v1. Doel, R.E., Levin, T.J., Marker, M.K., 2006. Extending modern cartography to the ocean depths: military patronage, Cold War priorities, and the Heezen Tharp mapping project, 1952 1959. J. Hist. Geogr. 32, 605 626. Domeier, M., Torsvik, T.H., 2014. Plate tectonics in the late Paleozoic. Geosci. Front. 5, 303 350. Exon, N., Kennett, J., Malone, M., Brinkhuis, H., Chaproniere, G., Ennyu, A., et al., 2002. Drilling reveals climatic consequences of Tasmanian gateway opening. Eos Trans. Am. Geophys. Union. 83, 253 259. Flament, N., Gurnis, M., Mu¨ller, R.D., 2013. A review of observations and models of dynamic topography. Lithosphere 5, 189 210. Flament, N., Gurnis, M., Williams, S., Seton, M., Skogseid, J., Heine, C., et al., 2014. Topographic asymmetry of the South Atlantic from global models of mantle flow and lithospheric stretching. Earth Planet. Sci. Lett. 387, 107 119. Forsyth, D., Uyeda, S., 1975. On the relative importance of the driving forces of plate motion. Geophys. J. Int. 43, 163 200. Frisch, W., Meschede, M., Blakey, R.C., 2011. Plate Tectonics Continental Drift and Mountain Building. Springer. Gaina, C., Mu¨ller, R., 2007. Cenozoic tectonic and depth/age evolution of the Indonesian gateway and associated back-arc basins. Earth-Sci. Rev. 83, 177 203. Gee, J.S., Kent, D.V., 2007. Source of oceanic magnetic anomalies and the geomagnetic polarity time scale. In: Treatise on Geophysics, vol. 5: Geomagnetism, pp. 455 507. Gibbons, A., Barckhausen, U., Van Den Bogaard, P., Hoernle, K., Werner, R., Whittaker, J., et al., 2012. Constraining the Jurassic extent of Greater India: tectonic evolution of the West Australian margin. Geochem. Geophys. Geosyst. 13, 25. Gibbons, A., Zahirovic, S., Mu¨ller, R., Whittaker, J., Yatheesh, V., 2015. A tectonic model reconciling evidence for the collisions between India, Eurasia and intra-oceanic arcs of the central-eastern Tethys. Gondwana Res. 28, 451 492 . Golonka, J., Krobicki, M., Pajak, J., Van Giang, N., Zuchiewicz, W., 2006. Global Plate Tectonics and Paleogeography of Southeast Asia. Faculty of Geology, Geophysics and Environmental Protection, AGH University of Science and Technology, Arkadia, Krakow. Gordon, R.G., Jurdy, D.M., 1986. Cenozoic global plate motions. J. Geophys. Res. Solid Earth (1978 2012) 91, 12389 12406. Gurnis, M., Turner, M., Zahirovic, S., Dicaprio, L., Spasojevic, S., Mu¨ller, R., et al., 2012. Plate tectonic reconstructions with continuously closing plates. Comput. Geosci. 38, 35 42. Hall, R., 1996. Reconstructing Cenozoic SE Asia, Geological Society, London, Special Publications, 106. pp. 153 184. Hall, R., 2002. Cenozoic geological and plate tectonic evolution of SE Asia and the SW Pacific: computer-based reconstructions, model and animations. J. Asian Earth Sci. 20, 353 431. Hall, R., 2012. Late Jurassic Cenozoic reconstructions of the Indonesian region and the Indian Ocean. Tectonophysics 570 571, 1 41. Hamilton, W.B., 1970. Tectonic Map of Indonesia: A Progress Report. US Geological Survey. Haq, B.U., Hardenbol, J., Vail, P.R., 1987. Chronology of fluctuating sea levels since the Triassic. Science 235, 1156 1167. Harrington, L., Zahirovic, S., Flament, N., Mu¨ller, R.D., 2017. The role of deep Earth dynamics in driving the flooding and emergence of New Guinea since the Jurassic. Earth Planet. Sci. Lett. 479, 273 283. Hassan, R., Mu¨ller, R.D., Gurnis, M., Williams, S.E., Flament, N., 2016. A rapid burst in hotspot motion through the interaction of tectonics and deep mantle flow. Nature 533, 239 242. Hay, W.W., 1996. Tectonics and climate. Geol. Rundsch. 85, 409 437.

Regional Geology and Tectonics

110

6. Resolving geological enigmas using plate tectonic reconstructions and mantle flow models

Heezen, B., Tharp, M., 1965. Tectonic fabric of the Atlantic and Indian Oceans and continental drift. Philos. Trans. R. Soc. Lond. A Math. Phys. Eng. Sci. 258, 90 106. Heine, C., Mu¨ller, R., Gaina, C., 2004. Reconstructing the lost eastern Tethys Ocean Basin: convergence history of the SE Asian margin and marine gateways. Geophys. Monogr. 149, 37 54. Hellinger, S., 1981. The uncertainties of finite rotations in plate tectonics. J. Geophys. Res. Solid Earth (1978 2012) 86, 9312 9318. Hess, H.H., 1954. Geological hypotheses and the Earth’s crust under the oceans. Proc. R. Soc. Lond. Ser. A Math. Phys. Sci. 222, 341 348. Hess, H.H., 1962. History of ocean basins. Pet. Stud. A Vol. Honor. A.F. Buddington 599 620. Hutchison, C.S., 1975. Ophiolite in Southeast Asia. Geol. Soc. Am. Bull. 86, 797 806. Jagoutz, O., Royden, L., Holt, A.F., Becker, T.W., 2015. Anomalously fast convergence of India and Eurasia caused by double subduction. Nat. Geosci. 8, 475. Johansson, L., Zahirovic, S., Muller, R.D., 2018. The interplay between the eruption and weathering of large igneous provinces and the deeptime carbon cycle. Geophys. Res. Lett. 45, 5380 5389. Jurdy, D., Van der Voo, R., 1975. True polar wander since the Early Cretaceous. Science 187, 1193 1196. Katili, J., 1975. Volcanism and plate tectonics in the Indonesian island arcs. Tectonophysics 26, 165 188. Katili, J.A., 1971. A review of the geotectonic theories and tectonic maps of Indonesia. Earth-Sci. Rev. 7, 143 163. Knopf, A., 1948. The geosynclinal theory. Geol. Soc. Am. Bull. 59, 649 670. Kreemer, C., Holt, W.E., Haines, A.J., 2003. An integrated global model of present-day plate motions and plate boundary deformation. Geophys. J. Int. 154, 8 34. Le Pichon, X., 1968. Sea-floor spreading and continental drift. J. Geophys. Res. 73, 3661 3697. Lee, T.-Y., Lawver, L.A., 1995. Cenozoic plate reconstruction of Southeast Asia. Tectonophysics 251, 85 138. Li, C., Van Der Hilst, R., Engdahl, E., Burdick, S., 2008. A new global model for P wave speed variations in Earth’s mantle. Geochem. Geophys. Geosyst. 9, 21. Lithgow-Bertelloni, C., Richards, M.A., 1998. The dynamics of Cenozoic and Mesozoic plate motions. Rev. Geophys. 36, 27 78. Matthews, K.J., Hale, A.J., Gurnis, M., Mu¨ller, R., Dicaprio, L., 2011a. Dynamic subsidence of Eastern Australia during the Cretaceous. Gondwana Res. 19, 373 383. Matthews, K.J., Mu¨ller, R.D., Wessel, P., Whittaker, J.M., 2011b. The tectonic fabric of the ocean basins. J. Geophys. Res. 116, B12109. Matthews, K.J., Maloney, K.T., Zahirovic, S., Williams, S.E., Seton, M., Mu¨ller, R.D., 2016. Global plate boundary evolution and kinematics since the late Paleozoic. Glob. Planet. Change 146, 226 250. Maus, S., Barckhausen, U., Berkenbosch, H., Bournas, N., Brozena, J., Childers, V., et al., 2009. EMAG2: a 2-arc min resolution Earth magnetic anomaly grid compiled from satellite, airborne, and marine magnetic measurements. Geochem. Geophys. Geosyst. 10, Q08005. McCourt, W., Crow, M., Cobbing, E., Amin, T., 1996. Mesozoic and Cenozoic plutonic evolution of SE Asia: evidence from Sumatra, Indonesia, Geological Society, London, Special Publications, 106. pp. 321 335. McKenzie, D., 1978. Some remarks on the development of sedimentary basins. Earth Planet. Sci. Lett. 40, 25 32. McKenzie, D., Parker, R., 1967. The North Pacific: an example of tectonics on a sphere. Nature 216, 1276 1280. Mendel, V., Munschy, M., Sauter, D., 2005. MODMAG, a MATLAB program to model marine magnetic anomalies. Comput. Geosci. 31, 589 597. Merdith, A.S., Collins, A.S., Williams, S.E., Pisarevsky, S., Foden, J.D., Archibald, D.B., et al., 2017. A full-plate global reconstruction of the Neoproterozoic. Gondwana Res. 50, 84 134. Metcalfe, I., 1988. Origin and assembly of south-east Asian continental terranes, Geological Society, London, Special Publications, 37. pp. 101 118. Morgan, W.J., 1972. Plate motions and deep mantle convection. Geol. Soc. Am. Mem. 132, 7 22. Mu¨ller, R., Royer, J., Lawver, L., 1993. Revised plate motions relative to the hotspots from combined Atlantic and Indian Ocean hotspot tracks. Geology 21, 275. Mu¨ller, R., Sdrolias, M., Gaina, C., Steinberger, B., Heine, C., 2008. Long-term sea-level fluctuations driven by ocean basin dynamics. Science 319, 1357. Mu¨ller, R., Dutkiewicz, A., Seton, M., Gaina, C., 2013. Seawater chemistry driven by supercontinent assembly, breakup, and dispersal. Geology 41, 907 910. Mu¨ller, R.D., Seton, M., Zahirovic, S., Williams, S.E., Matthews, K.J., Wright, N.M., et al., 2016. Ocean basin evolution and global-scale plate reorganization events since Pangea breakup. Annu. Rev. Earth Planet. Sci. 44, 107 138. Mu¨ller, R.D., Cannon, J., Qin, X., Watson, R.J., Gurnis, M., Williams, S., et al., 2018. GPlates: building a virtual earth through deep time. Geochem. Geophys. Geosyst. 19, 2243 2261. Mu¨ller, R.D., Zahirovic, S., Williams, S.E., Cannon, J., Seton, M., Bower, D.J., et al., 2019. A global plate model including lithospheric deformation along major rifts and orogens since the Triassic. Tectonics 38. Norvick, M.S., 2003. New Paleographic Maps of the Northern Margins of the Australian Plate: Updated Report. Geoscience Australia. O’Neill, C., Mu¨ller, D., Steinberger, B., 2005. On the uncertainties in hot spot reconstructions and the significance of moving hot spot reference frames. Geochem. Geophys. Geosyst. 6. Ogg, J.G., Ogg, G., Gradstein, F.M., 2016. A Concise Geologic Time Scale: 2016. Elsevier. Oreskes, N., 1988. The rejection of continental drift. Hist. Stud. Phys. Biol. Sci. 18, 311 348. Oreskes, N., 2003. Plate Tectonics An Insider’s History of the Modern Theory of the Earth. Westview Press, Boulder, CO. Pigram, C.J., Panggabean, H., 1984. Rifting of the northern margin of the Australian continent and the origin of some microcontinents in Eastern Indonesia. Tectonophysics 107, 331 353. Rangin, C., Jolivet, L., Pubellier, M., 1990. A simple model for the tectonic evolution of southeast Asia and Indonesia region for the past 43 my. Bull. Socie´te´ Ge´ologique Fr. 6, 889 905. Raymo, M., Ruddiman, W., 1992. Tectonic forcing of late Cenozoic climate. Nature 359, 117 122. Richards, M.A., Engebretson, D.C., 1992. Large-scale mantle convection and the history of subduction. Nature 355, 437.

Regional Geology and Tectonics

References

111

Rolland, J., Condamine, F.L., Beeravolu, C.R., Jiguet, F., Morlon, H., 2015. Dispersal is a major driver of the latitudinal diversity gradient of Carnivora. Glob. Ecol. Biogeogr. 24, 1059 1071. Romm, J., 1994. A new forerunner for continental drift. Nature 367, 407 408. Salles, T., Hardiman, L., 2016. Badlands: an open-source, flexible and parallel framework to study landscape dynamics. Comput. Geosci. 91, 77 89. Scotese, C.R., Gahagan, L.M., Larson, R.L., 1988. Plate tectonic reconstructions of the Cretaceous and Cenozoic ocean basins. Tectonophysics 155, 27 48. Seton, M., Mu¨ller, R., Zahirovic, S., Gaina, C., Torsvik, T., Shephard, G., et al., 2012. Global continental and ocean basin reconstructions since 200 Ma. Earth-Sci. Rev. 113, 212 270. Seton, M., Whittaker, J.M., Wessel, P., Mu¨ller, R.D., Demets, C., Merkouriev, S., et al., 2014. Community infrastructure and repository for marine magnetic identifications. Geochem. Geophys. Geosyst. 15, 1629 1641. Sijp, W.P., Anna, S., Dijkstra, H.A., Flo¨gel, S., Douglas, P.M., Bijl, P.K., 2014. The role of ocean gateways on cooling climate on long time scales. Glob. Planet. Change 119, 1 22. Solomon, S.C., Sleep, N.H., Jurdy, D.M., 1977. Mechanical models for absolute plate motions in the early Tertiary. J. Geophys. Res. 82, 203 212. Spasojevic, S., Liu, L., Gurnis, M., 2009. Adjoint models of mantle convection with seismic, plate motion, and stratigraphic constraints: North America since the Late Cretaceous. Geochem. Geophys. Geosyst. 10. Steinberger, B., Torsvik, T.H., 2008. Absolute plate motions and true polar wander in the absence of hotspot tracks. Nature 452, 620 623. Stoddard, P.R., Abbott, D., 1996. Influence of the tectosphere upon plate motion. J. Geophys. Res. Solid Earth (1978 2012) 101, 5425 5433. Torsvik, T., Mu¨ller, R., Van der Voo, R., Steinberger, B., Gaina, C., 2008. Global plate motion frames: toward a unified model. Rev. Geophys. 46. van der Meer, D.G., Van Hinsbergen, D.J., Spakman, W., 2018. Atlas of the underworld: slab remnants in the mantle, their sinking history, and a new outlook on lower mantle viscosity. Tectonophysics 723, 309 448. Van der Voo, R., Spakman, W., Bijwaard, H., 1999. Tethyan subducted slabs under India. Earth Planet. Sci. Lett. 171, 7 20. Vine, F., 1966. Spreading of the ocean floor: new evidence. Science 154, 1405 1415. Vine, F., Matthews, D., 1963. Magnetic anomalies over oceanic ridges. Nature 4897, 947 949. Wadati, K., 1928. Shallow and deep earthquakes. Geophys. Mag. 1, 2. Wegener, A., 1915. The Origin of Continents and Oceans. Dover, New York. Wernicke, B., 1981. Low-angle normal faults in the Basin and Range Province: nappe tectonics in an extending orogen. Nature 291, 645. Whitmore, T., 1982. Wallace’s Line: a result of plate tectonics. Ann. Mo. Bot.l Gard. 69, 668 675. Whittaker, J., Afonso, J., Masterton, S., Mu¨ller, R., Wessel, P., Williams, S., et al., 2015. Long-term interaction between mid-ocean ridges and mantle plumes. Nat. Geosci. 8, 479 483. Wignall, P.B., 2001. Large igneous provinces and mass extinctions. Earth-Sci. Rev. 53, 1 33. Wilson, J.T., 1963. A possible origin of the Hawaiian Islands. Can. J. Phys. 41, 863 870. Wilson, J.T., 1965a. Evidence from ocean islands suggesting movement in the Earth. Philos. Trans. R. Soc. Lond. Ser. A Math. Phys. Sci. 258, 145 167. Wilson, J.T., 1965b. Transform faults, oceanic ridges, and magnetic anomalies southwest of Vancouver Island. Science 150, 482 485. Yang, T., Gurnis, M., Zahirovic, S., 2016. Mantle-induced subsidence and compression in SE Asia since the early Miocene. Geophys. Res. Lett. 43, 1901 1909. Yang, T., Gurnis, M., Zahirovic, S., 2018. Slab avalanche-induced tectonics in self-consistent dynamic models. Tectonophysics 746, 251 265. Young, A., Flament, N., Maloney, K., Williams, S., Matthews, K., Zahirovic, S., et al., 2019. Global kinematics of tectonic plates and subduction zones since the late Paleozoic Era. Geosci. Front. 10, 989 1013. Zablocki, C.J., Tilling, R.I., 1976. Field measurements of apparent Curie temperatures in a cooling basaltic lava lake, Kilauea Iki, Hawaii. Geophys. Res. Lett. 3, 487 490. Zahirovic, S., Mu¨ller, R.D., Seton, M., Flament, N., Gurnis, M., Whittaker, J., 2012. Insights on the kinematics of the India-Eurasia collision from global geodynamic models. Geochem. Geophys. Geosyst. 13. Zahirovic, S., Seton, M., Mu¨ller, R., 2014. The Cretaceous and Cenozoic tectonic evolution of Southeast Asia. Solid Earth 5, 227 273. Zahirovic, S., Mu¨ller, R., Seton, M., Flament, N., 2015. Tectonic speed limits from plate kinematic reconstructions. Earth Planet. Sci. Lett. 418, 40 52. Zahirovic, S., Flament, N., Dietmar Mu¨ller, R., Seton, M., Gurnis, M., 2016a. Large fluctuations of shallow seas in low-lying Southeast Asia driven by mantle flow. Geochem. Geophys. Geosyst. 17, 3589 3607. Zahirovic, S., Matthews, K.J., Flament, N., Mu¨ller, R.D., Hill, K.C., Seton, M., et al., 2016b. Tectonic evolution and deep mantle structure of the eastern Tethys since the latest Jurassic. Earth-Sci. Rev. 162, 293 337.

Regional Geology and Tectonics

C H A P T E R

7 Tectonostratigraphic Megasequences and Chronostratigraphy Ian James Stewart Integrated Petroleum Exploration Ltd., Uplands, Pond Road, Woking, Surrey, United Kingdom

All sedimentary basins contain at least one, and typically several Megasequences that relate to the dominant tectonic process operating during deposition, or immediately prior to deposition, thus creating the basin topography over which sediments accumulated. A Megasequence is chronostratigraphic. That is, it represents a fixed period of geologic time. The boundaries may be diachronous, due to subsequent erosion or the often transgressive nature at their base, and their age is determined by their maximum stratigraphic development. To the petroleum geoscientists, Megasequences are the fundamental units of basin description and form the natural subdivision of a basin’s sedimentary fill. Their boundaries are most readily identified on regional seismic reflection data as an unconformity, or often composite erosion surfaces at basin margins, and as onlap or downlap surfaces in more basinal areas identifying condensed sedimentary section. The term ‘Tectonostratigraphic Megasequence’ was first used by Hubbard et al. (1985a,b, 1988), based on the interpretation of observations from reflection seismic data. To a large degree, they equate to the ‘super-sequence’ that was suggested by Larry Sloss (1963, 1964) in his classical descriptions of North American Phanerozoic basin margins. However, it is only in recent years that regional seismic data over some of these North American basins have been acquired or stitched together from older, shorter seismic line segments, and this has allowed the internal tectonostratigraphic architecture to be more completely established than by well data alone. Tectonic Megasequences are the foremost order of stratal subdivision and contain an internal geometry controlled largely by the rate of tectonic subsidence and the rate of sediment supply. They are typically referred to as first- and/or second-order sequences. Their accommodation develops through basin-margin loading in foreland basins, by initial rifting and subsequent thermal contraction in extensional settings or as pull-apart basins along releasing segments of an oblique-slip fault system. The duration of a tectonic Megasequence is around 10 80 Ma (e.g. North America average 56 Ma; Meyers and Peters, 2011). The shorter timeframes in the range are usually observed in the early relaxation of contractional terranes (e.g. the Early Permian age extension of the western European Variscides; McCann et al., 2006), or sometimes the rapid transition from a prerift through a syn-rift to a postrift Megasequence in both extensional and oblique-slip settings. The Megasequence architecture of any basin is readily deciphered from chronostratigraphic (time) diagrams (Wheeler, 1958, 1964) that can be created from calibrated outcrop and well data with or without regional seismic datasets (Fig. 7.1). The majority of basins are composite in nature and reflect in their Megasequence architecture multiple phases of basin formation. These are often punctuated by episodes of inversion or extension of earlier structural inhomogeneities due to tectonic events further afield. In fold belts and proximal foreland settings, the development of new crustal faults may cannibalize an older basins stratigraphy, as well as earlier sequences of the same foreland Megasequence. For such reasons, every basin is unique, and despite numerous attempts to classify or group sets of basins into categories based largely on tectonic location, lithospheric substrate and proximity to plate margin (e.g. Halbouty et al., 1970; Dickinson, 1974, 1976; Fischer, 1975; Klemme, 1980; Bally and Snelson, 1980; Kingston et al., 1983a,b; St. John et al., 1984; Busby and Ingersoll, 1995), the use of such classification schemes tends to be restricted to statistical studies in oil companies. However, grouping may provide analogues for immature or under-explored

Regional Geology and Tectonics. DOI: https://doi.org/10.1016/B978-0-444-64134-2.00011-0

113

© 2020 Elsevier B.V. All rights reserved.

114

7. Tectonostratigraphic Megasequences and Chronostratigraphy

FIGURE 7.1 From section to a chronostratigraphic diagram. (A) Simplified geoseismic line drawing from the eastern Nile Delta. Note that section is greatly exaggerated vertically. Identified Megasequences are shown by shading. (B) Exploding the depth or time section into the component Megasequences. Arrows by fault surfaces indicate sense of relative motion. (C) Plotting the Megasequences and their internal chronostratigraphic surfaces against a geological timescale. Annotation is important to describe the tectonic process acting upon each Megasequence. Detailed lithology can be added from outcrop, well data or from seismic observations as required.

basin evaluation, an objective of the scheme by Kingston et al. (1983a,b) and the basis of the Exxon tectonic map of the globe (Exxon Production Research Company, 1985). Basin classification schemes are not always adhered to by practising exploration geoscientists the simplest descriptive terms such as foreland, rift, passive margin, cratonic and pull-apart tend to be. For a recent summary of the various basin classification schemes, the reader should refer to Allen et al. (2015), Allen and Allen (2013), Ingersoll (2012) and Roberts and Bally (2012). Internally, Megasequences consist of sequences, usually arranged in sets, reflecting comparable depositional process. These sequences are usually referred to as third-order depositional sequences and reflect a duration of

Regional Geology and Tectonics

Tectonostratigraphic Megasequences and Chronostratigraphy

115

between 0.5 and 3 Ma (Haq et al., 1987), although the original definition of Vail et al. (1977a) carried a range of 1 10 Ma. Their description, and more importantly their interpretation, has occupied an inordinate amount of geoscientific literature since the publication of AAPG Memoir 26 (Payton, 1977) which, unburdened by a peerreview process (Miall and Miall, 2002), introduced global eustatic change as the primary controlling factor on the development of third-order erosional surfaces on basin margins and their correlative conformities in deeper water environments (Mitchum, 1977, Mitchum et al., 1977, Vail et al., 1977a,b,c). These papers also provided the unfortunate terms ‘Highstand’, ‘Lowstand’ and ‘Stillstand’ to genetically relate changes in accommodation space on a basin margin to a ‘global eustatic curve’ measured from preserved coastal onlap on seismic reflection data. However, the controls of accommodation space can simply be related to basin-specific mechanisms, as well as any global eustatic overprint. Deciphering a global signature from the basin-specific controls of rate of subsidence, rate of sediment supply and compaction remains an ongoing academic challenge. Perhaps the most contentious issues of the Exxon 1977 approach were the application of the ‘global cycle chart’ (later called the ‘Haq curve’; Haq et al., 1987) as a predictive tool in seismic and outcrop interpretation, and the assumed global synchroneity of third-order depositional sequences related to sea-level fluctuation. The original chart lacked detailed biostratigraphic calibration or details of its origin (although see Hardenbol et al., 1998, for north west Europe), but more significantly, it was compiled as a patchwork through time of curves derived from numerous ‘type-sections’ from various basin margins around the world (Carter et al., 1991, Carter, 1998). The ‘Haq curve’ is still used in some oil companies as a predictive tool without understanding its derivation; for the curve to have predictive value, it requires all basins to have the same subsidence history and the same sedimentation rate as the original type section for the particular age being studied, and for each individual basin to have a constant subsidence rate and sedimentation rate along its entire depositional strike. Seismic stratigraphic interpretation is an art that requires careful observation succeeded by interpretation. Observational description should embrace terms such as progradation, aggradation, retrogradation, onlap, etc., which are purely descriptive, rather than placing a model or a template on the data and then applying a genetic terminology. Observations can be easily enhanced by line drawings to accentuate and reveal the reflection patterns and relationships on the seismic profiles. It is one component of sequence stratigraphy that also includes well analysis, outcrop descriptions and chronostratigraphic calibration through palaeontological data (Catuneanu et al., 1988), and where appropriate isotope records. Catuneanu et al. (1988) described the significance of all stratigraphic surfaces [e.g. maximum flooding surfaces (Galloway, 1989), transgressive surfaces (Embry, 1993, 1995) and the erosional sequence boundary of the Exxon Group], and they recognized that each type of surface faces the issue of diachroneity towards basin margins. In deeper water areas, the most significant surfaces are condensed sections. These are usually correlative to more than one of the surface types observed at basin margins and approach the true chronostratigraphic significance of seismic reflections described by Vail et al. (1977d) for correlative conformities. They tend to be the most obvious continuous reflections seen on regional seismic sections and can form the natural subdivision of the Megasequence. They are frequently marked by high faunal abundance and may exhibit a marked well log deflection relating to high organic content or mineralogy and hence tend to have an obvious amplitude response on seismic data. They represent a significant period of time relative to the depositional units they partition and are commonly marked by seismic-scale onlap reflecting continued structural growth during periods of low sediment flux (e.g. in halokinetic or growth fault/fold provinces). Consequently, they are frequently misinterpreted as a structural unconformity. More importantly, condensed sections at Megasequence boundaries are commonly significant petroleum source rocks particularly at the syn- to postrift transition during the onset of thermal subsidence and at the base of a foreland Megasequence. Sediment starvation at these boundaries and their consequent organic enrichment and minimal clastic dilution relate to the time taken for depositional processes and drainage to reorganize after a particular tectonic phase. Regional seismic reflection data are now available to industry across most of the sedimentary basins of the globe. These seismic profiles provide a two-dimensional vertical ‘map’ of a basin, with well data, biostratigraphy and outcrop studies acting as the legend. More and more of these data are being released to academia through research and joint studies. It is increasingly apparent that just as each basin is unique, so is its sedimentary fill, and the geological history of that fill can only be resolved by careful observations to which an interpretation is then applied. Good observation is simply good science, and while observations rarely change interpretations frequently do. Basin analysis has moved forward from a period of applying a predictive template to seismic and outcrop data that in many cases led to interpretation overriding facts and observations being ignored.

Regional Geology and Tectonics

116

7. Tectonostratigraphic Megasequences and Chronostratigraphy

References Allen, P.A., Allen, J.R., 2013. Basin Analysis: Principles and Application to Petroleum Play Assessment, third ed. Wiley-Blackwell Publishing, Oxford, 632 pp. Allen, P.A., Eriksson, P.G., Alkmim, F.F., Betts, P.G., Catuneanu, O., Mazumder, R., et al., 2015. Classification of basins, with special reference to Proterozoic examples. In: Mazumder, R., Eriksson, P.G. (Eds.), Precambrian Basins of India: Stratigraphic and Tectonic Context, 43. Geological Society London, Memoirs, pp. 5 28. Bally, A.W., Snelson, S., 1980. Realms of subsidence. In: Miall, A.D. (Ed.), Facts and Principles of World Petroleum Occurrence, 6. Canadian Society of Petroleum Geologists Memoir, Calgary, pp. 9 94. Busby, C.J., Ingersoll, R.V. (Eds.), 1995. Tectonics of Sedimentary Basins. Blackwell Science Publishing, Cambridge, 579 pp. Carter, R.M., 1998. Two models: global sea-level change and sequence stratigraphic architecture. Sediment. Geol. 122, 23 36. Carter, R.M., Abbott, S.T., Fulthorpe, C.S., Haywick, D.W., Henderson, R.A., 1991. Application of global sea-level and sequence stratigraphic models in southern hemisphere Neogene strata from New Zealand. In: MacDonald, D.I.M. (Ed.), Sedimentation, Tectonics and Eustasy, 12. International Association of Sedimentologists Special Publication, pp. 41 65. Catuneanu, O., Willis, A.J., Miall, A.D., 1988. Temporal significance of sequence boundaries. Sediment. Geol. 121, 157 178. Dickinson, W.R., 1974. Plate tectonics and sedimentation. In: Dickinson, W.R. (Ed.), Tectonics and Sedimentation, 22. Society of Economic Paleontologists and Mineralogists Special Publication, pp. 1 27. Dickinson, W.R., 1976. Plate Tectonic Evolution of Sedimentary Basins. American Association of Petroleum Geologists Continuing Education Short Course Notes, 1. Embry, A.F., 1993. Transgressive regressive (T R) sequence analysis of the Jurassic succession of the Sverdrup Basin, Canadian Arctic Archipelago. Can. J. Earth Sci. 30, 301 320. Embry, A.F., 1995. Sequence boundaries and sequence hierarchies: problems and proposals. In: Steel, R.J., Felt, V.L., Johannessen, E.P., Mathieu, C. (Eds.), Sequence Stratigraphy on the Northwest European Margin, vol. 5. Norwegian Petroleum Society Special Publications., Elsevier, Amsterdam, pp. 1 11. Exxon Production Research Company, 1985. Tectonic Map of the World. World Mapping Project, Scale 1:5,000,000, 20 Panels. American Association of Petroleum Geologists, Tulsa, OK. Fischer, A.G., 1975. Origin and growth of basins. In: Fischer, A.G., Judson, S. (Eds.), Petroleum and Global Tectonics. Princeton University Press, Princeton, NJ, pp. 47 79. Galloway, W.E., 1989. Genetic stratigraphic sequences in basin analysis. I. Architecture and genesis of flooding-surface bounded depositional units. Am. Assoc. Pet. Geol. Bull. 73, 125 142. Halbouty, M.T., King, R.E., Klemme, H.D., Dott, R.H., Meyerhoff, A.A., 1970. World’s giant oil and gas fields, geologic factors affecting their formation, and basin classification, part II. In: Halbouty, M.T. (Ed.), Geology of Giant Petroleum Fields., vol. 14. American Association of Petroleum Geologists Memoir, pp. 528 555. Haq, B.U., Hardenbol, J., Vail, P.R., 1987. Chronology of fluctuating sea levels since the Triassic. Science. 235, 1156 1167. Hardenbol, J., Thierry, J., Farley, M.B., Jacquin, Th, de Graciansky, P.-C., Vail, P.R., 1998. Mesozoic and Cenozoic sequence chronostratigraphic framework of European basins. In: de Graciansky, P.C., et al., (Eds.), Mesozoic and Cenozoic Sequence Stratigraphy of European Basins, 60. SEPM Special Publication, Tulsa, OK, pp. 3 13. Charts 1 8. Hubbard, R.J., 1988. Age and significance of sequence boundaries on Jurassic and Early Cretaceous rifted continental margins. Bull. Am. Assoc. Pet. Geol. 72, 49 72. Hubbard, R.J., Pape, J., Roberts, D.G., 1985a. Depositional sequence mapping as a technique to establish tectonic and stratigraphic framework and evaluate hydrocarbon potential on a passive continental margin. In: Berg, O.R., Woolverton, D.G. (Eds.), Seismic Stratigraphy II, 39. American Association of Petroleum Geologists Memoir, pp. 79 92. Hubbard, R.J., Pape, J., Roberts, D.G., 1985b. Depositional sequence mapping to illustrate the evolution of a passive continental margin. In: Berg, O.R., Woolverton, D.G. (Eds.), Seismic Stratigraphy II: An Integrated Approach to Hydrocarbon Exploration, vol. 39. American Association of Petroleum Geologists Memoir, pp. 93 115. Ingersoll, R.V., 2012. Tectonics of sedimentary basins, with revised nomenclature. In: Busby, C., Azor Pe´rez, A. (Eds.), Tectonics of Sedimentary Basins: Recent Advances. Wiley Blackwell Publishing, pp. 3 43. Kingston, D.R., Dishroon, C.P., Williams, P.S., 1983a. Global basin classification system. Am. Assoc. Pet. Geol. Bull. 67, 2175 2193. Kingston, D.R., Dishroon, C.P., Williams, P.S., 1983b. Hydrocarbon plays and global basin classification. Am. Assoc. Pet. Geol. Bull. 67, 2194 2198. Klemme, H.D., 1980. Petroleum basins: classification and characteristics. J. Pet. Geol. 3, 187 207. McCann, T., Pascal, C., Timmerman, T.J., Krzywiec, P., Lopez-Gomez, J., Wetzel, A., et al., 2006. Post-Variscan (end Carboniferous Early Permian) basin evolution in Western and Central Europe. In: Gee, D.G., Stephenson, R.A. (Eds.), European Lithosphere Dynamics, vol. 32. Geological Society, London, Memoirs, pp. 355 388. Meyers, S., Peters, S., 2011. A 56million year rhythm in North American sedimentation during the Phanerozoic. Earth Planet. Sci. Lett. 303, 174 180. Miall, C.E., Miall, A.D., 2002. THE EXXON FACTOR: the roles of corporate and academic science in the emergence and legitimation of a new global model of sequence stratigraphy. Sociological Q. 42 (2), 3017 3334. Mitchum, R.M., 1977. Seismic stratigraphy and global changes of sea level, Part 1: glossary of terms used in seismic stratigraphy. In: Payton, C.E. (Ed.), Seismic Stratigraphy Applications to Hydrocarbon Exploration, vol. 26. American Association of Petroleum Geologists Memoir, pp. 205 212. Mitchum, R.M., Vail, P.R., Thompson III, S., 1977. Seismic stratigraphy and global changes of sea level, Part 2: the depositional sequence as a basic unit for stratigraphic analysis. In: Payton, C.E. (Ed.), Seismic Stratigraphy Applications to Hydrocarbon Exploration, vol. 26. American Association of Petroleum Geologists Memoir, pp. 53 62. Payton, C.W. (Ed.), 1977. Seismic Stratigraphy Applications to Hydrocarbon Exploration, vol. 26. American Association of Petroleum Geologists Memoir, Tulsa, OK.

Regional Geology and Tectonics

Further reading

117

Roberts, D.G., Bally, A.W., 2012. Some remarks on basins and basin classification and tectonostratigraphic Megasequences. In: first ed. Roberts, D.G., Bally, A.W. (Eds.), Regional Geology and Tectonics: Principles of Geologic Analysis, vol. 2012. Elsevier, pp. 77 92. , 2012. Sloss, L.L., 1963. Sequences in the cratonic interior of North America. Geol. Soc. Am. Bull. 74, 93 114. Sloss, L.L., 1964. Tectonic cycles of the North American Craton. In: Merriam, D.F. (Ed.), Symposium on Cyclic Sedimentation: Kansas Geological Survey Bulletin, 169. pp. 449 459. St. John, B., Bally, A.W., Klemme, H.D., 1984. Sedimentary Provinces of the World Hydrocarbon Productive and Non-Productive. American Association of Petroleum Geologists, Tulsa, OK, 1 Map, 35 pp. Vail, P.R., Mitchum Jr., R.M., Todd, R.G., Widmier, J.M., Thompson III., S., Sangree, J.B., et al., 1977a. Seismic stratigraphy and global changes in sea level. In: Payton, C.E. (Ed.), Seismic Stratigraphy Applications to Hydrocarbon Exploration, vol. 26. American Association of Petroleum Geologists Memoir, Tulsa, pp. 49 212. Vail, P.R., Mitchum, R.M., Thompson III, S., 1977b. Seismic Stratigraphy and Global Changes of Sea Level: Part 3. Relative Changes of Sea Level from Coastal Onlap: Section 2. Application of Seismic Reflection Configuration to Stratigraphic Interpretation., vol. 26. American Association of Petroleum Geologists Memoir, pp. 63 81. Vail, P.R., Mitchum, R.M., Thompson III, S., 1977c. Seismic Stratigraphy and Global Changes of Sea Level: Part 4. Global Cycles of Relative Changes of Sea Level.: Section 2. Application of Seismic Reflection Configuration to Stratigraphic Interpretation, vol. 26. American Association of Petroleum Geologists Memoir, pp. 83 97. Vail, P.R., Todd, R.G., Sangree, J.B., 1977d. Seismic Stratigraphy and Global Changes of Sea Level: Part 5. Chronostratigraphic Significance of Seismic Reflections: Section 2. Application of Seismic Reflection Configuration to Stratigraphic Interpretation, vol. 26. American Association of Petroleum Geologists Memoir, pp. 99 116. Wheeler, H.E., 1958. Time Stratigraphy, vol. 42. American Association of Petroleum Geologists Bulletin, pp. 1047 1063. Wheeler, H.E., 1964. Baselevel, lithosphere surface, and time-stratigraphy. Geol. Soc. Am. Bull 75 (7), 599 610.

Further reading Ashton, E., 2002. Transgressive-regressive (T-R) sequence stratigraphy. Trans. Gulf Coast. Assoc. Geol. Soc. 52, 151 172.

Regional Geology and Tectonics

C H A P T E R

8 Fault classification, fault growth and displacement Haakon Fossen Museum of Natural History/Department of Earth Science, University of Bergen, Bergen, Norway

Introduction Predicting fault locations and fault properties such as geometry, horizontal and vertical length, connectivity and displacement has been important to the mining industry for many centuries because of the ability of faults to conduct fluids and therefore host ore deposits, and for their tendency to complicate mining operations, particularly coal mining. More recently, it has become essential to understand faulting and to predict fault locations, geometries, orientations and properties in the context of oil and gas exploration and exploitation. As a consequence, much of our current understanding of faults is based on research driven by industrial needs, particularly since the 1980s when three-dimensional (3D) reflection seismic data sets became available. Since then we have seen an impressive improvement in seismic data quality and data enhancement methods that, together with field-based studies and physical and numerical modeling, has led us to the current understanding of faults and faulting. At a larger scale ( . 1
10 km) than that concerning hydrocarbon production and most mining operations, faults or fault zones affect the entire brittle crust and link with deeper lithospheric shear zones and may generate devastating earthquakes during their active lifetime. However, they all initiate as small features and grow into larger faults, fault zones and fault networks. The process by which faults form and grow and the related complications and fault geometries are the main focus of this chapter. This review is largely presented in the context of the normal fault regime, but most principles and properties apply also to strike-slip and thrust settings.

What is a fault? A fault represents a narrow physical discontinuity in rock and in the displacement field associated with the deformation, exhibiting predominantly shear (wall-parallel) displacement. Small-scale structures (e.g. magnified part of Fig. 8.1A) that fit this definition are usually referred to as shear fractures, while a fully developed fault is a composite structure that consists of a multitude of smaller-scale structures in a zone, together with one or more major slip surfaces and/or a fault core along which most of the offset is localized. Hence faults as observed in outcrop are rarely simple discrete “planes”, but rather irregular curvitabular volumes of variably deformed rocks. Typical elements found in such volumes are subsidiary faults, fractures, veins, gouge, breccia, deformation bands and volumes or lenses of less- or undeformed host rock. Large faults are well known to consist of multiple smaller fault elements in a zone (e.g. Braathen et al., 2009); hence, the term fault zone has for a long time been used to emphasize the fact that large faults are composite structures consisting of a multitude of smaller-scale faults and associated structures (Caine et al., 1996; Childs et al., 1996; Wibberley et al., 2008; Wibberley and Shipton, 2010).

Regional Geology and Tectonics. DOI: https://doi.org/10.1016/B978-0-444-64134-2.00007-9

119

© 2020 Elsevier B.V. All rights reserved.

120

8. Fault classification, fault growth and displacement

(A)

Damage zone

(B)

Fault zone

FW damage zone Fault core with lenses HW damage zone

FIGURE 8.1 Conceptualized illustrations of a complete fault and its different elements. (A) Slip localized on two or more narrow highstrain zones (slip surfaces or fault cores shown in black). A subsidiary footwall shear fracture is highlighted. (B) High-displacement fault showing a more extensively sheared central core with a surrounding low-strain damage zone. Source: (A) Figure inspired by Childs, C., Nicol, A., Walsh, J.J., Watterson, J., 1996. Growth of vertically segmented normal faults. J. Struct. Geol. 18, 1389
1397.

FIGURE 8.2 Example of a fault zone (reverse) in the Entrada Sandstone, Southern Utah. Rotation of layering in the zone and local normal drag along the bounding faults can be interpreted as evidence for breaching of a fault-propagation fold.

Fault anatomy The term fault zone is also used for smaller (outcrop-scale) faults that display composite structural elements, particularly where several slip surfaces can be discerned (Childs et al., 1996; 2009; Wibberley et al., 2008) (Fig. 8.1A). An example of a typical fault zone of this kind is shown in Fig. 8.2, where several closely arranged slip surfaces in a reverse fault zone have accommodated shortening of the Jurassic Entrada Sandstone in the Sevier foreland, Utah. The internal anatomy of many faults or fault zones fits the simple twofold classification of a central fault core and an enveloping damage zone (Caine et al., 1996) (Fig. 8.1B). Here, the fault core consists of highly sheared rocks that may be represented by fault gouge, cataclasite or breccia in which the original structure of the rock has been strongly masked or destroyed (Fig. 8.3). Its nature depends on the rocks involved; shale would easily create a zone of clay smear, while limestone, sandstone and igneous rocks typically develop different kinds of cataclasites and breccias. Large faults that have experienced late reactivation at shallow crustal levels may show a central core of noncohesive material within a more cohesive cataclastic outer core that formed during fault motions at greater depth (Fig. 8.3C). Variably deformed lenses of the wall rocks may form an integral part of the fault core (Fig. 8.1B), or when above a certain size, may be considered as a separate architectural element of a fault.

Regional Geology and Tectonics

What is a fault?

121

FIGURE 8.3 Three examples of fault cores in different lithologies. (A) Fault core with damage zone developed in Cretaceous fluvial sediments (near Salina, Utah, ca. 20 m offset). The core consists of crushed sandstone (cataclastic) and smeared clay-coal layers. (B) Fault with B100 m offset developed in shales of the Green River Formation, Utah. (C) Central noncohesive fault core surrounded by older flinty cataclasite as part of a several hundred meters
wide fault core in metamorphic rocks. Lærdal-Gjende fault, Norway, with several kilometers offset.

In the fault core-damage zone terminology, a fault core is completely surrounded by the damage zone, which is a zone of relatively low-displacement structures, notably shear fractures, but also veins (mineral filled extension fractures), short joints, deformation bands and/or stylolites (Fig. 8.1). Large faults may also contain smaller faults with their own damage zones, contained within the large damage zone of the first-order fault, as shown schematically in Fig. 8.4. Hence, the definition of a damage zone is to some extent scale-dependent. The relationship between fault displacement and damage zone width, however, seems to be rather scale-independent over a large range of sizes, meaning that the ratio between damage zone width and displacement is statistically the same for small and large faults. This emerges from plots such as the one shown in Fig. 8.5, which suggests a thickness
displacement ratio of 1:100 (displacement being 100 times the damage zone thickness). Note however that more specific data sets may deviate from this global rule (e.g. data from porous sandstones; Schueller et al., 2013), so establishing a relationship for specific areas and parts of the stratigraphy is always recommended. The scatter is also very large. Therefore, estimating displacement from damage zone thickness involves a large uncertainty. Field observations also show that damage zone width can vary greatly both vertically and laterally along a single fault due to variations in lithology, fault geometry and growth/linkage history. Fault core data from a variety of fault sizes also show a general increase in fault core thickness with increasing fault displacement, but with a similar two orders of magnitude uncertainty (Fig. 8.5). The anatomy of damage zones is also of interest, and their inner part generally contains a higher density of smallscale structures than do their peripheral part. An inner damage zone with higher density of structures and more complex structural relations can sometimes be distinguished from an outer low-strain damage zone (e.g. Cerveny et al., 2004; Berg and Skar, 2005). However, statistical evaluation of damage zones from extensional faults in porous sandstones (Schueller et al., 2013) has shown that most faults show a gradual decay in deformation band frequency away from the fault core, and that statistically this decrease can be described as logarithmic. Schueller et al. (2013) also suggest a scale-invariant growth process where the average density of deformation bands (15 6 9 bands/m) is statistically independent of fault displacement. Further, the distribution of deformation bands within the damage zone is qualitatively similar for small and large faults. Fault damage zones in nonporous or low-porosity rocks show a similar decay in fracture density away from the fault core (e.g. Caine et al., 1996; Faulkner et al., 2011). Furthermore, Savage and Brodsky (2011) suggested that the fracture density decay inside damage zones can be described by a power law with an average decay rate of approximately 0.8. Johri et al. (2014) numerically modelled

Regional Geology and Tectonics

122

8. Fault classification, fault growth and displacement

First-order fault 1–10 km

Seismically resolvable fault

~100 m

Subseismic fault

~1 m Fracture, def. band

~1 cm

FIGURE 8.4 Schematic illustration of fault hierarchy, from major first-order fault with several kilometers of displacement down to the scale of individual fractures or deformation bands. Three orders of damage zones are indicated, observable at different scales.

10,000

Displacement (D ) (m)

1:1000

1:100

1000 100 10 1:10 1 Damage zone 1:1

0.1 0.001

0.01

0.1

Fault core 1

10

100

1000

Thickness (m)

FIGURE 8.5 Fault displacement plotted against fault core and damage zone thickness in logarithmic diagram. Note that the fault core is on average two orders of magnitude thinner than the damage zone. Source: Modified from Fossen, H., 2016. Structural Geology, second ed. Cambridge University Press, Cambridge.

such a power-law decrease in fracture intensity away from the fault core. However, many such data sets are extracted from small faults with meter-scale displacement or less and do not reflect the complexity and variations associated with fault damage zones in nonporous rocks in general.

Fault drag The zone of fault-related folding along many faults, known as drag folding, is not considered as part of the fault itself, but nevertheless adds to its total displacement. Drag develops where a layering is oriented at an angle to the slip vector of the fault, for example horizontal beds affected by a normal or reverse fault. The term normal drag (normal in the sense of being common) is used about markers that are convex in the direction of slip. Similarly, reverse drag applies where markers are concave in the direction of slip (Grasemann et al., 2005). In other words, drag is considered to be normal when rotated into the fault (zone) in the same way that layers in

Regional Geology and Tectonics

What is a fault?

123

(A)

(B)

(C)

FIGURE 8.6 Formation of normal drag associated with a reverse fold as a result of fault-propagation folding. The fold forms ahead of the propagating fault tip (a-b) and is at some point dissected by the fault (lower part of B and C). Trishear modeling (Erslev, 1991). Source: Modified from Fossen, H., 2016. Structural Geology, second ed. Cambridge University Press, Cambridge.

metamorphic rocks are rotated into ductile shear zones (Ramsay, 1980). Note that normal and reverse drag are merely geometrically descriptive terms, so that both normal and reverse drag can be associated with normal faults, for example. Also note that normal drag along a reverse fault is geometrically similar to reverse drag along a normal fault. For instance, the drag along the reverse faults in Figs 8.2 and 8.6 is normal, while the km-scale drag related to the large normal fault in Fig. 8.7 is reverse. Also note that there is normal drag in a narrow zone along the main fault in the Jurassic section in Fig. 8.7. The drag zone can vary from less than a meter to several kilometers in width (Fig. 8.7) and typically varies vertically as layers of different mechanical properties get involved, but also laterally in many cases. In general, shales and clay-rich sequences tend to develop drag more easily than massive competent units (well-lithified sandstones and limestones). Drag zones that are wide enough to be imaged on seismic data are typically wider in the hanging wall than in the footwall. Drag may have several causes (Grasemann et al., 2005) and should only be used as a descriptive term about fault-adjacent layer rotation (folding). While friction along the fault core was typically called for in the older literature, fault-propagation folding is now considered to be a more common drag-forming mechanism. In the faultpropagation model, for which there is abundant evidence from many field examples, physical experiments and numerical models, a precursory fold forms by distributed deformation ahead of the propagating fault tip, and the fold becomes a drag fold the moment the tip propagates through the fold (Fig. 8.6). Impressive examples of Laramide-age fault-propagation monoclines underlain or cut by upward propagating basement faults are exposed on the Colorado Plateau (e.g. Zuluaga et al., 2014). Drag folds can also form along an existing fault due to fault bends and geometric complications caused by fault linkage processes. Because these complications tend to vary rapidly both laterally and vertically along faults, so does the appearance of drag. Rollover folds are a special case of reverse drag explained by listric normal fault geometry and are typically much larger than many other types of drag folds. Finally, differential compaction across major faults can also produce or add to largescale drag geometries. Drag folds are particularly important in hydrocarbon reservoirs where drag can significantly change the communication pattern across faults.

Regional Geology and Tectonics

124

8. Fault classification, fault growth and displacement

FIGURE 8.7 Example from the northern North Sea rift of how faults of different sizes typically appear on reflection seismic sections. The main fault offsets the rift basement by several kilometers, while its offset through the Jurassic section is only a few hundred meters, reducing to less than a hundred meters at the top Cretaceous (TC) level. Smaller antithetic faults in the hanging wall are magnified (upper left), as are nontectonic faults in the Cenozoic postrift package (upper right). These shallow faults are related to sediment compaction and dewatering. The magnified images show that there is room for different fault interpretations, with a zone of uncertainty of at least 100 m in width in this particular example that contains the fault damage zone and potential fault complications. Examples of normal and reverse drag are indicated. BCU, Base Cretaeous Unconformity. Source: Seismic data, courtesy of CGG.

Fault orientations, stress, strain and kinematics Relation between faults and stress Faults initiate with orientations that are largely controlled by the orientations of the principal stretching directions, which for structures involving small offsets and negligible block rotations can be correlated with principal stress directions (σ1, σ2, σ3). A simple relationship between stress and faults for isotropic rocks was used by Anderson (1951) as he defined his three tectonic regimes: normal, thrust and strike-slip (Fig. 8.8). Using the Coulomb criterion and a coefficient of friction of 0.6 (typically taken to be representative for common rock types, Byerlee’s Law) an angle of 30 degrees can be predicted between the maximum compressive stress (σ1) or shortening direction and the fault. This means that normal faults (σ1 5 σv) can be expected to dip at around 60 degrees, while reverse faults dip at around 30 degrees (σ1 5 σH), unless guided by preexisting structures. Strike-slip faults on the other hand are predicted to be vertical in this scheme. The simple Andersonian plane-strain model for faulting shown in Fig. 8.8 is founded on the assumption that the three principal stresses are always vertical or horizontal. In nature many, if not most, faults show evidence of oblique-slip, with components of both strike-slip and dip-slip displacement. Important reasons for this rotation of the principal stress axes are stress perturbations caused by mechanical strength variations, notably around weak faults and fractures, by slip along nonplanar faults, by slip along foliations, by rock anisotropy in general and by fault interaction in both the horizontal and vertical directions.

Regional Geology and Tectonics

125

Fault orientations, stress, strain and kinematics

σ1 σ3

60°

60°

σ3 60°

σ2

30°

σ1

σ2 Reverse faulting

Normal faulting

σ2 60°

σ1 σ3 Strike-slip faulting

FIGURE 8.8 Relation between the principal stress orientations and faults in the three Andersonian regimes, as illustrated by conjugate fault sets. These idealized figures show a close relationship between principal stress (or strain) axes and conjugate faults. Principal stress axes are indicated, but note that strictly speaking, these axes are instantaneous strain axes. Source: Modified from Fossen, H. 2016. Structural Geology, second ed. Cambridge University Press, Cambridge.

In naturally deformed rocks, the orientation of the principal stresses is well constrained where faults form two conjugate sets with opposite shear sense. Conjugate in this sense implies that the two sets were active at the same time, so that they locally or in a limited region show mutual crosscutting relationships. In this case, the shortening direction bisects the acute angle between the two sets of shear structures and can, for small displacements, be interpreted to represent σ1. Consequently, σ2 parallels their line of intersection, and σ3 bisects the obtuse angle between the shear structures, as shown in Fig. 8.8. Where fault slip data (fault orientation and direction and sense of slip, and if possible, amount of displacement) can be collected for a local fault population, paleostress or strain axes can be estimated by means of stress inversion methods (Angelier, 1979, 1984; Etchecopar et al., 1981) or kinematic analysis (Marrett and Allmendinger, 1990). Stress inversion analyses are based on the Wallace
Bott hypothesis, which makes the assumption that slip on a surface will occur in the direction of maximum resolved shear stress. Applying this hypothesis to measured fault slip data enables us to estimate the orientations of the principal stresses (Angelier, 1994). However, it can be argued that any fault analysis that is based on measurements of slip surfaces, slip directions and sense of slip are, strictly speaking, a kinematic approach that primarily gives the principal shortening (P) and extension (T) axes, as outlined by Marrett and Allmendinger (1990), and that stress can only be indirectly correlated with these axes, assuming no rotation of structures during deformation.

Strain and fault orientation patterns Simple conjugate sets of faults are compatible with plane strain, where the length of the intermediate strain axis Y remains unchanged during deformation. For the normal fault regime, the horizontal extension direction is then perpendicular to the (average) strike of the faults, and for a thrust belt setting, the principal horizontal shortening direction is perpendicular to the strike of the faults. Both natural fault populations and those formed during physical and numerical experiments show some variation in strike direction. Plane-strain experiments show such variations very well (Fig. 8.9B and C), and there are many examples of natural fault populations at different scales that contain faults or fault segments at low angles to the extension or shortening direction (e.g. Fig. 8.9A). In detail, we usually find fault bends and segments of somewhat different orientations, and zigzag-like geometries also occur in both numerical models (Cowie et al., 2000; Finch and Gawthorpe, 2017; Deng et al., 2017) and in nature. Two or more double sets of conjugate faults can also result from a single deformation episode, reflecting 3D (or triaxial) strain of the flattening type with extension along two principal strain axes (X and Y) (Oertel, 1965; Krantz, 1988; Reches, 1988; Healy et al., 2015) (Fig. 8.10) or doming with shortening in two directions. Hence,

Regional Geology and Tectonics

126

8. Fault classification, fault growth and displacement

Whakatane Graben, NZ

(A)

N

10 km

(B) Plaster model

(C) Clay model

5 cm

10 cm

FIGURE 8.9 Fault populations formed under approximate plane-strain conditions. (A) Whakatane Graben, New Zealand. This graben has been interpreted to have a slight transtensional character, but is close to pure extension. (B) Plaster model. (C) Detailed view of clay model. In all the models the faults trend nearly perpendicular to the extension direction. Some local variations in strike orientation can be related in most cases to their growth history. Source: (A) Modified from Lamarche, G., Barnes, P.M. Bull, J.M., 2006. Faulting and extension rate over the last 20,000 years in the offshore Whakatane Graben, New Zealand continental shelf. Tectonics, 25. doi:10.1029/2005tc001886. (B) Redrawn from Blækkan, I., 2016. Evolution of normal faults and fault-related damage: insights from physical experiments, Master thesis, University of Bergen. 86 pp. (C) redrawn from picture in Ackermann, R.V., Schlische, R.W. & Withjack, M.O., 2001. The geometric and statistical evolution of normal fault systems: an experimental study of the effects of mechanical layer thickness on scaling laws. J. Struct. Geol. 23, 1803
1819.

(A)

Plane stress and strain σ1

σ1

(C)

3-D stress and strain

σ3

σ2≈σ3

σ2

(B)

X

σ3

σ2≈σ3 Y

σ1 Z σ2

σ2

X ≈Y

(D)

σ3

σ1

(E)

X

σ1 Z

(F)

σ3

σ2 ≈σ3

X

X =Y

σ1 Z

σ3

σ2

FIGURE 8.10 Idealized relationship between fault patterns and strain (or stress), shown for the normal fault regime. Simple Andersonianstyle conjugate fault sets (A) result from plane strain, while orthorhombic (B) or polymodal pattern result where the resulting strain is nonplanar (i.e. 3D or triaxial strain). Figures (A, C, and D) show initial fracture pattern in relation to principal stress axes. Figures (B, D and E) show the appearance of the fractures (faults) in spherical projections and also show the strain axes (X $ Y $ Z).

Regional Geology and Tectonics

Displacement distributions on faults

127

3D strain of this kind results in a large variety in fault orientation caused by a single phase of deformation. This is the case where the different sets mutually crosscut each other; in contrast, if one set (orientation) systematically crosscuts another within a region, we are more likely looking at two phases or stages of deformation. Preexisting structures have been shown to influence fault orientation to various degrees, depending on their orientation (strike and dip), size, geometry and strength (e.g. Sibson, 1985). Two general cases can be envisioned in the context of a sedimentary basin: one where a sedimentary sequence is exposed to two phases of deformation with different extension directions, and a single phase where earlier structures occur in the basement beneath the basin. The first case has been explored through physical modeling by Henza et al. (2010), who looked at changes in extension direction at up to 45 degrees. Their experiments show that the preexisting structures are obliquely reactivated together with the formation of new faults that variably cut or terminate against older faults. The result is a complex fault pattern with a large variation in fault trends. The effect of the angle between the two extension directions would vary for natural cases, depending on fault properties, length, dip and planarity, but the experiments illustrate well how composite fault patterns may emerge from two deformation phases with extension directions differing by up to 45 degrees. Further complications occur when faults are reactivated in a different tectonic regime, for instance normal faults reactivated as reverse faults (Kelly et al., 1999; Marshak et al., 2000; Zuluaga et al., 2014). For instance, reverse faults preferentially form at a lower dip angle than normal faults (Fig. 8.8) and may utilize preexisting normal faults to a lesser extent, forming hanging-wall shortcuts in their upper parts (e.g. Amilibia et al., 2008). The second case, where an undeformed sedimentary sequence overlies a basement with preexisting structures, reactivation and upward propagation of basement faults can occur. Again, reactivation is highly dependent on orientation, geometry and strength of the preexisting basement structures. The strength of a preexisting shear zone or fault core in metamorphic basement rocks is difficult to predict and would also vary laterally and vertically. There are many examples where basement structures are reactivated in contraction, strike-slip and extension (e.g. Bailey et al., 2005; Bird et al., 2015; Phillips et al., 2016; Peace et al., 2018). In general, reactivation of basement structures involves a combination of upward propagation of the basement fault and nucleation of new faults above the basement structure that may link up and form a composite fault zone as strain accumulates (see section ‘Fault Growth’ below). Fault-propagation folding commonly occurs in the overlying sedimentary sequence during basement fault reactivation, both in extensional (Sharp et al., 2000) and contractional (Zuluaga et al., 2014) settings.

Displacement distributions on faults Isolated faults tend to show a gradual increase in displacement from the tipline towards a central point, and ideally the tipline is more or less elliptical, as shown in Fig. 8.11. This simple elliptical pattern of displacement contours is modified in mechanically stratified rocks. For horizontal layering, the elliptical shape is replaced by a more rectangular shape because of the vertical growth restriction imposed by the layering. For example, a close to circular tipline may be established as a fault initiates in a strong layer (Fig. 8.12A), but ellipticity changes dramatically when the radially propagating fault reaches the top and bottom of the strong layer: the layer boundaries impose restrictions on fault tip propagation (Fig. 8.12B). At some point, the fault will break through the restricting layer boundary, and the ellipticity decreases again. Further complications arise from fault linkage, as discussed below. Displacement profiles across faults show how displacement varies in the horizontal or vertical direction, and how the maximum displacement (Dmax) along such profiles generally increases with fault length (or height if measured in the vertical direction). This relation has been quantified by field investigations and seismic data interpretation. Global data that span many orders of magnitude show an approximately linear relationship between fault length and displacement (D  0.3L, Fig. 8.13). In detail, the data show a considerable spread, about 2
3 orders of magnitude, which may be due to fault growth by linkage (see below), crustal anisotropy (including layering) and 3D sampling effects. Hence, prediction of fault length from Dmax or vice versa is possible, but only with a considerable uncertainty, unless the relationship can be better constrained for the region or stratigraphic section in question.

Regional Geology and Tectonics

128

8. Fault classification, fault growth and displacement

(A)

(B)

A

B

50 m 37.5 25 12.5 0

500 m

(c) Gradient: 50/845 =0.06

50 m

D

25

Dmax

Gradient: 50/700 =0.07 B

0 500 m

A

1500 m

FIGURE 8.11 Displacement contours on a fault, idealized schematic model (A) and as interpreted from seismic data (B). The outer contour line (0 m) is the tipline of the fault. Figure (B) represents an isolated fault, as interpreted by Childs et al. (2003). (C) Displacement profile through the centre of the fault, with calculated average displacement gradients (0.06 and 0.07 for each side of the maximum). D
displacement.

Dmax

(A) No restriction D

Dmax

Height

L

L1 Length L1 (B) Vertically restricted Dmax D

(Underdisplaced)

L

L2

L2 (C) Horizontally restricted growth (abutting)

Dmax

D

(Overdisplaced)

L3

L

(D) Horizontal fault interaction

A

B

Displacement (D)

L3

Dmax A

B

Horizontal distance along fault

LA+B

L

FIGURE 8.12 Schematic models of fault growth in isotropic (A), vertically restricted (B) horizontally restricted (C) and fault overlap interactive (D) settings. Also shown are displacement profiles and Dmax-L evolution for each case. Source: In part from Fossen, H., 2016. Structural Geology, second edition. Cambridge University Press, Cambridge. Regional Geology and Tectonics

129

Fault initiation

105

104

0.028X

0.96

1000

100

10

Dmax

L D=

1

/1

0 00

/1

L D=

0.1

0.0017X

0.52

0.01

0.001

0.0001 0.01

Cataclastic deformation bands (F&H 1997) 0.1

1

10

100

1000

104

105

106

L FIGURE 8.13 Maximum displacement plotted against length for faults from different settings in logarithmic diagram (mostly normal faults). Cataclastic deformation bands are shown as a separate data set. Own data 1 data cited in Schultz et al. (2008). F&H
Fossen & Hesthammer. Note that straight correlation lines in log-log space represent power-law scaling relations, with the slope representing the exponent. In this case, the slope (0.96) is close to 1, in which case the relation between D and L is close to linear: D 5 0.3L. Source: From Fossen, H. & Hesthammer, J., 1997. Geometric analysis and scaling relations of deformation bands in porous sandstone. J. Struct. Geol., 19, 1479
1493.

Fault initiation Fault formation from scratch The initiation of faults in macroscopically homogeneous layers can be explored by physical modeling. Several studies demonstrate how a fault can develop from an array of minor precursor structures that define a brittle shear zone and are oriented at an angle to the initial shear zone boundaries and the resulting fault zone (Cloos, 1928; Riedel, 1929; Tchalenko, 1970). These incipient brittle structures tend to be oblique to the zone that they define, and based on their orientations and sense of slip they have been categorized into R (Riedel) and R0 (antitheric Riedel) shears. Ideally, R and R0 shears form conjugate sets that are bisected by the largest principal stress direction (σ1). An additional set of P shears that form at low angles to the zone can also occur (Fig. 8.14A). Examples are shown in Fig. 8.14 from sandstones (B and C), plaster experiment (D) and by a recent strike-slip earthquake surface rupture pattern (E), and also in gneisses in Fig. 8.15. As displacement accumulates, these precursor structures start to connect to form a continuous fault zone rather than a simple fault “plane” (Fig. 8.14F). In other cases, arrays of extensional en-echelon
arranged veins form (Fig. 8.16), particularly in strong rock layers. A component of ductile deformation is sometimes revealed by the rotation of the central and oldest portions of the veins, generating a sigmoidal vein geometry that at some point will be cut by new veins. Eventually, the zone will be breached, and a continuous fault zone forms (Fig. 8.15B), similar to the situation described above for R and R0 structures, and one or more continuous striated slip surfaces form (Fig. 8.15C). In carbonates, stylolites may form perpendicular to the veins (Fig. 8.16), and the orientations of vein tips and stylolites reveal the orientations of the instantaneous stretching directions (ISA), commonly equated to σ3 and σ1 for idealized

Regional Geology and Tectonics

130

8. Fault classification, fault growth and displacement

FIGURE 8.14 Early (A
E) and advanced (F) stages of faulting. (A) Principal sketch showing the orientation of different kinds of subsidiary structures: R
Riedel shears, R0
antithetic or conjugate Riedels, P
P-shears and ISA
instantaneous stretching axes. (B
C) Ladder structures in sandstone, composed of deformation bands. (D) Plaster experiment. (E) Surface rupture pattern during the 2010 Canterbury earthquake, New Zealand. (F) Fault in sandstone showing R shears (R) adjacent to the main slip surface M. Detail from Fig. 8.3A. Source: (E) Photo by New Zealand Ministry of Civil Defence & Emergency Management, used with permission.

low-strain situations (Fossen, 2016). ISA1 is the fastest stretching direction and would then correspond (by orientation, not magnitude) to σ3. σ1 would correspond to ISA3, which is the slowest (usually negative) stretching direction or, in terms of shortening, the fastest shortening direction. Most rocks are not homogeneous, but involve a metamorphic or depositional layering, which complicates fault formation and growth. In mechanically stratified sections that are exposed to layer-parallel stress, faults initiate in the stiffest or strongest layers or sequence of layers, that is layers with highest Young’s modulus. These are layers where stress is concentrated, and where the rock first yields (e.g. Gudmundsson, 2011). Hence, faults initiate in several different competent layers at more or less the same time in such layered rocks. Brittle deformation may initiate as shear fractures, extension fractures (joints, fissures or veins) or hybrid fractures. For example, turbidites, with strong sandy and sometimes calcareous layers alternating with mechanically weak shale (Fig. 8.17A), commonly show evidence of early extension fracturing and vein formation, followed by linkage as shown schematically in Fig. 8.17B
D. The different orientations of the extension structures (veins) and the resulting fault, together with any steps formed during linkage, create a zone of deformation rather than a simple slip plane.

Regional Geology and Tectonics

Fault initiation

131

FIGURE 8.15 (A) En-echelon fractures forming during incipient stages of faulting. (B) Veins are connected and filled with epidote. (C) Further shear on the fracture/vein creates striations on a smooth but curved surface. The axis of curvature indicates the slip direction. Devonian brittle deformation of caledonized Proterozoic gneiss, Øygarden Complex, SW Norway.

Faulting by activation of preexisting structures The importance of preexisting planar structures during fault initiation has been pointed out by several authors (Segall and Pollard, 1983; Martel et al., 1988; Bu¨rgmann and Pollard, 1994; Peacock, 2001; Crider and Peacock, 2004; Pollard and Fletcher, 2005). Preexisting structures that can localize strain and guide fault growth are shear zones, joints, veins, bedding and dike walls. Reactivation of earlier faults is not included here, as in this case a fault is already established. Joints are perhaps the most common structure (excluding preexisting faults) that influence fault nucleation. Segall and Pollard’s (1983) study from the Sierra Nevada, California, is a benchmark example of the importance of faulting by joint reactivation. Because joints tend to be steep, they are easily activated as strike-slip faults, given that their strike is favorably oriented with respect to the new active stress field. However, steep joints reactivated as subvertical faults or slip surfaces are also very common, generating steep faults or fault elements with complex geometries. Faulted joints are recognized primarily by striated joint surfaces and also give faults or fault segments an unusually planar geometry. They differ from primary faults by

Regional Geology and Tectonics

132

8. Fault classification, fault growth and displacement

FIGURE 8.16 En-echelon veins accompanied by stylolitic surfaces. ISA
instantaneous stretching axes.

having an initial length that is dictated by the length of the joint. Hence, they are characterized by low displacement/length (D/L) ratios (Wilkins et al., 2001). Faults formed by joint reactivation can result in a single sharp slip surface that lacks subsidiary structures such as Riedel shears, with virtually no fault core or damage zone. As the fault outgrows the joint, however, complications occur, and damage zone and fault core are established and grow. An outstanding example of faulting by joint reactivation is the fault population in the grabens area of Canyonlands National Park, Utah (McGill and Stromquists, 1979). This young fault population formed close to the surface in a B500 m thick sedimentary sequence containing sandstone layers by faulting of joints belonging to very regularly oriented and spaced joint sets (Moore and Schultz, 1999). Consequently, the faults are very straight, but locally take on zigzag geometries as they exploit different joint sets (Cartwright and Mansfield, 1998). Also, bedding and other lithologic contacts can be reactivated when favorably oriented, as in the example shown in Fig. 8.18. Many plastic (ductile) shear zones show evidence of brittle reactivation, particularly large shear zones with extensive length and width. The reason why faults preferentially initiate on shear zones is related to the mechanical anisotropy that occurs on a range of scales, from microfabrics through outcrop-scale foliation and mylonitic banding (weak mica-rich layers and stronger quartz-feldspar layers) and contacts between lithologic units with highly different properties, to crustal-scale anisotropy represented by major shear zones (e.g. White et al., 1986). Furthermore, large shear zones represent continuous tabular structures that cut through large portions of the crust. By nucleating on such shear zones, faults can avoid the complicating effect of mechanical stratification and irregularities that generally characterize the crust.

Regional Geology and Tectonics

Fault initiation

133

FIGURE 8.17 (A) Two conjugate incipient faults forming by linkage of veins formed preferentially in strong layers. (B
D) Schematic illustration of fault formation from rocks with alternating strong and weak layers: Veins form in strong layers (A) and faults form as shear fractures connect veins in different layers (B and C). Source: (B
D) Modified from Crider, J.A. Peacock, D.C.P., 2004. Initiation of brittle faults in the upper crust: a review of field observations. J. Struct. Geol., 26, 691
707.

Regional Geology and Tectonics

134

8. Fault classification, fault growth and displacement

FIGURE 8.18 Activation of bedding planes as slip surfaces in tilted carboniferous turbidites during Triassic rift-related faulting. The unconformity and Triassic sediments reveal the faulting (Near Sagres, Portugal).

Fault growth Faults grow by repeated seismic rupture and by more continuous aseismic creep. The San Andreas Fault, for example, has a long section that accumulates slip primarily by creep, flanked by seismically active segments (e.g. Scholz, 2002; Titus et al., 2006). In both cases, displacement accumulates over time, and faults tend to get longer and taller. This is reflected by the positive relationship between displacement and length that can be established for most fault populations (Fig. 8.13). Faults can grow from small fractures that propagate laterally and vertically as they accumulate slip, or they can grow by (re)activation of preexisting structures such as older joints or faults. The first case is an idealized case where (re)activation of preexisting structures is negligible, and where the faults grow in isolation until they incidentally interact. This model has been referred to as the isolated fault model (Walsh et al., 2003). Such growth can be studied in physical or numerical models devoid of preexisting structures. In nature, preexisting structures are always present and can significantly influence fault growth, as discussed above. For instance, reactivated joints or weak older faults easily accumulate slip along their entire length, creating early-stage faults that are long relative to their maximum displacement. Such underdisplaced faults will accumulate displacement without tip propagation, until they reach a D/L ratio that concentrates enough differential stress at the tip that propagation can occur. Hence, they can be expected to create vertical (constant L, increasing D) paths in diagrams such as Fig. 8.13 until they start to propagate beyond the tipline of the preexisting joint. Data supporting such a joint reactivation model in sandstone are provided by Wilkins et al. (2001). Low D/L ratios also characterize incipient and small faults in porous rocks, where faults form in or along deformation band clusters. In these cases, the deformation band clusters, which are then precursor structures, follow the trend of cataclastic deformation bands shown in Fig. 8.13, and once a continuous slip surface forms along this zone, it becomes an underdisplaced fault. Again, displacement can be expected to accumulate while the length remains unchanged until a normal D/L ratio is obtained (Fossen and Hesthammer, 1997). From this point onwards, their tips propagate within a tip damage zone of deformation bands that is maintained ahead of the slip surface (Shipton and Cowie, 2003; Fossen et al., 2007) and typically link up with adjacent structures. It is useful to know how fast displacement varies along a fault when predicting fault displacement or minimum fault length away from an observation point (Fig. 8.11C). The average displacement gradient is around 0.1
0.01 for most normal faults (Fig. 8.19). For a gradient of 0.1 or 0.01, moving 1 km along strike changes the displacement by 100 or 10 m, respectively, provided that we do not cross the Dmax point. As can be seen even from the simple fault presented in Fig. 8.11B and C, the gradient can change locally along a fault, and the average gradient can only be used as an approximate estimate of displacement variation, for instance away from a well location. It should also be noted that individual data sets tend to show a smaller range in displacement gradient than the global data set (Fig. 8.19). Hence, there are region-specific differences that may relate to mechanical stratigraphy, lithology, degree of linkage, strain (2D vs 3D) and tectonic regime that influence on the displacement gradient of the fault.

Regional Geology and Tectonics

135

Fault interaction and linkage

1.0

Displacement gradient (Dmax/0.5L)

Elliot (TF) Walsh & Watterson 87 (NF) Krantz 1988 (NF) Cart (NF) Peacock & Sanderson (NF)

0.1

Peacock Normal faults(NF) Dawers Normal faults (NF) Villemin strike-slip faults (SSF) Cococi Basin, Brazil (NF) Solite (NF)

0.01

0.001 0.01

0.1

1

10

100

103

104

105

106

Fault length (m)

FIGURE 8.19 Average displacement gradient (Dmax/(0.5L)) plotted against fault length for a variety of fault populations in logarithmic diagram. TF
thrust faults and NF
normal faults. Majority of the data show gradients between 0.1
0.01, but each fault population occupy a narrower range than the whole data set.

Fault interaction and linkage Linkage of faults and fractures occurs at almost any scale, from the linkage of microcracks to form mesoscopic shear fractures (Reches and Lockner, 1994; Crider, 2015) via the linkage of R
R0 structures shown in Fig. 8.14 to the linkage of large fault segments up to hundreds of kilometers long (Peacock et al., 2000). Linkage is a fundamental process of fault growth and can be observed in any tectonic regime and setting, including thrust (Nicol et al., 2002), strike-slip (Woodcock and Fisher, 1986) and various extensional settings. The latter will be the focus of the following discussion. Whether fault segments form in isolation or by reactivation of older structures, they will at some point interact with other faults and link up to form much longer faults. In the isolated fault model, this is considered a random process, but the position and growth of the linking segments can be controlled by preexisting structures. In either case, the linkage history starts when the fault tips get close enough that their zone of stress perturbation or elastic strain fields overlaps and influences their propagation paths. As the fault tips pass each other, a relay zone develops that is characterized by complex small-scale (subseismic) deformation structures, and layers are bent during the fault interaction. Steepening of the displacement profiles by up to 2.5 times the normal displacement gradient characterizes this stage, indicating a reduction in the tip propagation rate (Peacock and Sanderson, 1996; Gupta and Scholz, 2000). This goes together with the observation that the displacement profile of each fault becomes skewed, with maxima shifted towards the relay structure (Fig. 8.12D). The geometry of relay structures are scaleindependent, with a common length:width ratio of around 3
4 (Long and Imber, 2011) (Fig. 8.20). Fault growth by linkage is easily documented by simple physical experiments, such as the plaster experiment shown in Fig. 8.21. Here, several small segments (F2a
d in Fig. 8.21A) link up to form a longer fault (F2 in Fig. 8.21B), which after its formation accumulates displacement without lengthening (from Fig. 8.21B and C). Extension beyond the stage shown in Fig. 8.21C would break the ramps between segments F1
F3 and repeat the history of F2 at a larger scale to form a continuous curvilinear F1
F3 fault trace. A relay structure or relay zone represents an anomalously wide portion of the fault damage zone sheared by the two overlapping fault segments. The types of subsidiary structures developed in a relay zone depends largely on lithology and may encompass deformation bands, slip surfaces, extension fractures, stylolites, veins, dikes and minor faults, as described in several recent publications (e.g. Trudgill and Cartwright, 1994; Peacock and Sanderson, 1995; Acocella et al., 2000; Rotevatn et al., 2007; Bastesen and Rotevatn, 2012; Fossen and Rotevatn, 2016). Their density and connectivity depend on the maturity of the relay zone, that is the amount of strain or displacement accommodated in the zone. Eventually, the two overlapping faults will breach the relay ramp to

Regional Geology and Tectonics

136

8. Fault classification, fault growth and displacement

(A)

(B)

106

y = 3.3x0.99 R2 = 0.98

Overlap structure L

L (m)

104

W

W

102

L

1 L

Hanging-wall breach

Footwall breach

=

=

10

W

10-2 10-2

10-4

102

1 W (m)

104

Mean: 3.957

(C) 200

Median: 3.25 100

Central breach

Double breach

0

0

5

10

15

W/L

FIGURE 8.20 (A) Fault overlap structure (relay structure) created by two subparallel fault tips. The structure may be breached in different ways; four general scenarios are shown. (B) relay length scales with width over at least nine orders of magnitude, that is ramps are a selfsimilar type of structure. See Fossen and Rotevatn (2016) for data sources. (C) Statistical distribution for data plotted in (B), describing the range in L/W ratio reflected by the scatter in (B).

(A)

F2b F2a F2c F2d

F1b R

F1c1

F2

(B)

F2a F2b F2c

R

F1c2

F2d R F1A

F1b

F3

F1c

(C) R

R

F2

F1

F3 10 cm

FIGURE 8.21 Progressive evolution of fault zone produced in extensional plaster model (black arrows indicate the extension direction). Multiple small faults in (A) propagate and coalesce by linkage as strain accumulates. (B) Intermediate stage where several faults have coalesced to form three larger faults (F1
F3). Note that linkage points are preserved as pronounced jogs in fault trace. (C) At this stage, the tip of F1 has propagated by linkage of smaller faults, while F2 and F3 have experienced constant length (L) growth. If the experiment had continued, F1
F3 would have linked up to a single continuous fault. Several breached relays can be recognized by fault jogs. Source: Redrawn and modified from Blækkan, I., 2016. Evolution of normal faults and fault-related damage: insights from physical experiments, Master thesis, University of Bergen. p. 86.

Regional Geology and Tectonics

Fault interaction and linkage

137

form a continuous fault structure. Single-tip or double-tip breaching is possible, and further slip accumulation on the composite fault will leave much of the relay ramp inactive. The new and much longer fault will initially be underdisplaced, and a displacement minimum may exist for some time at the linkage point (Faure Walker et al., 2009). In the length-displacement diagram (Fig. 8.13), linkage results in rapid increase in length of the new combined fault followed by an increase in maximum displacement while the length remains unchanged, that is a horizontal, then vertical path (Fig. 8.12D) (Cartwright et al., 1995). Hence, growth by linkage can explain some of the scatter in length-displacement diagrams. For lateral linkage controlled by upward propagation of underlying structures, however, the minima may be erased at initial stages of linkage, and such fault systems are referred to as kinematically coherent (Walsh et al., 2003). In general, fault growth by linkage is considered the most efficient and common way for faults to grow in length. It also occurs in the vertical (dip) direction, as already indicated in Fig. 8.17, as well as in any other direction. In a more or less horizontally layered sequence, layering plays an important role. Mechanically contrasting layers may cause faults to initiate at different stratigraphic levels during strain accumulation, and extensive linkage occurs as they connect and grow into larger faults. Fig. 8.17 shows how this can influence the width of the fault zone. Field observations (Fig. 8.22) show a large variation in both fault core and fault damage zone width in the vertical direction that can be explained by vertical coalescence of fault segments. Hence, the resulting fault complications and variations in damage zone width and properties depend on lithology, the mechanical properties of the layers, their thickness, progressive fault rotation and fault displacement (e.g. van der Zee et al., 2008). Further work is needed on the role of these factors to obtain a useful algorithm for fault damage zone prediction. Indeed, faults with any orientation can interact, and they can do so in different ways (Fossen et al., 2005). Conjugate systems and subparallel faults are already covered above, but abutting situations where one fault terminates against another are also very common. In this case, a fault tip approaches an already existing fault and terminates against it. When both faults are active, the new fault will link up to form a kinematically coherent system of three blocks and a Y-type fault intersection. An example is shown in Fig. 8.23, where also another common feature is seen, known as fault tip deflection near an existing weak fault. This rotation reflects the stress rotation that occurs around weak structures (e.g. Dyer, 1988), and the fault propagating into the perturbated local stress field of an existing fault may result in a curved fault, as shown in Fig. 8.23B and C. Field examples show that these types of locations can display quite complicated patterns of small-scale structures with a multitude of orientations and even types of structures, in an anomalously wide damage zone. In the area covered by Fig. 8.23, comparison between single fault damage (e.g. location BC) and areas of fault interaction (location CR in Fig. 8.23) demonstrates this fact well, as illustrated somewhat schematically in Fig. 8.24 (see also Johansen et al., 2005).

FIGURE 8.22 Three faults at different stages of evolution, each consisting of several subsidiary elements. (A) Two faults linking up vertically, forming a releasing (extensional) stepover. Extension indicated by veins. Marble Canyon, Death Valley. (B) Fault zone established along minor (meter-scale) fault, cutting soft-sediment folds. Vertical linkage not completed yet. (C) Mature fault with approximately 100 m offset, with well-developed fault core and lenses. Latter two examples are from the Neogene-Quaternary Granada Basin, Spain.

Regional Geology and Tectonics

138

8. Fault classification, fault growth and displacement

Railr oa

d

191

? (A) Cover Cretaceous Morrison Fm Moab Mbr Slick Rock Mbr Dewey Bridge Mbr Navajo Fm Kayenta Fm Normal fault

n nyo ?

au bF

Arches N. P.

sh e

Seg ment

rF . S.

lt

Colorado River Moab

10 km

Fau l

tS

egm .

CR

1 km

(B)

Courthouse Branch Point

Courthouse Wash

oa

M

tt F au lt

use

Mill Canyo n

Ba rtl ett

Cou rtho

? Tu

Ba rtl e

t ul Fa

BC

Tusher Canyon

Hid

N

b oa M

den

Ca

Ca ny on

191

Courthouse Rock Utah Col. Plat.

(C)

CR Propagation direction

FIGURE 8.23

The northernmost part of the Moab Fault and a series of connecting subsidiary fault segments (A). The evolution of these segments is interpreted in B
C, and involves tip deflection related to stress perturbation near neighboring faults. Hence, the curvature of the southeastern Courthouse fault segment formed because of the already established Moab Fault. Also, the displacement of the Courthouse fault segment decreases towards the linkage point with the Moab Fault (CR), consistent with an abutting history. CR
Courthouse Rock locality and BC
Bartlett Canyon locality.

(A)

Single fault

Damage zone

Branch point (Y)

(B)

Branch point

MF

MF

20 m

CF FIGURE 8.24

(A) Single fault damage and (B) the more complicated Y-point (abutting) branch point situation. These two situations correspond to the BC and CR locations in the previous figure. The BC damage zone is illustrated in more detail in Fig. 9.13 in Fossen (2016), while the CR branch point situation is presented in detail by Johansen et al. (2005).

Fault populations Faults form populations that develop from a multitude of small faults to a more diverse population consisting of faults with a large range of displacements and lengths. As already emphasized, this evolution involves linkage of small faults to larger ones, where the largest faults take up most of the subsequent strain (e.g. Cowie, 1998) (Fig. 8.25). The early small faults that are not involved in such linkage will then become inactive. The extent of

Regional Geology and Tectonics

Fault populations

139

(A)

(B)

(C)

FIGURE 8.25

Schematic illustration of the evolution of a fault population in an extensional (rift) setting. (A) Initial population of minor isolated faults developing largely perpendicular to the extension direction. (B) Some growth by fault tip propagation creates zones of fault overlap. (C) Linkage of favorably arranged segments into long- and large-offset faults, with secondary formation of adjustment faults (red) between the large faults. Note that the late minor faults typically show a large variety of orientations, commonly trending at a high angle to the large faults.

linkage varies within the deformed region, and after some strain has been accumulated, there will be a distribution of fault sizes that in many cases can be described by a power-law relationship of the form N 5 aS2D, where S represents a fault size parameter such as displacement or length, N is the cumulative number of faults greater than or equal to S, a is a constant and D is the power-law coefficient or fractal dimension that characterizes the relative proportion of large and small faults in the population. D can be used to assess the amount of strain represented by different size ranges of a fault population, including subseismic faults. The amount of subseismic deformation depends on seismic resolution and can be substantial where regional data are considered (see Marrett and Allmendinger, 1992 and Walsh and Watterson, 1992 for further discussion). In terms of distribution, the largest faults in a faulted region commonly develop a regular pattern with a characteristic spacing. This spacing is in part controlled by the mechanical thickness of the relevant layer, which for very small faults could be the thickness of a competent sandstone or limestone layer. For larger faults, the relevant layer may comprise a supra-salt sequence, supradetachment hanging wall or thrust nappe, while for firstorder faults with several kilometers of displacement, the relevant layer may comprise the entire brittle crust. Soliva et al. (2006), who considered this relationship primarily for small faults, found that spacing is typically about half of the relevant layer thickness. For a 10 km thick brittle crust, this fits well with the B5 km average spacing observed in most rifts (Morellato et al., 2003). Even though a multitude of small faults form at early stages of rifting, most of which become inactive, small faults also form at later stages. Once larger faults are established, small faults and related deformation structures will potentially form in the fault blocks between these faults in response to complications during further deformation. In terms of stress, this can be explained by the way the existing faults perturb the regional stress field due to their geometry and relative movements. Hence, both the locations and orientations of new faults will be controlled by the existing faults and their geometries. Maerten et al. (2002) and Maerten and Maerten (2006) used geomechanical modeling to make predictions about such smaller-scale faults. They applied a 3D numerical model to determine the stress conditions in an area containing active N
S trending North Sea rift faults. The computed stress field around and between the larger faults was then combined with a Coulomb failure criterion to predict

Regional Geology and Tectonics

140

8. Fault classification, fault growth and displacement

Major fault

Minor fault

Calculated σ2 trend)

N 1 km

FIGURE 8.26 Numerically modelled stress field during E
W extension in the Oseberg Syd area, northern North Sea. The blue faults perturb the stress field, and many of the minor (red) faults are oriented in agreement with the local orientation of σ2. Hence, many minor faults and their great variety of orientations can be explained as having formed after the blue faults were established (but during the same phase of rifting). Source: Modified from Maerten, L., Gillespie, P. Pollard, D.D., 2002. Effects of local stress perturbation on secondary fault development. J. Struct. Geol. 24, 145
153. (A)

(B) N

58° 10’N

Bravo

Gullfaks Field 3° W

N

Beatrice Field

Alpha

2.5 km

o

2 10I

2.5 km

FIGURE 8.27 Two fault patterns from the North Sea rift, at two different stages of development. (A) The Beatrice Field (Inner Moray Firth Basin) where the majority of faults are subparallel and perpendicular to the extension direction. (B) The Gullfaks Field, where several small faults bound by the larger N
S trending faults have different orientations, making Y- and T-branch points with the N
S faults. Many of the red-colored faults may have formed at a relatively late stage of extension, due to kinematic complications caused by slip on the larger faults. Source: (A) Fault pattern extracted from Husmo, T., Hamar, G.P., Høiland, O., Johannesen, E.P., Rømuld, A., Spencer, A.M. et al., 2002. Lower and Middle Jurassic. In: Evans, D., Graham, C., Armour, A. & Bathurst, P. (Eds.), The Millennium Atlas: Petroleum Geology of the Central and Northern North Sea. Geological Society, London, pp. 129
155. (B) Fault pattern from Fossen, H., Rørnes, A. 1996. Properties of fault populations in the Gullfaks Field, northern North Sea. J. Struct. Geol. 18, 179
190.

the orientations and densities of smaller faults. The result (Fig. 8.26) shows a population of smaller faults with a large range in orientation, in part similar to the small-scale faults identified from seismic interpretation, and quite different from the N
S trending major faults. A somewhat similar fault pattern is observed in the Gullfaks Field fault population in the North Sea rift (Fig. 8.27B), where N
S trending domino-style rift faults separate smaller faults with many different orientations, many oblique or perpendicular to the larger faults (red faults in Fig. 8.27B). These small faults abut the larger faults

Regional Geology and Tectonics

141

Faults and fluids

5 km

T>20 m

5 km

1 FF, FF > CH

Unconventional systems tracts: HAST, LAST

Stratal stacking patterns

FIGURE 23.25 Classification of systems tracts. Stratal stacking patterns provide the basis for the definition of all units and surfaces of sequence stratigraphy. Downstream-controlled systems tracts are defined by stacking patterns associated with specific shoreline trajectories (i.e. NR, FR, T). Upstream-controlled systems tracts are defined by the dominant depositional elements. Systems tracts are independent of scale. Sequences of all scales include component systems tracts. Therefore, systems tracts can be assigned to different hierarchical orders that match the hierarchical level of the sequence to which they belong. NR normal regression; FR forced regression; T transgression; CH channels; FF overbank (dominated by floodplain fines); FSST falling-stage systems tract; LST lowstand systems tract; TST transgressive systems tract; HST highstand systems tract; HAST high-amalgamation systems tract; LAST low-amalgamation systems tract.

West

East

GR RES

GR RES

GR RES

GR RES

GR RES

GR RES

Datum (MFS)

30 m

Fluvial facies (main Almond)

Marine sandstone (upper Almond)

Marine shale (Lewis)

FIGURE 23.26

Long-term transgression punctuated by higher frequency transgressions and regressions (Campanian, Western Interior Seaway, Wyoming;). The cross-section is approximately 65 km long. GR gamma ray; RES resistivity; MFS maximum flooding surface. Source: From Catuneanu, O., Galloway, W.E., Kendall, C.G.St.C., Miall, A.D., Posamentier, H.W., Strasser, A., et al., 2011. Sequence stratigraphy: methodology and nomenclature. Newsl. Stratigr. 44 (3), 173 245. Modified after Weimer, R.J., 1966. Time-stratigraphic analysis and petroleum accumulations, Patrick Draw field, Sweetwater County, Wyoming. Bull. Am. Assoc. Pet. Geol. 50 (10), 2150 2175; and Martinsen, R.S., Christensen, G., 1992. A stratigraphic and environmental study of the Almond Formation, Mesaverde Group, Greater Green River Basin, Wyoming. In: Wyoming Geological Association Guidebook, Forty-third Field Conference, pp. 171 190.

(i.e. hierarchical orders) is consistent with the fact that depositional systems can also be observed at different scales (Figs 23.8, 23.19 and 23.21). The internal makeup of a systems tract may vary greatly with the scale of observation, from a succession of beds and bedsets (i.e. sedimentological cycles within the lowest rank depositional systems; Fig. 23.8) to a set of higher frequency sequences of lower hierarchical ranks (Figs 23.19, 23.20 and 23.22). The scale of the lowest rank systems tracts at any particular location defines the highest resolution that can be achieved with a sequence stratigraphic study. Systems tracts of all scales may include hiatal surfaces of equal and/or lower hierarchical ranks (e.g. a thirdorder transgressive systems tract may include a third-order wave-ravinement surface, as well as other types of unconformity of lower hierarchical ranks). At each stratigraphic scale, the 3D facies assemblages of systems tracts are linked by the dominant depositional trends associated with the defining stacking pattern. For example, a transgressive systems tract is defined by a retrogradational stacking pattern, even though the transgression may be interrupted by higher frequency regressions of lower hierarchical ranks (Figs 23.22 and 23.26). The change in stacking pattern across the systems tract boundary may be accompanied by a change in depositional system (e.g. a change from an estuary to a delta across a maximum flooding surface), or it may occur within a depositional system (e.g. a change from retrogradation to progradation within a shelf system across a maximum flooding surface). The highest frequency changes in depositional system associated with systems tract boundaries are recorded within the transit area of the shoreline (Figs 23.8, 23.19 and 23.22). In downstream-controlled settings, stratigraphic stacking patterns are defined by the trajectory of subaerial clinoform rollovers (i.e. stratigraphic shoreline trajectories), which can be observed at different scales. At any stratigraphic scale, the shoreline trajectory represents a trend that connects the maximum regressive shorelines of immediately lower hierarchical rank (e.g. a third-order shoreline trajectory connects the maximum regressive

Regional Geology and Tectonics

630

23. Sequence stratigraphy

Topset 2

1

RSLrise

3

Higher rank (e.g. seismic scale) shoreline trajectory Lower rank (e.g. subseismic scale) transgressions 1, 2, 3 Maximum regressive shorelines of lower hierarchical rank

FIGURE 23.27 Scale-independent stacking patterns: from deltas to shelf-slope systems. At any stratigraphic scale, the shoreline trajectory represents the trajectory of maximum regressive shorelines of immediately lower hierarchical rank. The lower rank shorelines transgress across the higher rank topset, affecting the preservation of continental deposits. The prevalent depositional system within the topset (continental vs marine) depends on the balance between continental aggradation and the subsequent ravinement erosion and marine aggradation during the higher frequency transgressions. Depending on the scale of observation, the shoreline transit time across the topset varies from diurnal (e.g. in the case of intertidal delta plains) to 104 105 yrs (e.g. in the case of continental shelves). In this context, a delta is a small-scale analogue of a siliciclastic shelf-slope system, whereby the progradation of clinoforms is enhanced during stages of lowstand in relative sea level, when the clinoform rollovers become subaerial and river-borne sediment is delivered directly to the clinoform surface. At the largest scale of observation, the trajectory of a shelf edge represents a first-order shoreline trajectory that separates a first-order topset (i.e. shelf setting) from a first-order foreset (i.e. slope setting). RSL relative sea level.

Fluvial system Coastline, older updip

Fluvial system

Coastline, younger downdip

Lacustine system

Lacustine system

1175 m

FIGURE 23.28 Diachronous limit between a lacustrine system (Pozo D-129 Formation) and a fluvial system (Castillo Formation), associated with the progradation of the shoreline during normal regression (Lower Cretaceous, Golfo San Jorge Basin, Argentina). The shingling of the seismic reflections is caused by the higher frequency transgressions that occurred during the long-term progradation of the shoreline. Source: Image courtesy of YPF Argentina.

shorelines of the fourth-order cycles; Fig. 23.27). Therefore, with the exception of the highest frequency shoreline shifts recorded at sedimentological scales (e.g. during tidal cycles), shoreline trajectories are trends defined by composite rather than single physical surfaces, observed at different stratigraphic scales (Figs 23.27 and 23.28). For example, the shoreline trajectory of a deltaic system can be mapped at the limit between topset and foreset, which connects the maximum regressive shorelines of lower hierarchical rank (Fig. 23.27). At the smallest stratigraphic scale, these maximum regressive shorelines are represented by the shoreline positions at low tide (i.e. the lowstand shorelines of the tidal cycles: the limit between intertidal and subtidal environments, which is the subaerial clinoform rollover of the lowest hierarchical rank; Fig. 23.27). At the opposite end of the spectrum, the trajectory of a shelf edge represents a first-order shoreline trajectory, which connects the lowstand shorelines of the second-order cycles (i.e. the subaerial clinoform rollovers of the highest hierarchical rank; Fig. 23.29; Catuneanu, 2019a,b). The transit area of the shoreline during higher frequency transgressions and regressions is located updip from the shoreline trajectory of higher hierarchical rank (Figs 23.27 and 23.29). The updip extent of this transit area is a function of topographic gradients, sediment supply, and the magnitude of relative sea-level changes (e.g. at the smallest scale of observation, the updip extent of the transit area within a deltaic environment depends primarily

Regional Geology and Tectonics

631

Sequence stratigraphic framework

Lowstand in RSL

Shallow water HST Continental

FSST

Shallow water

LST HST

Continental

FSST TST

LST Deepwater TST

Shoreline trajectory Shelf-edge trajectory Subaerial unconformity Correlative conformity Maximum regressive surface Maximum flooding surface Basal surface of forced regression

MarineTST (healing-phase wedge) Transgressive surface of erosion Continental-coastal TST Shoreline transit area within a higher rank topset

FIGURE 23.29 Shoreline versus shelf-edge trajectories in siliciclastic settings. Shoreline trajectories can be observed at different scales. At each scale of observation, the shoreline shifts across a transit area that is located updip of the shoreline trajectory of immediately higher hierarchical rank. At the largest scale of observation, the trajectory of the shelf edge is the first-order shoreline trajectory that marks the downdip limit of the second-order shoreline transit area. At each scale of observation, the width of the shoreline transit area depends on the gradient of the shelf, sediment supply, and the magnitude of RSL changes. Vertical scale: 102 103 m. LST lowstand systems tract; TST transgressive systems tract; HST highstand systems tract; FSST falling-stage systems tract; RSL relative sea level.

on the gradient of the delta plain and the tidal range). The transit time of the shoreline across the higher rank topsets varies with the scale of observation, from diurnal (tidal cycles, in the case of intertidal delta plains) to Milankovitch scales (104 105 yrs, in the case of continental shelves; Burgess and Hovius, 1998; Porebski and Steel, 2006). High-frequency sequences within higher rank topsets have been documented at timescales of 102 105 yrs, and thickness scales of 100 101 m (e.g. Pellegrini et al., 2017, 2018: 102 103 yrs, and 100 101 m; Nanson et al., 2013: 103 yrs, and 100 101 m; Ainsworth et al., 2017: 104 105 yrs, and 100 101 m). In upstream-controlled settings, the timescales involved in the formation of systems tracts depend on the rates of sedimentation of the depositional elements that define the stratigraphic stacking patterns, and the period of time over which the dominance of the diagnostic depositional elements can be maintained. In fluvial settings, the deposition of channel and overbank elements involves sedimentation rates of 1021 102 m/ka and minimum timescales of 102 yrs (Bridge and Leeder, 1979; Miall, 2015). This sets the lower limit of the temporal range required to form a fluvial stacking pattern and associated systems tract. How long this process can be sustained for (e.g. up to 105 yrs, 106 yrs, or even longer) depends on all variables that control sedimentation patterns in fluvial systems (i.e. the rates of floodplain aggradation, the frequency of avulsion, and the rates of lateral channel migration; Bristow and Best, 1993), including accommodation (i.e. subsidence rates in upstream-controlled settings), climate, source-area tectonism, and autogenic processes (Catuneanu, 2017, 2019a). A comparison between the sedimentation rates of fluvial systems (i.e. 1021 102 m/ka) and the average subsidence rates that operate at different timescales suggests that fluvial stacking patterns are most likely to develop within a 102 105 yrs time frame, which corresponds to the scale of high-frequency sequences (Miall, 2015). Within this time frame, the average subsidence rates in most tectonic settings match the typical rates of sedimentation of fluvial architectural elements, therefore providing suitable conditions for the development of unconventional systems tracts. In contrast, average subsidence rates on larger timescales (i.e. 106 yrs and longer) are orders of magnitude lower than the typical rates of sedimentation of the elements of the fluvial systems tracts (Miall, 2015). Nevertheless, fluvial sequences and component systems tracts are documented over a wide range of timescales, from short term (105 yrs and under; Fig. 23.30) to long term (106 yrs and over; Fig. 23.31). This indicates that controls other than subsidence need to be accounted for in order to explain sedimentation rates higher than the rates of subsidence on timescales of 106 yrs and longer. Indeed, subsidence is not the only control on sedimentation, and other factors may play a significant role in building the stratigraphic architecture. The shift from underfilled to overfilled stages in the evolution of a sedimentary basin indicates that upstream controls such as climate and/or source-area uplift can sustain long-term aggradation at rates higher than the rates of subsidence. In such cases, fluvial systems tracts can develop over timescales of 106 yrs or longer, independently of the average subsidence rates recorded during that time interval (e.g. Fig. 23.31).

Regional Geology and Tectonics

632

23. Sequence stratigraphy

GR 90 150 (API)

LLD 100

2 (ohm.m)

LAST

200 m

HAST

Depositional sequence

HAST

Amalgamated channel fills Dominantly overbank facies Subaerial unconformity

LAST HAST

Stratal stacking patterns of a fluvial succession in an overfilled basin, describing cyclicity at timescales of 105 yrs (Miocene, Assam Basin, India). Sedimentation rates were in a range of 100 101 m/ka. HAST high-amalgamation systems tract; LAST lowamalgamation systems tract. Source: Data courtesy of the Oil and Natural Gas Corporation, India.

FIGURE 23.30

Systems tracts in downstream-controlled settings In downstream-controlled settings, systems tracts form in relation to specific types of shoreline trajectory that define conventional stratal stacking patterns (Figs 23.25, 23.32 and 23.33). Conventional sequence stratigraphic frameworks may include the entire array of sequence stratigraphic surfaces and depositional systems (Fig. 23.34). Conventional systems tracts may be observed at different scales, depending on the resolution of the data available, from seismic scales (higher hierarchical ranks; Fig. 23.6), to the higher resolution scales afforded by welllog, core, and outcrop data (lower hierarchical ranks; Figs 23.8, 23.22 and 23.35). At any scale of observation (i.e. hierarchical level), sequences may consist of variable combinations of systems tracts, depending on the local conditions of accommodation and sedimentation (Csato and Catuneanu, 2012, 2014; Fig. 23.36). Therefore, systems tract successions are basin- or even subbasin-specific, and may or may not conform to the prediction of idealized models. For this reason, the construction of sequence stratigraphic frameworks of systems tracts and bounding surfaces needs to be performed case by case, based on local data rather than model assumptions. Falling-stage systems tract The falling-stage systems tract is defined by a forced regressive stacking pattern (Figs 23.32, 23.33, 23.37 and 23.38). Where the subaerial unconformity forms during forced regression, the falling-stage systems tract is bounded at the base by a marine basal surface of forced regression (i.e. the palaeo-seafloor at the onset of forced regression), and at the top by the subaerial unconformity and its correlative conformity (Fig. 23.34A). In this case, with the exception of floodplain terraces and lateral accretion macroforms that may accumulate and be preserved during forced regression in the fluvial environment, the falling-stage systems tract consists solely of marine deposits. Less commonly, forced regression may be accompanied by fluvial aggradation (Figs 23.34B and 23.39). In this case, the falling-stage systems tract is bounded at the base by a basal surface of forced regression with both continental and marine portions, and at the top by a conformity that marks the change in stacking pattern to the overlying lowstand systems tract (Fig. 23.34B). Forced regressions with fluvial topsets can be separated from normal regressions on the basis of several field criteria, including the downstepping versus upstepping of shoreface facies in a downdip direction, the compressed versus expanded development of shoreface successions, and the sharp-based versus gradationally based nature of shoreface profiles, respectively (Fig. 23.40). In areas where the

Regional Geology and Tectonics

Polarity chrons 26r

28n

Depositional sequence

LAST

28r

10 m 29n

K-T boundary

Amalgamated channel fills

HAST

29r

Dominantly overbank facies Subaerial unconformity Coal beds

CLAY

SILT

30n Vf F M C Vc

Sand

FIGURE 23.31 Stratal stacking patterns of a fluvial succession in an overfilled basin, describing cyclicity at timescales of 106 yrs (c. 3 My: 63 66 Ma, Alberta foredeep; Catuneanu and Sweet, 1999; Khidir and Catuneanu, 2003). This composite profile portrays the distal portion of a depositional sequence (referred to as Scollard, Coalspur or Willow Creek, depending on location), whose thickness increases to over 1000 m in the depocenter. Sedimentation rates were in a range of 1021 100 m/ka. The rates of accumulation of sequences and systems tracts in overfilled basins may outpace the rates of subsidence at syn-depositional time (see text for details). HAST high-amalgamation systems tract; LAST low-amalgamation systems tract.

Landward

Seaward Upstepping

RSL rise (+A)

NR

T Backstepping

RSL fall (–A)

Forestepping

No stratal stacking pattern

FR

Downstepping

FIGURE 23.32 Shoreline trajectories, as defined by combinations of lateral (forestepping, backstepping) and vertical (upstepping, downstepping) shoreline shifts. The stratal stacking patterns that define ‘conventional’ systems tracts in downstream-controlled settings are linked to shoreline trajectories: normal regression (forestepping and upstepping), forced regression (forestepping and downstepping), and transgression (backstepping and upstepping). All combinations are common in nature, except for transgression during relative sea-level fall. NR normal regression; FR forced regression; T transgression; RSL relative sea level; 1 A positive accommodation; 2 A negative accommodation.

634

23. Sequence stratigraphy

Systems tracts

Geometrical trends of the shoreline (i.e. shoreline trajectories)

Depositional trends at the shoreline (i.e. coastal processes)

HST

Forestepping and upstepping (NR following T)

Aggradation to progradation (convex-up shoreline trajectory)

Retrogradation with aggradation* (SU forms during FR)

TST

LST

FSST

Backstepping and upstepping (T)

Forestepping and upstepping (NR following FR)

Forestepping and downstepping (FR)

Retrogradation with degradation** (SU forms during T)

Progradation to aggradation (concave-up shoreline trajectory)

Progradation with degradation* (SU forms during FR) Progradation with aggradation** (SU forms during T)

FIGURE 23.33

Definition of systems tracts in downstream-controlled settings. The defining (diagnostic) criteria are provided by the geometrical trends of the shoreline (i.e. shoreline trajectories). Depositional trends may or may not follow the geometrical trends. At the shoreline, the depositional trends of progradation and retrogradation coincide with the geometrical trends of forestepping and backstepping, respectively; therefore, these terms can be used interchangeably. In contrast, the depositional trends of aggradation and degradation (controlled by changes in base level) may or may not follow the geometrical trends of upstepping and downstepping (controlled by changes in relative sea level). Irrespective of the timing of the subaerial unconformity (during forced regression or transgression), field criteria are available to separate forced from normal regressions, as well as lowstand from highstand normal regressions (see text for details). Occurrence:  , common;  , uncommon. FR forced regression; NR normal regression; T transgression; FSST falling-stage systems tract; LST lowstand systems tract; TST transgressive systems tract; HST highstand systems tract; SU subaerial unconformity.

overlying lowstand systems tract is missing, the conformity at the top of the falling-stage systems tract is replaced by surfaces associated with subsequent transgression (i.e. maximum regressive surface, transgressive surface of erosion, or subaerial unconformity). Where two or more sequence stratigraphic surfaces are superimposed, due to nondeposition or erosion, the name of the younger surface, which leaves the last imprint on the preserved contact, is typically used (Catuneanu, 2006). Lowstand systems tract The lowstand systems tract is defined by a normal regressive stacking pattern (Figs 23.32, 23.37 and 23.41) that follows a forced regression of the same hierarchical rank (Figs 23.34, 23.33 and 23.38). The lowstand systems tract is bounded at the base by the subaerial unconformity and/or the correlative conformity (Fig. 23.34). Where the lowstand systems tract is followed by transgression, the upper boundary is represented by the maximum regressive surface reworked in part by the transgressive surface of erosion (transgression accompanied by fluvial aggradation; Fig. 23.34A), or by a composite surface that includes the marine portion of the maximum regressive surface, the transgressive surface of erosion, and the subaerial unconformity (transgression accompanied by fluvial erosion; Fig. 23.34B). Where the lowstand systems tract is followed by forced regression, the upper boundary is represented by the subaerial unconformity and/or the basal surface of forced regression. Lowstand systems tracts typically include a continental topset and a marine foreset and bottomset, and tend to display a concave-up shoreline trajectory (Fig. 23.42). In the most common scenario, where the subaerial unconformity forms during forced regression (Fig. 23.34A), the lowstand topset includes the highest energy fluvial systems of a depositional sequence. In the less common cases where the subaerial unconformity forms during transgression (Fig. 23.34B), the lowstand topset includes the lowest energy fluvial systems of a depositional sequence. Transgressive systems tract The transgressive systems tract is defined by a retrogradational stratal stacking pattern (Figs 23.32, 23.33, 23.37 and 23.38). The transgressive systems tract is bounded at the base by the maximum regressive surface reworked

Regional Geology and Tectonics

(A) Timing of SU: during forced regression

(B) Timing of SU: during transgression

FSST to

pset

MFS TSE MRS CC SU RSME BSFR

TST LST FSST HST

Nonmarine Marine healing phase Nonmarine topset Marine Sharp-based shoreface Shelf Nonmarine topset Marine

Coastal onlap

FIGURE 23.34 Architecture of systems tracts and sequence stratigraphic surfaces in a shelf setting. (A) Forced regressive and transgressive shoreline trajectories are steeper than the fluvial profile, leading to the formation of the subaerial unconformity during forced regression. This scenario is most common and involves the dominance of downstream controls on the fluvial portion of systems tracts. In this case, the depositional trends of aggradation and degradation in fluvial to coastal environments follow the geometrical trends of shoreline upstepping and downstepping. (B) Fluvial profiles are steeper than the forced regressive and transgressive shoreline trajectories, leading to the formation of the subaerial unconformity during transgression. This scenario is less common and involves the dominance of upstream controls all the way to the shoreline. In this case, the entire stage of regression records progradation and aggradation in fluvial to coastal environments. However, criteria to separate forced from normal regressions are still available (see text for details). The strike variability in subsidence and sedimentation rates along the shoreline may result in the coeval development of different systems tracts between different areas of the same sedimentary basin (e.g. Catuneanu et al., 1998a, 1999, 2002). BSFR basal surface of forced regression; RSME regressive surface of marine erosion; SU subaerial unconformity; CC correlative conformity; MRS maximum regressive surface; TSE transgressive surface of erosion; MFS maximum flooding surface; HST highstand systems tract; FSST falling-stage systems tract; LST lowstand systems tract; TST transgressive systems tract. The TSE depicted in this diagram is a wave-ravinement surface, onlapped by fully marine transgressive ‘healing-phase’ deposits.

HST

TST

MFS MRS

LST

CC

FSST

FIGURE 23.35

Subseismic scale systems tracts in a shallow-water lacustrine setting (Lower Cretaceous, Araripe Basin, Brazil). In this example, the falling-stage systems tract consists of evaporites (gypsum), whereas the lowstand, transgressive and highstand systems tracts reflect the progradation (coarsening upward; red triangle) or the retrogradation (fining upward; blue triangle) of a mixed siliciclasticcarbonate shallow-water system. Climate exerted a dominant control on accommodation and sedimentation, with arid periods leading to evaporation (lake-level fall) and low terrigenous sediment influx and humid periods leading to precipitation (lake-level rise) and higher terrigenous influx into the lake. CC correlative conformity; MRS maximum regressive surface; MFS maximum flooding surface; FSST falling-stage systems tract; LST lowstand systems tract; TST transgressive systems tract; HST highstand systems tract.

Systems tracts

Conditions of development

FSST – LST – TST – HST

• Accommodation cycles: negative – positive • Sedimentation within the range of accommodation

FSST – TST – HST

• Accommodation cycles: negative – positive • Sedimentation within the range of accommodation • Accommodation starts with high rates (no LST)

FSST – TST

• Accommodation cycles: negative – positive • Accommodation > sedimentation at all times (no NR)

FSST – LST

• Accommodation cycles: negative – positive • Sedimentation > accommodation (no transgression)

TST – HST

• Positive accommodation only (no falling stage) • Sedimentation within the range of accommodation

FIGURE 23.36 Examples of systems tract combinations that can define sequences. There is no correlation between scale and the systems tract composition of sequences. Both small- and large-scale sequences may consist of similar combinations of systems tracts, and sequences of similar scales may vary in terms of number and types of component systems tracts. The scales and the systems tract composition of sequences are basinspecific, reflecting the local conditions of accommodation and sedimentation. FSST falling-stage systems tract; LST lowstand systems tract; TST transgressive systems tract; HST highstand systems tract; NR normal regression (lowstand and highstand systems tracts). Forced regression

Basinward

Offlap RSL fall

Stacking pattern: forestepping with downstepping Interpretation: progradation driven by relative sealevel fall (negative accommodation). The coastline is forced to regress, irrespective of sediment supply.

Subaerial unconformity Normal regression Topset

Stacking pattern: forestepping with upstepping RSL rise

Interpretation: progradation driven by sediment supply. Sedimentation rates outpace the rates of relative sealevel rise (positive accommodation) at the coastline. Shoreline trajectory

Transgression Stacking pattern: backstepping and upstepping RSL rise

Interpretation: retrogradation and upstepping driven by relative sea-level rise. Accommodation outpaces the sedimentation rates at the coastline. Transgressive surface of erosion

FIGURE 23.37

Stratal stacking patterns in downstream-controlled settings: forced regression, normal regression, and transgression. The stacking patterns illustrated here include the most common depositional trends; that is, fluvial bypass or erosion during forced regression, and fluvial aggradation during transgression. Exceptions from these trends are possible (see text for details). The amounts of upstepping of the coastline during normal regression and transgression, and downstepping of the coastline during forced regression, reflect the magnitude of relative sea-level changes at syn-depositional time. Source: From Catuneanu, O., Galloway, W.E., Kendall, C.G.St.C., Miall, A.D., Posamentier, H. W., Strasser, A., et al., 2011. Sequence stratigraphy: methodology and nomenclature. Newsl. Stratigr. 44 (3), 173 245.

HST

LST

HIATUS (SU)

TST

Fluvial onlap

Downlap

FSST

HST HIATUS (SU)

Condensed section

TST

Fluvial onlap

Condensed section

LST

Offlap

Condensed section

FSST HST

Condensed section

Basinward

FIGURE 23.38 Stratal stacking patterns in downstream-controlled settings, in a time domain (dip-oriented section, shelf setting). The degree of preservation of the sedimentary record increases in a downdip direction, from continental to marine systems. The lowstands and highstands in relative sea level can be reconstructed in a model-independent manner, on the basis of stratigraphic relationships: the LST is a normal regression that follows a forced regression; the HST is a normal regression that follows a transgression (Catuneanu, 2017, 2019a). FSST falling-stage systems tract; LST lowstand systems tract; TST transgressive systems tract; HST highstand systems tract.

637

Sequence stratigraphic framework

1

2

RSL fall (–A)

3

Depositional surface at the onset of FR (i.e. BSFR) Shoreline trajectory New fluvial profile during FR New seafloor profile during FR 1. Gradient of shoreline trajectory > landscape gradient: fluvial erosion during FR (most common) 2. Gradient of shoreline trajectory = landscape gradient: fluvial bypass during FR 3. Gradient of shoreline trajectory < landscape gradient: fluvial aggradation during FR

FIGURE 23.39 Depositional trends during forced regression. Diagnostic to forced regression is the combination of progradation and downstepping of the shoreline. Nondiagnostic depositional trends include fluvial processes of erosion (case 1), bypass (case 2) or aggradation (case 3), which depend on the gradient of the shoreline trajectory relative to the landscape gradient. FR forced regression; BSFR basal surface of forced regression; RSL relative sea level; 2 A negative accommodation. Source: Modified from Catuneanu, O., Zecchin, M., 2016. Unique vs. non-unique stratal geometries: relevance to sequence stratigraphy. Mar. Pet. Geol. 78, 184 195.

Atypical forced regression

Typical normal regression

Topset Topset RSL fall

RSL rise

RSME

Shoreline trajectory Downstepping upper shoreface facies Lower shoreface (sharp-based)

Shelf

Shoreline trajectory Upstepping upper shoreface facies Lower shoreface (gradationally based)

FIGURE 23.40 Criteria to separate atypical forced regressions with fluvial topsets from normal regressions that preserve their fluvial topsets. Forced regressions are characterized by downstepping and sharp-based shoreface facies and true downlap of shoreface clinoforms against the RSME. Normal regressions are characterized by upstepping and gradationally based shoreface facies, and apparent downlap (i.e. seismic artefact) of shoreface clinoforms against the shelf bottomset facies. The fall in relative sea level during forced regression also results in the deposition of ‘compressed’ shorefaces, with thicknesses limited by the depth of the fairweather wave base. In contrast, the rise in relative sea level during normal regression affords the development of ‘expanded’ shorefaces, with thicknesses that can exceed the depth of the fairweather wave base. RSL relative sea level; RSME regressive surface of marine erosion.

RSL rise (+A)

Depositional surface at the onset of NR Shoreline trajectory New fluvial profile during NR New sea-floor profile during NR

FIGURE 23.41 Depositional trends during normal regression. Diagnostic to normal regression is the combination of progradation and upstepping of the shoreline. This shoreline trajectory leads to a lowering of the fluvial gradient with time, which results in the aggradation of a fluvial topset. NR normal regression; RSL relative sea level; 1 A positive accommodation. Source: From Catuneanu, O., Zecchin, M., 2016. Unique vs. non-unique stratal geometries: relevance to sequence stratigraphy. Mar. Pet. Geol. 78, 184 195.

in part by the transgressive surface of erosion, and at the top by the maximum flooding surface (Fig. 23.34). Where transgression is accompanied by fluvial aggradation (Fig. 23.43), the fluvial portion of the transgressive systems tract may also rest directly on top of the subaerial unconformity, beyond the updip termination of the lowstand topset (Fig. 23.34A). Most commonly, transgressive systems tracts record the highest rates of fluvial aggradation (Shanley and McCabe, 1993; Wright and Marriott, 1993; Emery and Myers, 1996; Kerr et al., 1999),

Regional Geology and Tectonics

638

23. Sequence stratigraphy

Lowstand normal regression (accelerating RSLrise) RSL

Topset

The rates of progradation decrease with time, the rates of aggradation increase with time. Alowstand normal regression follows a forced regression of equal hierarchical rank.

Shoreline trajectory (concave up)

Highstand normal regression (decelerating RSLrise) Topset

RSL

The rates of progradation increase with time, the rates of aggradation decrease with time. Ahighstand normal regression follows a transgression of equal hierarchical rank.

Shoreline trajectory (convex up)

FIGURE 23.42 Architecture of lowstand versus highstand normal regressions. In both cases, progradation is driven by sediment supply during a period of relative sea-level rise (i.e. sedimentation outpaces accommodation at the shoreline). A lowstand normal regression records a change in depositional trend from dominantly progradational to dominantly aggradational (concave-up shoreline trajectory), whereas a highstand normal regression records a change in depositional trend from dominantly aggradational to dominantly progradational (convex-up shoreline trajectory). RSL relative sea level. Source: Modified from Catuneanu, O., 2006. Principles of Sequence Stratigraphy. Elsevier, Amsterdam, 375 pp (Figure 7.20, p. 306).

3

2

1

RSL rise (+A)

Depositional surface at the onset of T (i.e., MRS) Shoreline trajectory New fluvial profile during T New sea-floor profile during T (i.e. WRS) 1. Gradient of shoreline trajectory > landscape gradient: fluvial aggradation during T (most common) 2. Gradient of shoreline trajectory = landscape gradient: fluvial bypass during T SU forms during T 3. Gradient of shoreline trajectory < landscape gradient: fluvial erosion during T

FIGURE 23.43 Depositional trends during transgression. Diagnostic to transgression is the combination of retrogradation and upstepping of the shoreline. Nondiagnostic depositional trends include fluvial processes of aggradation (case 1), bypass (case 2) or erosion (case 3), which depend on the gradient of the shoreline trajectory relative to the landscape gradient. T transgression; MRS maximum regressive surface; WRS wave-ravinement surface; SU subaerial unconformity; RSL relative sea level; 1 A positive accommodation. Source: Modified from Catuneanu, O., Zecchin, M., 2016. Unique vs. non-unique stratal geometries: relevance to sequence stratigraphy. Mar. Pet. Geol. 78, 184 195.

leading to the lowest rates of river-borne sediment supply to the marine environment. This results in the sediment starvation of the seafloor and the formation of marine condensed sections, which are most often associated with transgression (Loutit et al., 1988). Where transgression is accompanied by fluvial erosion (Figs 23.34B, 23.43 and 23.44), the rates of river-borne sediment supply to the shoreline are the highest of the entire stratigraphic cycle. However, this does not necessarily translate into high sediment supply to the shelf edge, particularly in the case of wider shelves that can retain the sediment within the shallow-water systems. The efficiency and the mechanisms of sediment transfer from the shoreline to the shelf edge need to be assessed on a case-by-case basis, and depend on a variety of factors including the physiography of the basin (e.g. shelf width), the presence or absence of unfilled incised valleys

Regional Geology and Tectonics

Sequence stratigraphic framework

639 FIGURE 23.44 Fluvial incision and coastal erosion during transgression (Canterbury Plain, New Zealand). These processes are promoted by a landscape gradient that is steeper than the shoreline trajectory, due to the proximity of the provenance (i.e. the Southern Alps of New Zealand) to the shoreline. In this case, the upstream controls (i.e. source-area uplift) may dominate fluvial processes all the way to the shoreline.

across the shelf that could serve as conduits of sediment transport, the magnitude of storms (i.e. depth of the storm wave base), and the tidal range. River-borne sediment supply to the deepwater environment tends to remain lowest at the time of maximum flooding, despite the high supply to the shoreline, due to the location of the shoreline at the greatest distance from the shelf edge at the end of transgression (Fig. 23.45). In this scenario, the shoreline at the end of transgression marks the updip limit of the transgressive systems tract. Highstand systems tract The highstand systems tract is defined by a normal regressive stacking pattern (Figs 23.32, 23.37 and 23.41) that follows a transgression of the same hierarchical rank (Figs 23.33, 23.34 and 23.38). The highstand systems tract is bounded at the base by a maximum flooding surface with both continental and marine portions (transgression accompanied by fluvial aggradation; Fig. 23.34A), or by a marine maximum flooding surface and a subaerial unconformity (transgression accompanied by fluvial erosion; Fig. 23.34B). Where the highstand systems tract is followed by forced regression, the upper boundary is represented by a subaerial unconformity and/or the basal surface of forced regression (Fig. 23.34). Where the highstand systems tract is followed by transgression, the upper boundary is represented by the maximum regressive surface reworked in part by the transgressive surface of erosion (transgression accompanied by fluvial aggradation), or by a composite surface that includes the marine portion of the maximum regressive surface, the transgressive surface of erosion, and the subaerial unconformity (transgression accompanied by fluvial erosion). Highstand systems tracts typically include a continental topset and a marine foreset and bottomset, and tend to display a convex-up shoreline trajectory (Fig. 23.42). In the most common scenario, where the subaerial unconformity forms during forced regression (Fig. 23.34A), the highstand topset includes the lowest energy fluvial systems of a depositional sequence. In the less common cases where the subaerial unconformity forms during transgression (Fig. 23.34B), the highstand topset includes the highest energy fluvial systems of a depositional sequence.

Systems tracts in upstream-controlled settings In upstream-controlled settings, systems tracts form independently of relative sea/lake-level changes and shoreline shifts (Figs 23.25 and 23.46) and reflect the combined influence of accommodation, climate, source-area uplift and autogenic processes on sedimentation (Fig. 23.10). Upstream-controlled systems tracts may also be observed at different scales (Fig. 23.20), and are interpreted on the basis of stratal stacking patterns defined by the dominant depositional elements. In fluvial settings, the upstream-controlled stacking patterns are defined by the high versus low channel-to-overbank ratio (Figs 23.20 and 23.46 23.48). The nomenclature of ‘unconventional’ systems tracts is no longer tied to shoreline trajectories or the position of the relative sea level, nor to the inferred accommodation conditions at syn-depositional time (Catuneanu, 2017). Accommodation generated by basin subsidence is, in most cases, already overfilled during the development of upstream-controlled depositional sequences and component systems tracts (Fig. 23.11). While accommodation (i.e. rates of subsidence) may still play an important role, fluvial processes, including the rates of

Regional Geology and Tectonics

640

23. Sequence stratigraphy

Shelf edge

Shoreline Bathymetry Highstand normal regression

>SWB

Deep water Trajectory

Processes

Hemipelagic

Aggradation H

MFS >SWB Transgression 0 – SWB

Hemipelagic, Mudflows, slumps Low-density turbidites

MFS: highest biostratigraphic abundance and diversity

Aggradation or starvation Retrogradation (collapse) Aggradation

M, S

MRS Lowstand normal regression

MRS: cryptic LDT, G

0

Low-density turbidites, Grainflows

Progradation to aggradation

0 SWB – 0 >SWB

Grainflows High-density turbidites Mudflows, slumps Hemipelagic

Progradation with downstepping Aggradation to progradation Retrogradation (collapse) Aggradation

CC Forced regression

M, S

BSFR Highstand normal regression

>SWB

Hemipelagic

Aggradation

CC: shift in grain size

G, HDT

H

Erosional surface reworking and replacing the BSFR (base of submarine fan)

FIGURE 23.45

Shoreline versus shelf-edge trajectories and corresponding processes in the deepwater setting, in the case of highmagnitude changes in relative sea level (i.e. relative sea level below the elevation of the shelf edge at the end of forced regression). The shoreline is at the shelf edge during late forced regression and lowstand normal regression and updip from the shelf edge at all other times. Where forced regression is accompanied by the formation of the subaerial unconformity, offlap may be observed both on the shelf and at the shelf edge. Common mechanisms of sand delivery to the deepwater setting involve storm currents (shelf edge above SWB), shelf-edge deltas, and the collapse of coastal systems located at the shelf edge. Unfilled incised valleys across a submerged shelf may enhance the transfer of sediment from river mouths to the shelf edge even during stages of highstand in relative sea level. In such cases, as well as in the case of atypically narrow shelves, sediment supply (both volume and grain size) to the deepwater setting may increase overall; however, the trends that describe the relative changes in grain size during a stratigraphic cycle are still maintained. The deepwater processes in this diagram illustrate trends observed in the rock record (e.g. Posamentier and Kolla, 2003; van der Merwe et al., 2010; De Gasperi and Catuneanu, 2014). Deviations from these trends are possible (e.g. discussions in Catuneanu et al., 2009, 2011), which is why sequence stratigraphic frameworks are constructed on a case-by-case basis, guided by field observations rather than model assumptions. Ultimately, sediment supply to the deepwater setting depends on multiple variables, including the production of extrabasinal and intrabasinal sediment, and the efficiency of sediment transfer from fluvial systems to the shoreline (i.e. inversely proportional to the rates of fluvial aggradation) and from the shoreline to the shelf edge (i.e. inversely proportional to the capacity of shallow-water systems to retain sediment, which is lowest during forced regression, and to the width of the submerged shelf). BSFR basal surface of forced regression; CC correlative conformity; MRS maximum regressive surface; MFS maximum flooding surface; SWB storm wave base; H hemipelagic sediment; M mudflows; S slumps; HDT high-density turbidity flows; G grainflows; LDT low-density turbidity flows.

High CH/FF ratio: amalgamated channels

Low CH/FF ratio: isolated channels

Aggradation Graded profile

Degradation

Erosion (channel belt)

Incised valley Subaerial unconformity

Pedogenesis (interfluve area)

Channel fill (CH) Floodplain (FF)

FIGURE 23.46 Depositional trends in upstream-controlled settings. ‘Unconventional’ stratal stacking patterns depend on (1) the rates of floodplain aggradation (proportional to the size of arrows in the diagram); (2) the ability of channels to shift laterally, which is a function of fluvial style; and (3) the frequency of avulsion. The processes and the rates of aggradation and degradation depend on all factors that control sedimentation (i.e. sediment supply vs energy flux), including accommodation, climate, source-area tectonism, and the autogenic controls on sediment dispersal patterns. The aggrading side of the diagram illustrates the seven-step evolution of a channel under identical conditions of avulsion and lateral shift. In this case, the difference in stratigraphic architecture (i.e. amalgamated vs isolated channels) is the result of differences in the rates of floodplain aggradation.

Regional Geology and Tectonics

641

Sequence stratigraphic framework

cy

n

io

ls

vu fa

High

H ig h

(2)

o

Channel fill (CH) Floodplain (FF)

n ue

q re

Low

F w

(3)

(1)

Lo

High

Low (unconfined channels)

Channel fill (CH) Floodplain (FF)

Low

Rate of ff aggradation

High (confined channels)

High (confined channels)

Low (unconfined channels)

Degree of CH confinement

FIGURE 23.47 Fluvial architecture under variable conditions of floodplain aggradation, channel confinement, and avulsion frequency, as illustrated by the seven-step evolution of a channel. The rates of fluvial aggradation depend on all factors that control sedimentation, including accommodation, climate, source-area tectonism, and autocyclic changes in sediment distribution. The preservation of isolated channels is favoured by rapid aggradation coupled with frequent avulsion. The degree of channel amalgamation is proportional to the rate of lateral channel migration (higher in unconfined systems) and the frequency of avulsion (proportional to sediment supply), and inversely proportional to the rate of aggradation. (1), (2), and (3) indicate the most common succession of stacking patterns, from the base to the top of a fluvial depositional sequence. All these common stacking patterns assume avulsion, but are different in terms of rates of aggradation and/or lateral channel migration. Source: Modified from Bristow, C.S., Best, J.L., 1993. Braided rivers: perspectives and problems. In: Best, J.L., Bristow, C.S. (Eds.), Braided Rivers. Geological Society Special Publication No. 75, pp. 1 11

(A)

(B)

FIGURE 23.48 Fluvial stacking patterns in an upstream-controlled setting (Triassic, Karoo Basin). A Channel-dominated succession: high-amalgamation stacking pattern (Molteno Formation: higher energy, coarser-grained braided channels). B Floodplain-dominated succession: low-amalgamation stacking pattern (Burgersdorp Formation: lower energy, finer-grained meandering river system). The arrows mark the position of the subaerial unconformity that separates the two stacking patterns.

Regional Geology and Tectonics

642

23. Sequence stratigraphy

floodplain aggradation, are also influenced by all other controls that modify the balance between sediment supply and energy flux at any location (i.e. climate, source-area uplift, and autocyclicity; Fig. 23.10). Over a long term, the rates of fluvial aggradation exceed the rates of subsidence, leading to the shift from underfilled to overfilled basins (Fig. 23.11). The high-amalgamation (HAST) and low-amalgamation (LAST) systems tracts (Fig. 23.25; Catuneanu, 2017) were previously defined as ‘low-accommodation’ and ‘high-accommodation’ systems tracts, respectively, based on the assumption that accommodation is the main control on the degree of channel amalgamation (e.g. Boyd et al., 2000; Leckie and Boyd, 2003). However, it is becoming evident that accommodation alone cannot explain the development of fluvial stacking patterns across all temporal scales, particularly where the rates of accumulation of depositional elements do not match the rates of creation of accommodation (Miall, 2015). Therefore, the revised systems tract terminology emphasizes the observation of stratal stacking patterns rather than the interpretation of the underlying controls (Catuneanu, 2017). The only type of sequence stratigraphic surface that is common between the downstream- and upstreamcontrolled settings is the subaerial unconformity, which marks the boundary of depositional sequences. All other ‘conventional’ sequence stratigraphic surfaces (Fig. 23.34) require a marine/lacustrine environment, and therefore, are restricted to the downstream-controlled settings. Within upstream-controlled depositional sequences, systems tracts are separated by a ‘top-HAST’ surface, which marks the change from a high to a low degree of channel amalgamation. The change in the degree of channel amalgamation from one system tract to another may or may not be accompanied by a change in fluvial style. Within each systems tract, the fluvial style may or may not change between different locations within the basin (Catuneanu and Bowker, 2001; Catuneanu and Elango, 2001). Both subaerial unconformities and top-HAST surfaces can be observed at different scales (i.e. hierarchical orders; Fig. 23.20). Systems tracts in upstream-controlled settings may be observed over a wide range of physical and temporal scales, generally within 100 102 m and 102 106 yrs. As the rates of fluvial aggradation can vary over at least four orders of magnitude (1021 102 m/ka; Bridge and Leeder, 1979; Miall, 2015), the ratio between the thickness and the timescale of systems tracts can also be highly variable (e.g. 101 m over 106 yrs vs 102 m over 105 yrs; Figs 23.30 and 23.31). High-amalgamation systems tract The high-amalgamation systems tract (formerly termed ‘low-accommodation’ systems tract) is defined by a high degree of channel amalgamation (i.e. high channel-to-overbank ratio; Figs 23.20, 23.30, 23.31 and 23.48A). The formation of high-amalgamation systems tracts is promoted by (1) low rates of floodplain aggradation; (2) unconfined fluvial channels; and (3) a high frequency of channel avulsion (Fig. 23.47; Bristow and Best, 1993). To a lesser degree, confined channels may also amalgamate under conditions of low rates of floodplain aggradation and high frequency of avulsion (Shanley and McCabe, 1993; Wright and Marriott, 1993; Emery and Myers, 1996). High-amalgamation systems tracts form commonly in relation to the highest energy river systems of a fluvial depositional sequence, and are therefore found at the base of sequences, overlying the subaerial unconformity (Figs 23.20, 23.30, 23.31, 23.49 and 23.50). The processes that lead to channel amalgamation depend on all factors that control fluvial sedimentation, including accommodation (i.e. rates of subsidence), climate, source-area uplift, and autocyclicity (Fig. 23.10). API 0

150

Lithostratigraphy Dinosaur Park formation

(1)

SU Upper siltstone unit

(2)

25 m

(1)

SU (2)

Oldman formation Comrey sandstone Taber coal zone

Foremost formation

FIGURE 23.49 Unconformity-bounded fluvial sequence in a downstream-controlled foredeep setting (Oldman Formation, Belly River Group; Upper Cretaceous, Western Canada Sedimentary Basin). (1) Amalgamated braided channels; (2) floodplain-dominated meandering system. SU subaerial unconformity; API gamma-ray scale in American Petroleum Institute units.

Regional Geology and Tectonics

643

Sequence stratigraphic framework

Stacking patterns of a typical fluvial depositional sequence:

SP

SU (3)

SU (3) (2)

}

FF dominated

(2)

Low-amalgamation stacking pattern 50 m

(1) Channel fill (CH) Floodplain (FF)

CH dominated SU

(1)

High-amalgamation stacking pattern

SU

FIGURE 23.50 Common stratigraphic architecture of a fluvial sequence, consisting of (1) a section of high-energy unconfined channels, accumulated under conditions of low rates of floodplain aggradation; (2) a section of low-energy confined channels and floodplains, accumulated under conditions of high rates of floodplain aggradation; and (3) a section of low-energy confined channels and floodplains, accumulated under conditions of low rates of floodplain aggradation. Section (3) includes the lowest energy river systems and therefore the finest channel fills of the fluvial sequence and has the lowest preservation potential due to the development of the subaerial unconformity at the top. Deviations from this common architecture are possible, with end-members represented entirely by either high-energy unconfined systems or low-energy confined systems. In the latter case, the degree of channel amalgamation may be low even at the base of the fluvial sequence. SU subaerial unconformity; SP spontaneous potential. The well-log example illustrates a fluvial succession in the Golfo San Jorge Basin, Argentina. Source: Data courtesy of YPF Argentina.

Low-amalgamation systems tract The low-amalgamation systems tract (formerly termed ‘high-accommodation’ systems tract) is defined by the dominance of floodplain deposits (i.e. low channel-to-overbank ratio; Figs 23.20, 23.30, 23.31 and 23.48B). The formation of low-amalgamation systems tracts is promoted by (1) high rates of floodplain aggradation; (2) confined fluvial channels; and (3) a low frequency of channel avulsion (Fig. 23.47; Bristow and Best, 1993). However, this stacking pattern may also form under conditions of low rates of floodplain aggradation and high frequency of avulsion, where channels display a high degree of confinement and the pattern of avulsion maintains a low channel-to-overbank ratio (Fig. 23.50). Low-amalgamation systems tracts form commonly in relation to the lower energy river systems of a fluvial sedimentation cycle, which dominates the upper part of depositional sequences (Figs 23.20, 23.30, 23.31, 23.49 and 23.50). The processes that lead to a low degree of channel amalgamation depend on all factors that control fluvial sedimentation, including accommodation (i.e. rates of subsidence), climate, source-area uplift, and autocyclicity (Fig. 23.10).

Nomenclature of systems tracts Systems tracts are the stratigraphic building blocks of sequences, which can be observed at all stratigraphic scales (Fig. 23.18). On practical grounds, the identification of systems tracts is more important than the delineation of sequences. Systems tracts afford an understanding of the sediment dispersal patterns at different stages during the development of sequences, which is most significant to the distribution of natural resources. The construction of a framework of systems tracts and bounding surfaces fulfils the practical purpose of sequence stratigraphy. Beyond this, the delineation of sequences becomes a matter of model-dependent organization of systems tracts into stratigraphic cycles (Fig. 23.3). Objectivity in the nomenclature of systems tracts requires emphasis on observational rather than interpretive criteria. As such, systems tracts in both downstream- and upstreamcontrolled settings are defined on the basis of observed stratal stacking patterns and stratigraphic relationships. All standard sequence stratigraphic models (Fig. 23.3) account for the presence of a seaway within the basin under analysis and are centred on shoreline trajectories that control the timing of systems tracts and bounding surfaces. Under these circumstances, the presence of a palaeoshoreline justifies a systems tract nomenclature that makes reference to transgressions and regressions, with the latter being classified into ‘forced’ and ‘normal’, depending on the specific stacking pattern that accompanies the shoreline regression (i.e. downstepping vs upstepping subaerial clinoform rollovers, respectively; Fig. 23.32). Shoreline trajectories are controlled by the interplay of relative sea-level changes (i.e. accommodation) and base-level changes (i.e. sedimentation) at the shoreline. Both elements of this dual control are critically important, and none is a constant during the development of stratigraphic cycles. Changes in relative sea level control the geometrical trends of shoreline downstepping (i.e. forced regressions, during relative sea-level fall) and upstepping (i.e. normal regressions or

Regional Geology and Tectonics

644

23. Sequence stratigraphy

Stratigraphic architecture Geometrical trends of the shoreline

Depositional trends at the shoreline

(i.e. shoreline trajectories)

(i.e. coastal processes)

Vertical trends: upstepping, downstepping

A

Lateral trends: forestepping, backstepping

A and S

S

Vertical trends: aggradation, degradation Lateral trends: progradation, retrogradation

FIGURE 23.51 Controls on the stratigraphic architecture: accommodation (A) and sedimentation (S). The architecture of the downstreamcontrolled stratigraphic framework may be described in terms of geometrical trends of the shoreline (i.e. shoreline trajectories) and/or in terms of depositional trends at the shoreline (i.e. coastal processes that accompany the shoreline shifts). The lateral components of the two types of trend are equivalent to each other (i.e. forestepping 5 progradation; backstepping 5 retrogradation), as they are both controlled by the interplay of relative sea-level changes (A) and base-level changes (S). However, the vertical components of the two types of trend may be offset relative to one another, as they are controlled by different processes (A vs S). Relative sea-level changes (A) control the upstepping (relative sea-level rise) or the downstepping (relative sea-level fall) of the shoreline (i.e. the vertical shifts of subaerial clinoform rollovers), whereas base-level changes (S) control the depositional trends of aggradation or degradation that accompany the shoreline shifts. While the temporary base level in fluvial and coastal environments commonly follows the change in relative sea level (i.e. degradation during shoreline downstepping, and aggradation during shoreline upstepping), exceptions may occur. Diagnostic to the definition of systems tracts are the geometrical trends that describe shoreline trajectories, irrespective of the depositional trends of aggradation or degradation that accompany the shoreline shifts.

transgressions, during relative sea-level rise; Figs 23.32 and 23.51), which affect the efficiency of sediment transfer from fluvial to deepwater systems (i.e. highest during forced regression, when the shelf capacity to retain sediment is minimum). For this reason, and notwithstanding the importance of sediment supply, the relative sea level remains a key concept in downstream-controlled settings, with impact on the stratigraphic architecture and the patterns of sediment distribution across a sedimentary basin. With emphasis on shoreline trajectories, transgression defines the transgressive systems tract, and forced regression defines the falling-stage systems tract. Any intervening stages of normal regression are classified into ‘lowstand’ (i.e. a normal regression that follows a forced regression of equal hierarchical rank) and ‘highstand’ (i.e. a normal regression that follows a transgression of equal hierarchical rank). The lowstand and highstand normal regressions define the lowstand and highstand systems tracts, respectively. It can be noted that these conventional systems tracts are defined by the observation of stratal stacking patterns and stratigraphic relationships, and account for the interplay of relative sea-level changes and sedimentation at the shoreline. The reference to the lowstand and highstand positions of relative sea level does not detract from the objective definition of the systems tracts. Ignoring the relative sea level would, in fact, be counterproductive. The identification of lowstand and highstand systems tracts is important in terms of sediment distribution on the shelf (e.g. relative locations of lowstand and highstand deltas) and sediment supply to the deepwater setting (e.g. ‘lowstand shedding’ in the case of siliciclastic settings, and ‘highstand shedding’ in the case of carbonate settings). These differences relate to the landward relocation of the shoreline during intervening stages of transgression (Fig. 23.52). Exceptions from the common trends of sediment dispersal are possible under special circumstances and need to be rationalized on a case-by-case basis (e.g. the possibility of progradation of highstand deltas to the shelf edge in the case of narrow shelves and/or high sediment supply). An alternative systems tract nomenclature emphasizes the depositional trends recorded in coastal environments during the shifts of the shoreline: that is, progradation to aggradation (‘PA’) instead of lowstand normal regression; retrogradation (‘R’) instead of transgression; aggradation to progradation (‘AP’) instead of highstand normal regression; and degradation (‘D’) instead of forced regression (Neal and Abreu, 2009; Abreu et al., 2010). While this terminology was meant to uncouple the nomenclature of systems tracts from any reference to the relative sea level, its proponents still assumed that depositional trends are controlled by accommodation, as ‘D’ is attributed to relative sea-level fall (i.e. negative accommodation), and ‘PA’, ‘R’, and ‘AP’ are attributed to stages of relative sea-level rise (i.e. positive accommodation). This assumption defeats the purpose of a nomenclatural change, as the terms ‘PA’ and lowstand, as well as ‘AP’ and highstand, become interchangeable (Neal and Abreu, 2009). The flaw of this assumption is that the depositional trends in coastal environments may not follow the geometrical trends of the shoreline (i.e. shoreline trajectories), as the vertical components of the two types of trend are controlled by different processes (i.e. base-level changes vs relative sea-level changes, respectively; Fig. 23.51).

Regional Geology and Tectonics

645

Sequence stratigraphic framework

6

7

1 2 3

4

5

TST/MFS of “n” hierarchical order TST/MFS of “n+1” hierarchical order (i.e. lower hierarchical rank) Subaerial unconformity of “n” hierarchical order

FIGURE 23.52 Stratigraphic architecture of falling-stage (timesteps 1 3), lowstand (timesteps 3 5), transgressive (timesteps 5 6) and highstand (timesteps 6 7) systems tracts, as defined by the shoreline trajectory of ‘n’ hierarchical order. A transgression of ‘n 1 1’ hierarchical order does not change the location of the shoreline observed at the ‘n’ hierarchical level. A transgression of ‘n’ hierarchical order leads to the relocation of the shoreline observed at the ‘n’ hierarchical level. This provides a criterion to separate transgressions of different hierarchical ranks, even where the transgressive systems tracts fall below the resolution of the data available. TST transgressive systems tract; MFS maximum flooding surface.

Geometrical trends of the shoreline

Progradation with downstepping

Stratal stacking patterns

Forced regression (1) with fluvial/coastal degradation

Depositional trends in fluvial/coastal systems Progradation with degradation

Forced regression (2) with fluvial/coastal aggradation Progradation with upstepping

Progradation with aggradation Normal regression

FIGURE 23.53 Architecture of forced versus normal regressions. Stratal stacking patterns may be described in terms of geometrical trends of the shoreline (i.e. shoreline trajectories) or depositional trends in fluvial to coastal systems. The terms that make reference to geometrical trends (i.e. ‘progradation with downstepping’ and ‘progradation with upstepping’) are better descriptors of systems tracts and accommodation conditions at syn-depositional time. The recognition of forced versus normal regressions may be based on facies data in high-resolution studies (e.g. to demonstrate the sharp- vs gradationally based nature of shoreface deposits) or on the observation of the stratigraphic architecture in lower resolution studies (e.g. the downstepping vs upstepping of subaerial clinoform rollovers). (1) Most common in the stratigraphic record. (2) Less common in the stratigraphic record.

The systems tract nomenclature based on coastal depositional trends attempts to be objective, by removing the reference to relative sea-level changes, while recognizing at the same time the importance of relative sea-level changes (i.e. the ‘accommodation succession method’ that describes the ExxonMobil methodology; Neal and Abreu, 2009). This contradiction is rooted in the confusion between shoreline trajectories and coastal depositional trends, which triggers several model-driven errors: it overemphasizes the role of accommodation on the development of depositional trends (e.g. it assumes that subaerial unconformities form only during stages of negative accommodation, which is not always true: Fig. 23.34); it relies on ideal shoreline trajectories (i.e. concave-up vs convex-up; Figs 23.29 and 23.42), which may or may not be observed in the rock record; and it assumes that ‘A’ (aggradation) requires creation of accommodation, which may or may not be the case (i.e. continental to coastal aggradation may occur during both stages of positive and negative accommodation; Figs 23.39, 23.41 and 23.43). The latter error hinders the distinction between forced and normal regressions, which is a significant pitfall of this approach. The original nomenclature (i.e. ‘lowstand’ and ‘highstand’) remains more accurate in terms of definition of systems tracts and separation between the deposits of negative versus positive accommodation (Figs 23.51 and 23.53). The relative sea level plays a significant role, along with sedimentation, in the formation of downstreamcontrolled systems tracts, and the distinction between lowstand and highstand normal regressions, as well as between normal regressions and forced regressions, remain critical to understand the sediment distribution

Regional Geology and Tectonics

646

23. Sequence stratigraphy

across a basin and to reconstruct the relative sea-level changes at syn-depositional time. Changes in relative sea level modify the efficiency of sediment transfer from the shelf to the deepwater setting, as well as the production of intrabasinal sediment, and therefore are important to recognize and reconstruct. The identification of lowstands and highstands in relative sea level is based on local stratigraphic relationships, and not on correlations with global sea-level changes. Global cycle charts are no longer part of the sequence stratigraphic workflow and methodology. In overfilled basins, as well as in the upstream-controlled portion of underfilled basins, the definition and nomenclature of systems tracts is based on the dominant depositional elements (e.g. channels vs floodplain deposits in fluvial settings; Figs 23.20 and 23.46). In such cases, the relative sea level no longer plays any role in the formation of systems tracts, and the temporary base level becomes the sole control on the development of unconformity-bounded depositional sequences (Catuneanu, 2019a). Accommodation is only one of the several controls that interplay to generate stacking patterns in fully continental settings (Fig. 23.10), and therefore any direct link between accommodation and the nomenclature of systems tracts should be avoided. This led to a change in nomenclature from terms that make reference to inferred accommodation conditions (i.e. low- vs highaccommodation) to terms that describe the observed stacking patterns (i.e. high- vs low-amalgamation; Catuneanu, 2017, 2019a).

Stratigraphic sequences: definition and scales Sequences are stratigraphic cycles of change in stratal stacking pattern, defined by the recurrence of the same types of sequence stratigraphic surface in the sedimentary record (Fig. 23.2). As a variety of allogenic and autogenic processes may interplay and contribute to the development of stratigraphic cyclicity (Figs 23.9 and 23.10), the definition of sequences is based on the observation of stratal stacking patterns and not on the interpreted origin of cycles (Catuneanu and Zecchin, 2013). The separation between observation and interpretation in the definition of a sequence is fundamental. Traditionally, a sequence was equated to an allocycle (e.g. an accommodation cycle; Neal and Abreu, 2009). However, this view is an oversimplification; in reality, sediment supply, which depends in part on autogenic processes, can affect the timing of formation of all main types of ‘sequence boundary’ (i.e. subaerial unconformity, maximum flooding surface, and maximum regressive surface; Catuneanu and Zecchin, 2016). The distinction between allogenic and autogenic controls on sedimentation and the development of unconformities is sometimes evident (e.g. when a relation to the relative sea level can be established), but it cannot be generalized. For example, subaerial unconformities are often assumed to be the product of negative accommodation in both downstream- and upstream-controlled settings. However, such unconformities can also form during transgression in downstream-controlled settings (i.e. interplay of positive accommodation and sedimentation; Fig. 23.34B), and in response to a variety of processes in upstream-controlled settings, some of which are unrelated to accommodation (Fig. 23.10). Stratigraphic cyclicity can be observed at different scales, depending on the purpose of the study or the resolution of the data available. The range of stratigraphic scales extends from 102 to 108 years. Schlager (2010) also recognized stratigraphic cycles at timescales of 100 101 yrs, but these periods of time become too short for the formation of geologically recognizable soil horizons as evidence of subaerial unconformities, and the accumulation of depositional elements that define depositional systems. Therefore, realistic timescales for a sequence stratigraphic analysis start with 102 103 yrs (Catuneanu, 2017, 2019a,b; Fig. 23.17). At each scale of observation (i.e. hierarchical level), sequences consist of systems tracts and component depositional systems (Fig. 23.18). A stratigraphic cycle assumes no repetition of systems tracts and sequence stratigraphic surfaces within the sequence, at the selected scale of observation (i.e. systems tracts and bounding surfaces of the hierarchical rank of the host sequence cannot be repeated within the sequence). Stratigraphic cycles may be symmetrical or asymmetrical, and the corresponding sequences may include a variable number of distinct systems tracts, up to four in the case of ideal ‘conventional’ sequences that develop and preserve all systems tracts (Figs 23.25 and 23.36). Therefore, a sequence is not defined by its internal makeup, but by the recurrence of the sequence stratigraphic surface that marks its boundaries. Not all types of stratal stacking patterns (i.e. systems tracts) and bounding sequence stratigraphic surfaces may occur in the succession under analysis. For example, stratigraphic cyclicity may be defined by the repetition of transgressions and highstand normal regressions, without intervening stages of forced regression and lowstand normal regression; or, by the repetition of forced regressions and lowstand normal regressions, without intervening stages of transgression

Regional Geology and Tectonics

Sequence stratigraphic framework

647

and highstand normal regression (Fig. 23.36). Therefore, the types of recurring stacking patterns and bounding surfaces that define stratigraphic cyclicity may vary with the case study, which underlines the need for a modelindependent approach to the sequence stratigraphic analysis. Sequences of all scales may include unconformities of equal hierarchical rank (e.g. a depositional sequence may include unconformities associated with transgression; a genetic stratigraphic sequence may include a subaerial unconformity). Therefore, sequences may or may not be relatively conformable successions, but they always consist of genetically related strata that belong to the same cycle of change in stratigraphic stacking pattern (Catuneanu et al., 2009). For example, a genetic stratigraphic sequence that includes a subaerial unconformity of equal hierarchical rank cannot be described as a ‘relatively conformable succession’ within the area of development of the unconformity, but it does consist of ‘genetically related strata’ that belong to the same stratigraphic cycle. The subaerial unconformity has a smaller geographic extent than the genetic sequence hosting it, which maintains the integrity of the sequence as one stratigraphic cycle (Galloway, 1989). Beyond the termination of the unconformity, the correlative conformity preserves the continuity of sedimentation and the genetic relationship between the underlying and the overlying strata within the sequence. Unconformities with basinwide extent become sequence boundaries by default, at the scale of the underlying and overlying sequences. Stratigraphic sequences of lower hierarchical ranks are nested within higher rank systems tracts, as illustrated by outcrop-scale sequences (100 101 m) that commonly build seismic-scale (101 102 m) systems tracts (Figs 23.18 23.20). The scale of observation is set by the resolution of the data available or by the purpose of the study (e.g. basin analysis: 102 103 m; petroleum exploration: 101 102 m; petroleum production development: 100 101 m; cyclostratigraphy of astronomical and solar radiation cycles: potentially reaching submeter scales). The architecture of sequences becomes increasingly complex with the increase in the scale of observation (Fig. 23.18). Beyond this general trend, there are no standards for the scale and internal makeup of sequences of any hierarchical rank; the scale, the systems tract composition, and the relative development of systems tracts within sequences vary with the tectonic, climatic, and depositional settings (e.g. Fielding et al., 2006, 2008; Martins-Neto and Catuneanu, 2010; Csato and Catuneanu, 2012). The scale and the internal makeup of sequences are basin- or even subbasin-specific, reflecting the influence of local controls on accommodation and sedimentation. In downstream-controlled settings, high-frequency sequences (and component systems tracts and depositional systems) are commonly observed at scales of 100 101 m and 102 105 yrs (e.g. Tesson et al., 1990, 2000; Lobo et al., 2004; Amorosi et al., 2005, 2009, 2017; Bassetti et al., 2008; Nanson et al., 2013; Nixon et al., 2014; Magalhaes et al., 2015; Ainsworth et al., 2017; Pellegrini et al., 2017, 2018; Zecchin et al., 2017a,b), which defines the scope of high-resolution sequence stratigraphy (Figs 23.17 and 23.54). The stacking pattern of high-frequency sequences defines systems tracts and component depositional systems of higher hierarchical ranks in lower resolution studies (Figs 23.6, 23.19, 23.22 and 23.27). In upstream-controlled settings, stratigraphic cyclicity can also be observed at different scales, starting from 102 yrs (Bridge and Leeder, 1979; Miall, 2015), with lower rank depositional sequences nested within higher rank systems tracts (Fig. 23.20). In fluvial settings, sequences of different scales show consistent features, including a change from high- to low-amalgamation stacking patterns along with fining-upward trends, indicating that the decline in energy that characterizes a fluvial sequence is independent of scale and subsidence rates. At each hierarchical level, stages of increase in fluvial energy result in the formation of sequence-bounding unconformities, and stages of decrease in fluvial energy result in the deposition of sequences. These energy cycles that define stratigraphic sequences in upstream-controlled settings may correspond to tectonic cycles, climate cycles, or cycles of autogenic migration of alluvial channel belts (Catuneanu and Bowker, 2001; Catuneanu and Elango, 2001; Hajek et al., 2010; Hofmann et al., 2011; Miall, 2015; Catuneanu, 2017, 2019a,b).

Types of stratigraphic sequence The concept of ‘sequence’ evolved since the 1940s, in terms of (1) the scale of a sequence and (2) the nature of the sequence boundary. The original definition regarded the sequence as a large-scale unit bounded by the most significant, basin-scale unconformities (Longwell, 1949; Sloss et al., 1949). Subsequent increases in stratigraphic resolution led to the definition of sequences at progressively smaller scales (Fig. 23.2). As a result, unconformities of smaller magnitude, as well as conformities, were employed as sequence boundaries (Wheeler, 1964). At the same time, the types of surface selected as ‘sequence boundary’ diversified as well, from surfaces associated with subaerial exposure (e.g. Sloss, 1963; Mitchum et al., 1977; Posamentier et al., 1988) to maximum flooding surfaces (e.g. Frazier, 1974; Galloway, 1989) and maximum regressive surfaces (e.g. Johnson and Murphy, 1984; Embry

Regional Geology and Tectonics

648

23. Sequence stratigraphy

Resolution

Alternative nomenclature

Sequence rank

Scale Control Temporal (years)

10

First order: sedimentary basin fill (tectonic setting)

8

Physical(m)

103 1

Megasequence 107

2

1

High-resolution sequence stratigraphy

Sequence

Fourth order and lower ranks: Parasequence sub-seismic scale Small-scale cycle (production development) High-frequency sequence

102

3

3

105 Autocyclicity

Third order: seismic scale (petroleum exploration)

2

106

Supersequence Allocyclicity

Low-resolution sequence stratigraphy

Second order: continental scale (Sloss-type sequence)

104

4

10

1

4 10

3

102

100

FIGURE 23.54 Classification of stratigraphic sequences.‘Megasequence’ and ‘supersequence’ nomenclature from Krapez (1996). Fourthorder and lower rank sequences are also termed ‘cyclothems’ (Wanless and Weller, 1932), ‘cycles’ (Heckel, 1986) or ‘simple sequences’ (Vail et al., 1991; Schlager, 2010); third-order sequences are also termed ‘mesothems’ (Ramsbottom, 1979), ‘megacyclothems’ (Heckel, 1986) or ‘standard sequences’ (Vail et al., 1991; Schlager, 2010); second-order sequences are also termed ‘composite sequences’ (Abreu et al., 2010). A scaleindependent nomenclature remains the most objective approach to terminology (see text for details). Source: Temporal and physical scales compiled from Vail, P.R., Mitchum, R.M. Jr., Thompson, S., III, 1977b. Seismic stratigraphy and global changes of sea level, part 4: global cycles of relative changes of sea level. In: Payton, C.E. (Ed.), Seismic Stratigraphy Applications to Hydrocarbon Exploration. American Association of Petroleum Geologists Memoir 26, pp. 83 97; Vail, P.R., Audemard, F., Bowman, S.A., Eisner, P.N., Perez-Cruz, C., 1991. The stratigraphic signatures of tectonics, eustasy and sedimentology an overview. In: Einsele, G., Ricken, W., Seilacher, A., (Eds.), Cycles and Events in Stratigraphy. Springer-Verlag. pp. 617 659; Williams, D.F., 1988. Evidence for and against sea-level changes from the stable isotopic record of the Cenozoic. In: Wilgus, C.K., Hastings, B.S., Kendall, C.G.St.C., Posamentier, H.W., Ross, C.A., Van Wagoner, J.C. (Eds.), Sea Level Changes An Integrated Approach, SEPM Special Publication, 42, pp. 31 36; Van Wagoner, J.C., Mitchum Jr., R.M., Campion, K.M., Rahmanian, V.D., 1990. Siliciclastic Sequence Stratigraphy in Well Logs, Core, and Outcrops: Concepts for High-Resolution Correlation of Time and Facies. American Association of Petroleum Geologists Methods in Exploration Series 7. 55 pp; Carter, R.M., Abbott, S.T., Fulthorpe, C.S., Haywick, D.W., Henderson, R.A., 1991. Application of global sea-level and sequence-stratigraphic models in southern hemisphere Neogene strata. In: Macdonald, D.I.M., (Ed.), Sedimentation, Tectonics and Eustasy: Sea-Level Changes at Active Margins. International Association of Sedimentologists Special Publication 12, pp. 41 65; Einsele, G., Ricken, W., Seilacher, A. (Eds.), 1991. Cycles and Events in Stratigraphy. Springer-Verlag, 955 pp; Reid, S.K., Dorobek, S.L., 1993. Sequence stratigraphy and evolution of a progradational foreland carbonate ramp, Lower Mississippian Mission Canyon Formation and stratigraphic equivalents, Montana and Idaho. In: Loucks, R.G., Sarg, J.F. (Eds.), Carbonate Sequence Stratigraphy, Recent Developments and Applications. American Association of Petroleum Geologists Memoir 57, pp. 327 352; Duval, B., Cramez, C., Vail, P.R., 1998. Stratigraphic cycles and major marine source rocks. In: De Graciansky, P.C., Hardenbol, J., Jacquin, T., Vail, P.R. (Eds.), Mesozoic and Cenozoic Sequence Stratigraphy of European Basins, Society for Sedimentary Geology, Special Publication 60, pp. 43 51; Lehrmann, D.J., Goldhammer, R.K., 1999. Secular variation in parasequence and facies stacking patterns of platform carbonates: a guide to application of stackingpatterns analysis in strata of diverse ages and settings. In: Harris, P.M., Saller, A.H., Simo, J.A. (Eds.), Advances in Carbonate Sequence Stratigraphy: Application to Reservoirs, Outcrops and Models. SEPM (Society for Sedimentary Geology) Special Publication 63, pp. 187 225; Schlager, W., 2004. Fractal nature of stratigraphic sequences. Geology 32, 185 188; Schlager, W., 2010. Ordered hierarchy versus scale invariance in sequence stratigraphy. Int. J. Earth Sci. 99, S139 S151; and Miall, A.D., 2010. The geology of stratigraphic sequences, second ed. Springer-Verlag, Berlin, 522 pp.

and Johannessen, 1992) (Fig. 23.4). In this process, conformable surfaces have become increasingly important to the delineation of sequences, to the point that some types of high-frequency sequence (e.g. genetic stratigraphic) no longer require unconformities at the sequence boundary. Nevertheless, sequences of any scale may include unconformities of equal or lower hierarchical ranks, while still maintaining the genetically related character of the comprising strata. The revisions to the definition of a ‘sequence’ (Fig. 23.2) reflect (1) the gradual increase in the resolution of stratigraphic studies and (2) the need for an inclusive definition that accommodates all existing approaches. The current definition provides the flexibility to apply the sequence stratigraphic methodology in a manner that is independent of model and at all scales afforded by the data available. Different types of stratigraphic sequence can be defined, depending on the sequence stratigraphic surface that is selected as the sequence boundary (Fig. 23.3). Depositional sequence The depositional sequence is a stratigraphic sequence bounded by subaerial unconformities or their correlative conformities (Mitchum, 1977). Different types of depositional sequence have been defined (Fig. 23.4), which can be classified into two groups as a function of the timing of the marine portion of the sequence boundary (i.e. the

Regional Geology and Tectonics

Sequence stratigraphic framework

649

‘correlative conformity’; Fig. 23.3): one group considers the correlative conformity as the palaeo-seafloor at the onset of forced regression (i.e. the correlative conformity sensu Posamentier et al., 1988; herein referred to as the ‘basal surface of forced regression’); and another group considers the correlative conformity as the palaeoseafloor at the end of forced regression (i.e. the correlative conformity sensu Van Wagoner et al., 1988; herein referred to as the ‘correlative conformity’). Other differences between the various depositional sequence models relate to the definition and nomenclature of the component systems tracts (Fig. 23.3). The most recent depositional sequence model (Hunt and Tucker, 1992) presents the advantage of separating and assigning forced and normal regressions to different systems tracts. This distinction is necessary in view of the different sediment dispersal patterns associated with the two types of regression, which is significant on several grounds (e.g. the distribution of petroleum plays; Posamentier et al., 1992a). All types of depositional sequence, as originally defined, assume full cycles of change in accommodation and relate the sequence boundary to stages of negative accommodation. It should be noted, however, that subaerial unconformities may also form during stages of relative sea/lake-level rise and transgression (Figs 23.43 and 23.34B). Therefore, the interpretation of the underlying controls that are responsible for the formation of depositional sequences needs to be separated from the methodological workflow that relies on the observation of stratal stacking patterns (Catuneanu and Zecchin, 2016). The concept of depositional sequence applies to both downstream- and upstream-controlled settings, where subaerial unconformities may form. Genetic stratigraphic sequence The genetic stratigraphic sequence is a stratigraphic sequence bounded by maximum flooding surfaces (Galloway, 1989). As maximum flooding surfaces form during stages of positive accommodation, the formation of genetic stratigraphic sequences does not require stages of negative accommodation. At any scale of observation, a genetic stratigraphic sequence corresponds to a regressive-transgressive cycle, which may occur during a full cycle of change in accommodation or during a stage of positive accommodation. In the latter case, the genetic stratigraphic sequence does not include falling-stage and lowstand systems tracts, nor any sequence stratigraphic surface that is exclusively associated with forced regression (i.e. the basal surface of forced regression, the regressive surface of marine erosion, and the correlative conformity; Fig. 23.34). Often, however, genetic stratigraphic sequences do include subaerial unconformities of equal hierarchical rank (Frazier, 1974; Galloway, 1989). In this case, the genetic stratigraphic sequence cannot be described as a ‘relatively conformable succession’ within the area of development of the subaerial unconformity, although it does consist of ‘genetically related strata’ that belong to the same stratigraphic cycle; therefore, satisfying the definition of a stratigraphic sequence (Fig. 23.2). The subaerial unconformity has a smaller areal extent than the host genetic sequence, thus maintaining the integrity of the sequence as one stratigraphic cycle. Beyond the termination of the subaerial unconformity, the correlative conformity preserves the continuity of sedimentation and the genetic relationship between the underlying and the overlying strata within the sequence (i.e. the ‘continuity’ surface of Wheeler, 1964). The genetic stratigraphic sequence approach does not rely on the development and recognition of subaerial unconformities and correlative conformities; this presents an advantage where (1) stratigraphic cyclicity develops within conformable successions and (2) seismic data are unavailable, preventing the observation of stratal terminations that facilitate the recognition of subaerial unconformities and correlative conformities. Instead, the physical record of transgression provides ‘readily recognized regionally correlative, easily and accurately datable, and robust sequence boundaries’ (Galloway, 1989). The concept of genetic stratigraphic sequence applies to downstream-controlled settings, where maximum flooding surfaces may form. Transgressive-regressive sequence The transgressive-regressive (T-R) sequence is a stratigraphic sequence bounded by maximum regressive surfaces (Johnson and Murphy, 1984). In its original definition, the formation of T-R sequences does not require stages of negative accommodation. As maximum regressive surfaces of any hierarchical rank may form during stages of positive accommodation, the T-R sequences may be generated either during full cycles of change in accommodation or during periods of positive accommodation. In the latter case, the T-R sequence consists solely of transgressive and highstand systems tracts. The T-R sequence boundary is also termed ‘transgressive surface’, where emphasis is placed on the onset of transgression rather than the end of regression. A proposal to modify the original definition of the T-R sequence was made to include the subaerial unconformity as the continental portion of the sequence boundary (Embry and Johannessen, 1992). However, as the maximum regressive surface is most commonly younger than the subaerial unconformity, the marine portion of the maximum regressive surface may not meet with the basinward termination of the subaerial unconformity

Regional Geology and Tectonics

650

23. Sequence stratigraphy

(Embry and Johannessen, 1992). In this case, the maximum regressive surface and the subaerial unconformity may occur at different stratigraphic levels in one vertical section, separated by the topset of the lowstand systems tract (Fig. 23.34A). Therefore, the original definition of Johnson and Murphy (1984) is still recommended, as it provides a foolproof approach. The concept of T-R sequence applies to downstream-controlled settings, where maximum regressive surfaces may form.

Surfaces of sequence stratigraphy Stratigraphic surfaces in the rock record may be classified into ‘sequence stratigraphic’ and ‘within-trend facies contacts’ (e.g. discussion in Catuneanu, 2006). Sequence stratigraphic surfaces define the sequence stratigraphic framework of sequences and component systems tracts, whereas within-trend facies contacts mark lithological changes within the systems tracts. Both types of stratigraphic surface are important to the delineation of facies and depositional systems within a stratigraphic section. However, the workflow of sequence stratigraphic analysis requires the identification of sequence stratigraphic surfaces first, followed by the placement of facies contacts within the framework of systems tracts and bounding sequence stratigraphic surfaces (Catuneanu, 2006; Zecchin and Catuneanu, 2013). A sequence stratigraphic surface is a type of stratigraphic contact that can serve, at least in part, as a systems tract boundary (Catuneanu et al., 2011). As systems tract boundaries, sequence stratigraphic surfaces mark changes in stratal stacking pattern between the units below and above the contact (Fig. 23.34). This defining attribute separates a sequence stratigraphic surface from any other type of stratigraphic contact and provides the basis for sequence stratigraphic correlation. In contrast, the within-trend facies contacts are relevant as limits of architectural units defined on lithological criteria and provide the basis for lithostratigraphic or allostratigraphic correlations. Such architectural units may develop across systems tract boundaries, and are significant for petroleum reservoir studies, particularly at the scale of high-frequency sequences (Zecchin and Catuneanu, 2013, 2015). For example, condensed sections may form discrete architectural units that include both transgressive and highstand sediment, bounded by flooding surfaces at the base (within transgressive systems tracts) and downlap surfaces at the top (at the limit between bottomsets and foresets within highstand systems tracts) (Zecchin and Catuneanu, 2013). In this example, the maximum flooding surface (i.e. the systems tract boundary) lies within the condensed section (Posamentier and Allen, 1999; Catuneanu, 2006). Within-trend facies contacts have been defined in all systems tracts, in relation to normal regressions (e.g. within-trend normal regressive surfaces), forced regressions (e.g. within-trend forced regressive surfaces), and transgressions (e.g. within-trend flooding surfaces) (Catuneanu, 2006). Sequence stratigraphic surfaces may or may not have a lithological expression. Conformable sequence stratigraphic surfaces are prone to be lithologically ‘cryptic’ (e.g. maximum flooding surfaces within marine condensed sections), as the change in stratal stacking pattern, rather than grain size, is the primary defining feature. In the case of maximum flooding surfaces, the overlying highstand systems tract includes the upper part of the condensed section, as well as the clinoforms that display the typical prograding and upstepping stacking pattern. The limit between the condensed section and the overlying clinoforms is a within-trend facies contact that may have a stronger physical expression than the maximum flooding surface (Zecchin and Catuneanu, 2013). Nevertheless, the maximum flooding surface is more important for regional correlation and the construction of the sequence stratigraphic framework, including as a datum, and it displays a lower degree of diachroneity. Wherever sediment supply plays a role in the timing of their formation, the origin of sequence stratigraphic surfaces may relate to any combination of allogenic and autogenic controls (Catuneanu and Zecchin, 2013). This is the case with all sequence stratigraphic surfaces whose timing is linked to shoreline transgression (i.e. maximum regressive surfaces, transgressive surfaces of erosion, maximum flooding surfaces, and some subaerial unconformities; Fig. 23.34). Only surfaces associated strictly with forced regression (i.e. the basal surface of forced regression, the regressive surface of marine erosion, and the correlative conformity; Fig. 23.34) have their timing controlled exclusively by allogenic factors. All sequence stratigraphic surfaces are potentially diachronous (Catuneanu, 2006, p. 321). Seven types of sequence stratigraphic surface have been defined in the context of ‘conventional’ sequence stratigraphy (Fig. 23.34). With the exception of the subaerial unconformity, all other ‘conventional’ sequence stratigraphic surfaces require a marine/lacustrine environment to form, and therefore, are restricted to downstreamcontrolled settings. Only four of these surfaces (i.e. the basal surface of forced regression, the correlative conformity, the maximum regressive surfaces, and the maximum flooding surface) extend into the deepwater system. Therefore, the most complete set of sequence stratigraphic surfaces is found in the continental to shallow-water systems of the downstream-controlled settings (Fig. 23.34).

Regional Geology and Tectonics

651

Sequence stratigraphic framework

FIGURE

23.55 Subaerial unconformity (arrow) at the limit between the floodplaindominated fluvial succession of the Las Cabras Formation and the overlying channeldominated fluvial system of the lower Potrerillos Formation (Triassic, Cuyana Basin, Argentina). Diagnostic for the subaerial unconformity is the preservation of continental deposits on top.

Subaerial unconformity The subaerial unconformity (Sloss et al., 1949) is a hiatal surface that forms in continental environments as a result of fluvial erosion or bypass, pedogenesis, wind degradation, or karstification (Fig. 23.55). Subaerial unconformities may form in both downstream- and upstream-controlled settings, under various accommodation conditions (Figs 23.20 and 23.34). Where the subaerial unconformity forms during forced regression, it is commonly found at the top of the highstand topset and of the falling-stage foreset (Fig. 23.34A). However, in cases where the exposure is associated with significant erosion (e.g. leading to the formation of incised valleys), the underlying highstand and falling-stage systems tracts may not be preserved, and the subaerial unconformity may truncate any older systems tracts. Under particular circumstances (e.g. where the landscape profile is steeper than the shoreline trajectory), the subaerial unconformity may also form during periods of positive accommodation and transgression (Figs 23.34B and 23.43). Alternative terms include ‘lowstand unconformity’ (Schlager, 1992) and ‘regressive surface of fluvial erosion’ (Plint and Nummedal, 2000). However, the term ‘subaerial unconformity’ is preferred because it does not link, nor restrict, the formation of this surface to stages of lowstand or regression. The identification of a subaerial unconformity in the rock record requires the preservation of continental deposits (fluvial or eolian) on top (Fig. 23.55). The facies preserved below a subaerial unconformity may range from continental to marine, and therefore, they are not diagnostic to the identification of this surface. Subaerial unconformities may be subsequently reworked and replaced by younger erosional surfaces, in which case the composite unconformity takes the name of the younger surface. For example, a subaerial unconformity that is reworked by waves during a subsequent transgression is replaced by a transgressive surface of erosion. In this case, the deposits on top of the composite unconformity are transgressive marine, diagnostic of the younger surface (i.e. the subaerial unconformity is not preserved as a physical surface, and its temporal attributes are transferred to the transgressive surface of erosion). Basal surface of forced regression The basal surface of forced regression (Hunt and Tucker, 1992) is a sequence stratigraphic surface that marks a change in stratal stacking pattern from normal regression (below) to forced regression (above) (Figs 23.56 and 23.57). Most commonly, the underlying normal regression is highstand, in the case of sequences that include a transgressive systems tract (Fig. 23.34), but it can also be lowstand in the case of sequences that consist only of falling-stage and lowstand systems tracts. In either case, the basal surface of forced regression marks the onset of forced regression; therefore, the criteria that afford the distinction between forced and normal regressive deposits (e.g. Posamentier and Morris, 2000; Catuneanu and Zecchin, 2016; Figs 23.40, 23.58) are critical to the identification of this surface. Field examples that illustrate the physical expression of the basal surface of forced regression have been documented on different data sets, from seismic to core and outcrop (e.g. Posamentier and Allen, 1999; Catuneanu, 2006; Catuneanu et al., 2011; MacEachern et al., 2012; Figs 23.6, 23.56 and 23.57). Where the subaerial unconformity forms during forced regression (Fig. 23.34A), the basal surface of forced regression is a marine surface (i.e. the palaeo-seafloor at the onset of forced regression) truncated at the top by the subaerial unconformity. Where the subaerial unconformity forms during transgression (Fig. 23.34B), the basal

Regional Geology and Tectonics

Top

BSFR

BSFR

Base

3 cm

FIGURE 23.56 Lithological expression of the basal surface of forced regression (BSFR) in a siliciclastic shallow-water setting (Early Cretaceous Viking Formation, Judy Creek Field). Within a shelf succession, the BSFR marks an increase in grain size that accompanies the shift from highstand normal regression (lower rates of progradation, expanded shoreface development, and lower sediment supply to the shelf) to forced regression (higher rates of progradation, compressed shoreface development, and higher sediment supply to the shelf). In shallowwater settings, the proximal portion of the BSFR is commonly reworked by shoreface waves during forced regression and replaced by the regressive surface of marine erosion, while the distal portion (e.g. this example) may be preserved as a conformable surface where the gradient of the shoreline trajectory during forced regression is lower than the shelf gradient. In deepwater settings, the BSFR is commonly reworked by early forced regressive slumps and mudflows triggered by the lowering of the wave base during relative sea-level fall. Source: From MacEachern, J.A., Dashtgard, S.E., Knaust, D., Catuneanu, O., Bann, K.L., Pemberton, S.G., 2012. Sequence stratigraphy. In: Knaust, D., Bromley, R.G. (Eds.), Trace Fossils as Indicators of Sedimentary Environments. Developments in Sedimentology, vol. 64. Elsevier, pp. 157 194.

E TST

prograding, sharp-based

Shoreface sandstone

WRS

D FSST C RSME

B BSFR

A

HST carbonate platform

FIGURE 23.57 Lithological expression of the basal surface of forced regression (BSFR) in a mixed carbonate-siliciclastic shelf setting (Aptian, Araripe Basin, Brazil). Facies: A clean limestone, horizontally laminated (outer shelf); B marls, horizontally stratified (inner shelf); C shoreface clinoforms downlapping the regressive surface of marine erosion (RSME); D massive shoreface sandstone, concretionary towards the top; E transgressive silts and shale. The carbonate factory is shut down by siliciclastic sediment influx during forced regression. As a result, limestones are replaced by marls in the inner shelf and by sandstone in the shoreface. The BSFR marks the limit between the clean highstand limestone (stable carbonate platform) and the overlying falling-stage sediments that record an increase in the rates of progradation and in the amount of terrigenous influx to the lake. WRS wave-ravinement surface; HST highstand systems tract; FSST falling-stage systems tract; TST transgressive systems tract.

653

Sequence stratigraphic framework

Forced regression

Normal regression

A. Syn-depositional time:

A. Syn-depositional time:

Offlap

Topset RSL rise

RSL fall

RSME

Shelf

Subaerial unconformity Downstepping upper shoreface facies

Shoreline trajectory Upstepping upper shoreface facies

B. Following erosion: A

B. Following erosion: C

B

D

RSME

A

Shelf

Truncation surface Upstepping upper shoreface facies

Truncation surface Downstepping upper shoreface facies

C

B

Marine or nonmarine

Marine or nonmarine Sharp-based shoreface

Shelf

20 m

Shelf

D

Gradationally based shoreface

Upper shoreface Lower shoreface

Shelf

20 m

Shelf

FIGURE 23.58 Criteria to separate forced from normal regressions, where erosion removes the evidence of offlap and fluvial topsets. Forced regressions are characterized by downstepping and sharp-based shoreface facies and true downlap of shoreface clinoforms against the RSME. Normal regressions are characterized by upstepping and gradationally based shoreface facies and apparent downlap (i.e. seismic artefact) of shoreface clinoforms against the shelf bottomset facies. RSL relative sea level; RSME regressive surface of marine erosion; A, B, C and D wells, illustrated with gamma-ray logs.

surface of forced regression includes both marine and continental portions. The basal surface of forced regression is also known as the ‘correlative conformity’ in the sense of Posamentier et al. (1988) (i.e. the depositional sequence II model in Fig. 23.3). Correlative conformity The correlative conformity (Van Wagoner et al., 1988; Hunt and Tucker, 1992) is a sequence stratigraphic surface that marks a change in stratal stacking pattern from forced regression (below) to lowstand normal regression (above) (Figs 23.34, 23.59 and 23.60). This type of sequence stratigraphic surface was first recognized by Wheeler (1964, p. 606), as a ‘continuity’ surface that correlates with the downdip termination of a subaerial unconformity. The continuity surface of Wheeler (1964) was renamed as the ‘correlative conformity’ in the 1970s, when it was incorporated in the revised definition of a sequence (Mitchum, 1977; Fig. 23.2). Only subaerial unconformities that form during forced regression have correlative conformities sensu stricto (Fig. 23.34A). In the case of subaerial unconformities that form during transgression, the true correlative conformities are represented by maximum regressive surfaces, and the physical connection between subaerial unconformities and their correlative maximum regressive surfaces is made by the transgressive surface of erosion (Fig. 23.34B). However, in a broader sense, the term ‘correlative conformity’ can be retained for the surface that marks the end of forced regression, irrespective of the timing of the subaerial unconformity, for nomenclatural consistency and simplicity (Fig. 23.34). Where the subaerial unconformity forms during forced regression (Fig. 23.34A), the correlative conformity is a marine surface (i.e. the palaeo-seafloor at the end of forced regression) that connects physically with the downdip termination of the subaerial unconformity at the location of the shoreline at the end of forced regression. This

Regional Geology and Tectonics

654

23. Sequence stratigraphy

TST

Fluvial topset

FSST

LST

Top

15 cm

CC

WRS/MRS

Base

FIGURE 23.59

Lithological expression of the correlative conformity (CC) in a siliciclastic shallow-water system (Early Cretaceous Viking Formation, Judy Creek Field). Within this shoreface succession, the CC marks a decrease in grain size that accompanies the shift from forced regression (higher rates of progradation and higher sediment supply to the shoreline) to lowstand normal regression (lower rates of progradation and lower sediment supply to the shoreline). The lowstand shoreface (coarsening-upward foreset) is overlain by the fluvial topset of the LST. WRS/MRS wave-ravinement surface reworking and replacing a maximum regressive surface; FSST falling-stage systems tract; LST lowstand systems tract; TST transgressive systems tract. Source: From MacEachern, J.A., Dashtgard, S.E., Knaust, D., Catuneanu, O., Bann, K. L., Pemberton, S.G., 2012. Sequence stratigraphy. In: Knaust, D., Bromley, R.G. (Eds.), Trace Fossils as Indicators of Sedimentary Environments. Developments in Sedimentology, vol. 64. Elsevier, pp. 157 194.

B

CC

A

FIGURE 23.60 Lithological expression of the correlative conformity (CC) in a mixed carbonate-siliciclastic deepwater setting (Triassic, The Dolomites, Italy). A siliciclastic high-density turbidites (coarser grained, with a high sand-to-mud ratio, dominated by the divisions A and B of the Bouma sequence); B clastic carbonate low-density turbidites (finer grained, with a low sand-to-mud ratio, dominated by the divisions C, D and E of the Bouma sequence). The CC marks a decrease in grain size and a change from siliciclastic river-borne sediment (relative sealevel fall: carbonate factory shut down) to carbonate sediment (relative sea-level rise: carbonate factory switched on). Source: From Catuneanu, O., Galloway, W.E., Kendall, C.G.St.C., Miall, A.D., Posamentier, H.W., Strasser, A., et al., 2011. Sequence stratigraphy: methodology and nomenclature. Newsl. Stratigr. 44 (3), 173 245.

scenario is most common in the stratigraphic record. Where the subaerial unconformity forms during transgression (Fig. 23.34B), the ‘correlative conformity’ becomes a conformable surface with both marine and continental portions, without a physical and temporal relationship with the subaerial unconformity. In either case, the correlative conformity marks the end of forced regression and the beginning of lowstand normal regression

Regional Geology and Tectonics

655

Sequence stratigraphic framework

(Fig. 23.34); therefore, the criteria that afford the distinction between forced and normal regressive deposits (e.g. Posamentier and Morris, 2000; Catuneanu and Zecchin, 2016; Figs 23.40 and 23.58) are critical to the identification of this surface. Field examples that illustrate the physical expression of the correlative conformity have been documented on different data sets, from seismic to core and outcrop (e.g. Posamentier and Allen, 1999; Catuneanu, 2006; Catuneanu et al., 2011; MacEachern et al., 2012; Figs 23.6, 23.35, 23.59 and 23.60). Maximum regressive surface The maximum regressive surface (Helland-Hansen and Martinsen, 1996) is a sequence stratigraphic surface that marks a change in the direction of shoreline shift from regression to subsequent transgression (Fig. 23.61). Most commonly, the underlying deposits display a normal regressive stacking pattern (Fig. 23.34). The normal regression may be lowstand, in the case of sequences that include a falling-stage systems tract (i.e. a stage of negative accommodation during the formation of the sequence; Fig. 23.34), or highstand, in the case of sequences that consist only of transgressive and highstand systems tracts without intervening stages of negative accommodation (i.e. sequences that form during continuous positive accommodation, as a result of variations in the rates of accommodation and sedimentation at the shoreline) (e.g. Csato and Catuneanu, 2012, 2014). Where sequences record both stages of positive and negative accommodation, but the lowstand systems tract is missing, the maximum regressive surface coincides with the correlative conformity at the end of forced regression (e.g. in extensional settings where accommodation is generated rapidly by the reactivation of faults; MartinsNeto and Catuneanu, 2010). Therefore, maximum regressive surfaces may top any type of regressive deposit (i.e. lowstand normal regressive, highstand normal regressive, or forced regressive), so the actual stratigraphic relationship in each study needs to be determined on a case-by-case basis. Sequences that consist only of falling-stage and lowstand systems tracts do not include a maximum regressive surface, nor any other surfaces associated with transgression (i.e. transgressions are suppressed by sediment supply at the scale of observation that defines the hierarchical rank of the sequence). The maximum regressive surface is the palaeo-seafloor at the end of regression, and its correlative depositional surface within the continental setting. At least part of the continental portion of the maximum regressive surface is reworked and replaced by the transgressive surface of erosion during subsequent transgression (Fig. 23.34). The marine portion of the maximum regressive surface has a better preservation potential (Fig. 23.61), and it is typically onlapped by transgressive marine or lacustrine ‘healing-phase’ deposits (i.e. coastal onlap: Figs 23.62 and 23.63; Posamentier and Allen, 1999). The maximum regressive surface is also known as the ‘transgressive surface’ (Posamentier and Vail, 1988). The term ‘maximum regressive surface’ is recommended where emphasis is placed on the end of regression (e.g. top of a coarsening-upward trend in a shallow-water system); the term ‘transgressive surface’ is recommended where emphasis is placed on the onset of transgression (e.g. the base of the oldest facies of an estuarine system). FIGURE 23.61 Wave-ravinement (WRS), maximum Outer shelf

MFS

MRS

Inner shelf

MFS

flooding (MFS), and maximum regressive (MRS) surfaces (Triassic rift, upper Potrerillos to Cacheuta formations, Cuyana Basin, Argentina). The wave-ravinement surface is unconformable (scouring above the fairweather wave base) and has diagnostic transgressive ‘healing-phase’ (lacustrine here) deposits on top. The maximum regressive and maximum flooding surfaces may be conformable (deposition below the fairweather wave base). In this example, the wave-ravinement and the maximum regressive surfaces coincide with flooding surfaces (episodes of abrupt water deepening). The maximum flooding surfaces are lithologically cryptic, within condensed sections. This cyclicity describes the internal architecture of a larger scale (higher hierarchical rank) transgressive systems tract.

WRS Shoreface

Regional Geology and Tectonics

656

23. Sequence stratigraphy

Toplap Offlap

Onlap Truncation Downlap

FIGURE 23.62 Types of stratal terminations: truncation, toplap, onlap, downlap, and offlap.

Young

Truncation: termination of strata against an overlying erosional surface. Truncation refers to the strata below an erosional surface, typically an angular unconformity.

Old

Old

Young

Toplap: termination of inclined strata against an overlying lower angle surface that shows no evidence of erosion; commonly, a foreset overlain by a surface of sediment bypass (no deposition, no erosion). Lapouts mark the updip depositional limits of the sedimentary units and become progressively younger basinward.

Toplap implies a graded fluvial profile at the top of clinoforms during progradation. Typically, a seismic artefact: a topset below the seismic resolution, or truncation with indiscernible evidence of erosion.

Young Old

Onlap: termination of low-angle strata against an underlying steeper surface, in which each successively younger unit extends beyond the limit of the older unit on which it lies. A lapout with upstepping in the younging direction, which marks the updip termination of a sedimentary unit at its depositional limit.

Typically, the product of sedimentation during relative sea-level rise (either transgression or normal regression). Types of onlap: • Marine onlap: deepwater sediment onlapping the continental slope, most commonly during transgression (e.g.‘slope aprons’ of Galloway, 1989; ‘healing-phase’ deposits of Posamentier and Allen, 1993) • Coastal onlap: coastal to shallow-water sediment onlapping the transgressive surface of erosion (tidal- or wave-ravinement surfaces) during transgression • Fluvial onlap: fluvial sediment onlapping the basin margin as the area of fluvial sedimentation expands landward during relative sea-level rise (normal regression or transgression)

Old

Young

Downlap: termination of inclined strata against an underlying lower angle surface. A baselap which marks the real or apparent downdip termination of a sedimentary unit at its depositional limit; a change from foreset to bottomset in a subaqueous environment.

Typically, the product of sediment progradation during relative sea-level rise or fall. A seismic artifact in the case of gradationally based prograding units with bottomsets that fall below the seismic resolution; a real termination in the case of sharp-based, forced regressive prograding shorefaces.

Old

Young

Offlap: termination of low-angle strata against an underlying steeper surface, in which each successively younger unit leaves exposed a portion of the older unit on which it lies. A lapout with downstepping in the younging direction, which marks the updip termination of a sedimentary unit at its depositional limit.

Typically, the product of sedimentation during relative sea-level fall, diagnostic of forced regression.

FIGURE 23.63 Stratal terminations: definitions and diagnostic features. Source: Modified after Mitchum, R.M., Jr., 1977. Seismic stratigraphy and global changes of sea level, part 11: glossary of terms used in seismic stratigraphy. In: Payton, C.E. (Ed.), Seismic Stratigraphy Applications to Hydrocarbon Exploration. American Association of Petroleum Geologists Memoir 26, pp. 205 212.

The two terms are interchangeable, as the end of regression and the beginning of transgression are essentially coincident at stratigraphic timescales. However, depending on the nature of the facies that are preserved and/or observed (i.e. regressive vs transgressive), one term or the other may become more relevant within the context of a particular case study.

Regional Geology and Tectonics

Sequence stratigraphic framework

657

Maximum flooding surface The maximum flooding surface (Frazier, 1974; Posamentier et al., 1988; Van Wagoner et al., 1988; Galloway, 1989) is a sequence stratigraphic surface that marks a change in stratal stacking pattern from transgressive (below) to highstand normal regressive (above) (Figs 23.34 and 23.61). It is the palaeo-seafloor at the end of transgression, and, where transgression is accompanied by fluvial or eolian aggradation, its correlative depositional surface within the nonmarine setting (Fig. 23.34A). The continental portion of the maximum flooding surface is commonly associated with the highest water table relative to the topographic profile, and therefore, it may be marked by the development of regional coal seams (e.g. Gastaldo et al., 1993; Wright and Marriott, 1993; Hamilton and Tadros, 1994; Shanley and McCabe, 1994; Bohacs and Suter, 1997; Holz et al., 2002; Fanti and Catuneanu, 2010). Within marine sections, the maximum flooding surface is often lithologically cryptic, at the heart of condensed sections (e.g. Carter et al., 1998; Posamentier and Allen, 1999; Catuneanu, 2006; Figs 23.35 and 23.61). Nevertheless, the identification of maximum flooding surfaces is still possible on the basis of various independent methods that involve seismic data (e.g. the observation of ‘downlap surfaces’ at the limit between low-angle reflections associated with coastal onlap and the overlying highstand clinoforms; Fig. 23.34), well logs calibrated with lithologs (e.g. highest radioactivity in fine-grained sediments), and biostratigraphic data (e.g. highest abundance and diversity of microfossils; Gutierrez Paredes et al., 2017). Notably, downlap surfaces interpreted as maximum flooding surfaces are typically observed on lowresolution seismic transects, where the bottomsets of highstand clinoforms fall bellow the vertical seismic resolution (Figs 23.62 and 23.63). In high-resolution studies, a sedimentological downlap surface may be defined within the highstand systems tract, at the limit between the hemipelagic condensed section, within which the maximum flooding surface resides, and the overlying coarser-grained clinoforms (Zecchin and Catuneanu, 2013). This limit between highstand bottomset and foreset is a within-trend facies contact that may have a stronger physical expression than the maximum flooding surface, but also a higher rate of diachroneity that matches the rate of shoreline progradation. Despite its potentially cryptic physical expression, the maximum flooding surface remains a more important marker and datum for regional correlation. Alternative terms include ‘final transgressive surface’ (Nummedal et al., 1993) and ‘maximum transgressive surface’ (Helland-Hansen and Martinsen, 1996). The term ‘maximum flooding surface’ is still recommended, as it is strongly entrenched in the literature and has historical priority. It is important to avoid confusion between maximum flooding surfaces and flooding surfaces. The distinction is not a matter of scale (e.g. major vs minor transgressions; NB: there are more than two scales in stratigraphy), but a matter of definition: a maximum flooding surface marks a change in stratal stacking pattern (i.e. it is a sequence stratigraphic surface), and it may or may not be associated with a lithological contrast; a flooding surface is a lithological discontinuity (i.e. an allostratigraphic surface), which may or may not mark a change in stratal stacking pattern. Both flooding surfaces and maximum flooding surfaces may form at the same scale of observation (i.e. hierarchical rank), in relation to the same transgressions (Fig. 23.64). Transgressive surface of erosion The transgressive surface of erosion (Posamentier and Vail, 1988) is an unconformity that forms during transgression by means of wave scouring (i.e. ‘wave-ravinement surface’; Swift, 1975; Fig. 23.61) or tidal scouring (i.e. ‘tidal-ravinement surface’; Allen and Posamentier, 1993; Fig. 23.65) in subtidal to intertidal environments. The two types of transgressive ravinement surface merge into a single contact in open-shoreline settings (i.e. the interfluves of estuaries), where backstepping beaches are reworked by both waves and tides during transgression; in this case, the composite surface is described as wave-ravinement, as waves in the subtidal environment leave the last imprint on the composite contact. In river-mouth settings, the two transgressive ravinement surfaces may be preserved as distinct contacts separated by the estuary-mouth complex (i.e. with the tidal-ravinement below, and the wave-ravinement above; Posamentier and Allen, 1999; Catuneanu, 2006); in this case, the tidal-ravinement surface forms within the estuary, updip relative to the area of development of the wave-ravinement surface, by the amalgamation of channel scours in the process of landward migration of tidal channels (Dalrymple et al., 1992; Kitazawa and Murakoshi, 2016). The tidal-ravinement surface is commonly found at the base of tidal bars in tide-dominated estuaries, and at the base of barrier islands or flood-tidal deltas in wave-dominated estuaries. Irrespective of setting, the wave-ravinement surface has a wider spread development at the base of all fully marine transgressive deposits, both in front of estuaries and along open shorelines, and it is therefore accounted for as the primary transgressive surface of erosion (Fig. 23.34).

Regional Geology and Tectonics

658

23. Sequence stratigraphy

SP (mV) –120

Surfaces

Cycles

20

Shoreline trajectory

NR MFS/FS/MRS

50 m

NR MFS FS MRS MFS FS

T NR T

MRS NR MFS T MRS

FIGURE 23.64

Stratigraphic cycles in a shallow-water setting: parasequences (red arrows), genetic stratigraphic sequences (blue arrows), and transgressive-regressive sequences (black arrows). Flooding surfaces may or may not coincide with systems tract boundaries. In the latter case, flooding surfaces are allostratigraphic contacts. T transgression; NR normal regression; MRS maximum regressive surface; MFS maximum flooding surface; FS flooding surface.

TRS

FIGURE 23.65 Tidal-ravinement surface (TRS) at the base of a backstepping estuary-mouth complex. Source: Photo courtesy of H.W. Posamentier; Muddy Formation, Upper Cretaceous, Colorado.

All transgressive ravinement surfaces young in an updip direction (Nummedal and Swift, 1987), with a rate of diachroneity that matches the rate of shoreline transgression (Catuneanu, 2006). The oldest portion of any transgressive ravinement surface invariably reworks part of the maximum regressive surface, thus becoming a systems tract boundary (Catuneanu et al., 2011; Fig. 23.34); for this reason, the transgressive surface of erosion is included on the list of sequence stratigraphic surfaces. The amount of erosion associated with transgressive ravinement processes is proportional to the energy of waves and tides along the transgressive coastline (i.e. the depth of the wave base and the magnitude of the tidal range). In most cases, an average of 10 20 m of section is removed by ravinement processes during transgression (Demarest and Kraft, 1987; Abbott, 1998). This amount can increase significantly in areas of exceptionally high energy (e.g. c. 40 m along the Canterbury Plains shoreline in New Zealand; Leckie, 1994). In cases where transgression is punctuated by stages of short-term progradation (Figs 23.26 and 23.66), the long-term backstepping of the shoreline is accompanied by the formation of multiple transgressive ravinement

Regional Geology and Tectonics

Sequence stratigraphic framework

(A) Transgression with fluvial aggradation 1. Uninterrupted transgression punctuated by flooding events Sea level 4 3

Sea floor

4 2

3 1

2 1

2.Transgression punctuated by progradation and flooding events Sea level 4

4 3

4

Sea floor

3 2

3

2 1

2 1 1

(B) Transgression with fluvial erosion 1. Uninterrupted transgression punctuated by flooding events

Sea level

FSSTto

pset

Sea floor

4 3 2

1

2.Transgression punctuated by progradation and flooding events Sea level

FSSTto

pset

4 4

Sea floor

3 3

659 FIGURE 23.66 Evolution of coastal systems during stepped transgression punctuated by episodes of abrupt water deepening. A Shoreline trajectory steeper than the fluvial profile; transgressive fluvial and coastal systems aggrade and may be preserved below the transgressive surface of erosion (TSE). B Fluvial profile steeper than the shoreline trajectory; the TSE reworks and replaces the subaerial unconformity, and no transgressive fluvial and coastal systems are preserved below the TSE. Both A and B: the episodes of flooding may or may not be followed by short-term stages of progradation. In the former case, the TSE is a composite of multiple lower rank TSEs, and the backstepping parasequences are high-frequency genetic stratigraphic sequences; in the latter case, the TSE is a single physical surface, and the backstepping parasequences are bedsets within the transgressive systems tract. The TSEs depicted in this diagram are wave-ravinement surfaces. Where the backstepping coastal system is an estuary, tidalravinement surfaces may also form at the base of the estuary-mouth complex. BSFR basal surface of forced regression; RSME regressive surface of marine erosion; SU subaerial unconformity; CC correlative conformity; MRS maximum regressive surface; MFS maximum flooding surface; HST highstand systems tract; FSST falling-stage systems tract; LST lowstand systems tract; TST transgressive systems tract.

2 2 1 1

MFS TSE MRS CC SU RSME BSFR

TST LST FSST HST

Nonmarine Marine healing phase Nonmarine topset Marine Sharp-based shoreface Shelf Nonmarine topset Marine

TSTshoreline: Backstepping coastal system Prograding coastline-shoreface Coastal onlap

surfaces of lower hierarchical rank separated by the prograding facies (Fig. 23.66). In this situation, the shortterm transgressions and regressions define high-frequency genetic stratigraphic sequences within the larger scale transgressive systems tract. Where the transgression is not interrupted by shorter term stages of progradation, the transgressive surface of erosion forms as a single, through-going physical surface at the hierarchical level of the associated transgression (Fig. 23.66). Diagnostic to the transgressive surface of erosion is the presence of backstepping estuary-mouth complex (in the case of tidal-ravinement surfaces) or marine ‘healing-phase’ (in the case of wave-ravinement surfaces) deposits on top (Posamentier and Allen, 1993, 1999; Catuneanu, 2006). The occurrence of a transgressive lag and/or a concentration of onlapping shell beds immediately above the contact is common (Fig. 23.67; Kidwell, 1991; Zecchin and Catuneanu, 2013). Notably, the facies preserved below a transgressive surface of erosion may range from continental to marine and therefore are not diagnostic to the identification of this surface. Regressive surface of marine erosion The regressive surface of marine erosion (Plint, 1988) is an unconformity that forms by means of wave scouring during forced regression, as a result of the lowering of the wave base in parallel to the fall in relative sea/ lake level (Figs 23.57 and 23.68). This subaqueous unconformity youngs in a downdip direction, with a rate of diachroneity that matches the rate of shoreline regression (Catuneanu, 2006). The oldest portion of the regressive surface of marine erosion invariably reworks part of the basal surface of forced regression, thus becoming a

Regional Geology and Tectonics

660

23. Sequence stratigraphy

FIGURE 23.67 Wave-ravinement surface (arrow), commonly overlain by transgressive lags and/or onlapping shell beds (Eocene, Talara Basin, Peru). The amount of erosion associated with the process of wave ravinement is limited to the depth of the fairweather wave base of the transgressive shoreface. The thickness of the transgressive lags and/or onlapping shell beds is variable, depending on wave energy and the nature of the substrate that is being eroded, commonly in a range of cm to dm, but exceptionally in a range of metres (e.g. Catuneanu and Biddulph, 2001).

Shoreface

Shelf

FIGURE 23.68 Regressive surface of marine erosion (arrow) at the limit between shelf facies and the overlying ‘sharp-based’ forced regressive shoreface (Triassic upper Potrerillos Formation, Cuyana Basin, Argentina). Diagnostic for the regressive surface of marine erosion is the preservation of ‘sharp-based’ prograding shoreface deposits on top.

systems tract boundary (Fig. 23.34); for this reason, the regressive surface of marine erosion is a sequence stratigraphic surface. The degree of preservation of the basal surface of forced regression on the shelf depends on the gradient of the forced regressive shoreline trajectory relative to the seafloor gradient. Where the shoreline trajectory is steeper than the seafloor gradient (unlike the situation depicted in Fig. 23.34), the basal surface of forced regression may be entirely reworked and replaced by the regressive surface of marine erosion within the shelf setting. The amount of scouring associated with the formation of the regressive surface of marine erosion is proportional to the magnitude of relative sea/lake-level fall. Alternative terms include ‘regressive ravinement surface’ (Galloway, 2001) and ‘regressive wave ravinement’ (Galloway, 2004). Diagnostic to the regressive surface of marine erosion is the presence of ‘sharp-based’ prograding shoreface deposits on top (Plint, 1988; Figs 23.58 and 23.68), whose maximum thickness is constrained by the depth of the fairweather wave base (Catuneanu, 2006). Below this surface, there is typically an abrupt shift to finer-grained shelf facies that may belong to either the falling-stage or the highstand systems tracts (Fig. 23.34).

Regional Geology and Tectonics

Sequence stratigraphic framework

661

Three-dimensional stratigraphic architecture All sequence stratigraphic surfaces are potentially diachronous along dip and/or strike directions (see summary in Fig. 7.31 of Catuneanu, 2006). Surfaces that form in relation to the four events in Fig. 23.3 are prone to be diachronous along strike, due to changes in the rates of accommodation and sedimentation along the shoreline. Among these surfaces, those that form independently of sedimentation (i.e. the basal surface of forced regression at the onset of relative sea-level fall, and the correlative conformity at the end of relative sea-level fall) are typically closer to time lines. In contrast, surfaces associated with transgression (i.e. the maximum regressive surface at the onset of transgression, and the maximum flooding surface at the end of transgression) are typically more diachronous, as their timing depends both on accommodation and sedimentation. Surfaces that form during stages of forced regression (i.e. the regressive surface of marine erosion, and some subaerial unconformities; Fig. 23.34A) or transgression (i.e. the transgressive surface of erosion, and some subaerial unconformities; Fig. 23.34B) are invariably diachronous along dip, with rates that reflect the rates of shoreline shift. The diachronous development of sequence stratigraphic surfaces along dip and/or strike directions implies that different types of systems tracts can accumulate at the same time within a sedimentary basin. A common example of stratigraphic variability is the coeval progradation and retrogradation along a shoreline, as documented both in the geological record (e.g. the ammonite work of Gill and Cobban, 1973, along the proximal shoreline of the Cretaceous Western Interior seaway) and in the present-day environment (e.g. Figs 23.69 and 23.70). Changes in the direction of shoreline shift along the same coastline are more common than depicted by the early sequence stratigraphic models and reflect along-strike variations in the rates of subsidence and sedimentation. As a consequence, different systems tracts may form at the same time in juxtaposition, separated by structural elements (e.g. active faults that limit subbasins with different subsidence rates; Fig. 23.69), sediment entry points (e.g. deltas; Fig. 23.70), or merely the points of balance between accommodation and sedimentation along the coastline. More evident in the latter case, conditions are favourable for the development of diachronous sequence stratigraphic surfaces, as the location of the points of balance between accommodation and sedimentation may shift through time along the coastline (Catuneanu et al., 1998a). Diachronous surfaces may be expressed as discrete contacts (e.g. a maximum flooding surface that can be mapped as a single physical surface; Ito and O’Hara, 1994), or as composite surfaces that consist of staggered higher frequency contacts of the same type (e.g. a regional regressive surface of marine erosion that consists of multiple regressive surfaces of

30 km

A

B Low subsidence

High subsidence

C

FIGURE 23.69 Active fault separating an area of low subsidence and coastal progradation (A delta; B prograding strandplain) from an area of high subsidence and coastal backstepping (C transgressive lagoon-barrier island system) (Pennar Basin, eastern India).

Regional Geology and Tectonics

662

23. Sequence stratigraphy

Pro

gra din (S g sho > A re line )

Backstepping shoreline (A > S)

FIGURE 23.70 Coeval transgression and regression along the coastline of the Adriatic Sea, Italy, driven by differences in sediment supply. The sediment derived from the Po River delta is transported to the South by longshore currents, leading to progradation to the South and retrogradation to the North. HNR highstand normal regression; T transgression; S sedimentation rate; A accommodation rate (i.e. rate of relative sealevel rise).

Sediment supply

10 km

marine erosion of lower hierarchical rank: Plint and Nummedal, 2000; or a regional transgressive waveravinement surface that consists of multiple wave-ravinement surfaces of lower hierarchical rank: Magalhaes et al., 2015). At larger scales, contrasting trajectories may also be observed between the proximal and distal shorelines of an interior seaway, due to regional variations in accommodation and sedimentation across the basin. For example, retroarc foreland systems are characterized by out-of-phase flexural tectonics between the foredeep and the forebulge flexural provinces, which result in ‘reciprocal’ stratigraphic frameworks across the flexural hingelines (e.g. Catuneanu et al., 1999). In such settings, tectonism exerts a fundamental control on the development of proximal versus distal reciprocal stratigraphies, which refer to out-of-phase depositional sequences and subaerial unconformities in overfilled forelands, and the coeval deposition of different conventional systems tracts in underfilled forelands (e.g. summaries in Catuneanu et al., 1999; Catuneanu, 2019c). Contrasting stratigraphic frameworks between different subbasins of a sedimentary basin may also develop in any other tectonic setting, in relation to variations in subsidence and sedimentation regimes. The 3D architecture of the sequence stratigraphic framework is well documented (e.g. Wehr, 1993; Martinsen and Helland-Hansen, 1995; Catuneanu et al., 1998b, 1999, 2002; Catuneanu, 2004b, 2006; Helland-Hansen and Hampson, 2009). As summarized by Catuneanu and Zecchin (2016, p. 189), ‘Variations in accommodation and sedimentation conditions along a shoreline explain the diachroneity of sequence stratigraphic surfaces and the coeval deposition of different systems tracts along strike’ (Catuneanu et al., 1998a; Posamentier and Allen, 1999; Catuneanu, 2006; Csato and Catuneanu, 2014). Nevertheless, the sequence stratigraphic methodology is still typically illustrated with 1D and 2D data sets (e.g. well data, outcrop profiles, seismic lines) that afford the observation of stratal stacking patterns that are diagnostic to the definition of stratal units and bounding surfaces (Figs 23.34, 23.62 and 23.63). The 3D variability of the sequence stratigraphic framework does not change the methodological workflow (Figs 23.71 and 23.72), which commonly starts with the analysis of 1D and/or 2D data. Once sufficient 1D and 2D data sets are observed and interpreted, the full 3D architecture can be assembled by integrating the entire information available. The strategy of selection of the most relevant 2D sections within a 3D data set depends on the depositional setting under analysis. In the case of coastal to shallow-water systems, it is advisable to start the analysis with the interpretation of dip-oriented 2D profiles, which afford the observation of clinoforms, stratal terminations and stacking patterns, followed by the integration of the 2D profiles into the 3D stratigraphic architecture (e.g. Catuneanu et al., 1998a). The construction of the 3D stratigraphic framework also assumes integration of all other types of data available, among which the fossil record plays a particularly important role in terms of constraining palaeo-ecological conditions and age-stratigraphic relationships (Fig. 23.73). In the case of depositional systems dominated by dip-oriented sediment fairways (e.g. fluvial incised valleys, or deepwater gravity-driven processes), it is advisable to start the analysis with strike-oriented lines, which afford the best visualization of the

Regional Geology and Tectonics

Sequence stratigraphic framework

663

Sequence stratigraphic methodology (data-driven construction of sequence stratigraphic frameworks)

Diagnostic

Systems tracts, sequence stratigraphic surfaces

Nondiagnostic

Depositional trends that are common among systems tracts

Calibrated

Practical applications, realistic

Uncalibrated

Theoretical applications, potentially unrealistic

Observation of stratal stacking patterns

Interpretation of underlying controls

Sequence stratigraphic modelling (model-driven testing of controls on sequence development)

FIGURE 23.71

Sequence stratigraphic modelling versus methodology. The construction of a sequence stratigraphic framework is based on the observation of stratal stacking patterns, irrespective of the interpretation of the underlying controls. Modelling may be used to test the possible controls on sequence development, but it does not play any role in the methodological workflow. In downstream-controlled settings, diagnostic stacking patterns relate to shoreline trajectories, whereas nondiagnostic stacking patterns refer to depositional trends that can accompany the formation of any systems tract (e.g. the aggradation of fluvial topsets). In upstream-controlled setting, diagnostic stacking patterns relate to the dominant depositional elements (e.g. channels vs floodplains in fluvial systems), irrespective of fluvial style or the interpreted accommodation conditions at syn-depositional time. Muddling the distinction between modelling and methodology leads to unnecessary confusion and even a reversal of the progress made in the development of sequence stratigraphy as a data-driven methodology.

1. Model-independent methodology

2. Model-dependent choices

Stratal stacking patterns (facies, geometries), bounding surfaces

Choice of surface(s) that should be selected as ‘sequence boundaries’

Framework of systems tracts and sequence stratigraphic surfaces

Delineation of specific types of stratigraphic ‘sequence’

FIGURE 23.72 Methodological workflow of sequence stratigraphy. Stratal stacking patterns provide the basis for the definition of all units and surfaces of sequence stratigraphy. The sequence stratigraphic methodology is independent of scale; the same types of stratal stacking patterns may be observed at different scales, depending on the resolution of the data available and/or the purpose of the study. Source: Modified from Catuneanu, O., Galloway, W.E., Kendall, C.G.St.C., Miall, A.D., Posamentier, H.W., Strasser, A., et al., 2011. Sequence stratigraphy: methodology and nomenclature. Newsl. Stratigr. 44 (3), 173 245.

sediment-transport systems. Where 3D seismic volumes are available, the strike-oriented 2D images are best complemented by the observation of depositional elements in plan view (i.e. seismic geomorphology on horizon slices; Posamentier, 2004; Davies et al., 2007). The 3D variability of the stratigraphic architecture can be observed at different scales. A sequence stratigraphic framework constructed at a specific scale of observation (i.e. as constrained by the scope of the study and/or the resolution of the data available) reflects the physical and temporal relationships of stratal units and bounding surfaces that develop at that particular scale. The degree of diachroneity of the stratigraphic framework may vary with the scale of observation (i.e. hierarchical level), depending on the variability in the development of stratal stacking patterns at different scales. The 3D nature of the stratigraphic architecture demonstrates that sequence stratigraphic frameworks are basin-specific, reflecting the importance of local controls on accommodation and sedimentation. This reiterates the fact that the methodology needs to be decoupled from global standards, and be based on data acquired from within the basin under analysis.

Regional Geology and Tectonics

664

23. Sequence stratigraphy

Stratal geometries (stratal terminations, stratal architecture)

Calibration

Palaeo-environments (regional data, facies data, seismic geomorphology)

Calibration

Vertical profiles (e.g., fining- vs coarsening-upward)

Stratal stacking patterns (physical architecture, relative chronology)

Sequence stratigraphic framework (sequences, systems tracts, bounding surfaces)

Bounding surfaces (timing, hiatuses, diachroneity)

Calibration

Age data (biostratigraphy, radiochronology)

Calibration

Stratal units (timing, duration, diachroneity)

FIGURE 23.73 Construction of the sequence stratigraphic framework, based on the integration and mutual calibration of independent data sets. The significance of stratal stacking patterns is best constrained by placing stratal geometries and vertical trends in a palaeogeographic context. Not all types of data may be available for sequence stratigraphic analysis. The reliability of the constructed framework depends on the amount and quality of the data available. In offshore ‘frontier’ basins (i.e. prior to drilling), where only seismic data are available, stratal stacking patterns are observed at scales above the seismic resolution, on the basis of seismic reflection terminations and architecture (i.e. seismic stratigraphy). Age data are not required to identify stratal stacking patterns at specific locations (e.g. in outcrops or boreholes) or at larger scales (e.g. on 2D seismic lines or well-log cross-sections). Time control enhances the reliability of correlations, but the lack thereof (e.g. in most Precambrian and many Phanerozoic case studies) does not prevent the application of sequence stratigraphy. In the absence of age data, regional stratigraphic markers (e.g. volcanic ash beds, coal seams, maximum flooding surfaces) may provide physical reference horizons to constrain the correlations. The end result of the sequence stratigraphic methodology is the construction of a framework of systems tracts and bounding surfaces at the scales afforded by the data available.

Hierarchy in sequence stratigraphy Hierarchy refers to the classification of stratal units and bounding surfaces on the basis of their absolute or relative scales. The classification of stratigraphic elements that develop at different scales is necessary in order to define the relationship between the nested cycles of the sequence stratigraphic framework. Several hierarchy systems have been proposed since the 1970s, based on criteria that emphasize different attributes of sequences, from their temporal scales (e.g. Vail et al., 1977b, 1991) to their physical features (e.g. Embry, 1995) or their ‘relatively conformable’ character (e.g. Mitchum and Van Wagoner, 1991; Sprague et al., 2003). None of these hierarchy systems has received universal acceptance or validation by data in all depositional and tectonic settings. Hierarchy systems typically require a reference (i.e. an ‘anchor’), relative to which smaller or larger units can be defined. Classifications that employ a nomenclature based on hierarchical orders typically use the ‘first-order’ units and bounding surfaces as the anchor for the hierarchy system. In sedimentology, ‘first order’ designates the smallest units and bounding surfaces that develop at bedform scales, and the hierarchy ranks scale up from bedforms to macroforms, architectural elements, and depositional systems (e.g. Miall, 1996). In stratigraphy, ‘first order’ designates the largest genetic units and bounding surfaces of a sedimentary basin, and the hierarchy ranks scale down to progressively smaller stratigraphic cycles, as far as permitted by the resolution of the data available (e.g. Vail et al., 1977b, 1991; Embry, 1995; Posamentier and Allen, 1999; Catuneanu, 2006). The contrast between sedimentological and stratigraphic approaches reflects the different types of data that are commonly used to classify sedimentary cycles. Sedimentological classifications are based on the observation of facies (e.g. in outcrops, core, or modern environments), whereas the reconstruction of the larger scale stratigraphic architecture requires more regional data, such as seismic profiles. The latter approach is particularly evident in frontier basins where only seismic data are available (e.g. offshore basins prior to drilling). The construction of stratigraphic frameworks in mature basins, where all types of data are available, can be approached from both ends, either starting from sedimentological units (e.g. beds and bedsets) and scaling up or starting from the basin fill and scaling down. Anchors for a hierarchy system can also be selected at intermediate

Regional Geology and Tectonics

Hierarchy in sequence stratigraphy

665

scales (e.g. shoreline shelf-transit cycles, or ‘relatively conformable’ sequences), in which case the designation of hierarchical ranks is more logically based on names (e.g. parasequence, sequence, composite sequence, etc.) rather than numerical orders (i.e. a reference selected at an intermediate scale cannot be of ‘first’ order). Irrespective of approach, a fundamental fact is that stratigraphic frameworks are basin-specific (e.g. potentially nothing in common between an interior basin and a coeval continental margin), reflecting the unique accommodation and sedimentation conditions of each tectonic and depositional setting. Therefore, a system of classification of stratal units and bounding surfaces that can be applied universally must be independent of any features that are specific to a tectonic or depositional setting. For example, the shelf-transit cycle of a shoreline at Milankovitch scales of 104 105 yrs (Burgess and Hovius, 1998; Porebski and Steel, 2006; Ainsworth et al., 2018) may provide a reference for the definition of smaller and larger units in shelf settings dominated by climate cycles, but may have no relevance or expression in overfilled basins or ramp settings controlled tectonic cycles (Catuneanu and Elango, 2001; Martins-Neto and Catuneanu, 2010). Given the uniqueness of each sedimentary basin, stratigraphic features observed at intermediate scales fail to provide a reproducible anchor for the classification of stratigraphic cycles worldwide. The only anchor for stratigraphic classification that can be used universally in all geological settings, and in both frontier and mature basins (i.e. irrespective of tectonic setting, depositional setting, and the types and resolution of the data available), is the ‘first-order’ sedimentary basin cycle, relative to which stratigraphic cycles of lower hierarchical ranks can be defined in the context of each basin. In this case, sequences of any hierarchical order are basin-specific in terms of origin, timing, scales, and internal makeup.

Approaches to stratigraphic classification: absolute versus relative scales The classification of stratigraphic cycles may be approached from two different perspectives: (1) absolute scales (i.e. temporal duration of cycles) and (2) relative scales (i.e. scale of cycles relative to each other). Both approaches have been used to develop hierarchy systems for sequence stratigraphic units and bounding surfaces. Classifications based on absolute scales have been applied since the dawn of modern sequence stratigraphy (e.g. Vail et al., 1977b, 1991; Krapez, 1996; Duval et al., 1998). Even though each of the several controls on stratigraphic cyclicity may record natural periodicities (Miall, 2010; Fig. 23.9), this approach is both impractical and artificial, as shown by two significant pitfalls: (1) age data are not always available, and the duration of stratigraphic cycles becomes increasingly difficult to measure in older successions with the decline in stratigraphic resolution; and (2) the scales of stratigraphic cycles are basin-specific, reflecting the interplay of multiple local and global controls. The scales of any specific natural processes may or may not be expressed in a local stratigraphic framework, as several processes with different natural periodicities interplay to generate a unique architecture. As a result, the scale of stratigraphic cycles worldwide is statistically random across a continuum of time spans and physical dimensions (Carter et al., 1991; Drummond and Wilkinson, 1996). No natural breaks can be replicated in all sedimentary basins, so any arbitrary limits (e.g. 1 My between third- and fourth-order cycles; Vail et al., 1977b) are bound to be artificial. Numerous case studies demonstrate the variability of the stratigraphic record across the entire spectrum of scales (e.g. Posamentier et al., 1992b; Thorne, 1995; Fouke et al., 2000; D’Argenio, 2001; Van Wagoner et al., 2003; Schlager, 2004, 2010; Miall, 2010; Csato et al., 2014). This is further enhanced by the fact that the rates and periodicity of natural processes changed over time, from the Precambrian to the Phanerozoic (e.g. Eriksson et al., 2004, 2005, 2013; Catuneanu et al., 2005, 2012). Ultimately, stratigraphic frameworks are basin-specific in terms of scales and architecture, reflecting unique combinations of local and global controls on accommodation and sedimentation. Classifications based on relative scales require a reference unit (‘anchor’) relative to which larger or smaller units can be defined, irrespective of their absolute temporal and physical scales. The ‘relatively conformable’ sequence of seismic stratigraphy (Mitchum, 1977) introduced a reference relative to which smaller units (i.e. parasequences) and larger units (i.e. sequence sets, composite sequences, composite sequence sets, and megasequences) have been defined (Sprague et al., 2003). The flaw of this approach is that ‘relatively conformable’ sequences can be observed at different scales, depending on the resolution of the data available (e.g. ‘relatively conformable’ sequences at seismic scales are typically unconformable at the subseismic scales of high-resolution sequence stratigraphy; Fig. 23.5). In this context, an increase in the resolution of a stratigraphic study results in a decrease in the scale of ‘relatively conformable’ sequences, which triggers changes to the nomenclature of all previously defined units. Therefore, nomenclature needs to be dissociated from arbitrary variables such as data resolution, to avoid the description of stratigraphic units in superfluous terms. This can be accomplished by selecting

Regional Geology and Tectonics

666

23. Sequence stratigraphy

a reference that is independent of data resolution; that is, the basin fill, which represents the largest genetic unit of any stratigraphic framework. Within the basin fill, the nested architecture of smaller cycles may be unique in terms of temporal and physical scales to each individual basin, and so it needs to be defined on a case-by-case basis. Criteria to discriminate between the relative stratigraphic significance of cycles of different hierarchical ranks have been proposed for specific settings (e.g. Embry, 1995, for rift basins), as well as in a broader sense for all tectonic settings (Catuneanu, 2006, p. 330 334).

Hierarchy systems: approaches to nomenclature In terms of nomenclature of hierarchical ranks, one approach is to assign numerical orders to stratigraphic cycles of different scales (e.g. Vail et al., 1977b, 1991; Embry, 1995; Krapez, 1996; Posamentier and Allen, 1999; Catuneanu, 2003, 2006), although the criteria employed to define the different orders of cyclicity may vary from temporal to physical standards. In this approach, the nomenclature of stratal units and bounding surfaces is consistent at all scales, while their relative stratigraphic significance is indicated by hierarchical orders (Figs 23.18 and 23.20). For example, maximum flooding surfaces can be observed at multiple scales, in relation to transgressions of different magnitudes (Figs 23.19, 23.21 and 23.22). A third-order maximum flooding surface is more significant than a fourth-order maximum flooding surface, but the name of the surface remains the same at all scales of observation (i.e. hierarchical ranks). The same is true for sequences, systems tracts, and depositional systems: for example, a third-order transgression is more significant than a fourth-order transgression, but estuaries and transgressive systems tracts can be observed at both scales (Figs 23.19 and 23.22). Therefore, this approach to nomenclature is independent of scale. An alternative approach is to use different names for units that develop at different scales, starting with the ‘relatively conformable’ sequence as a reference for the hierarchy system (e.g. sequences , sequence sets , composite sequences , composite sequence sets , megasequences; Van Wagoner et al., 1990; Mitchum and Van Wagoner, 1991; Sprague et al., 2003; Neal and Abreu, 2009). This approach leads to ambiguity in the nomenclatural designation of stratal units, as the scale of a ‘relatively conformable’ sequence depends on the resolution of the data available, and hence, it is variable (Fig. 23.5). As a result, the complex terminology promoted by this hierarchy system leads not only to nomenclatural confusion but also to nomenclatural inconsistency between basins or subbasins where different types of data are available (e.g. a seismic-scale ‘sequence’ in one area may be equivalent to a ‘composite sequence’ in another area where well data are available). Another drawback of this hierarchy system is that the classification focuses only on stratal units, but it overlooks the nomenclature of bounding surfaces (e.g. maximum flooding surfaces) that develop at different scales, in relation to stratigraphic cycles of different magnitudes.

Hierarchy systems: orderly versus variable patterns The definition of a ‘relatively conformable’ sequence as an anchor to hierarchy, with different types of unit observed at different scales (Mitchum, 1977; Van Wagoner et al., 1990; Mitchum and Van Wagoner, 1991), implies that the stratigraphic record is organized according to an orderly pattern in which each type of unit has a specific internal makeup (i.e. parasequences are the building blocks of systems tracts and sequences, sequences are the building blocks of sequence sets and composite sequences, etc.; Sprague et al., 2003; Neal and Abreu, 2009; Abreu et al., 2010). This model also ties the stratigraphic architecture to specific accommodation conditions, by linking processes of coastal aggradation to stages of positive accommodation, and the formation of sequence boundaries to stages of negative accommodation (Neal and Abreu, 2009; Abreu et al., 2010). The flaws of these assertions have been pointed out by independent data-based and numerical modelling studies (see Catuneanu, 2019a, for a recent summary). An underlying assumption of the orderly pattern model is that the ‘standard’ sequences of seismic stratigraphy (statistically within the third-order range of cyclicity, by both temporal and physical standards: 105 106 yrs and 101 102 m; Vail et al., 1991, Duval et al., 1998; Schlager, 2010; Figs 23.18 and 23.54) form during full cycles of accommodation and include lowstand systems tracts, whereas any lower rank stratigraphic cycles (i.e. parasequences) form during periods of positive accommodation, therefore missing lowstand deposits (Van Wagoner et al., 1990; Duval et al., 1998). In this approach, hierarchical ranks are tied both to scale and internal makeup, predicting that ‘standard’ seismic-scale sequences include all systems tracts, whereas their building blocks at subseismic scales (i.e. parasequences) consist only of transgressive and highstand deposits. The pitfall of this model

Regional Geology and Tectonics

Hierarchy in sequence stratigraphy

667

is that accommodation cycles are recorded at all scales, starting from the sedimentological scales of tidal cycles, and exposure surfaces are as common as flooding surfaces in the rock record (Vail et al., 1991; Schlager, 2004, 2010; Sattler et al., 2005). Therefore, parasequences may consist of any types of deposit, and the scales of sequences and parasequences are not mutually exclusive; that is, the two types of unit are merely different alternatives for the definition of stratigraphic cycles at the scales of high-resolution sequence stratigraphy (Catuneanu et al., 2010, 2011). As summarized by Schlager (2010), ‘data on sequences of 103 107 years duration, the interval most relevant to practical application of sequence stratigraphy, do not conform well to the ordered-hierarchy model. Particularly unsatisfactory is the notion that the building blocks of classical sequences (approximate domain 105 106 years) are parasequences bounded by flooding surfaces’ (Van Wagoner et al., 1990; Duval et al., 1998). Several pitfalls of the parasequence concept prevent its dependable usage as the building block of sequences and component systems tracts, as originally intended (see discussion in Catuneanu, 2019a). Parasequences have become obsolete with the advent of high-resolution sequence stratigraphy, as sequences that develop at parasequence scales provide a better and more reliable alternative for correlation. As a result, parasequences are no longer part of the methodological workflow of sequence stratigraphy. The nomenclatural and conceptual pitfalls of the orderly pattern model call for a more realistic approach to stratigraphic classification, in line with the natural variability of the sedimentary record. Sequences of all scales can include all or any combinations of systems tracts (e.g. Posamentier et al., 1992a,b; Catuneanu et al., 2011; Csato and Catuneanu, 2012, 2014). This variability requires the construction of stratigraphic frameworks on the basis of local data, free of any model-dependent assumptions (Catuneanu, 2019a,b).

Model-independent hierarchy: basin-specific stratigraphic frameworks Statistical surveys show that sequences in sedimentary basins worldwide are not organized into discrete classes of temporal or physical scales, but are rather part of a stratigraphic continuum (Carter et al., 1991; Drummond and Wilkinson, 1996). This variability of stratigraphic sequences in terms of time spans and physical dimensions is the result of the complex interplay of multiple local and global controls on accommodation and sedimentation. Indeed, the interplay of various processes within a sedimentary basin can alter or override the orderly patterns that may be expected from the natural periodicities of any specific controls on stratigraphic cyclicity (Fig. 23.9). Therefore, the definition of any temporal or physical standards for a universal classification of stratigraphic sequences in all tectonic and depositional settings becomes unrealistic, as it would merely provide an arbitrary subdivision of a stratigraphic continuum (Drummond and Wilkinson, 1996; Schlager, 2010; Catuneanu, 2017). Every sedimentary basin displays a potentially unique architecture of nested stratigraphic cycles, with scales that may or may not follow global standards. Stratigraphic cyclicity is basin- or subbasin-specific, reflecting the importance of local controls on accommodation and sedimentation. The stratigraphic framework of a particular basin reflects the unique evolution of that basin and may differ from the stratigraphic frameworks of other sedimentary basins in terms of timing and duration of cycles, the geometry (thickness, geographic extent) of sequences, and the underlying controls. Stratigraphic frameworks may also differ between subbasins of the same sedimentary basin, as a result of changing accommodation and sedimentation conditions across subbasin boundaries (e.g. Catuneanu et al., 1999, 2002; Catuneanu, 2001, 2004a,b, 2019c; Miall et al., 2008). The natural variability of the stratigraphic record indicates that the classification of sequences and bounding surfaces is best approached on a case-by-case basis (i.e. basin-specific), rather than using global standards or reference cycle charts. The stratigraphic framework of each sedimentary basin needs to be constructed on the basis of local data rather than information extrapolated from other basins. Subsequently, global correlations can still be tested once local frameworks are in place. Within each sedimentary basin, the nested architecture of sequences defines their relative stratigraphic significance (Figs 23.5, 23.18 23.20, 23.22 and 23.74 23.78). In downstreamcontrolled settings, the scale of stratigraphic cycles reflects the scale of observation of coastal depositional systems (Figs 23.19 and 23.22). The cyclicity of shoreline trajectories is commonly expressed in the stratigraphic architecture on either side of the coastline (e.g. cycles of change in the types of gravity flows in deepwater settings: Fig. 23.24; Posamentier and Kolla, 2003; van der Merwe et al., 2010; De Gasperi and Catuneanu, 2014). In upstream-controlled settings, the scale of stratigraphic cycles reflects the scale of observation of depositionalelement associations (e.g. channel- vs floodplain-dominated in fluvial settings) within the same depositional system (Fig. 23.20).

Regional Geology and Tectonics

668

23. Sequence stratigraphy

Depositional sequences Tectonic setting Third order Second order First order

Foreland basin 55 Ma Tectonic quiescence(3)

HST

(1)

64 Ma (2)

Rift basin

Tectonic quiescence (Yacoraite Formation)

66 Ma

HST

71 Ma

TST

73.5 Ma

Tectonic activity (1)

TST

Tectonic activity

MFS WRS

LST

130 Ma

Basement

First-order unconformity Second-order unconformity Third-order unconformity

Sequence hierarchy in the Northwest Basin, Argentina (). Not to scale. Dominant depositional systems: (1)continental; lacustrine-marine; (3)lacustrine. The first-order rift sequence records a duration of 75 My and a thickness of 103 m. The second-order sequences developed over timescales of 106 107 yrs, with thicknesses of 102 103 m. The Yacoraite Formation (73.5 64 Ma, c. 250 m thick) is well exposed and affords the observation of third-order (106 yrs and 101 m scales), fourth-order (105 yrs and c. 10 m scales), and fifth-order (104 yrs and 1 3 m scales) sequences. Fifth-order sequences can be mapped over distances of 101 km; all higher rank sequences can be mapped across the entire basin (102 km scale). Stratigraphic cyclicity was controlled mainly by the interplay of tectonism and climate. The role of tectonism becomes increasingly dominant at larger scales, and it is evident at the third, second and first hierarchical orders. The role of climate is evident at smaller scales (fifth, fourth, and third orders). WRS wave-ravinement surface; MFS maximum flooding surface; LST lowstand systems tract; TST transgressive systems tract; HST highstand systems tract. Source: Data courtesy of GEOMAP Argentina. Modified after Hernandez, R.M., Gomez Omil, R., Boll, A., 2008. Estratigrafia, tectonica y potencial petrolero del rift Cretacico en la provincia de Jujuy. Relatorio del XVII Congreso Geologico Argentino, Jujuy 2008, pp. 207 232.

FIGURE 23.74 (2)

Second-order unconformity

Fourth-order unconformity Fifth-order unconformity

Fourth-order MFS

FIGURE 23.75 Stratigraphic cyclicity within the Yacoraite succession of interbedded lacustrine limestones and shales (Northwest Basin, Argentina). The unconformities originated as subaerial (with evidence of exposure: karstification, desiccation, and pedogenesis), now preserved as wave-ravinement surfaces. MFS maximum flooding surface.

Regional Geology and Tectonics

Hierarchy in sequence stratigraphy

(A)

Third-order TST

Fourth-order sequence

669

Third-order systems tracts (101 m scale) within the Yacoraite Formation, defined by trends of change in bathymetry and the limestone-to-shale ratio. (A) Red arrows indicate fourth-order unconformities. (B) Arrows indicate an angular third-order unconformity (depositional sequence boundary). TST transgressive systems tract; HST highstand systems tract.

FIGURE 23.76

Fourth-order sequence

(B)

Third-order HST

FIGURE 23.77 Fourth-order systems tracts within the Yacoraite Formation, defined by trends of change in bathymetry and the limestone-to-shale ratio over # 10 m scales. MFS maximum flooding surface; WRS/SU wave-ravinement surface reworking and replacing a subaerial unconformity; HST highstand systems tract; TST transgressive systems tract.

MFS

Fourth-order TST

WRS/SU

WRS/SU

Fourth-order HST

MFS

Regional Geology and Tectonics

670

23. Sequence stratigraphy

FIGURE 23.78 Fifth-order sequences and component systems tracts within the Yacoraite Formation, defined by trends of change in bathymetry and the limestone-to-shale ratio over 1 3 m scales. MFS maximum flooding surface; WRS/SU wave-ravinement surface reworking and replacing a subaerial unconformity; HST highstand systems tract; TST transgressive systems tract. WRS/SU

Fourth-order HST

HST WRS/SU Fifth-order sequence

MFS TST

Hierarchical orders may be assigned within the context of each sedimentary basin, starting with the ‘firstorder’ basin fill as a reference (Figs 23.74 23.78). The sedimentary basin cycle defines the largest genetic unit of a stratigraphic framework, and therefore provides a reference that is independent of the types and resolution of the data available. The acquisition of higher resolution data in more advanced studies affords the recognition of smaller scale units and unconformities, without affecting the definition and nomenclature of the larger scale framework already established with lower resolution data. A first-order sequence is the fill of a sedimentary basin that accumulated within a specific tectonic setting (i.e. with accommodation controlled by a related set of subsidence mechanisms; Fig. 23.74). In the case of polyphase basins, first-order sequence boundaries mark changes in the tectonic setting, which are the most significant changes that can be observed within a sedimentary succession (Figs 23.21 and 23.74). For example, the post-Gondwana basins in Colombia include a backarc stage (c. 160 65 Ma) followed by a foreland stage (c. 65 Ma to the present day; Fig. 23.21). These stages should not be lumped together into one first-order sequence simply because they add up to a convenient duration (e.g. in excess of 100 My), nor should they be regarded as second-order sequences because their durations fit an arbitrary time span (e.g. 10 100 My; Vail et al., 1977b). Irrespective of their temporal and physical scales, each stage corresponds to a first-order sequence defined by a specific tectonic setting, and the limit between them is a first-order sequence boundary because it marks a change in the tectonic setting. First-order sequences can be subdivided into lower rank sequences according to the nested architecture that defines their stratigraphic relationships (Figs 23.18 23.20 and 23.74 23.78). In this approach, hierarchical orders have no time or thickness connotations, but only a relative stratigraphic significance in relation to each other within the context of each basin. Where regional data are insufficient to observe the entire basin fill, sequences of different scales within the stratigraphic interval of interest can be referred to in relative terms (e.g. higher vs lower frequency sequences of lower vs higher hierarchical ranks, respectively), without the designation of hierarchical orders (Fig. 23.79). The specification of numerical orders is not required for sequence stratigraphic analysis, as the same methodology applies to all scales of observation; Catuneanu, 2019a,b). The relative ranking of sequences of different scales is defined by their stacking patterns, as lower rank sequences are nested within higher rank systems tracts (Figs 23.18 23.20, 23.22, 23.52 and 23.74 23.79). The possibility that stratigraphic frameworks may correlate from one basin to another in response to global controls on accommodation or sedimentation may not be excluded, particularly in the case of tectonically ‘passive’ settings, but also, it cannot be generalized. The importance of local controls on accommodation and sedimentation is demonstrated by the 3D variability of the sequence stratigraphic framework in terms of the coeval deposition of different systems tracts and the development of diachronous sequence stratigraphic surfaces (Catuneanu, 2006, 2019a). For this reason, hierarchical orders are only meaningful within the context of the sedimentary basin in which they are defined, and their scales and significance may change from one basin to another. For example, ‘third-order’ sequences may occur in any sedimentary basin, but they may differ from one basin to another in terms of timing, duration, thickness, geographic extent, and underlying controls. The ‘third-order’ connotation is only meaningful within the context of the sedimentary basin in which it was defined, relative to the lower and higher rank stratigraphic cycles within the same basin.

Regional Geology and Tectonics

671

Discussion

(A) Architecture of lower frequency (higher rank) depositional sequences and systems tracts NW

SE

Valanginian unconformity

Datum 1

Datum 2

10 km 150 m Late Jurassic unconformity

(B) Architecture of higher frequency (lower rank) depositional sequences and systems tracts SE

NW 11 10

9 8 7

6 5

4

3 2

10 km

1

150 m

Subaerial unconformities and/or correlative conformities Maximum flooding surfaces Maximum regressive and/or wave-ravinement surfaces Basal surfaces of forced regression

Highstand systems tract Transgressive systems tract Lowstand systems tract Falling-stage systems tract

FIGURE 23.79 Sequence stratigraphic framework of the Vaca Muerta Quintuco system (Late Jurassic Early Cretaceous, Neuque´n Basin, Argentina), based on 2D and 3D seismic data calibrated with well data. Line thicknesses are proportional to the hierarchical rank of surfaces (e.g. note depositional sequence boundaries of three magnitudes). The trajectory of the shelf edge defines the first-order shoreline trajectory. At any smaller scales, the shoreline shifted within transit areas across the shelf. Changes in relative sea level modify the efficiency of sediment transfer from the shelf to the deepwater setting, as well as the production of intrabasinal sediment. Therefore, the distinction between lowstand and highstand systems tracts, as well as the recognition of forced regressions and transgressions, remains critically important. The identification of lowstand and highstand systems tracts is based on local stratigraphic relationships (i.e. normal regressions that follow forced regressions vs transgressions, respectively) and not on correlations with the global sea level. Global cycle charts are no longer part of the sequence stratigraphic workflow and methodology (Catuneanu, 2019a). Source: Modified after Dominguez, R.F., Continanzia, M.J., Mykietiuk, K., Ponce, C., Pe´rez, G., Guerello, R., et al., 2016. Organic-rich stratigraphic units in the Vaca Muerta Formation, and their distribution and characterization in the Neuque´n Basin (Argentina). In: Unconventional Resources Technology Conference (URTeC), 2016, pp. 1 12. ,https://doi.org/10.15530-urtec-2016-#2456851.; and Dominguez, R.F., Catuneanu, O., 2017. Regional stratigraphic framework of the Vaca Muerta Quintuco system in the Neuque´n Embayment, Argentina. In: The 20th Geological Congress of Argentina, 7 11 August 2017, Tucuman, Argentina, pp. 1 10.

Statistical data indicate broad trends in the stratigraphic record, whereby most stratigraphic sequences develop at scales of 103 m and 106 108 yrs (first order), 102 103 m and 105 107 yrs (second order), 101 102 m and 104 106 yrs (third order), 100 101 m and 103 105 yrs (fourth order), and # 100 m and 102 104 yrs (fifth order and lower ranks) (Figs 23.18, 23.54 and 23.74 23.78). However, sedimentary basins need to be examined on a case-by-case basis, and exceptions from statistical trends should be considered as a safe norm.

Discussion Sequence stratigraphy in the context of the ‘modelling revolution’ The ‘modelling revolution’ refers to the employment of numerical simulations in stratigraphic research. Forward modelling, in particular, involves a series of assumptions with respect to the mode of sediment transport (e.g. geometric, diffusive, or process-based) and the relative contributions of the tested controls on sequence development (e.g. eustasy, subsidence, sediment supply). The outcome of the model depends on the selection of input parameters, and multiple combinations of input parameters can lead to similar results.

Regional Geology and Tectonics

672

23. Sequence stratigraphy

Uncalibrated modelling can ‘demonstrate’ any stratigraphic scenario, whether realistic or unrealistic; therefore, calibration with real data is essential for any meaningful conclusions (see Catuneanu and Zecchin, 2016, for a discussion). Notwithstanding the potential benefits of numerical modelling, especially when used in conjunction with field data (e.g. Euzen et al., 2004; Rabineau et al., 2005, 2006; Csato et al., 2013, 2015; Leroux et al., 2014), one must remain aware of its limitations and of the difference between modelling and methodology in sequence stratigraphy (Fig. 23.71). Muddling the distinction between the two lines of stratigraphic research leads to unnecessary confusion and even a reversal of the progress made in the development of sequence stratigraphy as a datadriven methodology that relies on observations rather than model-driven assumptions. Numerical modelling plays no role in the sequence stratigraphic methodology, and it may continue indefinitely after the construction of a sequence stratigraphic framework. The separation of modelling and methodology is an important ‘first amendment’ in sequence stratigraphy (Catuneanu and Zecchin, 2016). Mixing the methodology with the interpretation of controls on sequence development was a pitfall since the inception of sequence stratigraphy, which took decades of work to correct. The early models favoured eustasy as the dominant control, which led to the assumption of global correlations in the 1970s and the 1980s, while the role of sediment supply on par with accommodation was only fully recognized starting with the 1990s (i.e. the ‘dual control’ of Schlager, 1993). Between these end-members, any combinations are possible. Linking the methodology to any particular control on sequence development is ultimately misleading, as it is always a combination of controls that defines the stratigraphic architecture. For this reason, the methodology needs to remain neutral with respect to the interpretation of underlying controls, and any reference to a specific control (e.g. ‘tectono-sequence stratigraphy’) must be avoided. The latest trend in numerical modelling is the shift from an overemphasis on accommodation to an overemphasis on sediment supply. This leads to an extreme view whereby all aspects of the stratigraphic architecture can be explained by variations in sediment influx or even solely by autocyclicity. While possible in the virtual world of uncalibrated numerical models, the overemphasis on sediment supply loses touch with reality and brings more confusion than clarification. On practical grounds, this conceptual setback undermines the predictive power of sequence stratigraphy, by downplaying the role of accommodation as an underlying control on the stratigraphic architecture (e.g. downstepping vs upstepping shoreline trajectories). This promotes randomness over predictability (i.e. a ‘chaos theory’), which is both misleading and counterproductive. While any extreme views promoted by uncalibrated numerical models are equally deceptive, the methodology remains grounded on field data and the model-independent observation of stratal stacking patterns and stratigraphic relationships. The pitfalls of uncalibrated modelling can be exacerbated by basic misconceptions incorporated in the numerical models (e.g. confusions between ‘sediment supply’ and ‘sedimentation’, or between ‘accommodation’ and ‘sedimentation’), as well as by the incorrect application of the methodology. For example, the 3D variability of accommodation rates across an underfilled basin makes it necessary to specify that only the relative sea-level changes at the shoreline are relevant to the timing of stratigraphic sequences. Failure to do so (e.g. by employing the curve of relative sea-level changes in the deepwater setting as a reference for the development of sequences) leads to overblown and ultimately false conclusions. Such ‘developments’ leave the practitioner confused and therefore at a loss. Source-to-sink studies expand the scope of numerical modelling to larger scales, by simulating sediment production and the delivery systems that link the source areas to the depocenters. This involves an assessment of extrabasinal sediment sources, weathering efficiency in relation to palaeoclimates, distances and means of sediment transport, and the location of sediment entry points into the marine or lacustrine basins. Intrabasinal sediment sources are equally important, as they explain facies and sedimentation patterns that are unrelated to extrabasinal sediment supply. In this context, the sequence stratigraphic framework provides an anchor for the calibration of model results with field data (Fig. 23.71). The development of source-to-sink models does not change nor replace the need for sequence stratigraphic work. Sequence stratigraphy continues to provide the methodology to analyze the stratigraphic relationships within a sedimentary basin, in a data-driven manner that is independent of model assumptions. The results of sequence stratigraphic analysis provide a reality check for numerical simulations, and need to be used to constrain realistic input parameters for the source-to-sink models. Beyond modelling, sequence stratigraphy remains a data-based methodology centred on the observation of stratal stacking patterns in the stratigraphic record. The sequence stratigraphic framework depicts the architecture of sequences and systems tracts in an objective manner that is independent of the interpretation of the underlying controls. While the sequence stratigraphic methodology follows a standard workflow that leads to the construction of basin-specific frameworks, which reflect the local conditions of accommodation and

Regional Geology and Tectonics

Discussion

673

sedimentation at syn-depositional time, the modelling and testing of the underlying controls on accommodation and sedimentation may continue indefinitely. It is therefore important to recognize methodology and modelling as two independent lines of research that follow different workflows and serve different purposes (Fig. 23.71).

Workflow of sequence stratigraphy The basin-specific nature of the sequence stratigraphic framework makes the acquisition of local data (i.e. collected from the basin under analysis) the essential first step in the sequence stratigraphic workflow. The resolution and reliability of the constructed sequence stratigraphic framework depend on the amount and quality of the data available. It is advisable to acquire as many different types of data as possible, as each type of data contributes with specific insights towards the construction of the sequence stratigraphic framework (Fig. 23.73). Data integration is critical in sequence stratigraphy, as the mutual calibration of different data sets (e.g. outcrop, core, well-log and seismic) leads to the most comprehensive and reliable results (Fig. 23.73). For example, information from scattered outcrops benefits greatly from integration with the continuous subsurface imaging provided by seismic data, wherever possible. At the same time, the use of seismic data without calibration with well data can lead to erroneous interpretations, especially where the architecture of seismic reflections is not diagnostic of any particular depositional system. Similarly, the lack of calibration of well logs with rock data (cuttings, core, or nearby outcrops), and their correlation without the support provided by seismic imaging, can also lead to erroneous interpretations (e.g. similar log motifs may be encountered in different depositional systems). The integration of all available data sets is therefore key to the most effective and reliable application of sequence stratigraphy. Stratal stacking patterns and changes thereof can be identified without time control, both at specific locations (e.g. in individual outcrops or boreholes) and at larger scales (e.g. on 2D seismic lines or well-log cross-sections) (Fig. 23.73). Time control enhances the reliability of correlations, but the lack thereof (e.g. in most Precambrian and many Phanerozoic case studies) does not prevent the application of sequence stratigraphy. The lack of age data may be compensated by the identification of reliable stratigraphic markers (e.g. volcanic ash beds, regional coal seams, or regional maximum flooding surfaces) and/or a good knowledge of the facies relationships within the study area. The minimum amount of data that affords a sequence stratigraphic analysis is typically encountered in offshore ‘frontier’ basins (i.e. prior to drilling), where only seismic data are available. In such cases, stratal stacking patterns are observed at scales above the seismic resolution, on the basis of seismic reflection terminations and architecture (i.e. seismic stratigraphy). The best practice requires a 3D control of the stratigraphic architecture, which integrates the observation of vertical profiles afforded by outcrops and well data with section views (e.g. seismic reflection terminations and architecture on 2D lines) and plan views (e.g. geomorphological features on seismic horizon slices). This approach is possible where sufficient data are available, including 3D seismic surveys. Following the initial lowresolution screening of the available data, smaller areas of interest can be defined for higher resolution studies. Improvements in the quality of subsurface data acquisition and processing afford not only an imaging of the geometry of depositional elements but also insights into depositional processes. Owing to its genetic approach, sequence stratigraphy relies on process sedimentology, which is a prerequisite for a process-based approach to stratigraphy, as well as on integration with other disciplines including all types of classical stratigraphy, geophysics, geomorphology, isotope geochemistry, and basin analysis. For example, the identification of the different types of unconformity requires knowledge of the nature of the overlying facies (e.g. a wave-ravinement surface is defined by the presence of transgressive shallow-water deposits on top). The construction of a sequence stratigraphic framework is best approached from the ‘big picture’ to the detail, as the former provides the context to rationalize the smaller scale elements of the sedimentary record. For example, knowledge of the strike and dip directions within the basin at syn-depositional time helps devising the best strategy to observe shallow-water systems (i.e. as clinoforms are best visualized along dip directions), fluvial incised valley (i.e. best visualized along strike directions), or leveed channels of turbidity flows in a deepwater setting (i.e. best visualized along strike directions). The natural progression in the petroleum exploration of a sedimentary basin, from a ‘frontier’ stage when only seismic data are available to a ‘mature’ stage when well data become available as well, exemplifies the practice of starting with the larger scale stratigraphic framework, and gradually increasing the degree of detail as higher resolution data become available. The workflow of sequence stratigraphy is devised to minimize the risk of errors in the interpretation of data: Step 1 clarification of the tectonic setting, to provide the context for the sequence stratigraphic study (i.e. type

Regional Geology and Tectonics

674

23. Sequence stratigraphy

of sedimentary basin, basin physiography, subsidence mechanisms, dip and strike directions); Step 2 clarification of the depositional setting, to constrain the interpretation of depositional trends (e.g. the meaning of coarsening- vs fining-upward trends, which varies with the depositional system); and Step 3 construction of the sequence stratigraphic framework (i.e. identification of systems tracts and bounding surfaces; Figs 23.71 23.73). Preliminary studies of the tectonic setting (including the location of the study area relative to the main elements of the host basin, such as the basin margin, the shelf edge, etc.) and depositional setting (depositional environments and palaeogeography) ensure that the data are understood in the correct geological context and that the constructed sequence stratigraphic framework honours all available data sets.

Standard methodology and nomenclature The stratigraphic community needs guidelines on methodology and nomenclature, to enable a consistent application of sequence stratigraphy irrespective of geological setting and the types of data available. The same types of stratal units and bounding surfaces can be observed across a wide range of stratigraphic scales, depending on objective (e.g. geological setting), subjective (e.g. purpose of study), and arbitrary (e.g. types of data available) variables (Fig. 23.18). The sequence stratigraphic methodology and nomenclature require consistency at all stratigraphic scales, for an objective approach that is independent of any local variables (Catuneanu, 2019a,b). The sequence stratigraphic framework consists of sequences and component systems tracts, which may be observed at multiple scales (Figs 23.19, 23.20, 23.22 and 23.74 23.78). Systems tracts may be subdivided further into sequence stratigraphic cycles (higher frequency sequences), allostratigraphic cycles (parasequences), or sedimentological cycles (beds and bedsets). The sequence stratigraphic methodology involves the identification of sequence stratigraphic units and bounding surfaces that develop at different scales, based on the observation of stratal stacking patterns (Figs 23.71 23.73). The scales of observation may be defined by the purpose of the study (e.g. larger scales of petroleum exploration vs smaller scales of petroleum production development) and/or the resolution of the data available (e.g. lower resolution seismic data vs higher resolution well-log data). Within a nested architecture of stratigraphic cycles, the relative ranking of sequences is defined by their stacking patterns (Figs 23.20, 23.79 and 23.80). Stratal stacking patterns that define stratigraphic sequences and their component systems tracts in downstream-controlled settings (Fig. 23.32) and upstream-controlled settings (Fig. 23.46) are independent of scale. The same types of stacking patterns can be observed at different scales (i.e. hierarchical levels), depending on the

Ma

SW

NE

1.5 2.0

Pleistocene

1.0

Condensed sections

2.5

(3)

(2)

(1)

3.5 4.0

Pliocene

3.0

4.5 5.0

Turbidites

5.5

Seismic-scale sequences:

Land Shelf Deep water

FIGURE 23.80 Nested architecture of stratigraphic cycles in the deepwater setting of the Gulf of Mexico (see map for location). Sequences of all scales record an increase in sediment supply during deposition, driven by forced regressions of corresponding magnitudes that dominated the icehouse regime of the Plio-Pleistocene. Coeval with the forced regressions on the shelf (i.e. stages of relative sea-level fall), the deepwater setting was subject to higher rates of subsidence and consequent relative sea-level rise. Only the relative sea-level changes at the shoreline are relevant to the timing of the observed sequences. Source: Modified from Weimer, P., 1990. Sequence stratigraphy, facies geometries, and depositional history of the Mississippi Fan, Gulf of Mexico. Bull. Am. Assoc. Pet. Geol. 74, 425 453; Weimer, P., Dixon, B.T., 1994. Regional sequence stratigraphic setting of the Mississippi Fan complex, northern deep Gulf of Mexico: implications for evolution of the northern Gulf basin margin. In: Weimer, P., et al. (Eds.), Submarine Fans and Production Characteristics, Society for Sedimentary Geology (SEPM), Gulf Coast Section, 15th Annual Research Conference, Houston, Texas, pp. 373 381; Madof et al. (2019); Catuneanu, O., 2019b. Scale in sequence stratigraphy. Mar. Pet. Geol. 106, 128 159.

200 km

(1) Sequences with timescales of 106 years (2) Sequences with timescales of 105 years (3) Sequences with timescales of 104–105 years

Regional Geology and Tectonics

Discussion

675

purpose of study and/or the resolution of the data available (e.g. forced regressions: Figs 23.79 and 23.80; normal regressions: Figs 23.19, 23.22 and 23.74 23.79; transgressions: Figs 23.8, 23.19, 23.21, 23.22 and 23.74 23.79; and shoreline-independent, in upstream-controlled settings: Fig. 23.20). The development of stratigraphic stacking patterns is basin-specific, with timing and scales controlled by local conditions of accommodation and sedimentation. Therefore, sequence stratigraphic frameworks are also basin-specific, with a nested architecture of stratigraphic cycles that reflects changes in stratal stacking patterns at different scales. The existence of several competing approaches (Figs 23.3 and 23.4) hindered, for decades, the definition of a standard methodology and the inclusion of sequence stratigraphy in international stratigraphic guides. These competing approaches differ in terms of nomenclature of systems tracts and bounding surfaces (Figs 23.3 and 23.4), the selection of the ‘sequence boundary’ (Fig. 23.3), the classification and nomenclature of stratigraphic cycles, and the assertions of the dominant controls on sequence development (e.g. global eustasy: Vail et al., 1977a,b, 1991; Haq et al., 1987; tectonism: Embry, 1995; accommodation: Neal and Abreu, 2009; or the interplay of accommodation and sediment supply: Schlager, 1993). The standard methodology transcends these differences and relies on the core principles that underlie all competing approaches. The model-independent guidelines are simpler than the requirements of any particular model, thus promoting greater flexibility in the application of the method. Significant progress has been made in outlining the common ground in sequence stratigraphy (Catuneanu et al., 2009, 2010), which led to the publication of formal recommendations by the International Commission of Stratigraphy (Catuneanu et al., 2011): ‘The definition of the common ground in sequence stratigraphy should promote flexibility with respect to the choice of approach that is best suited to a specific set of conditions as defined by tectonic setting, depositional setting, data available, and scale of observation’ (Catuneanu et al., 2011, p. 176); ‘A standard methodology can be defined based on the common ground between the different approaches, with emphasis on the observation of stratal stacking patterns in the rock record’ (Catuneanu et al., 2011, p. 233). It has become clear that none of the competing models provides the ‘best practice’ under all circumstances, as defined by different geological settings and types of data available. The application of sequence stratigraphy requires the acquisition of local data (i.e. derived from the basin under analysis). Global standards (e.g. information derived from other sedimentary basins) should not be used for the construction of local sequence stratigraphic frameworks; conversely, global correlations can be tested once local frameworks are in place. At each scale of observation (i.e. hierarchical level), systems tracts are defined by specific stacking patterns, and changes in the stacking pattern mark the position of sequence stratigraphic surfaces. The construction of a framework of systems tracts and bounding surfaces, at scales selected by the practitioner or afforded by the resolution of the data available, fulfils the practical purpose of sequence stratigraphy. Within this framework, the practitioner can explain and predict the patterns of sediment distribution between the different depositional environments within a sedimentary basin, at the scale(s) of the study. Beyond the construction of a model-independent framework of systems tracts and bounding surfaces, the practitioner can make model-dependent choices with respect to the selection of the ‘sequence boundary’ (Fig. 23.72). Such choices are often guided by the mappability of the different types of surface that are present within the studied section, which depends in part on the data available (e.g. well logs vs seismic lines). The flexibility of this model-independent workflow frees the practitioner from the expectation to fulfil model-driven predictions (e.g. the need to find a specific type of surface as ‘sequence boundary’ or to identify orderly patterns in the stratigraphic record and ideal successions of systems tracts). This promotes objectivity in the construction of sequence stratigraphic frameworks, which may consist of variable successions and combinations of systems tracts (e.g. Csato and Catuneanu, 2012, 2014; Fig. 23.36). A key aspect of the methodology and nomenclature is the scale at which sequences can be defined. In the context of seismic stratigraphy, the definition of a sequence as a ‘relatively conformable succession’ (Mitchum, 1977; Fig. 23.2) inadvertently linked the scale of a sequence to the resolution of the data available. The subsequent definition of other types of stratigraphic cycle at smaller and larger scales (e.g. parasequences, composite sequences, megasequences; Van Wagoner et al., 1988, 1990; Mitchum and Van Wagoner, 1991; Sprague et al., 2003; Neal and Abreu, 2009; Abreu et al., 2010) led to nomenclatural inconsistency, since the scale of the reference unit (i.e. the ‘relatively conformable’ sequence) varies with the resolution of the data available (Fig. 23.5; see discussion in section ‘Stratigraphic Resolution’). In reality, sequences do not occupy any specific niche within a framework of nested stratigraphic cycles. Sequences can be observed at all stratigraphic scales, depending on the geological setting (i.e. local conditions of accommodation and sedimentation), the resolution of the data available (e.g. seismic vs well-log data), and/or the scope of the study (e.g. petroleum exploration vs production development). This also renders parasequences obsolete in a sequence stratigraphic context, as they only provide an allostratigraphic alternative to sequences that develop at similar scales (see Catuneanu, 2019a, for a recent discussion).

Regional Geology and Tectonics

676

23. Sequence stratigraphy

Standardization of sequence stratigraphy is feasible because, despite the variability of the stratigraphic architecture, there is only a limited number of stratal stacking patterns that are diagnostic to the definition of systems tracts and bounding surfaces. The observation of the diagnostic stacking patterns, at scales selected by the practitioner or imposed by the resolution of the data available, affords a standard approach to sequence stratigraphy. The construction of the sequence stratigraphic framework is independent of the interpretation of the underlying controls. Therefore, it is important to separate methodology from modelling in sequence stratigraphy (Catuneanu and Zecchin, 2016; Fig. 23.71). A standard methodology does not prevent future developments in the field of stratigraphic modelling, as the interpretation and testing of the possible controls on sequence development may continue indefinitely following the construction of a sequence stratigraphic framework. Despite the scale invariance of stratal stacking patterns, the stratigraphic architecture is not truly fractal, as sequences of different hierarchical ranks may have different controls and internal makeup of systems tracts (Fig. 23.36). Moreover, there is no correlation between scale and the systems tract composition of sequences; both smaller and larger scale sequences may consist of similar combinations of systems tracts, and sequences of similar scales may vary in terms of number and types of component systems tracts. The scales and the systems tract composition of sequences are basin-specific, reflecting the local conditions of accommodation and sedimentation. Therefore, the observation of stratal stacking patterns takes precedence over any model assumptions in the process of constructing a sequence stratigraphic framework. The methodology and nomenclature must remain independent of scale and all local variables (e.g. geological setting and the resolution of the data available), for a consistent and objective application of sequence stratigraphy.

Conclusions Sequence stratigraphy is a type of stratigraphy that relies on stratal stacking patterns for the definition of stratal units and bounding surfaces. The same types of stacking patterns can be observed at all stratigraphic scales. At each scale of observation, the architecture of sequence stratigraphic units and bounding surfaces describes changes in the distribution of sediment between different depositional environments during the evolution of the sedimentary basin. In building the stratigraphic architecture, sediment is as important as the space that it requires to accumulate. Multiple local and global controls interplay to define the accommodation and sedimentation conditions at the time of deposition (Figs 23.9 and 23.10). The separation between accommodation and sedimentation (i.e. relative sea-level changes vs base-level changes, respectively) is important in downstream-controlled settings, where shoreline trajectories are controlled by the interplay of these two variables. The separation between accommodation and sedimentation becomes less meaningful in upstream-controlled settings, where stratigraphic cyclicity is controlled by base-level changes. The sequence stratigraphic framework records a nested architecture of stratigraphic cycles that develop at different scales. At each scale of observation, depositional systems are the building blocks of systems tracts, and systems tracts are the building blocks of sequences. Sequences and component systems tracts are the stratal units of sequence stratigraphy. A sequence corresponds to a stratigraphic cycle of change in stratal stacking pattern, defined by the recurrence of the same type of sequence stratigraphic surface in the sedimentary record. Systems tracts are subdivisions of sequences, defined by specific stratal stacking patterns and bounding surfaces. Sequence stratigraphic surfaces are palaeo-depositional surfaces that serve, at least in part, as systems tract boundaries. At the smallest stratigraphic scales, systems tracts and component depositional systems consist solely of sedimentological cycles (i.e. beds and bedsets; Figs 23.15 23.18). The scale of the lowest rank systems tracts at any location defines the highest resolution that can be achieved with a stratigraphic study. At any larger stratigraphic scales, systems tracts and component depositional systems consist of higher frequency (lower rank) stratigraphic cycles (i.e. sequences). Stratigraphic cyclicity is basin- or even subbasin-specific in terms of timing and scales, reflecting the importance of local controls on accommodation and sedimentation. There are no standards for the physical or temporal scales of sequences and component systems tracts and depositional systems. The scale of the smallest stratigraphic sequences at any location depends on the geological setting (i.e. local conditions of accommodation and sedimentation). The smallest stratigraphic scale that can be identified at any location depends on the resolution of the data available. For example, the smallest ‘sequence’ that can be identified in the context of

Regional Geology and Tectonics

Discussion

677

seismic stratigraphy is typically not the smallest stratigraphic cycle within the study area, but the smallest stratigraphic cycle that is above the resolution of the seismic data (Fig. 23.5). The inability to identify the smallest sequence in every study indicates that the classification of stratigraphic cycles is best approached from the largest (i.e. the ‘first-order’ basin fill) rather than the smallest stratigraphic cycle as an anchor for the definition of hierarchical ranks. The scale gap between sedimentology and seismic stratigraphy was a ‘grey area’ subject to different approaches to the definition of stratigraphic cycles at subseismic scales. From a sequence stratigraphic perspective, the definition of cycles based on sequence stratigraphic criteria (i.e. changes in stratal stacking pattern), as opposed to allostratigraphic criteria (i.e. lithological discontinuities), provides the adequate solution for a consistent methodological approach. The high-frequency sequences of high-resolution sequence stratigraphy, which develop typically at scales of 100 101 m and 102 105 yrs, render parasequences obsolete. Depositional systems play a key role in the scale-independent definition of sequences and systems tracts. The formation of depositional systems, and implicitly of systems tracts, requires typically minimum timescales of 102 yrs, and it may be sustained for as long as the defining environments are maintained as dominant sediment fairways (Fig. 23.21). Within the transit area of a shoreline, where changes in depositional environment are most frequent, only the lowest rank depositional systems consist strictly of process-related facies accumulated in specific environments; these depositional systems sensu stricto develop commonly at scales below the resolution of seismic stratigraphy (Figs 23.8 and 23.22). At larger scales (higher hierarchical ranks), depositional systems sensu lato reflect dominant depositional trends but may record higher frequency changes in depositional environment (Figs 23.19 and 23.22). The distinction between depositional systems sensu stricto and sensu lato becomes less meaningful outside of the shoreline transit area, where stratigraphic cyclicity may develop without changes in depositional environment (Figs 23.20 and 23.80). Confusions between modelling and methodology triggered unwarranted questions about the ‘future of sequence stratigraphy’. In reality, the future is already here in terms of a standard methodology. Despite the variability of the stratigraphic architecture, which includes a mixture of geometrical and depositional trends (Figs 23.51 and 23.53), there is only a limited number of stratal stacking patterns that are diagnostic to the definition of sequence stratigraphic units and bounding surfaces, which can be observed at all stratigraphic scales. The standard methodology involves the identification of diagnostic stacking patterns, at scales defined by the purpose of study and/or by the resolution of the data available, in a manner that is independent of the interpretation of underlying controls. Therefore, it is important to separate methodology from modelling in sequence stratigraphy, as the testing of underlying controls can continue indefinitely after the construction of a sequence stratigraphic framework. A standard methodology does not prevent future developments in the field of stratigraphic modelling. Scale is a key issue in sequence stratigraphy, with implications for methodology, nomenclature, and practical applications. The scale of observation may be selected by the practitioner according to the purpose of study (e.g. petroleum exploration vs production development), or it may be constrained by local parameters such as geological setting (e.g. local rates of subsidence and sedimentation) and the resolution of the data available. The same types of stratal units and bounding surfaces may be observed at different scales (i.e. hierarchical levels), and the scales associated with any particular hierarchical rank may vary between and within sedimentary basins. This scale variance reflects the natural variability in local accommodation and sedimentation conditions (Fig. 23.9), and the diachronous development of sequence stratigraphic surfaces (Catuneanu et al., 1998a; Catuneanu, 2006, 2019a). For these reasons, the methodology and nomenclature must remain independent of scale and all local variables, for a consistent application of sequence stratigraphy across the entire range of geological settings, stratigraphic scales, and types of data available. The 3D variability of the stratigraphic architecture in terms of timing and scales of stratal units and bounding surfaces justifies the paradigm shift from a model-driven approach, underlain by assumptions of global correlations, to a data-driven approach that honours the reality of each sedimentary basin. Local data take precedence over any model-driven assumptions and lead to the construction of potentially unique sequence stratigraphic frameworks. Despite the nested nature of stratigraphic cycles, the stratigraphic framework is not truly fractal, as sequences of similar or different scales may have different controls and internal architecture of systems tracts. The methodology is now decoupled from global standards and any preconceived assumptions regarding the architecture of the stratigraphic framework and its underlying controls. Global correlations can still be tested following the construction of local frameworks.

Regional Geology and Tectonics

678

23. Sequence stratigraphy

Acknowledgements This work builds on the joint effort that led to the publication of formal recommendations on sequence stratigraphic methodology and nomenclature by the International Subcommission on Stratigraphic Classification (ISSC) of the International Commission on Stratigraphy (Catuneanu et al., 2011). I thank Bill Galloway, Chris Kendall, Andrew Miall, Henry Posamentier, Andre Strasser, and Maurice Tucker for their contributions and support as members of the ISSC task group on sequence stratigraphy. Additional insights were provided by many experts over the years, including V. Abreu, J.P. Bhattacharya, M.D. Blum, I. Csato, R.W. Dalrymple, A.F. Embry, P.G. Eriksson, C.R. Fielding, W.L. Fisher, P. Gianolla, M.R. Gibling, K.A. Giles, J.M. Holbrook, R. Jordan, B. Macurda, O.J. Martinsen, J.E. Neal, D. Nummedal, L. Pomar, B.R. Pratt, J.F. Sarg, K.W. Shanley, R.J. Steel, C. Winker, and M. Zecchin.

References Abbott, S.T., 1998. Transgressive systems tracts and onlap shellbeds from Mid-Pleistocene sequences, Wanganui Basin, New Zealand. J. Sediment. Res. 68, 253 268. Abreu, V., Neal, J.E., Bohacs, K.M., Kalbas, J.L., 2010. Sequence Stratigraphy of Siliciclastic Systems the ExxonMobil methodology. SEPM Concepts in Sedimentology and Paleontology 9, 226 pp. Ainsworth, R.B., Vakarelov, B.K., MacEachern, J.A., Rarity, F., Lane, T.I., Nanson, R.A., 2017. Anatomy of a shoreline regression: implications for the high-resolution stratigraphic architecture of deltas. J. Sediment. Res. 87, 1 35. Ainsworth, R.B., McArthur, J.B., Lang, S.C., Vonk, A.J., 2018. Quantitative sequence stratigraphy. Am. Assoc. Pet. Geol. Bull. 102, 1913 1939. Allen, G.P., Posamentier, H.W., 1993. Sequence stratigraphy and facies model of an incised valley fill: the Gironde estuary, France. J. Sediment. Petrol. 63 (3), 378 391. Amorosi, A., Centineo, M.C., Colalongo, M.L., Fiorini, F., 2005. Millennial-scale depositional cycles from the Holocene of the Po Plain, Italy. Mar. Geol. 222 223, 7 18. Amorosi, A., Ricci Lucchi, M., Rossi, V., Sarti, G., 2009. Climate change signature of small-scale parasequences from Lateglacial-Holocene transgressive deposits of the Arno valley fill. Palaeogeogr. Palaeoclimatol. Palaeoecol. 273, 142 152. Amorosi, A., Bruno, L., Campo, B., Morelli, A., Rossi, V., Scarponi, D., et al., 2017. Global sea-level control on local parasequence architecture from the Holocene record of the Po Plain, Italy. Mar. Pet. Geol. 87, 99 111. Arnott, R.W.C., Zaitlin, B.A., Potocki, D.J., 2002. Stratigraphic response to sedimentation in a net-accommodation-limited setting, Lower Cretaceous Basal Quartz, south-central Alberta. Bull. Can. Pet. Geol. 50 (1), 92 104. Barrell, J., 1917. Rhythms and the measurements of geological time. Geol. Soc. Am. Bull. 28, 745 904. Bartek, L.R., Vail, P.R., Anderson, J.B., Emmet, P.A., Wu, S., 1991. Effect of Cenozoic ice sheet fluctuations in Antarctica on the stratigraphic signature of the Neogene. J. Geophys. Res. 96, 6753 6778. Bartek, L.R., Andersen, J.L.R., Oneacre, T.A., 1997. Substrate control on distribution of subglacial and glaciomarine seismic facies based on stochastic models of glacial seismic facies deposition on the Ross Sea continental margin, Antarctica. Mar. Geol. 143, 223 262. Bassetti, M.A., Berne, S., Jouet, G., Taviani, M., Dennielou, B., Flores, J.A., et al., 2008. The 100-ka and rapid sea level changes recorded by prograding shelf sand bodies in the Gulf of Lions (western Mediterranean Sea). Geochem. Geophys. Geosyst. 9, 1 27. Bastia, R., Radhakrishna, M., Das, S., Kale, A.S., Catuneanu, O., 2010. Delineation of the 85 E ridge and its structure in the Mahanadi Offshore Basin, Eastern Continental Margin of India (ECMI), from seismic reflection imaging. Mar. Pet. Geol. 27, 1841 1848. Bhattacharya, J.P., 1993. The expression and interpretation of marine flooding surfaces and erosional surfaces in core: examples from the Upper Cretaceous Dunvegan Formation in the Alberta foreland basin. In: Summerhayes, C.P., Posamentier, H.W. (Eds.), Sequence Stratigraphy and Facies Associations. International Association of Sedimentologists Special Publication, pp. 125 160., no.18. Blackwelder, E., 1909. The valuation of unconformities. J. Geol. 17, 289 299. Blum, M.D., 1994. Genesis and architecture of incised valley fill sequences: a Late Quaternary example from the Colorado River, Gulf Coastal Plain of Texas. In: Weimer, P., Posamentier, H.W. (Eds.), Siliciclastic Sequence Stratigraphy: Recent Developments and Applications, 58. American Association of Petroleum Geologists Memoir, pp. 259 283. Blum, M.D., Price, D.M., 1998. Quaternary alluvial plain construction in response to glacio-eustatic and climatic controls, Texas Gulf coastal plain. In: Shanley, K.W., McCabe, P.J. (Eds.), Relative Role of Eustasy, Climate, and Tectonism in Continental Rocks, 59. SEPM Special Publication, pp. 31 48. Blum, M.D., To¨rnqvist, T.E., 2000. Fluvial responses to climate and sea-level change: a review and look forward. Sedimentology 47, 2 48. Bohacs, K., Suter, J., 1997. Sequence stratigraphic distribution of coaly rocks: fundamental controls and paralic examples. Am. Assoc. Pet. Geol. Bull. 81 (10), 1612 1639. Bond, G., Broecker, W., Johnsen, S., McManus, J., Labeyrie, L., Jouzel, J., et al., 1993. Correlations between climate records from North Atlantic sediments and Greenland ice. Nature 365, 143 147. Boyd, R., Diessel, C.F.K., Wadsworth, J., Leckie, D., Zaitlin, B.A., 2000. Organization of non marine stratigraphy. In: Boyd, R., Diessel, C.F.K., Francis, S. (Eds.), Advances in the Study of the Sydney Basin. Proceedings of the 34th Newcastle Symposium, University of Newcastle, Callaghan, New South Wales, Australia, pp. 1 14. Bridge, J.S., Leeder, M.R., 1979. A simulation model of alluvial stratigraphy. Sedimentology 26, 617 644. Bristow, C.S., Best, J.L., 1993. Braided rivers: perspectives and problems. In: Best, J.L., Bristow, C.S. (Eds.), Braided Rivers. Geological Society Special Publication, pp. 1 11., No. 75. Broecker, W.S., Andree, M., Wolfi, W., Oeschger, H., Bonani, G., Kennett, J., et al., 2010. The chronology of the last Deglaciation: Implications to the cause of the Younger Dryas Event. Paleoceanography 3, 1 19. Brown Jr., L.F., Fisher, W.L., 1977. Seismic stratigraphic interpretation of depositional systems: examples from Brazilian rift and pull apart basins. In: Payton, C.E. (Ed.), Seismic Stratigraphy Applications to Hydrocarbon Exploration, 26. American Association of Petroleum Geologists Memoir, pp. 213 248.

Regional Geology and Tectonics

References

679

Bruno, L., Bohacs, K.M., Campo, B., Drexler, T.M., Rossi, V., Sammartino, I., et al., 2017. Early Holocene transgressive palaeogeography in the Po coastal plain (northern Italy). Sedimentology 64, 1792 1816. Burgess, P.M., Hovius, N., 1998. Rates of delta progradation during highstands: consequences for timing of deposition in deep-marine systems. J. Geol. Soc. Lond. 155, 217 222. Butcher, S.W., 1990. The nickpoint concept and its implications regarding onlap to the stratigraphic record. In: Cross, T.A. (Ed.), Quantitative Dynamic Stratigraphy. Prentice-Hall, Englewood Cliffs, pp. 375 385. Cantalamessa, G., Di Celma, C., Ragaini, L., 2005. Sequence stratigraphy of the middle unit of the Jama Formation (early Pleistocene, Ecuador): insights from integrated sedimentologic, taphonomic and paleoecologic analysis of molluscan shell concentrations. Palaeogeogr. Palaeoclimatol. Palaeoecol. 216, 1 25. Cantalamessa, G., Di Celma, C., Ragaini, L., Valleri, G., Landini, W., 2007. Sedimentology and high-resolution sequence stratigraphy of the late Middle to Late Miocene Angostura Formation (western Borbon Basin, northwestern Ecuador). J. Geol. Soc. Lond. 164, 653 665. Carter, R.M., Abbott, S.T., Fulthorpe, C.S., Haywick, D.W., Henderson, R.A., 1991. Application of global sea-level and sequence-stratigraphic models in southern hemisphere Neogene strata. In: Macdonald, D.I.M. (Ed.), Sedimentation, Tectonics and Eustasy: Sea-Level Changes at Active Margins, 12. International Association of Sedimentologists Special Publication, pp. 41 65. Carter, R.M., Fulthorpe, C.S., Naish, T.R., 1998. Sequence concepts at seismic and outcrop scale: the distinction between physical and conceptual stratigraphic surfaces. Sediment. Geol. 122, 165 179. Catuneanu, O., 2001. Flexural partitioning of the Late Archaean Witwatersrand foreland system, South Africa. Sediment. Geol. 141 142, 95 112. Catuneanu, O., 2003. Sequence stratigraphy of clastic systems, Geological Association of Canada Short Course Notes, 16. 248 pp. Catuneanu, O., 2004a. Basement control on flexural profiles and the distribution of foreland facies: the Dwyka Group of the Karoo Basin, South Africa. Geology 32, 517 520. Catuneanu, O., 2004b. Retroarc foreland systems evolution through time. J. Afr. Earth Sci. 38, 225 242. Catuneanu, O., 2006. Principles of Sequence Stratigraphy. Elsevier, Amsterdam, 375 pp. Catuneanu, O., 2017. Sequence stratigraphy: guidelines for a standard methodology. In: Montenari, M. (Ed.), Stratigraphy and Timescales, Vol. 2. Academic Press, UK, pp. 1 57. Catuneanu, O., 2019a. Model-independent sequence stratigraphy. Earth-Sci. Rev. 188, 312 388. Catuneanu, O., 2019b. Scale in sequence stratigraphy. Mar. Pet. Geol. 106, 128 159. Catuneanu, O., 2019c. First-order foreland cycles: interplay of flexural tectonics, dynamic loading, and sedimentation. J. Geodyn. 129, 290 298. Catuneanu, O., Biddulph, M.N., 2001. Sequence stratigraphy of the Vaal Reef facies associations in the Witwatersrand foredeep, South Africa. Sediment. Geol. 141 142, 113 130. Catuneanu, O., Bowker, D., 2001. Sequence stratigraphy of the Koonap and Middleton fluvial formations in the Karoo foredeep, South Africa. J. Afr. Earth Sci. 33, 579 595. Catuneanu, O., Elango, H.N., 2001. Tectonic control on fluvial styles: the Balfour Formation of the Karoo Basin, South Africa. Sediment. Geol. 140, 291 313. Catuneanu, O., Sweet, A.R., 1999. Maastrichtian-Paleocene foreland basin stratigraphies, Western Canada: a reciprocal sequence architecture. Can. J. Earth Sci. 36, 685 703. Catuneanu, O., Zecchin, M., 2013. High-resolution sequence stratigraphy of clastic shelves II: controls on sequence development. Mar. Pet. Geol. 39, 26 38. Catuneanu, O., Zecchin, M., 2016. Unique vs. non-unique stratal geometries: relevance to sequence stratigraphy. Mar. Pet. Geol. 78, 184 195. Catuneanu, O., Martins-Neto, M., Eriksson, P.G., 2005. Precambrian sequence stratigraphy. Sediment. Geol. 176, 67 95. Catuneanu, O., Martins-Neto, M.A., Eriksson, P.G., 2012. Sequence stratigraphic framework and application to the Precambrian. Mar. Pet. Geol. 33, 26 33. Catuneanu, O., Sweet, A.R., Miall, A.D., 1997a. Reciprocal architecture of Bearpaw T-R sequences, uppermost Cretaceous, Western Canada Sedimentary Basin. Bull. Can. Pet. Geol. 45 (1), 75 94. Catuneanu, O., Beaumont, C., Waschbusch, P., 1997b. Interplay of static loads and subduction dynamics in foreland basins: reciprocal stratigraphies and the “missing” peripheral bulge. Geology 25 (12), 1087 1090. Catuneanu, O., Willis, A.J., Miall, A.D., 1998a. Temporal significance of sequence boundaries. Sediment. Geol. 121, 157 178. Catuneanu, O., Hancox, P.J., Rubidge, B.S., 1998b. Reciprocal flexural behaviour and contrasting stratigraphies: a new basin development model for the Karoo retroarc foreland system, South Africa. Basin Res. 10, 417 439. Catuneanu, O., Sweet, A.R., Miall, A.D., 1999. Concept and styles of reciprocal stratigraphies: Western Canada foreland basin. Terra Nova 11, 1 8. Catuneanu, O., Sweet, A.R., Miall, A.D., 2000. Reciprocal stratigraphy of the Campanian-Paleocene Western Interior of North America. Sediment. Geol. 134, 235 255. Catuneanu, O., Hancox, P.J., Cairncross, B., Rubidge, B.S., 2002. Foredeep submarine fans and forebulge deltas: orogenic off-loading in the underfilled Karoo Basin. J. Afr. Earth Sci 35 (4), 489 502. Catuneanu, O., Abreu, V., Bhattacharya, J.P., Blum, M.D., Dalrymple, R.W., Eriksson, P.G., et al., 2009. Towards the standardization of sequence stratigraphy. Earth-Sci. Rev. 92, 1 33. Catuneanu, O., Bhattacharya, J.P., Blum, M.D., Dalrymple, R.W., Eriksson, P.G., Fielding, C.R., et al., 2010. Sequence stratigraphy: common ground after three decades of development. First Break. 28, 21 34. Catuneanu, O., Galloway, W.E., Kendall, C.G.St.C., Miall, A.D., Posamentier, H.W., Strasser, A., et al., 2011. Sequence stratigraphy: methodology and nomenclature. Newsl. Stratigr. 44 (3), 173 245. Cloetingh, S., 1988. Intraplate stress: a new element in basin analysis. In: Kleinspehn, K., Paola, C. (Eds.), New Perspectives in Basin Analysis. Springer-Verlag, New York, pp. 205 230. Cloetingh, S., McQueen, H., Lambeck, K., 1985. On a tectonic mechanism for regional sea level variations. Earth Planet. Sci. Lett. 75, 157 166.

Regional Geology and Tectonics

680

23. Sequence stratigraphy

Cloetingh, S., Kooi, H., Groenewoud, W., 1989. Intraplate stresses and sedimentary basin evolution. In: Price, R.A. (Ed.), Origin and Evolution of Sedimentary Basins and Their Energy and Mineral Resources, vol. 48. American Geophysical Union, Geophysical Monographs, pp. 1 16. Cross, T.A., 1991. High-resolution stratigraphic correlation from the perspectives of base-level cycles and sediment accommodation. In: Dolson, J. (Ed.), Unconformity Related Hydrocarbon Exploration and Accumulation in Clastic and Carbonate Settings. Rocky Mountain Association of Geologists, pp. 28 41., Short Course Notes. Csato, I., Catuneanu, O., 2012. Systems tract successions under variable climatic and tectonic regimes: a quantitative approach. Stratigraphy 9 (2), 109 130. Csato, I., Catuneanu, O., 2014. Quantitative conditions for the development of systems tracts. Stratigraphy 11 (1), 39 59. Csato, I., Granjeon, D., Catuneanu, O., Baum, G.R., 2013. A three-dimensional stratigraphic model for the Messinian crisis in the Pannonian Basin, eastern Hungary. Basin Res. 25, 121 148. Csato, I., Catuneanu, O., Granjeon, D., 2014. Millennial-scale sequence stratigraphy: numerical simulation with Dionisos. J. Sediment. Res. 84, 394 406. Csato, I., Toth, S., Catuneanu, O., Granjeon, D., 2015. A sequence stratigraphic model for the Upper Miocene-Pliocene basin fill of the Pannonian Basin, eastern Hungary. Mar. Pet. Geol. 66, 117 134. Dalrymple, R.W., Zaitlin, B.A., Boyd, R., 1992. Estuarine facies models: conceptual basis and stratigraphic implications. J. Sediment. Pet. 62, 1130 1146. D’Argenio, B., 2001. From megabanks to travertines the independence of carbonate rock growth-forms from scale and organismal templates through time. Proc. Int. School Earth Planet. Sci. 109 130. Siena. Davies, R.J., Posamentier, H.W., Wood, L.J., Cartwright, J.A. (Eds.), 2007. Seismic Geomorphology: Applications to Hydrocarbon Exploration and Production. Geological Society, London, Special Publication 277, 274 pp. De Gasperi, A., Catuneanu, O., 2014. Sequence stratigraphy of the Eocene turbidite reservoirs in Albacora field, Campos Basin, offshore Brazil. Am. Assoc. Pet. Geol. (AAPG) Bull. 98 (2), 279 313. Demarest, J.M., Kraft, J.C., 1987. Stratigraphic record of Quaternary sea levels: implications for more ancient strata. In: Nummedal, D., Pilkey, O.H., Howard, J.D. (Eds.), Sea-Level Fluctuation and Coastal Evolution, 41. SEPM Special Publication, pp. 223 239. Di Celma, C., Cantalamessa, G., 2007. Sedimentology and high-frequency sequence stratigraphy of a forearc extensional basin: the Miocene Caleta Herradura Formation, Mejillones Peninsula, northern Chile. Sediment. Geol. 198, 29 52. Dominguez, R.F., Catuneanu, O., 2017. Regional stratigraphic framework of the Vaca Muerta Quintuco system in the Neuque´n Embayment, Argentina. In: The 20th Geological Congress of Argentina, 7 11 August 2017, Tucuman, Argentina, pp. 1 10. Dominguez, R.F., Continanzia, M.J., Mykietiuk, K., Ponce, C., Pe´rez, G., Guerello, R., et al., 2016. Organic-rich stratigraphic units in the Vaca Muerta Formation, and their distribution and characterization in the Neuque´n Basin (Argentina). In: Unconventional Resources Technology Conference (URTeC), 2016, pp. 1 12. ,https://doi.org/10.15530/URTEC-2016-2456851.. D’Orbigny, A., 1842. Paleontologie Francaise. Description zoologique et geologique des tous les animaux mollusques et rayonnes fossiles de France. 2 (Gastropoda). Paris: Victor Masson, 456 pp. Drummond, C.N., Wilkinson, B.H., 1996. Stratal thickness frequencies and the prevalence of orderedness in stratigraphic sequences. J. Geol. 104, 1 18. Duval, B., Cramez, C., Vail, P.R., 1998. Stratigraphic cycles and major marine source rocks. In: De Graciansky, P.C., Hardenbol, J., Jacquin, T., Vail, P.R. (Eds.), Mesozoic and Cenozoic Sequence Stratigraphy of European Basins, 60. Society for Sedimentary Geology, pp. 43 51., Special Publication. Einsele, G., Ricken, W., Seilacher, A. (Eds.), 1991. Cycles and Events in Stratigraphy. Springer-Verlag, 955 pp. Embry, A.F., 1995. Sequence boundaries and sequence hierarchies: problems and proposals. In: Steel, R.J., Felt, V.L., Johannessen, E.P., Mathieu, C. (Eds.), Sequence Stratigraphy on the Northwest European Margin, 5. Norwegian Petroleum Society (NPF), pp. 1 11., Special Publication. Embry, A.F., Johannessen, E.P., 1992. T R sequence stratigraphy, facies analysis and reservoir distribution in the uppermost Triassic-Lower Jurassic succession, western Sverdrup Basin, Arctic Canada. In: Vorren, T.O., Bergsager, E., Dahl-Stamnes, O.A., Holter, E., Johansen, B., Lie, E., Lund, T.B. (Eds.), Arctic Geology and Petroleum Potential, vol. 2. Norwegian Petroleum Society (NPF), pp. 121 146. (Special Publication). Emery, D., Myers, K.J. (Eds.), 1996. Sequence Stratigraphy. Blackwell Science, 297 pp. Eriksson, P.G., Condie, K.C., Tirsgaard, H., Mueller, W.U., Altermann, W., Miall, A.D., et al., 1998. Precambrian clastic sedimentation systems. Sediment. Geol. 120, 5 53. The precambrian Earth: tempos and events. In: Eriksson, P.G., Altermann, W., Nelson, D.R., Mueller, W.U., Catuneanu, O. (Eds.), Developments in Precambrian Geology, 12. Elsevier, Amsterdam, 941 pp. Eriksson, P.G., Catuneanu, O., Nelson, D., Popa, M., 2005. Controls on Precambrian sea-level change and sedimentary cyclicity. Sediment. Geol. 176 (1 2), 43 65. Eriksson, P.G., Mazumder, R., Catuneanu, O., Bumby, A.J., Ilondo, B.O., 2006. Precambrian continental freeboard and geological evolution: a time perspective. Earth-Sci. Rev. 79, 165 204. Eriksson, P.G., Banerjee, S., Catuneanu, O., Corcoran, P.L., Eriksson, K.A., Hiatt, E.E., et al., 2013. Secular changes in sedimentation systems and sequence stratigraphy. Gondwana Res. 24, 468 489. Euzen, T., Joseph, P., du Fornel, E., Guillocheau, F., 2004. Three-dimensional stratigraphic modelling of the Gres d’Annot system, EoceneOligocene, SE France. Geol. Soc. Lond. Spec. Publ. 221 (1), 161 180. Fanti, F., Catuneanu, O., 2010. Fluvial sequence stratigraphy: the Wapiti Formation, west-central Alberta, Canada. J. Sediment. Res. 80, 320 338. Fiduk, J.C., Weimer, P., Trudgill, B.D., Rowan, M.G., Gale, P.E., Phair, R.L., et al., 1999. The Perdido fold belt, Northwestern Deep Gulf of Mexico, part 2: seismic stratigraphy and petroleum systems. Am. Assoc. Pet. Geol. Bull. 83 (4), 578 612. Fielding, C.R., Naish, T.R., Woolfe, K.J., Lavelle, M.A., 2000. Facies analysis and sequence stratigraphy of CRP-2/2A, Victoria Land Basin, Antarctica. Terra Antartica 7, 323 338.

Regional Geology and Tectonics

References

681

Fielding, C.R., Naish, T.R., Woolfe, K.J., 2001. Facies architecture of the CRP-3 drillhole, Victoria Land Basin, Antarctica. Terra Antartica 8, 217 224. Fielding, C.R., Bann, K.L., MacEachern, J.A., Tye, S.C., Jones, B.G., 2006. Cyclicity in the nearshore marine to coastal, Lower Permian, Pebbley Beach Formation, southern Sydney Basin, Australia: a record of relative sea-level fluctuations at the close of the Late Palaeozoic Gondwanan ice age. Sedimentology 53, 435 463. Fielding, C.R., Frank, T.D., Birgenheier, L.P., Rygel, M.C., Jones, A.T., Roberts, J., 2008. Stratigraphic record and facies associations of the late Paleozoic ice age in eastern Australia (New South Wales and Queensland). In: Fielding, C.R., Frank, T.D., Isbell, J.L. (Eds.), Resolving the Late Paleozoic Ice Age in Time and Space. Geological Society of America, pp. 41 57., Special Paper 441. Fisher, W.L., McGowen, J.H., 1967. Depositional systems in the Wilcox group of Texas and their relationship to occurrence of oil and gas. Gulf Coast. Assoc. Geol. Soc. Trans. 17, 105 125. Fouke, B.W., Farmer, J.D., Des Marais, D.J., Pratt, L., Sturchio, N.C., Burns, P.C., et al., 2000. Depositional facies and aqueous-solid geochemistry of travertine-depositing hot springs (Angel Terrace, Mammoth Hot Springs, Yellowstone National Park, USA). J. Sediment. Res. 70, 265 285. Frazier, D.E., 1974. Depositional Episodes: Their Relationship to the Quaternary Stratigraphic Framework in the Northwestern Portion of the Gulf Basin. University of Texas at Austin Bureau of Economic Geology, Geological Circular 71-1, 28 pp. Galloway, W.E., 1989. Genetic stratigraphic sequences in basin analysis, I. Architecture and genesis of flooding-surface bounded depositional units. Am. Assoc. Pet. Geol. Bull. 73, 125 142. Galloway, W.E., 2001. The many faces of submarine erosion: theory meets reality in selection of sequence boundaries. In: AAPG Hedberg Research Conference on “Sequence Stratigraphic and Allostratigraphic Principles and Concepts,” Dallas, 26 29 August, Program and Abstracts Volume, pp. 28 29. Galloway, W.E., 2004. Accommodation and the sequence stratigraphic paradigm. Reserv. Can. Soc. Pet. Geol. 31 (5), 9 10. Gastaldo, R., Denko, T., Liu, Y., 1993. Application of sequence and genetic stratigraphic concepts to carboniferous coal-bearing strata: an example from the Black Warrior Basin, U.S.A. Geol. Rundsch. 82, 212 226. Gill, J.R., Cobban, W.A., 1973. Stratigraphy and Geologic History of the Montana Group and Equivalent Rocks, Montana, Wyoming, and North and South Dakota: United States Geological Survey, Professional Paper 776, 73 pp. Grabau, A.W., 1905. Physical characters and history of some New York formations. Science 22, 528 535. Grabau, A.W., 1906. Types of sedimentary overlap. Geol. Soc. Am. Bull. 17, 567 636. Gressly, A., 1838. Observations geologiques sur le Jura Soleurois. Nouv. Mem. Soc. Helv. Sci. Nat 2, 1 112. Gutierrez Paredes, H.C., Catuneanu, O., Romano, U.H., 2017. Sequence stratigraphy of the Miocene section, southern Gulf of Mexico. Mar. Pet. Geol. 86, 711 732. Hajek, E.A., Heller, P.L., Sheets, B.A., 2010. Significance of channel-belt clustering in alluvial basins. Geology 38, 535 538. Hamilton, D.S., Tadros, N.Z., 1994. Utility of coal seams as genetic stratigraphic sequence boundaries in non-marine basins: an example from the Gunnedah Basin, Australia. Am. Assoc. Pet. Geol. Bull. 78, 267 286. Haq, B.U., Hardenbol, J., Vail, P.R., 1987. Chronology of fluctuating sea levels since the Triassic (250 million years ago to present). Science 235, 1156 1166. Heckel, P.H., 1986. Sea-level curve for Pennsylvanian eustatic marine transgressive-regressive depositional cycles along midcontinent outcrop belt, North America. Geology 14, 330 334. Hedberg, H.D., 1951. Nature of time-stratigraphic units and geologic time units. Am. Assoc. Pet. Geol. Bull. 76-1c, 191 193. Report of Activities. Helland-Hansen, W., Gjelberg, J.G., 1994. Conceptual basis and variability in sequence stratigraphy: a different perspective. Sediment. Geol. 92, 31 52. Helland-Hansen, W., Hampson, G.J., 2009. Trajectory analysis: concepts and applications. Basin Res. 21, 454 483. Helland-Hansen, W., Martinsen, O.J., 1996. Shoreline trajectories and sequences: description of variable depositional-dip scenarios. J. Sediment. Res. 66 (4), 670 688. Hernandez, R.M., Gomez Omil, R., Boll, A., 2008. Estratigrafia, tectonica y potencial petrolero del rift Cretacico en la provincia de Jujuy. Relatorio del XVII Congreso Geologico Argentino, Jujuy 2008 207 232. Hofmann, M.H., Wroblewski, A., Boyd, R., 2011. Mechanisms controlling the clustering of fluvial channels and the compensational stacking of cluster belts. J. Sediment. Res. 81, 670 685. Holbrook, J.M., 1996. Complex fluvial response to low gradients at maximum regression: a genetic link between smooth sequence-boundary morphology and architecture of overlying sheet sandstone. J. Sediment. Res. 66 (4), 713 722. Holbrook, J., Scott, R.W., Oboh-Ikuenobe, F.E., 2006. Base-level buffers and buttresses: a model for upstream versus downstream control on fluvial geometry and architecture within sequences. J. Sediment. Res. 76, 162 174. Holz, M., Kalkreuth, W., Banerjee, I., 2002. Sequence stratigraphy of paralic coal-bearing strata: an overview. Int. J. Coal Geol. 48, 147 179. Homewood, P.W., Eberli, G.P. (Eds.), 2000. Genetic Stratigraphy on the Exploration and Production Scales: Case Studies from the Upper Devonian of Alberta and the Pennsylvanian of the Paradox Basin, 24. Elf Exploration Production Editions Memoirs, 290 pp. Homewood, P.W., Guillocheau, F., Eschard, R., Cross, T.A., 1992. Correlations haute resolution et stratigraphie genetique: une demarche integree. Bull. Cent. Rech. Explor.-Prod. Elf-Aquitaine 16, 357 381. Hunt, D., Tucker, M.E., 1992. Stranded parasequences and the forced regressive wedge systems tract: deposition during base-level fall. Sediment. Geol. 81, 1 9. Hunt, D., Tucker, M.E., 1995. Stranded parasequences and the forced regressive wedge systems tract: deposition during base-level fall reply. Sediment. Geol. 95, 147 160. Hutton, J., 1788. Theory of the Earth: or an investigation of the laws observable in the composition, dissolution, and restoration of land upon the globe. Trans. R. Soc. Edinb. 1, 209 304. Hutton, J., 1795. Theory of the Earth with proofs and illustrations, vols. I and II. Edinburgh.

Regional Geology and Tectonics

682

23. Sequence stratigraphy

Isbell, J.L., Cole, D.I., Catuneanu, O., 2008. Carboniferous-Permian glaciation in the main Karoo Basin, South Africa: stratigraphy, depositional controls, and glacial dynamics. In: Fielding, C.R., Frank, T.D., Isbell, J.L. (Eds.), Resolving the Late Paleozoic Ice Age in Time and Space. Geological Society of America, pp. 71 82., Special Paper 441. Ito, M., O’Hara, S., 1994. Diachronous evolution of systems tracts in a depositional sequence from the middle Pleistocene palaeo-Tokyo Bay, Japan. Sedimentology 41, 677 697. Jameson, R., 1805. A mineralogical description of the County of Dumfries. Edinburgh. Jervey, M.T., 1988. Quantitative geological modeling of siliciclastic rock sequences and their seismic expression. In: Wilgus, C.K., Hastings, B. S., Kendall, C.G.St.C., Posamentier, H.W., Ross, C.A., Van Wagoner, J.C. (Eds.), Sea Level Changes An Integrated Approach, 42. Society of Economic Paleontologists and Mineralogists (SEPM) Special Publication, pp. 47 69. Johnson, J.G., Murphy, M.A., 1984. Time-rock model for Siluro-Devonian continental shelf, western United States. Geol. Soc. Am. Bull. 95, 1349 1359. Kerr, D., Ye, L., Bahar, A., Kelkar, B.G., Montgomery, S., 1999. Glenn Pool Field, Oklahoma: a case of improved prediction from a mature reservoir. Am. Assoc. Pet. Geol. Bull. 83 (1), 1 18. Khidir, A., Catuneanu, O., 2003. Sedimentology and diagenesis of the Scollard sandstones in the Red Deer Valley area, central Alberta. Bull. Can. Pet. Geol. 51 (1), 45 69. Kidwell, S.M., 1991. Condensed deposits in siliciclastic sequences: expected and observed features. In: Einsele, G., Ricken, W., Seilacher, A. (Eds.), Cycles and Events in Stratigraphy. Springer-Verlag, Berlin, pp. 682 695. Kidwell, S.M., 1997. Anatomy of extremely thin marine sequences landward of a passive margin hinge-zone: Neogene Calvert Cliffs succession, Maryland, USA. J. Sediment. Res. 67, 322 340. Kitazawa, T., Murakoshi, N., 2016. Tidal ravinement surfaces in the Pleistocene macrotidal tide-dominated Dong Nai estuary, southern Vietnam. In: Tessier, B., Reynaud, J.-Y. (Eds.), Contributions to Modern and Ancient Tidal Sedimentology: Proceedings of the Tidalites 2012 Conference. International Association of Sedimentologists, John Wiley & Sons, pp. 233 241. Krapez, B., 1996. Sequence-stratigraphic concepts applied to the identification of basin-filling rhythms in Precambrian successions. Aust. J. Earth Sci. 43, 355 380. Leckie, D.A., 1994. Canterbury Plains, New Zealand implications for sequence stratigraphic models. AAPG Bull. 78, 1240 1256. Leckie, D.A., Boyd, R., 2003. Towards a nonmarine sequence stratigraphic model, American Association of Petroleum Geologists Annual Convention, Salt Lake City, vol. 12., p. A101. 11 14 May 2003, Official Program. Leckie, D.A., Wallace-Dudley, K.E., Vanbeselaere, N.A., James, D.P., 2004. Sedimentation in a low-accommodation setting: nonmarine (Cretaceous) Mannville and marine (Jurassic) Ellis Groups, Manyberries Field, southeastern Alberta. Am. Assoc. Pet. Geol. Bull. 88 (10), 1391 1418. Lehrmann, D.J., Goldhammer, R.K., 1999. Secular variation in parasequence and facies stacking patterns of platform carbonates: a guide to application of stacking-patterns analysis in strata of diverse ages and settings. In: Harris, P.M., Saller, A.H., Simo, J.A. (Eds.), Advances in Carbonate Sequence Stratigraphy: Application to Reservoirs, Outcrops and Models, 63. SEPM (Society for Sedimentary Geology) Special Publication, pp. 187 225. Leroux, E., Rabineau, M., Aslanian, D., Granjeon, D., Droz, L., Gorini, C., 2014. Stratigraphic simulations of the shelf of the Gulf of Lions: testing subsidence rates and sea-level curves during the Pliocene and Quaternary. Terra Nova 26, 230 238. Lobo, F.J., Tesson, M., Gensous, B., 2004. Stratal architectures of late Quaternary regressive-transgressive cycles in the Roussillon Shelf (SW Gulf of Lions, France). Mar. Pet. Geol. 21, 1181 1203. Longwell, C.R., 1949. Sedimentary facies in geologic history. Geol. Soc. Am. Mem. 39, 171 pp. Loutit, T.S., Hardenbol, J., Vail, P.R., Baum, G.R., 1988. Condensed sections: the key to age-dating and correlation of continental margin sequences. In: Wilgus, C.K., Hastings, B.S., Kendall, C.G.St.C., Posamentier, H.W., Ross, C.A., Van Wagoner, J.C. (Eds.), Sea Level Changes An Integrated Approach, 42. SEPM Special Publication, pp. 183 213. Lyell, C., 1833. Principles of Geology, Being an Attempt to Explain the Former Changes of the Earth’s Surface, by Reference to Causes now in Operation, vol. 3. John Murray, London, 109 pp. MacEachern, J.A., Dashtgard, S.E., Knaust, D., Catuneanu, O., Bann, K.L., Pemberton, S.G., 2012. Sequence stratigraphy. In: Knaust, D., Bromley, R.G. (Eds.), Trace Fossils as Indicators of Sedimentary Environments, vol. 64. Elsevier, pp. 157 194., Developments in Sedimentology. Madof, A.S., Harris, A.D., Baumgardner, S.E., Sadler, P.M., Laugier, F.J., Christie-Blick, N., 2019. Geology 47 (6), 545 549. Magalhaes, A., Raja Gabaglia, G., Scherer, C., Ballico, M., Guadagnin, F., Bento Freire, E., et al., 2015. Sequence hierarchy in the Mesoproterozoic Tombador Formation, Chapada Diamantina, Brazil. Basin Res. 1 40. Maravelis, A.G., Boutelier, D., Catuneanu, O., Seymour, K.St, Zelilidis, A., 2016. A review of tectonics and sedimentation in a forearc setting: Hellenic Thrace Basin, North Aegean Sea and Northern Greece. Tectonophysics 674, 1 19. Maravelis, A.G., Catuneanu, O., Nordsvan, A., Landerberger, B., Zelilidis, A., 2017. Interplay of tectonism and eustasy during the Early Permian icehouse: Southern Sydney Basin, southeast Australia. Geol. J. 1 32. Available from: https://doi.org/10.1002/gj.2962. Martins-Neto, M.A., Catuneanu, O., 2010. Rift sequence stratigraphy. Mar. Pet. Geol. 27, 247 253. Martinsen, R.S., Christensen, G., 1992. A stratigraphic and environmental study of the Almond Formation, Mesaverde Group, Greater Green River Basin, Wyoming. In: Wyoming Geological Association Guidebook, Forty-third Field Conference, pp. 171 190. Martinsen, O.J., Helland-Hansen, W., 1995. Strike variability of clastic depositional systems: does it matter for sequence stratigraphic analysis? Geology 23, 439 442. Mawson, M., Tucker, M.E., 2009. High-frequency cyclicity (millennial-scale and Milankovitch) in slope carbonates (Zechstein) Permian, NE England. Sedimentology 56, 1905 1936. Menegazzo, M., Catuneanu, O., Chang, H.K., 2016. The South American retroarc foreland system: the development of the Bauru Basin in the back-bulge province. Mar. Pet. Geol. 73, 131 156. Miall, A.D., 1986. Eustatic sea-level change interpreted from seismic stratigraphy: a critique of the methodology with particular reference to the North Sea Jurassic record. Am. Assoc. Pet. Geol. Bull. 70, 131 137.

Regional Geology and Tectonics

References

683

Miall, A.D., 1991. Stratigraphic sequences and their chronostratigraphic correlation. J. Sediment. Petrol. 61, 497 505. Miall, A.D., 1992. The Exxon global cycle chart: an event for every occasion? Geology 20, 787 790. Miall, A.D., 1996. The geology of fluvial deposits: sedimentary facies, basin analysis and petroleum geology. Springer-Verlag Inc, Heidelberg, 582 pp. Miall, A.D., 2010. The geology of stratigraphic sequences, second ed Springer-Verlag, Berlin, 522 pp. Miall, A.D., 2015. Updating uniformitarianism: stratigraphy as just a set of “frozen accidents”. In: Smith, D.G., Bailey, R.J., Burgess, P., Fraser, A. (Eds.), Strata and Time, 404. Geological Society, London, pp. 11 36., Special Publications. Miall, A.D., 2016. The valuation of unconformities. Earth-Sci. Rev. 163, 22 71. Miall, A.D., Catuneanu, O., Vakarelov, B.K., Post, R., 2008. The Western Interior Basin. In: Miall, A.D. (Ed.), The Sedimentary Basins of the United States and Canada. Sedimentary Basins of the World. Elsevier, Amsterdam, pp. 329 362. (Series Editor: K.J. Hsu). Milankovitch, M., 1930. Mathematische klimalehre und astronomische theorie der klimaschwankungen. In: Koppen, W., Geiger, R. (Eds.), Handbuch der Klimatologie, I (A). Gebru¨der Borntraeger, Berlin. Mitchum Jr., R.M., 1977. Seismic stratigraphy and global changes of sea level, part 11: glossary of terms used in seismic stratigraphy. In: Payton, C.E. (Ed.), Seismic Stratigraphy Applications to Hydrocarbon Exploration, 26. American Association of Petroleum Geologists Memoir, pp. 205 212. Mitchum Jr., R.M., Van Wagoner, J.C., 1991. High-frequency sequences and their stacking patterns: sequence stratigraphic evidence of highfrequency eustatic cycles. Sediment. Geol. 70, 131 160. Mitchum Jr., R.M., Vail, P.R., Thompson III, S., 1977. Seismic stratigraphy and global changes of sea level, part 2: the depositional sequence as a basic unit for stratigraphic analysis. In: Payton, C.E. (Ed.), Seismic Stratigraphy Applications to Hydrocarbon Exploration, 26. American Association of Petroleum Geologists Memoir, pp. 53 62. Moore, C.H., 2001. Carbonate reservoirs porosity evolution and diagenesis in a sequence stratigraphic framework, Developments in Sedimentology, 55. Elsevier, 460 pp. Muto, T., Steel, R.J., 2002. Role of autoretreat and A/S changes in the understanding of deltaic shoreline trajectory: a semi-quantitative approach. Basin Res. 14, 303 318. Naish, T., Kamp, P.J.J., 1997. Foraminiferal depth palaeoecology of Late Pliocene shelf sequences and systems tracts, Wanganui Basin, New Zealand. Sediment. Geol. 110, 237 255. Naish, T.R., Woolfe, K.J., Barrett, P.J., Wilson, G.S., Atkins, C., Bohaty, S.M., et al., 2001. Orbitally induced oscillations in the East Antarctic ice sheet at the Oligocene/Miocene boundary. Nature 413, 719 723. Nanson, R.A., Vakarelov, B.K., Ainsworth, R.B., Williams, F.M., Price, D.M., 2013. Evolution of a Holocene, mixed-process, forced regressive shoreline: the Mitchell River delta, Queensland, Australia. Mar. Geol. 339, 22 43. Neal, J., Abreu, V., 2009. Sequence stratigraphy hierarchy and the accommodation succession model. Geology 37 (9), 779 782. Neuendorf, K.K.E., Mehl Jr., J.P., Jackson, J.A. (Eds.), 2005. Glossary of Geology. fifth ed American Geological Institute, Alexandria, VA, 779 pp. Nixon, F.C., England, J.H., Lajeunesse, P., Hanson, M.A., 2014. Deciphering patterns of postglacial sea level at the junction of the Laurentide and Innuitian Ice Sheets, western Canadian High Arctic. Quat. Sci. Rev. 91, 165 183. Nummedal, D., Swift, D.J.P., 1987. Transgressive stratigraphy at sequence-bounding unconformities: some principles derived from Holocene and Cretaceous examples. In: Nummedal, D., Pilkey, O.H., Howard, J.D. (Eds.), Sea-level Fluctuation and Coastal Evolution, 41. Society of Economic Paleontologists and Mineralogists (SEPM) Special Publication, pp. 241 260. Nummedal, D., Riley, G.W., Templet, P.L., 1993. High-resolution sequence architecture: a chronostratigraphic model based on equilibrium profile studies. In: Posamentier, H.W., Summerhayes, C.P., Haq, B.U., Allen, G.P. (Eds.), Sequence Stratigraphy and Facies Associations, 18. International Association of Sedimentologists Special Publication, pp. 55 68. Oppel, A., 1856. Die Juraformation Englands, Frankreichs und des sudwestlichen Deutschlands. Wurttemb. Naturwiss. Verein Jahresh., vol. xii, Stuttgart, 438 pp. Payton, C.E. (Ed.), 1977. Seismic Stratigraphy Applications to Hydrocarbon Exploration, 26. American Association of Petroleum Geologists Memoir, 516 pp. Pellegrini, C., Maselli, V., Gamberi, F., Asioli, A., Bohacs, K.M., Drexler, T.M., et al., 2017. How to make a 350-m-thick lowstand systems tract in 17,000 years: the Late Pleistocene Po River (Italy) lowstand wedge. Geology 45 (4), 327 330. Pellegrini, C., Asioli, A., Bohacs, K.M., Drexler, T.M., Feldman, H.R., Sweet, M.L., et al., 2018. The Late Pleistocene Po River lowstand wedge in the Adriatic Sea: controls on architecture variability and sediment partitioning. Mar. Pet. Geol. 96, 16 50. Peper, T., Cloetingh, S., 1995. Autocyclic perturbations of orbitally forced signals in the sedimentary record. Geology 23, 937 940. Peper, T., Beekman, F., Cloetingh, S., 1992. Consequences of thrusting and intraplate stress fluctuations for vertical motions in foreland basins and peripheral areas. Geophys. J. Int. 111, 104 126. Pitman, W.C., Golovchenko, X., 1988. Sea-level changes and their effect on the stratigraphy of Atlantic-type margins. In: Sheridan, R.E., Grow, J.A. (Eds.), The Geology of North America, vol. I-2, The Atlantic Continental Margin, United States. Geological Society of America, Boulder, CO, pp. 545 565. Plint, A.G., 1988. Sharp-based shoreface sequences and “offshore bars” in the Cardium Formation of Alberta; their relationship to relative changes in sea level. In: Wilgus, C.K., Hastings, B.S., Kendall, C.G.St.C., Posamentier, H.W., Ross, C.A., Van Wagoner, J.C. (Eds.), Sea Level Changes An Integrated Approach, 42. SEPM Special Publication, pp. 357 370. Plint, A.G., Nummedal, D., 2000. The falling stage systems tract: recognition and importance in sequence stratigraphic analysis. In: Hunt, D., Gawthorpe, R.L. (Eds.), Sedimentary Response to Forced Regression, 172. Geological Society of London Special Publication, pp. 1 17. Porebski, S.J., Steel, R.J., 2006. Deltas and sea-level change. J. Sediment. Res. 76, 390 403. Posamentier, H.W., 2004. Seismic geomorphology: imaging elements of depositional systems from shelf to deep basin using 3D seismic data: implications for exploration and development. In: Davies, R.J., Cartwright, J.A., Stewart, S.A., Lappin, M., Underhill, J.R. (Eds.), 3D Seismic Technology: Application to the Exploration of Sedimentary Basins, 29. Geological Society, London, pp. 11 24., Memoir. Posamentier, H.W., Allen, G.P., 1993. Variability of the sequence stratigraphic model: effects of local basin factors. Sediment. Geol. 86, 91 109.

Regional Geology and Tectonics

684

23. Sequence stratigraphy

Posamentier, H.W., Allen, G.P., 1999. Siliciclastic sequence stratigraphy: concepts and applications. SEPM Concepts in Sedimentology and Paleontology (7), 210 pp. Posamentier, H.W., Kolla, V., 2003. Seismic geomorphology and stratigraphy of depositional elements in deep-water settings. J. Sediment. Res. 73 (3), 367 388. Posamentier, H.W., Morris, W.R., 2000. Aspects of the stratal architecture of forced regressive deposits. In: Hunt, D., Gawthorpe, R.L. (Eds.), Sedimentary Responses to Forced Regressions, 172. Geological Society Special Publication, pp. 19 46. Posamentier, H.W., Vail, P.R., 1985. Eustatic controls on depositional stratal patterns. In: Soc. Econ. Paleont. Mineral. (SEPM) Research Conference No. 6, Sea Level Changes An Integrated Approach. Houston, October 20 23, 1985 (Abstract and Poster). Posamentier, H.W., Vail, P.R., 1988. Eustatic controls on clastic deposition. II. Sequence and systems tract models. In: Wilgus, C.K., Hastings, B.S., Kendall, C.G.St.C., Posamentier, H.W., Ross, C.A., Van Wagoner, J.C. (Eds.), Sea Level Changes An Integrated Approach, vol. 42. SEPM Special Publication, pp. 125 154. Posamentier, H.W., Jervey, M.T., Vail, P.R., 1988. Eustatic controls on clastic deposition. I. Conceptual framework. In: Wilgus, C.K., Hastings, B.S., Kendall, C.G.St.C., Posamentier, H.W., Ross, C.A., Van Wagoner, J.C. (Eds.), Sea Level Changes An Integrated Approach, vol. 42. SEPM Special Publication, pp. 110 124. Posamentier, H.W., Allen, G.P., James, D.P., Tesson, M., 1992a. Forced regressions in a sequence stratigraphic framework: concepts, examples, and exploration significance. Am. Assoc. Pet. Geol. Bull. 76, 1687 1709. Posamentier, H.W., Allen, G.P., James, D.P., 1992b. High resolution sequence stratigraphy the East Coulee Delta, Alberta. J. Sediment. Pet. 62 (2), 310 317. Rich, J.L., 1951. Three critical environments of deposition and criteria for recognition of rocks deposited in each of them. Geol. Soc. Am. Bull. 62, 1 20. Qayyum, F., Betzler, C., Catuneanu, O., 2017. The Wheeler diagram, flattening theory, and time. Mar. Pet. Geol. 86, 1417 1430. Rabineau, M., Berne´, S., Aslanian, D., Olivet, J.-L., Joseph, P., Guillocheau, F., et al., 2005. Sedimentary sequences in the Gulf of Lion: a record of 100,000 years climatic cycles. Mar. Pet. Geol. 22, 775 804. Rabineau, M., Berne´, S., Olivet, J.-L., Aslanian, D., Guillocheau, F., Joseph, P., 2006. Paleo sea levels reconsidered from direct observation of paleoshoreline position during Glacial Maxima (for the last 500,000 yr). Earth Planet. Sci. Lett. 252, 119 137. Ramsbottom, W.H.C., 1979. Rates of transgression and regression in the Carboniferous of NW Europe. J. Geol. Soc. Lond. 136, 147 153. Rashid, H., Polyak, L., Mosley-Thompson, E. (Eds.), 2011. Abrupt Climate Change: Mechanisms, Patterns, and Impacts. American Geophysical Union, Geophysical Monograph 193, 242 pp. Reid, S.K., Dorobek, S.L., 1993. Sequence stratigraphy and evolution of a progradational foreland carbonate ramp, Lower Mississippian Mission Canyon Formation and stratigraphic equivalents, Montana and Idaho. In: Loucks, R.G., Sarg, J.F. (Eds.), Carbonate Sequence Stratigraphy, Recent Developments and Applications, 57. American Association of Petroleum Geologists Memoir, pp. 327 352. Sarg, J.F., 1988. Carbonate sequence stratigraphy. In: Wilgus, C.K., Hastings, B.S., Kendall, C.G.St.C., Posamentier, H.W., Ross, C.A., Van Wagoner, J.C. (Eds.), Sea Level Changes An Integrated Approach, 42. SEPM Special Publication, pp. 155 182. Sarkar, S., Banerjee, S., Eriksson, P.G., Catuneanu, O., 2005. Microbial mat control on siliciclastic Precambrian sequence stratigraphic architecture. Sediment. Geol. 176, 195 209. Sarmiento Rojas, L.F., 2001. Mesozoic rifting and Cenozoic basin inversion history of the Eastern Cordillera, Colombian Andes. Ph.D. thesis, Vrije University, Amsterdam, The Netherlands, 319 pp. Sattler, U., Immenhauser, A., Hillgartner, H., Esteban, M., 2005. Characterization, lateral variability and lateral extent of discontinuity surfaces on a carbonate platform (Barremian to Lower Aptian, Oman). Sedimentology 52, 339 361. Saul, G., Naish, T.R., Abbott, S.T., Carter, R.M., 1999. Sedimentary cyclicity in the marine Pliocene-Pleistocene of the Wanganui Basin (New Zealand): sequence stratigraphic motifs characteristic of the past 2.5 m.y. Geol. Soc. Am. Bull. 111, 524 537. Schlager, W., 1992. Sedimentology and sequence stratigraphy of reefs and carbonate platforms. Continuing Education Course Note Series #34, American Association of Petroleum Geologists, 71 pp. Schlager, W., 1993. Accommodation and supply a dual control on stratigraphic sequences. In: Cloetingh, S., Sassi, W., Horvath, F., Puigdefabregas, C. (Eds.), Basin Analysis and Dynamics of Sedimentary Basin Evolution, 86. Sedimentary Geology, pp. 111 136. Schlager, W., 2004. Fractal nature of stratigraphic sequences. Geology 32, 185 188. Schlager, W., 2010. Ordered hierarchy versus scale invariance in sequence stratigraphy. Int. J. Earth Sci. 99, S139 S151. Schumm, S.A., 1993. River response to baselevel change: implications for sequence stratigraphy. J. Geol. 101, 279 294. Shanley, K.W., McCabe, P.J., 1993. Alluvial architecture in a sequence stratigraphic framework: a case history from the Upper Cretaceous of southern Utah, U.S.A. In: Flint, S., Bryant, I. (Eds.), Quantitative Modeling of Clastic Hydrocarbon Reservoirs and Outcrop Analogues, 15. International Association of Sedimentologists, pp. 21 55., Special Publication. Shanley, K.W., McCabe, P.J., 1994. Perspectives on the sequence stratigraphy of continental strata. Am. Assoc. Pet. Geol. Bull. 78, 544 568. Sloss, L.L., 1962. Stratigraphic models in exploration. Am. Assoc. Pet. Geol. Bull. 46, 1050 1057. Sloss, L.L., 1963. Sequences in the cratonic interior of North America. Geol. Soc. Am. Bull. 74, 93 114. Sloss, L.L., Krumbein, W.C., Dapples, E.C., 1949. Integrated facies analysis. In: Longwell, C.R. (Ed.), Sedimentary Facies in Geologic History, 39. Geological Society of America Memoir, pp. 91 124. Smith, W., 1819. Strata identified by organized fossils. London (1816 1819). Sprague, A.R., Patterson, P.E., Sullivan, M.D., Campion, K.M., Jones, C.R., Garfield, T.R., et al., 2003. Physical stratigraphy of clastic strata: a hierarchical approach to the analysis of genetically related stratigraphic elements for improved reservoir prediction. South Tex. Geol. Soc. Bull. 44, 7. Stefani, M., Vincenzi, S., 2005. The interplay of eustasy, climate and human activity in the late Quaternary depositional evolution and sedimentary architecture of the Po Delta system. Mar. Geol. 222 (1), 19 48. Steno, N., 1669. De solido intra solidum naturaliter contento dissertationis prodromus. Firenze. Strasser, A., 2016. Hiatuses and condensation: an estimation of time lost on a shallow carbonate platform. Depos. Rec. 1 (2), 91 117.

Regional Geology and Tectonics

References

685

Strasser, A., 2018. Cyclostratigraphy of shallow-marine carbonates limitations and opportunities. In: Montenari, M. (Ed.), Stratigraphy & Timescales, vol. 3. Academic Press, UK, pp. 151 187. Summerfield, M.A., 1985. Plate tectonics and landscape development on the African continent. In: Morisawa, M., Hack, J. (Eds.), Tectonic Geomorphology. Allen and Unwin, Boston, pp. 27 51. Swift, D.J.P., 1975. Barrier-island genesis: evidence from the central Atlantic shelf, eastern U.S.A. Sediment. Geol. 14, 1 43. Tesson, M., Gensous, B., Allen, G.P., Ravenne, C., 1990. Late Quaternary deltaic lowstand wedges on the Rhoˆne continental shelf, France. Mar. Geol. 91, 325 332. Tesson, M., Posamentier, H.W., Gensous, B., 2000. Stratigraphic organization of Late Pleistocene deposits of the western part of the Rhone shelf (Languedoc shelf) from high resolution seismic and core data. Am. Assoc. Pet. Geol. Bull. 84, 119 150. Thorne, J., 1995. On the scale independent shape of prograding stratigraphic units. In: Barton, C.B., La Pointe, P.R. (Eds.), Fractals in Petroleum Geology and Earth Processes. Plenum Press, New York, pp. 97 112. Udden, J.A., 1912. Geology and mineral resources of the Peoria Quadrangle. U.S. Geol. Surv. Bull. 506, 103 pp. Vail, P.R., Mitchum Jr., R.M., Thompson III, S., 1977a. Seismic stratigraphy and global changes of sea level, part 3: relative changes of sea level from coastal onlap. In: Payton, C.E. (Ed.), Seismic Stratigraphy Applications to Hydrocarbon Exploration, 26. American Association of Petroleum Geologists Memoir, pp. 63 81. Vail, P.R., Mitchum Jr., R.M., Thompson III, S., 1977b. Seismic stratigraphy and global changes of sea level, part 4: global cycles of relative changes of sea level. In: Payton, C.E. (Ed.), Seismic Stratigraphy Applications to Hydrocarbon Exploration, 26. American Association of Petroleum Geologists Memoir, pp. 83 97. Vail, P.R., Audemard, F., Bowman, S.A., Eisner, P.N., Perez-Cruz, C., 1991. The stratigraphic signatures of tectonics, eustasy and sedimentology an overview. In: Einsele, G., Ricken, W., Seilacher, A. (Eds.), Cycles and Events in Stratigraphy. Springer-Verlag, pp. 617 659. van der Merwe, W.C., Flint, S.S., Hodgson, D.M., 2010. Sequence stratigraphy of an argillaceous, deepwater basin-plain succession: Vischkuil Formation (Permian), Karoo Basin, South Africa. Mar. Pet. Geol. 27, 321 333. Van Wagoner, J.C., 1995. Sequence stratigraphy and marine to nonmarine facies architecture of Foreland Basin Strata, Book Cliffs, Utah, U.S. A. In: Van Wagoner, J.C., Bertram, G.T. (Eds.), Sequence Stratigraphy of Foreland Basin Deposits, 64. American Association Petroleum Geologists Memoir, pp. 137 223. Van Wagoner, J.C., Posamentier, H.W., Mitchum, R.M., Vail, P.R., Sarg, J.F., Loutit, T.S., et al., 1988. An overview of sequence stratigraphy and key definitions. In: Wilgus, C.K., Hastings, B.S., Kendall, C.G.St.C., Posamentier, H.W., Ross, C.A., Van Wagoner, J.C. (Eds.), Sea Level Changes An Integrated Approach, vol. 42. SEPM Special Publication, pp. 39 45. Van Wagoner, J.C., Mitchum Jr., R.M., Campion, K.M., Rahmanian, V.D., 1990. Siliciclastic Sequence Stratigraphy in Well Logs, Core, and Outcrops: Concepts for High-Resolution Correlation of Time and Facies. American Association of Petroleum Geologists, Methods in Exploration Series 7, 55 pp. Van Wagoner, J.C., Hoyal, D.C.J.D., Adair, N.L., Sun, T., Beaubouef, R.T., Deffenbaugh, M., et al., 2003. Energy Dissipation and the Fundamental Shape of Siliciclastic Sedimentary Bodies. American Association of Petroleum Geologists, Search Discovery, Article #40080. Wadsworth, J., Boyd, R., Diessel, C., Leckie, D., Zaitlin, B., 2002. Stratigraphic style of coal and non-marine strata in a tectonically influenced intermediate accommodation setting: the Mannville Group of the Western Canadian Sedimentary Basin, south-central Alberta. Bull. Can. Pet. Geol. 50 (4), 507 541. Wadsworth, J., Boyd, R., Diessel, C., Leckie, D., 2003. Stratigraphic style of coal and non-marine strata in a high accommodation setting: Falher Member and Gates Formation (Lower Cretaceous), western Canada. Bull. Can. Pet. Geol. 51 (3), 275 303. Walther, J., 1894. Einleitung in Die Geologie Als Historische Wissenschaft: Beobachtungen Uber die Bildung der Gesteine und Ihrer Organischen Einschlusse. Jena G. Fischer, 1055 pp. Wanless, H.R., Shepard, E.P., 1936. Sea level and climatic changes related to Late Paleozoic cycles. Geol. Soc. Am. Bull. 47, 1177 1206. Wanless, H.R., Weller, J.M., 1932. Correlation and extent of Pennsylvanian cyclothems. Geol. Soc. Am. Bull. 43, 1003 1016. Wehr, F.L., 1993. Effects of variations in subsidence and sediment supply on parasequence stacking patterns. In: Weimer, P., Posamentier, H. W. (Eds.), Siliciclastic Sequence Stratigraphy Recent Developments and Applications, 58. American Association of Petroleum Geologists Memoir, pp. 369 378. Weimer, R.J., 1966. Time-stratigraphic analysis and petroleum accumulations, Patrick Draw field, Sweetwater County, Wyoming. Bull. Am. Assoc. Pet. Geol. 50 (10), 2150 2175. Weimer, P., 1990. Sequence stratigraphy, facies geometries, and depositional history of the Mississippi Fan, Gulf of Mexico. Bull. Am. Assoc. Pet. Geol. 74, 425 453. Weimer, P., Dixon, B.T., 1994. Regional sequence stratigraphic setting of the Mississippi Fan complex, northern deep Gulf of Mexico: implications for evolution of the northern Gulf basin margin. In: Weimer, P., et al. (Eds.), Submarine Fans and Production Characteristics, Society for Sedimentary Geology (SEPM), Gulf Coast Section, 15th Annual Research Conference, Houston, Texas, pp. 373 381. Weller, J.M., 1930. Cyclical sedimentation of the Pennsylvanian period and its significance. J. Geol. 38, 97 135. Wheeler, H.E., 1958. Time stratigraphy. Am. Assoc. Pet. Geol. Bull. 42, 1047 1063. Wheeler, H.E., 1964. Baselevel, lithosphere surface, and time-stratigraphy. Geol. Soc. Am. Bull. 75, 599 610. Wilgus, C.K., Hastings, B.S., Kendall, C.G.St.C., Posamentier, H.W., Ross, C.A., Van Wagoner, J.C. (Eds.), 1988. Sea Level Changes An Integrated Approach, vol. 42. SEPM Special Publication, 407 pp. Williams, D.F., 1988. Evidence for and against sea-level changes from the stable isotopic record of the Cenozoic. In: Wilgus, C.K., Hastings, B. S., Kendall, C.G.St.C., Posamentier, H.W., Ross, C.A., Van Wagoner, J.C. (Eds.), Sea Level Changes An Integrated Approach, 42. SEPM Special Publication, pp. 31 36. Willis, B., 1910. Principles of palaeogeography. Science 33, 248 251. Wright, V.P., Marriott, S.B., 1993. The sequence stratigraphy of fluvial depositional systems: the role of floodplain sediment storage. Sediment. Geol. 86, 203 210. Zachos, Z., Pagani, M., Sloan, L., Thomas, E., Billups, K., 2001. Trends, rhythms, and aberrations in global climate 65 Ma to present. Science 292, 686 693.

Regional Geology and Tectonics

686

23. Sequence stratigraphy

Zaitlin, B.A., Potocki, D., Warren, M.J., Rosenthal, L., Boyd, R., 2000. Sequence stratigraphy in low accommodation foreland basins: an example from the lower Cretaceous basal Quartz Formation of southern Alberta (Abstracts). In: GeoCanada 2000 Conference. Canadian Society of Petroleum Geologists, CD-ROM. Zaitlin, B.A., Warren, M.J., Potocki, D., Rosenthal, L., Boyd, R., 2002. Depositional styles in a low accommodation foreland setting: an example from the Basal Quartz (Lower Cretaceous), southern Alberta. Bull. Can. Pet. Geol. 50 (1), 31 72. Zecchin, M., Catuneanu, O., 2013. High-resolution sequence stratigraphy of clastic shelves I: units and bounding surfaces. Mar. Pet. Geol. 39, 1 25. Zecchin, M., Catuneanu, O., 2015. High-resolution sequence stratigraphy of clastic shelves III: applications to reservoir geology. Mar. Pet. Geol. 62, 161 175. Zecchin, M., Catuneanu, O., 2017. High-resolution sequence stratigraphy of clastic shelves VI: mixed siliciclastic-carbonate systems. Mar. Pet. Geol. 88, 712 723. Zecchin, M., Catuneanu, O., Rebesco, M., 2015. High-resolution sequence stratigraphy of clastic shelves IV: high-latitude settings. Mar. Pet. Geol. 68, 427 437. Zecchin, M., Catuneanu, O., Caffau, M., 2017a. High-resolution sequence stratigraphy of clastic shelves V: criteria to discriminate between stratigraphic sequences and sedimentological cycles. Mar. Pet. Geol. 85, 259 271. Zecchin, M., Caffau, M., Catuneanu, O., Lenaz, D., 2017b. Discrimination between wave-ravinement surfaces and bedset boundaries in Pliocene shallow-marine deposits, Crotone Basin, southern Italy: an integrated sedimentological, micropaleontological and mineralogical approach. Sedimentology 64, 1755 1791. Zuhlke, R., 2004. Integrated cyclostratigraphy of a model Mesozoic carbonate platform the Latemar (Middle Triassic, Italy). In: D’Argenio, B., Fischer, A.G., Premoli Silva, I., Weissert, H., Ferreri, V. (Eds.), Cyclostratigraphy: Approaches and Case Histories. SEPM, pp. 183 211., Special Publication No. 81.

Regional Geology and Tectonics

C H A P T E R

24 Concepts of conventional petroleum systems Jan de Jager De Jager Geological Consultancy, The Hague Area, Netherlands

Introduction The petroleum system concept was initially developed by Wallace Dow (1972, 1974) to assist the study of the occurrence of oil accumulations in the Williston Basin and was later formalized by Magoon and Dow (1994). This latter publication includes the history of the development of petroleum system concept. The petroleum system concepts links accumulations of hydrocarbons to a source rock, thus placing emphasis on the origin of the hydrocarbons. The elements of a petroleum system include a source rock, reservoir, trap and seal. Geological processes involved are trap formation and generation-migration-accumulation. The formalization of the petroleum system concept by Magoon and Dow (1994) has greatly assisted in making explorers aware of the fundamental logic of where and how hydrocarbons accumulations occur in the subsurface and thus in making hydrocarbon exploration more effective. Petroleum systems as defined by Magoon and Dow (1994) are rooted in the source rock and charge system and are formally named after the source rock and, separated by a hyphen, the reservoir formation in which hydrocarbons have accumulated. This practice has changed somewhat over time. Many investigators now only use the name of the source rock to define a petroleum system, for example the Kimmeridge petroleum system of the central and northern North Sea. This practice acknowledges the fact that a specific source rock can, and often has, charged reservoirs at different stratigraphic levels. Hydrocarbons generated from the Kimmeridgian source rocks in the North Sea have, for example, provided charge for a very large number of reservoir levels: from Devonian to Cenozoic. In other instances, hydrocarbon accumulations may have been charged from more than one source rock level. In that case, the most important source rock level should be used for the name of the petroleum system. The term hydrocarbon play has been used in the oil industry in a rather informal sense for a very long time. The term is used for groups of hydrocarbon accumulations and prospects that share geological characteristics. The history and common uses of the play have been summarized by Doust (2010). As prospects belonging to a certain play share geological characteristics, they commonly also share certain risks and they can be evaluated with a broadly similar approach and application of technologies. In practice, all prospects and accumulations of a hydrocarbon play occur at a single reservoir level, after which the play is named, for example the Brent play of the northern North Sea
which is one of the many plays of the Kimmeridge petroleum system. Defining a play at a single reservoir level facilitates making play maps that show the main geological elements impacting the prospectivity of a play: structuration, charge, reservoir and seal distribution. Play-based exploration (PBE) is now widely adopted by the petroleum industry as the best-practice workflow to evaluate petroleum systems. The essence of this workflow is illustrated in the PBE pyramid (Fig. 24.1). The workflow includes three tranches: basin evaluation, followed by play evaluation and finally prospect evaluation. The play-based workflow is not new and has, with different terminology, been the way exploration has been executed in many oil companies for many decades. However, in the nineties of the last century, with the advent of three-dimensional (3D) exploration and the accompanied increased use of workstations, explorers have tended to focus in on the prospect, ignoring its geological context. The PBE workflow aims to correct for this by forcing the explorer to base prospect exploration on a solid understanding of how the basin developed and an understanding of the characteristics and risks of the hydrocarbon play to which the prospect belongs. Regional Geology and Tectonics. DOI: https://doi.org/10.1016/B978-0-444-64134-2.00022-5

687

© 2020 Elsevier B.V. All rights reserved.

688

24. Concepts of conventional petroleum systems

Prospect evaluation

Play evaluation

FIGURE 24.1 The play-based exploration (PBE) pyramid. The PBE workflow intends to ensure that prospect evaluation is based on a sound understanding of the basin evolution, petroleum systems within the basin and of the characteristics and risks of the play to which the prospect belongs. PBE consists of three tranches: (1) Basin evaluation aims at arriving at an understanding of the basin evolution, its basin cycles and its petroleum systems; (2) play evaluation aims at identification and understanding of potential plays and leads that belong to those play; (3) prospect evaluation concerns the detailed evaluation of the geology of the prospect, assessment of risks and potential hydrocarbon volumes and an assessment of its commercial value.

Basin evaluation

FIGURE 24.2

Hydrocarbon accumulations can only occur in the subsurface if all essential ingredients are in place at a single location: (1) trap: for example an anticlinal structure of fault-bounded closed geometry, (2) reservoir: for example a sandstone or carbonate rock with sufficient porosity and permeability, (3) seal: a lithology above the reservoir with very low permeability, preventing hydrocarbons to migrate (leak) to the surface, and (4) Charge: a supply of hydrocarbons that can accumulate in the pores of the reservoir in the trap.

In this chapter, first the essential ingredients (charge, reservoir, seal and trap) will be discussed (Fig. 24.2). Then the PBE workflow is described together with its typical products, followed by how in the industry the risks and potential volumes of undrilled prospects are assessed.

Clarification of terminology The terms petroleum and hydrocarbons are here used as synonyms and refer to oil, gas and condensate. Formally also solid hydrocarbons (tar) are included in the definition. Generally, however, to avoid confusion, it is explicitly indicated when tar is meant or included. Hydrocarbon accumulation is used for any volume of trapped hydrocarbons, with sufficiently high saturations to be considered ‘free’ hydrocarbons and potentially producible. Hydrocarbon field is used for accumulations from which hydrocarbons are being, or have been, produced or that are considered to be commercial and have firm development plans. An accumulation is thus a more general term for trapped hydrocarbons and does not imply that the volume is sufficient for commercial exploitation. The hydrocarbon volume of a field is considered to be commercial. The term reserves is used for the volume of recoverable hydrocarbons. The total volume of trapped hydrocarbons is referred to as in-place volumes. The recovery factor is the fraction of the in-place volumes that are deemed to be commercially recoverable under current conditions and with current technologies. A prospect is an identified subsurface geometry in which hydrocarbons may have accumulated. A lead is a loosely defined term for a potential trap (prospect) that needs further evaluation to allow a meaningful assessment of its risks and potential volumes. Potentially recoverable hydrocarbon volumes in an undrilled prospect or lead are prospective resources or, simply, recoverable oil, gas or condensate volumes. The essential ingredients of a working petroleum system are: trap, reservoir, seal and charge. No hydrocarbon accumulation can be present if at the location of a prospect any of these is missing (Fig. 24.2).

Regional Geology and Tectonics

The essential ingredients

689

Conventional petroleum systems, the topic of this chapter, are petroleum systems where hydrocarbons occur in well-delineated structural or stratigraphic traps with a hydrocarbon
water contact below which the reservoir is water-bearing. In unconventional petroleum systems, hydrocarbons are not limited to conventional structural or stratigraphic traps. They are of regional or subregional extent and, contrary to conventional buoyancy-driven traps, they do not have well-defined hydrocarbon
water contacts. They generally occur in low-permeability reservoirs (,1 mD) that produce at low rates, generally require hydraulic stimulation to improve production and they require large numbers of wells to develop. The main types are basin-centred gas systems, coal bed methane systems and shale gas or shale oil systems.

The essential ingredients Many, but not all, sedimentary basins contain hydrocarbon accumulations. For a hydrocarbon accumulation to be present in the subsurface, the following ingredients need to be present simultaneously (Fig. 24.2): • • • •

A trap, A reservoir rock, A sealing rock and Hydrocarbon charge. If any of these is lacking, no conventionally trapped and producible hydrocarbons can be present (Fig. 24.3).

Trap Traps are 3D subsurface geometries in which hydrocarbons can be trapped if a reservoir is present that is overlain by a seal and has access to hydrocarbon charge. In the simplest classification, traps can be subdivided into structural and stratigraphic traps. Structural traps can be simple anticlinal structures dipping down in all directions or faulted structures. There is an endless variation of more complex structural trapping geometries. Fig. 24.4 shows some cartoons of some of the more simple trap types in cross section and map view together with names that are commonly used for them. Fig. 24.5 shows some stratigraphic trap types. Simple anticlinal traps (four-way dip closures) carry least risks. The presence of faults and fractures increases the risks that hydrocarbons will leak out of the trap. With increasing curvature (more prominent folding), extensional fractures may develop over the crest of the anticline, increasing the risk increases that the top seal is breached. It is no wonder that the world’s biggest oil field is a simple anticlinal trap with limited curvature: the Ghawar oil field of Saudi Arabia with initial oil reserves of reportedly c. 100 billion barrels oil equivalent (Sorkhabi, 2010). In fault-bounded traps, not only the top seal, but also the fault seal must be effective
either by favourable juxtaposition or the fault plane itself must be sealing. This is discussed in more detail in the section on Seals. For stratigraphic traps in addition to the top seal also, the seat seal (or bottom seal) must ‘work’ (Dolson et al., 1999). In a truncation trap, the number of sequences that must be sealing goes up to three: seat seal, top seal (stratigraphically directly overlying the reservoir) and the truncation seal (sequence deposited above FIGURE 24.3 Schematic representation of how the four essential ingredients (trap, reservoir, seal and charge) can result in a hydrocarbon accumulation. Note that here the reservoir directly overlies the source rock. That is only the case in some instances; in most cases, other lithologies separate the reservoir from the source rock, requiring expelled hydrocarbons to migrate through them to reach the reservoir.

Regional Geology and Tectonics

690

24. Concepts of conventional petroleum systems

Trap Names Dip closure 4-way closure Anticlinal trap

FIGURE 24.4 Cartoons of some simple structural trap types in cross section and map view with names as they are commonly referred to. Structural traps may be much more complex than depicted schematically here.

3-way closure Fault-controlled trap Tilted fault block

Trap-door trap

Horst block trap

Footwall trap

FIGURE 24.5 Cartoons of some simple stratigraphic traps in cross section.

Pinch-out trap

Truncation trap

Top seal Eagle Ford shale

Channel sand trap

Carbonate stratigraphic trap

FIGURE 24.6 The stratigraphic trap of the East Texas Oil field. For a trap like these three seals must all be effective at the same time: the top seal, seat seal and truncation seal (sketch after Allan et al., 2006).

Truncation seal Austin Chalk

Reservoir Woodbine Fm

Seat seal Washita Fm

the unconformity). Nevertheless, such traps have been found to work. An impressive example is the East Texas oil field with 6 billion barrels of oil reserves (Allen et al., 2006) (Fig. 24.6). Different basin types have specific trap types. Traps in rift basins are related to the tectonic phases that rift basins are typically subject to: extensional faulting during the syn-rift cycle and, often oblique, compressional structuration during the postrift phase. In addition, drape structures may develop above high fault blocks (Fig. 24.7). Trap formation in deltas, such as the Niger delta, is related to the formation of gravitationally induced thinskinned growth faults. Traps typically have multiple reservoir-seal pairs, often with a shift of the culminations at the different reservoir levels. Sometimes, there are several 10s of hydrocarbon-bearing sands in a structure, each with their own hydrocarbon
water contact and often with a variable mix of oil and gas (Fig. 24.8). In deltas, socalled direct hydrocarbon indicators (DHIs) are regularly observed on seismic data and provide indications that a hydrocarbon accumulation may be present. DHIs result from the fact that the presence gas and/or oil in a reservoir rock changes the acoustic impedance of the lithology and thus its acoustic impedance contrast with the overlying lithology, whereby the acoustic impedance is the product of the density and sonic velocity of the formation. A higher acoustic impedance contrast between two lithologies results in a stronger reflection. It must be remembered, however, that (residual) gas saturations of less than 10% result in a similar acoustic impedance and similar reflectivity. Gas-bearing siltstone with very low porosities and permeabilities (tight or ineffective

Regional Geology and Tectonics

The essential ingredients

Tilted faultblock

Drape structure

Simple horstblock

Inversion structure

Simple roll-over

Roll-over with antithetic fault

Downfaulted trap

Flower structure

Hanging-wall trap

Stratigraphic trap

Roll-over with multiple gross-faults

691 FIGURE 24.7 Typical trap types in rift basins may be caused by extensional faulting (upper row) or by inversion during the postrift cycle (inversion and flower structures). Drape structures over older high fault blocks and subcrop unconformities (for example below the end-of-rift unconformity) occur as well.

FIGURE 24.8 Typical trap styles in deltas result from the development of growth faults (after Weber et al., 1978). Often traps in deltas have many more hydrocarbon-bearing sands than indicated schematically here.

Roll-over with collapsed crest

reservoir) also may appear on seismic virtually the same as a sandstone with good reservoir quality and high-saturation producible hydrocarbons. The thin-skinned deformation of a delta is normally compensated for by thin-skinned compressional deformation in the outboard deepwater domain where a deepwater foldbelt develops with turbidite sandstones as reservoir (Corredor et al., 2005; Cullen, 2010). Simple elongated anticlinal structures with reverse thrust faults at the deeper levels are the most important traps in this setting and can normally very easily be interpreted from seismic profiles (Fig. 24.9). Also, in this setting, trapped hydrocarbons often result in DHIs. Above traps gas chimneys may be seen on seismic. These can be of such dimensions that a total wipe-out of the seismic signal results. Traps displaying such gas chimneys are obviously leaking gas, but generally still retain oil and gas as well. In some cases, all gas has leaked out and oil is retained. Subthrust trapping geometries are possible in deepwater foldbelts but more difficult to interpret from seismic due to the steep dips on the steep flank of these structures. Stratigraphic traps in this setting may be due to the pinch-out of turbidites onto highs developing syndepositionally, but are not common. Anticlines may be crossed by channelized turbidite systems. Reservoir in such settings may not be developed over the entire trap, but only within the channels. Such traps are often called stratigraphic, but as long as the hydrocarbon
water contact is controlled by a normal spill point, as is generally the case, it is more useful to refer to such traps as structural. Obviously, thin sandstones or siltstones must be present outside the channel systems for the water contact to coincide with the traps spill point. In the deepwater setting of the Gulf of Mexico, true stratigraphic traps due to turbidite pinch-out against steep salt domes are quite common (Godo, 2006; Fig. 24.10). In thrust- and foldbelts compressional structuration may have resulted in the formation of so-called duplex structures (Fig. 24.11). Due to the terrain conditions in these settings, seismic data are often scarce and of poor quality. The often complex structuration can then only be unravelled by carefully linking outcrop data with Regional Geology and Tectonics

692

24. Concepts of conventional petroleum systems

FIGURE 24.9 Typical traps in deepwater foldbelts. Auger

Macaroni

0

Depth (km)

2

4

6

8

FIGURE 24.10 Examples of traps in deepwater turbidites in the mini-basins of the Gulf of Mexico (after Reilly and Flemings, 2010). Decollement

Hidden (Footwall)

Duplex

FIGURE 24.11 Trap types typical for the compressional setting of fold-and thrustbelts.

information from seismic profiles. Palinspastic reconstructions are often required to make sure the interpreted structural geometries are realistic and geometrically possible. Some basins have salt sequences sufficiently thick for halokinesis to occur. The upward movement of salt leads to the development of salt swells and piercing salt domes. This results in a large variety of potential trapping geometries. Halokinesis affects different basins in different ways (Fig. 24.12). In intracratonic settings such as the Southern Permian Basin if NW Europe, stretching from the UK to Poland, salt doming results in turtleback anticlines in between rising salt structures and rim synclines close to piercing salt domes. Traps may be within the turtleback anticlines, against the salt, above salt domes and below unconformities resulting from halokinesis related uplift (Fig. 24.13). Ductile salt sequences deposited on the passive margins of Africa and Southern America act as a detachment level above which younger sequences are sliding gravitationally towards the oceanic domain. In this process, the overburden breaks up into so-called rafts while deepwater deposition is ongoing. This leads to pseudo turtleback anticlines and associated trap types (Fig. 24.14). Salt in the Gulf of Mexico behaves extraordinarily as it is being pushed forward and up under the weight of the sediments of the advancing Mississippi delta. Different canopies of salt developed, while above salt withdrawal areas in between rising salt walls so-called mini-basins developed with ponding of deepwater turbidites. Salt withdrawal may locally be (virtually) complete leaving only a so-called ‘salt-weld’. The range of resulting Regional Geology and Tectonics

The essential ingredients

693

(A)

(B)

(C)

(D)

FIGURE 24.12 Cartoons depicting typical structuration in different types of salt basins: (A) Intracratonic basin where salt swells and salt domes develop with associated turtleback anticlines and rim synclines, (B) passive margin with gravitationally induced raft tectonics, (C) the Gulf of Mexico where the outbuilding Mississippi delta resulted in salt canopies at different stratigraphic levels and formation of mini-basins and (D) thrust and foldbelt with salt acting as an important detachment level.

trapping geometries in this setting is almost endless. Sometimes, traps are similar to those shown in Fig. 24.12; other more typical geometries for the Gulf of Mexico are shown in Figs 24.10 and 24.15. Spill and leak points The hydrocarbon
water contact in most hydrocarbon accumulations is controlled by either a spill point or a leak point. The spill point of a trap defines the deepest contour to which a trap can be hydrocarbon bearing. Spill points occur shallower, usually where the top seal is breached at a fault or fracture (Fig. 24.16), sometimes where the top seal is breached by a channel cutting through the top seal. In fault-bounded traps, leak points are normally controlled by the precise juxtaposition of sealing and permeable lithologies across the fault plane. Hydrocarbons may leak out of a trap where there is sand-to-sand juxtaposition (unfavourable juxtaposition). A fault juxtaposition diagramme (Allen diagramme) can visually display the depth of the shallowest sand-to-sand juxtaposition (Fig. 24.17). Faults with sand-to-sand juxtaposition can also be sealing. There are a number of mechanisms that can make faults sealing: (1) clay smear, (2) shale gouge, (3) cataclasis, (4) diagenesis and (5) slivers of sealing lithologies (such as claystone) caught up in the fault zone. Empirical formulas have been developed that give an indication of the likelihood that a fault is sealing. This will be discussed further in the section dealing with Seals.

Reservoir Reservoirs are porous and permeable lithologies. The porosity creates space for hydrocarbons; the permeability allows hydrocarbons to flow to production wells and to be produced. Some confusion may arise between

Regional Geology and Tectonics

FIGURE

24.13

Trap

styles

related

to

halokinesis.

Cap rock

Rim syncline Turtleback anticline

FIGURE 24.14

Typical trap geometries associated with raft tectonics (after Jackson and Houdec, 2017).

HCs may leak along weld

Cemented HCs may leak across weld into mini-basin HCs may be trapped against weld

FIGURE 24.15 Some typical trap geometries of the Gulf of Mexico (after Jackson and Houdec, 2017).

695

The essential ingredients

Leak point

Spill point

Spill point

Leak point

FIGURE 24.16 The spill point of a trap defines the deepest contour to which a trap can be filled, normally at a saddle point in the contours at top reservoir. Potential leak points occur at shallower depths, generally where the seal is breached by a fault (as in the figure) or a fracture. Erosional channels or unconformities may also breach the top seal and result in leakage shallower than the spill point. Note that the schematic cross section in this figure does not run along a straight line; it runs north
south through the leak point and then to the east through the spill point. (A)

Cross section (D)

(B)

Block diagramme

Block diagramme

(E)

Allen diagramme Footwall

(C)

Allen diagramme

FIGURE 24.17 On a cross section, (A) it can be seen how in some of the sandstones in the upthrown (or footwall) fault-block hydrocarbons can be trapped, while others are in unfavourable sand-to-sand juxtaposition. (B) represents a block diagramme where throw (offset) of the fault is constant. (C) is a juxtaposition plot (Allen diagramme) of the same configuration. The lighter yellow colour shows the projection of the sands in the footwall onto the fault plane, and the darker orange colour the projection of the sands of the hanging wall onto the same fault plane. Where the sands of the two fault blocks overlap, there is sand-to-sand juxtaposition, indicated in light purple. In reality, the throw of faults is never constant as shown in (D), and then the juxtaposition diagramme will show a more complex pattern of sand-to-sand juxtaposition, as displayed in (E).

Hanging wall

Footwall

Hanging wall

explorationists and reservoir engineers. For the former, a reservoir is any porous and permeable lithology irrespective of its fluid content; the latter only speak of reservoirs if they contain hydrocarbons
that is if they occur in a hydrocarbon field. Water-bearing lithologies are then called aquifers. In this chapter, the exploration practice is followed. In most petroleum basins, sandstones are the main reservoirs. Nevertheless, as is often stated, more oil is trapped in carbonates than in clastics. This is because the richest petroleum basin by far is the Middle East where carbonates are the dominant reservoir. Other lithologies than sandstone or limestone can also act as reservoir, as long as they are porous and permeable. For example fractured or weathered basement (crystalline or metamorphic rocks; Koning, 2007) and volcanic lithologies such as tuff or basalt (Yang et al., 2017). The reservoir parameters important for the assessment of the hydrocarbon volumes of a (potential) hydrocarbon accumulation are indicated in Table 24.1. Many of the reservoir parameters can be assessed from wells using petrophysical logs. The geometry of the reservoir may be visible on seismic profiles and it may be evident from well log correlations of the reservoir sequence that a reservoir is not sheet-like. For undrilled exploration prospects, the reservoir parameters can be inferred from offset data in nearby wells, provided that there is a good understanding of the depositional setting of the reservoir sequence. Reservoir porosity is independent of grain size; a fine-grained sandstone may have the same porosity as a coarse-grained sandstone. Sorting, on the other hand, strongly affects the porosity. In poorly sorted reservoirs,

Regional Geology and Tectonics

696 TABLE 24.1

24. Concepts of conventional petroleum systems

Reservoir parameters.

Reservoir parameter

Remarks

Gross thickness

The total thickness of the reservoir sequence

Net-to-gross ration (N/G)

The fraction of the total gross thickness that represents porous and permeable reservoir. This excludes nonreservoir intervals such as intercalated shale layers

Geometry and continuity

The shape of the reservoir, for example sheet-like, wedging or ribbon-like and the lateral and vertical continuity of reservoir units

Porosity (ϕ)

The fraction of the total net reservoir volume that is void (nonsolid)

Permeability (k)

The property of a reservoir that allows pore fluids to flow through its pores

Hydrocarbon saturation (Shc)

The fraction of the total pore space that is filled with hydrocarbons 2 the remainder is filled with water. Water saturation 1 hydrocarbon saturation 5 100%

Recovery factor

The fraction of the in-place hydrocarbon volume that can be produced commercially

the smaller grains may fill the pores between the larger grains, thus reducing the porosity. Initial porosities of sediments that have just been deposited can be very high. Burial and mechanical compaction, under the pressure of the overburden, will reduce porosity. Diagenetic processes such as cementation will also result in porosity reduction, with precipitated cements partly filling the pore space. In general, the greater the quartz content, the greater the mechanical stability (less compaction); the higher the variety of minerals (‘dirtier’ sandstones with more clay particles), the more cementation and compaction. In general, better sorted sandstones are more quartzrich and maintain good porosities to greater depth (Hartmann et al., 2000). In carbonate rocks, porosities are much more difficult to predict than in clastics as diagenesis plays a much stronger role because of the solubility of limestone in water. Strong diagenetic processes in carbonates may occur already at very shallow depth. Carbonates that had initially good porosities may become completely tight because due to carbonate cements filling the pore space. Other diagenetic processes such as dolomitization and karst processes (dissolution of carbonates) may drastically increase porosities. Reservoir permeability is measured in units of Darcy and is generally expressed in units of Darcy or milliDarcy (mD). Permeabilities in reservoirs may range from 1 mD to several Darcies. Reservoirs with permeabilities of less than 1 mD are referred to as tight; permeabilities of more than 1 Darcy are excellent. The permeability of a reservoir is controlled primarily by the size of the pore throats connecting the pores. Because of the layered nature of most lithologies, reservoirs may display a marked difference in vertical and horizontal permeability. Vertical permeability may suffer from stratigraphic heterogeneity, such as the presence of intercalated shale layers. If these shale intervals are very extensive, this can be an important factor influencing the recovery of a field and the optimal development strategy. In most reservoirs, reservoir porosity and permeability are reasonably well correlated. This means that reservoirs with higher porosity also have higher permeability. But there are important exceptions. The smaller the grain size of a reservoir, the lower the permeability. With small grains, also the pores and pore throats are very small, making it more difficult for formation fluids (including oil and gas) to move through the reservoir. An example is chalk, an important reservoir in the Central North Sea, where porosities may be up in the high 20s while permeabilities are just several milliDarcies. Another factor influencing permeability is sorting. Better sorted sandstones have higher permeability than more poorly sorted sandstones, where the smallest grains may occupy space in the pores and pore throats of the larger grains. The net-to-gross ratio (N/G) of a reservoir can be defined in two different ways: 1. Lithology cut-off: Then simply all sandstones are included as reservoir and mudstones or shales are excluded. Sandstones can in most cases be identified on a gamma-ray log by their very low gamma-ray values whereas shales show up as intervals with high gamma-ray values. 2. Porosity cut-off: Then all intervals with porosities below a specified cut-off porosity value are excluded. This means that not alone all mudstones and shales are excluded, but in addition sandstones with very low porosity; so-called tight sandstones. The porosity cut-off will result in a lower N/G as some low-porosity sandstones will not be included as net reservoir. The reason to use a porosity cut-off to define the N/G if it is concluded, or assumed, that rocks with lower porosities than the cut-off will not contribute to the production of hydrocarbons. Some care has to be taken.

Regional Geology and Tectonics

697

The essential ingredients

FIGURE 24.18 Summary of reservoir characteristics of several clastic reservoir types.

Gamma-Ray log character

Fluvial channel sands–good to moderate: Well to poorly sorted; fine to coarse grained; straight-sinuous; connections variable