Geodynamics of Terrane Accretion Within Southern Alaska Benjamin Patrick Hooks

The University of Maine DigitalCommons@UMaine

Electronic Theses and Dissertations Fogler Library

2009 Geodynamics of Terrane Accretion within Southern Alaska Benjamin Patrick Hooks

Follow this and additional works at: http://digitalcommons.library.umaine.edu/etd Part of the Tectonics and Structure Commons

Recommended Citation Hooks, Benjamin Patrick, "Geodynamics of Terrane Accretion within Southern Alaska" (2009). Electronic Theses and Dissertations. 98. http://digitalcommons.library.umaine.edu/etd/98

This Open-Access Dissertation is brought to you for free and open access by DigitalCommons@UMaine. It has been accepted for inclusion in Electronic Theses and Dissertations by an authorized administrator of DigitalCommons@UMaine. GEODYNAMICS OF TERRANE ACCRETION WITHIN SOUTHERN ALASKA

Benjamin Patrick Hooks

B.S. Allegheny College, 2001

M.S. University of Maine, 2003

A THESIS

Submitted in Partial Fulfillment of the

Requirements for the Degree of

Doctor of Philosophy

(in Earth Sciences)

The Graduate School

The University of Maine

August, 2009

Advisory Committee:

Peter O. Koons, Professor of Earth Sciences, Advisor

Phaedra Upton, Faculty Associate in Earth Sciences

Scott E. Johnson, Professor of Earth Sciences

Daniel R. Lux, Professor of Earth Sciences

Terry L. Pavlis, Professor of Geological Sciences, University of Texas at El Paso Library Rights Statement

In presenting this thesis in partial fulfillment of the requirements for an advanced degree at The University of Maine, I agree that the Library shall make it freely available for inspection. I further agree that permission for "fair use" copying of this thesis for scholarly purposes may be granted by the Librarian. It is understood that any copying or publication of this thesis for financial gain shall not be allowed without my written permission.

Signature:

Date:

GEODYNAMICS OF TERRANE ACCRETION

WITHIN SOUTHERN ALASKA

By Benjamin P. Hooks

Thesis Advisor: Dr. Peter O. Koons

An Abstract of the Dissertation Presented

in Partial Fulfillment of the Requirements for the

Degree of Doctor of Philosophy

(in Earth Sciences)

August, 2009

The subduction and accretion of an exotic terrane at the southern margin of

Alaska is driving uplift of the St. Elias and Alaska Ranges, and is responsible for some of the largest strain releases in history. Here are presented results from numerical models conditioned by geological observations that reproduce the tectonic landscape, deformation, and strain patterns at macro- (1000-km) and meso- (<100 km) scales. These models utilize completely coupled thermal and mechanical solutions that account for the development of heterogeneities to both the thermal and rheological structure of the lithosphere. Perturbation to the thermal structure related to flattening of the buoyant down-going slab offsets the hot mantle wedge flow, cooling the fore-arc region of the orogen developing a thin sliver of material that behaves frictionally. This frictional sliver provides a primary control on the transfer of strain to the over-riding crust and influences the observed deformation patterns. Strengthening of the fore-arc causes a large-scale discontinuous jump in the deformation front. Initial deformation consists of the development of the Alaska Range orogenic wedge and dextral Denali Fault system. The deformation pattern reorganizes most of the strain captured by the St. Elias orogenic wedge forming above the down-dip limit of the frictional sliver. These model results are consistent with the observed slip on the Denali Fault indicating the partitioning of northwestward translation of the accreting terrane into the fold-thrust belt of the Alaska

Range, relatively fast uplift within the St. Elias Range, and the temporal shift in deformation patterns observed within the thermochronological and stratigraphic records.

The mesoscale model strain patterns, including the effects of evolving topography and erosion, are consistent with the geological observations; the St. Elias Range thin-skinned fold-thrust belt develops with uplift reaching a maximum within the kinematic tectonic corner. The basic strain pattern is controlled by the tectonic geometry, with the surface conditions providing a secondary influence on the rates and magnitudes of deformation.

The results of this study indicate that the non-linear feedback between rheology, temperature, and geometry provide a primary control on strain patterns during orogenesis. DISSERTATION

ACCEPTANCE STATEMENT

On behalf of the Graduate Committee for Benjamin P. Hooks, I affirm that this manuscript is the final and accepted dissertation. Signatures of all committee members are on file with the Graduate School at the University of Maine, 42 Stodder Hall, Orono,

Maine.

______

Dr. Peter O. Koons, Professor of Earth Sciences

iii

Acknowledgments

This project has benefitted from the support of the St. Elias Erosion and Tectonics Project and a research grant from the National Science Foundation Continental Dynamics

Program (EAR #0409162; principle investigators - P.O. Koons and P. Upton).

Table of Contents

Acknowledgments...... iv

List of Figures ...... ix

List of Tables ...... xii

Chapter 1 : Introduction ...... 1

1.1 Purpose ...... 1

1.2 ST. Elias Erosion/Tectonics Project ...... 2

1.3 Geologic Background ...... 3

1.4 Theoretical Background ...... 11

1.5 General Geological Implications ...... 14

1.6 Introduction to the Project ...... 16

Chapter 2 : Numerical Methods ...... 17

2.1 Introduction ...... 17

2.2 Thermal Solutions ...... 23

2.3 Mechanical Solutions ...... 26

2.4 Model Megathrust ...... 29

2.5 Sensitivity Analysis and Limitations...... 30

Chapter 3 : The Influence of Thermal Perturbations on the Characteristic Signal of

Terrane Accretion ...... 34

3.1 Introduction ...... 34

3.2 Numerical Methods ...... 37

3.3 Results and Discussion ...... 38

3.3.1 Thermal Evolution ...... 38

3.3.2 Thermo-Mechanical Evolution ...... 43

3.4 Petrology and Strain Transfer ...... 46

3.5 Petrological Implications...... 54

3.6 Conclusions ...... 59

Chapter 4 : Macroscale Geodynamics of Southern Alaska ...... 60

4.1 Introduction ...... 60

4.2 Geodynamics of Southern Alaska ...... 60

4.3 Numerical Modeling Methods ...... 65

4.4 Results ...... 67

4.4.1 Reference Model Results ...... 67

4.4.2 Tectonic Model Results ...... 72

4.4.3 Strain-Softening Model Results...... 78

4.4.4 Erosional Tectonic Model Results ...... 84

4.4.5 Erosional Strain-Softening Model Results ...... 87

4.5 Discussion ...... 91

4.5.1 Discussion of Model Results ...... 91

4.5.2 Comparison with Geological Observations ...... 94

4.5.3 Influence of Thermal Advection, Rheological Weakening, and Erosion ...... 103

4.5.4 Petrological Implications ...... 104

4.5.5 Tectonics Implications ...... 106

4.6 Conclusions ...... 109

Chapter 5 : Mesoscale Geodynamics of Southern Alaska ...... 110

5.1 Introduction ...... 110

5.2 Numerical Modeling Methods ...... 111

5.3 Mesoscale Modeling Results ...... 112

5.3.1 Mechanical Results ...... 112

5.3.2 Petrologic Implications ...... 121

5.4 Discussion of Model Results ...... 125

5.4.1 Role of Erosion ...... 125

5.4.2 Comparison with Thermochronology ...... 127

5.4.3 Comparison with Structural Geology ...... 130

5.5 Tectonic Aneurysm ...... 134

5.6 Conclusions ...... 136

vii

Chapter 6 : Geodynamics of Terrane Accretion ...... 137

Chapter 7 : Conclusions ...... 147

7.1 Influence of the Thermal Structure ...... 147

7.2 Role of Heterogeneity at the Macroscale ...... 149

7.3 Non-linear Feedbacks at the Mesoscale ...... 151

7.4 Suggestions for Future Work ...... 152

References Cited ...... 155

Appendix A: Numerical modeling Code ...... 175

Appendix B: Numerical Model Sensitivity Analysis...... 176

Appendix C: Animations of Model Results ...... 183

Appendix D: Explanation of Variables ...... 186

Biography of the Author ...... 188

viii

List of Figures

Figure 1.1. Map of Geodynamics of southern Alaska...... 4

Figure 1.2. Late Cretaceous geologic map of southern Alaska...... 6

Figure 1.3. Mid-Eocene geologic map of southern Alaska...... 8

Figure 1.4. Present geologic map of southern Alaska...... 10

Figure 1.5. Schematic cross section of a subduction zone...... 12

Figure 2.1. Spatial location of numerical models...... 18

Figure 2.2. Macroscale model geometry...... 19

Figure 2.3. Mesoscale model geometry...... 20

Figure 2.4. 2D Macroscale model geometry...... 22

Figure 3.1. 2D thermal results...... 40

Figure 3.2. 3D thermal results...... 41

Figure 3.3. Model Rheology ...... 42

Figure 3.4. Reference model mechanical results...... 44

Figure 3.5. Fully-coupled model mechanical results...... 45

Figure 3.6. P-T Facies Diagram...... 47

Figure 3.7. Seismic hazard of south-central Alaska...... 51

Figure 3.8. Megathrust strain patterns...... 53

Figure 3.9. Derived P-T paths...... 56

Figure 3.10. Spatial distribution of metamorphic facies...... 58

Figure 4.1. Geodynamics of southern Alaska...... 62

Figure 4.2. Reference model strain patterns...... 68

Figure 4.3. Uplift and thermal results for the Reference model...... 70

Figure 4.4. Temporal uplift patterns in the Reference model...... 71

Figure 4.5. Spatial development of the frictional sliver...... 73

Figure 4.6. Strain patterns associated with the megathrust...... 74

Figure 4.7. Tectonic model strain patterns...... 75

Figure 4.8. Uplift and temperature results for the Tectonic model...... 77

Figure 4.9. Temporal uplift patterns in the Tectonic model...... 79

Figure 4.10. Strain-softening model strain patterns...... 80

Figure 4.11. Uplift and temperature results for the Strain-softening model...... 82

Figure 4.12. Temporal uplift patterns in the Strain-softening model...... 83

Figure 4.13. Erosional Tectonic model strain and uplift patterns...... 85

Figure 4.14. Effect of erosion on crustal strength...... 88

Figure 4.15. Erosional Strain-softening model strain and uplift results...... 89

Figure 4.16. Geological uplift patterns...... 95

Figure 4.17. Structural comparisons...... 96

Figure 4.18. Seismic hazard map of southern Alaska...... 99

Figure 4.19. Model P-T paths...... 105

Figure 4.20. Tectonics of terrane accretion...... 107

Figure 5.1. Mesoscale surface boundary conditions...... 113

Figure 5.2. Tectonic model strain patterns...... 115

Figure 5.3. Tectonic model uplift and thermal patterns...... 116

Figure 5.4. Topographic model strain patterns...... 118

Figure 5.5. Topographic model uplift and thermal patterns...... 119

Figure 5.6. Comparison of vertical stresses...... 120

Figure 5.7. Erosion model strain patterns...... 122

Figure 5.8. Erosion model uplift and thermal patterns...... 123

Figure 5.9. Model P-T paths...... 124

Figure 5.10. Maximum modeled metamorphic conditions...... 126

Figure 5.11. Plot of thermochronological data...... 129

Figure 5.12. Comparison of interpreted and modeled faults...... 135

Figure 6.1. Initial stages of convergence...... 139

Figure 6.2. Initial flat slab subduction...... 141

Figure 6.3. Reorganization of deformation patterns...... 142

Figure 6.4. Resumption of subduction...... 146

Figure B.1. Sensitivity analysis for pressure-dependent rheology...... 176

Figure B.2. Sensitivity analysis for C and φ...... 177

Figure B.3. Sensitivity analysis for thermally-dependent rheology...... 178

Figure B.4. Sensitivity analysis for thermally-dependent rheological variations...... 179

Figure B.5. Sensitivity analysis for initial kinematics...... 180

Figure B.6. Sensitivity analysis for discretization and absence of interface...... 181

Figure B.7. Sensitivity analysis for interface properties...... 182

List of Tables

Table 2.1. Material Properties...... 25

xii

Chapter 1 : Introduction

1.1 Purpose

The overarching goal of this project is to develop a quantitative model that reproduces observed geomorphic and geologic patterns of Southern Alaska within the context of the basic physics of terrane accretion and allows for the study of how and why those features form. To achieve this goal, the numerical models developed for this study must incorporate geological relationships, tectonics, and surface processes in order to describe the regional deformation patterns within the orogen. The models presented here explore the characteristic signal of accretionary tectonics by utilizing dynamic coupling between the thermal evolution and rheological structure of the lithosphere, and its effect on overall deformation patterns.

The combination of realistic thermo-mechanical numerical models and appropriate geological data and expertise can provide valuable insights into the evolution of the Earth at a range of time and spatial scales (e.g. Buck and Poliakov, 1998;

Beaumont et al., 2001; Koons et al., 2002; Sobolev et al., 2005). The results of this research project have broad implications for understanding the geodynamics of oblique collision, the growth of the crustal lithosphere through terrane accretion, and the role of nonlinear feedbacks on the evolution of collisional orogens (e.g. Behn et al., 2002; Nagel and Buck, 2004; Naylor et al., 2005; Gerbault et al., 2005). As such, these findings can provide valuable insight into the evolution of deformation pattern observed in other terrane accretion events, such as the Laramide and Acadian orogens.

1.2 ST. Elias Erosion/Tectonics Project

This project is a portion of a larger multi-disciplinary study addressing the evolution of southern Alaska and northwestern Canada; the St. Elias Erosion/Tectonics

Project, or STEEP. The St.Elias orogen was formed as the Yakutat block, a continental- oceanic terrane, obliquely collided and accreted to North America at the eastern syntaxis of the Aleutian subduction zone. The resulting mountain range is the product of the balance between rapid uplift driven by crustal accretion and rapid exhumation by a regional glaciation. The development of a comprehensive model for the St. Elias orogen accounting for the interaction of tectonic and climatic processes is the overarching goal of this project. As such, three fundamental questions will be addressed:

• What is the relative role of tectonics and surface processes in influencing

regional deformation patterns?

• How is deformation partitioned within the orogen? How does erosion

influence the partitioning of deformation?

• What is the role of flat-slab subduction with respect to the large-scale

deformation patterns? What is the role of thermal advection?

This study has broad implications for understanding the geodynamics of oblique collision, including the role of the coupling between erosion and deformation.

Furthermore, the results of this work can be extrapolated to both modern and ancient oblique collisions and provide important information on the growth of continents through assembly of allochthonous terranes.

1.3 Geologic Background

The active tectonic margin of southern Alaska, located at the northern limit of

North American Cordillera, was assembled through the accretion of several

allochthonous terranes since the late Mesozoic (Figure 1.1; Bruns, 1985; Plafker and

Berg, 1994; Plafker et al., 1994; Bruhn et al., 2004; Pavlis et al., 2004). Over the last ~10

Ma, deformation and uplift within the St. Elias and Alaska Ranges was driven by the

ongoing collision of the thickened oceanic plateau and associated sedimentary cover of the Yakutat terrane with North America at the eastern syntaxis of the Aleutian subduction

zone (Figure 1.1; Bruns, 1985; Plafker et al., 1994). Today the heavily glaciated St. Elias

Range has the highest coastal relief on Earth (e.g. Mt. Logan: 5959 m and Mt. St. Elias:

5489 m) and the glacially derived sedimentary flux into the Gulf of Alaska has been

interpreted as the driving force for structural reorganization of the orogen (Berger et al.,

2008a; Berger et al., 2008b).

The southern margin of Alaska can be divided into four composite terranes based

upon their geologic histories and compositions. The Yukon terrane, located to the north

and northeast of the Denali Fault, consists of material that had accreted to North America

during the middle Mesozoic and were affected by Jurassic contraction and extension in

the Early Cretaceous (Pavlis et al., 1993). For the purpose of this study the Yukon

terrane is considered part of cratonic North America (Figure 1.1; Plafker et al., 1994).

South of the Denali fault suture zone is the Wrangell Composite Terrane (WCT), a late

Paleozoic volcanic arc, early Mesozoic mafic magmatic assemblage, and associated

sedimentary rock sequences (Nokleberg et al., 1994). During the late Jurassic to Early

Figure 1.1. Map of Geodynamics of southern Alaska.

Figure 1.1. Map of the Geodynamics of southern Alaska. This is a composite map showing the approximate terrane boundaries, major fault traces, and locations of important geological features. Additionally, the yellow shaded region shows the approximate boundaries of the subducted portion of the Yakutat terrane (Eberhart-Philips et al., 2006). The vectors for the Pacific Plate and Yakutat terrane, black arrows, are taken from published sources (DeMets et al., 1994; Fletcher and Freymueller, 1999; DeMets and Dixon, 1999). Additionally, the composite terranes of Alaska (after Plafker and Berg, 1994) are show. See text for details.

Cretaceous sedimentary basins developed along the leading edge of the WCT, into which

the sediment shed adjacent arc terranes, forming the Kahiltna/Gravina flysch (Nokleberg

et al., 1994). South of the WCT is the Southern Margin Composite Terrane (SMCT),

inclusive of Prince William and Chugach Terranes. The SMCT formed as an

accretionary prism trenchward of the WCT and is a variably deformed oceanic

assemblage derived from off-scraping of the down-going Kula plate. Finally, the Yakutat

terrane is actively accreting to southern Alaska at the eastern syntaxis of the Aleutian

subduction zone, further details are discussed below.

During the Late Cretaceous to early Tertiary, the geodynamics of southern Alaska

were characterized by the longstanding northward subduction of the Kula plate beneath

North America. The subduction-associated coastal plutonic complex and volcanic arc

developed parallel to the trench along the ancestral Denali Fault zone (Figure 1.2; Plafker et al., 1994). Deformation patterns were related to convergent tectonics to the northwest, including early broad uplift and exhumation of metamorphic bedrock within the Alaska

Range (Cole et al., 2000), while to the southeast obliquity in the convergence angle drives transform motion on the ancestral Denali fault system (Nokleberg et al., 1986;

Nokleberg et al., 1994). The current Denali Fault is a major tectonic boundary and has been interpreted as the reactivated suture zone between cratonic North America and terranes accreted during the Mesozoic (Cole et al., 2002; Ridgway et al., 2007).

By the Middle Eocene, changes to the stress state between North America and the

Pacific plate potentially due to the subduction of the Kula spreading ridge caused a 30 -

Figure 1.2. Late Cretaceous geologic map of southern Alaska.

Figure 1.2. Late Cretaceous geologic map of southern Alaska. Tectonic history of southern Alaska during Late Cretaceous to Early Tertiary (After Plafker and Berg, 1994). This period of time within southern Alaska was characterized by northward subduction of the Kula plate. See text for additional discussion.

50º rotation of western North America (Figure 1.3; Byrne, 1979; Engebretson et al.,

1985; Plafker et al., 1989; Csejtey, 1990; Plafker et al., 1994), anomalous heat flow

within the forearc leading to high-temperature/low-pressure metamorphism, and near trench anatectic plutonism (Sisson and Hollister, 1988; Dusel-Bacon et al., 1993; Sisson and Pavlis, 1993; Pavlis and Sisson, 1995; Bradley et al., 2003). At ~30 Ma, the structural reorganization of the subduction zone culminated in a eastward jump in the oblique transform boundary between North America and the Pacific plate, forming the proto-Yakutat terrane from the Cretaceous accretionary prism to the east of the

Dangerous River Zone (DRZ) and near-trench Paleogene mafic intrusive basement rocks to the west of the DRZ (Bruns, 1983; Plafker et al., 1994; Chapman et al., 2008).

Following the excising of the Yakutat terrane from North America, it was translated northward along the Queen Charlotte-Fairweather transform fault system and ultimately accreted to southern Alaska. The basement of the Yakutat terrane was covered with a sequence of Paleogene marine to fluvial sediments shed from the western margin of southern Alaska and British Columbia. The lower sedimentary units, the Kulthieth and

Poul Creek Formations, are overlain by the synorogenic Miocene to Pleistocene fluvial, glacial, and glacial-marine Yakutaga Formation formed in response to the uplift of the St.

Elias Range (Eyles et al., 1991; Plafker et al., 1994; Jaeger et al., 1998; Gulick et al.,

2007). The Yakutaga rocks are currently uplifting and deforming within the offshore fold and thrust belt of the St. Elias orogen.

Figure 1.3. Mid-Eocene geologic map of southern Alaska.

Figure 1.3. Mid Eocene geologic map of southern Alaska. Tectonic history of southern Alaska during the Paleocene to Mid Eocene (After Plafker and Berg, 1994). This period of time was characterized by the counterclockwise rotation and structural reorganization of southern Alaska. See text for additional discussion.

The basement of the Yakutat terrane has been accreting to and subducting at a

relatively shallow angle beneath the North American plate at the eastern syntaxis of the

Aleutian subduction zone over the last ~10 Ma (Figure 1.4; Bruns, 1985; Fuis and

Plafker, 1991; Plafker et al., 1994; Pavlis et al., 2004; Fuis et al., 2008). This subduction and collision event is driving fast uplift within southern Alaska (Sheaf et al., 2003;

Berger et al., 2008a; Berger et al., 2008b; Enkelmann et al., 2009), creating the largest coastal mountains on Earth, the St. Elias Range, and further uplift within the Alaska

Range. The St. Elias orogen has formed at the transition from transform motion associated with the dextral Fairweather fault to fold and thrust compression associated

with subduction and terrane accretion (Chapman et al., 2008; ). Due to the high

topography and proximity to a large source of moisture associated with the maritime

climate, the St.Elias Range experiences high precipitation, mostly snow that is later

incorporated into the abundant glaciers (e.g. Péwé, 1975). The abundant, warm based

glaciers provide a large erosive flux of sediment into the Gulf of Alaska. This flux of

mass from the orogen into the offshore basins has been suggested to drive the tectonic

signature of the orogen (e.g. Hallet et al., 1996; Meigs and Sauber, 2000; Berger et al.,

2008a; Berger et al., 2008b).

Figure 1.4. Present geologic map of southern Alaska.

Figure 1.4. Present geologic map of southern Alaska. Current tectonics of southern Alaska (After Plafker and Berg, 1994; Pavlis et al., 2004; Bruhn and Haeussler, 2004). This period of time was characterized by the counterclockwise rotation and structural reorganization of southern Alaska. Note the outline of the present day Alaskan coastline for spatial reference. See text for additional discussion.

1.4 Theoretical Background

The theoretical basis of this study is based upon the relatively simple physical principles and assumption that plate motion derived from mantle kinematics drives the deformation within the overriding crust (Figure 1.5). The kinematics and deformation of a subduction zone are controlled by the gravitational body force and shear coupling between the thermo-gravitationally induced mantle flow and the overriding and subducting plates (Conrad and Lithgow-Bertelloni, 2002; Funiciello et al., 2003; Conrad et al., 2004; Billen, 2008). The gravitational body force is a function of the integrated density of the lithosphere, and is strongly dependent upon the thickness and temperature of the crustal component. Thick or relatively hot crust is associated with higher buoyancy, resulting in flat-slab subduction. Shear coupling is dependent upon the viscous or frictional properties of the interfaces between the interacting plates, which is a function of temperature and material properties, and acts to transfer strain between the down-going mantle and over-riding plates.

Flow within the mantle exerts a viscous shear upon the base of the oceanic lithosphere that drives both the subducting and overriding plate toward the subduction zone from outside of the deforming zone (Conrad and Lithgow-Bertelloni, 2002). The viscous shear at the base of the oceanic lithosphere should be proportional to the strain transferred across the megathrust to the overriding plate. This transfer of viscous strain derived from the mantle flow through the down-going slab and across a frictional shear coupled interface, seismogenic megathrust, deforms the overriding plate.

Figure 1.5. Schematic cross section of a subduction zone.

Figure 1.5. Schematic cross section of a subduction zone. This 2D cartoon schematic drawing shows the subduction of oceanic lithosphere (green) below continental lithosphere (brown). The seismogenic portion of the interface between the two, the megathrust, is shown in red. See text for additional discussion.

The megathrust seismic zone is the expression of coupling between the

subducting and overriding plates. The extent and relative strength of this seismically

coupled zone are functions of the thermal structure, composition, metamorphic fluid

production, and subducted sediment abundance (Peacock and Wang, 1999; Rossi et al.,

2006; Reyners and Eberhart-Phillips, 2009). The seismogenic zone is controlled at the up-dip limit by the high temperature stability range of clays and at the down-dip limit by metasomatic production of talc within the slab-mantle interface (Peacock and Hyndman,

1999). A well-coupled interface transfers more strain to the overriding plate resulting in a narrow, high relief zone of deformation. Conversely, weak coupling favors wide, low relief zones of deformation. The relative strength of the coupling, and therefore the transfer of strain, can be directly related to the slab dip, where the strength is a function of temperature.

Transfer of strain across the plate interface causes deformation within the crust and is a function of the basal strength of the interface. The deformation within the crust has been described using the critical wedge theory relying on a balance between horizontal normal forces and the basal shear of the deforming material (Davis et al.,

1983; Dahlen, 1984; Koons, 1994; Koons, 1995);

, 1.1

where σxx is the horizontal normal stress applied over the depth of the wedge (z) and τxz

is the basal shear stress along the length of the wedge (x; Figure 1.6). Changes to either

the strength or width of the frictional interface between the down-going and over-riding plates results in a change in the integrated basal shear stress, and if the wedge is to remain

critical, also the normal stress. This study departs from the critical wedge theory

treatment in it is not bound to the requirement that the wedge be always and everywhere at failure, does not assume spatial and temporal homogeneity, and uses a three

dimensional geometry.

1.5 General Geological Implications

Flat subduction and/or accretion of exotic terranes have important implications for

the overall deformation patterns of the overriding plate where the relationship between

the dynamic evolution of material properties and tectonics becomes evident (Uyeda and

Kanamori, 1979; Jarrard, 1986; Isacks, 1988; Gutscher et al., 2000a). The offset of the mantle wedge flow and vertical advection of cold oceanic lithosphere conspire to cool and strengthen the forearc region (Munoz, 2005; Abers et al., 2006) driving a spatial and temporal structural reorganization of the resultant orogen (Chapters 3 and 4; Koons et al.,

2009). Tectonic refrigeration of the megathrust zone causes broader areas of material to behave in brittle fashion, allowing for larger accumulation of strain resulting in some of the largest recorded earthquakes (e.g. Plafker, 1965; Estabrook et al, 1992; Carver and

Plafker, 2008; Plafker and Thatcher, 2008). Within southern Alaska, the thickened crust of an oceanic plateau increases the buoyancy of the Yakutat terrane causing the angle of subduction to shallow (e.g Ferris et al., 2003; Fuis et al., 2008; Eberhart-Phillips et al.,

2006). This shallowing offsets the relatively hot mantle wedge flow, develops a wider seismogenic zone, and creates a relatively cool lower crust related to downward vertical advection of relatively cold oceanic lithosphere (Munoz, 2005; Eberhart-Philips et al.,

2006; Abers et al., 2006; Rossi et al., 2006). The perturbations to both the rheological

and thermal structure are characteristic of flat slab subduction culminating in terrane

accretion and tectonic reorganization (e.g. Royden, 1996; Liu and Yang, 2003; Hall et al.,

1998; Murphy et al., 1999; Koons et al., 2009).

One of the more important nonlinear feedback mechanisms in geosciences is the

interaction between climate and tectonics (Willett, 1992; Zeitler et al., 2001; Koons et al.,

2002; Naylor et al., 2005; Enkelmann et al., 2009; Hooks et al., in review). Orographic

lift causes relatively high precipitation rates on the windward side of large mountain

belts, driving regional asymmetric erosion patterns and locally focused fluvial and glacial

erosion (Meigs et al., 2008). Focused erosion influences the evolution of local

deformation patterns and development of focused slip partitioning during orogenesis.

Thinning of the upper crust due to focused erosion perturbs the material flow paths

through the lithosphere and drives a thermal weakening feedback that ultimately results in localization of fast exhumation and high strain rates (e.g. Beaumont et al., 1996; Batt and Braun, 1999). This was recognized in the eastern and western syntaxes of the

Himalaya (Zeitler et al., 2001; Koons et al, 2002; Beaumont et al., 2004) where the coupled erosion-uplift feedback within the plate corners was termed 'tectonic aneurysm' behavior. The aneurysm behavior has some important petrologic implications; vertical advection causes a zone of high heat flow that drives local high temperature, low pressure metamorphism and crustal anatexis. Southern Alaska records the early stages of the development of this behavior, as evidenced by relatively fast uplift rates observed within the Seward Glacier region (Chapter 5; Barker, 2007; Enkelmann et al., 2009). A model of the evolution of plate corners can be developed through a combination of the modeling

of the early stages of aneurysm behavior from this work with the well-developed, mature aneurysm of the syntaxes of the Himalaya.

1.6 Introduction to the Project

The initial step of this project investigates the influence of the thermal structure of the lithosphere on the overall deformation patterns in the overriding plate and is framed within respect to the interpreted flat slab subduction zone of southern Alaska (Chapter 3).

Vertical thermal advection alters the strength of the crust and facilitates a spatial and temporal reorganization of the large-scale deformation patterns by creating evolving zones of strong, frictional material associated with the down-dip extent of the seismogenic megathrust. This chapter also explores the petrologic and neotectonic implications of flat slab subduction and terrane accretion. That initial thermal structure and resulting perturbations to the rheological structure are then utilized to develop a fully coupled thermo-mechanical model for the large scale (Chapter 4) and regional-scale

(Chapter 5) geomorphic and structural evolution of the St. Elias orogen. The development the thermal and mechanical heterogeneities provide a first order control of the evolution of the model orogen. These models reproduce the strain patterns, thermochronological record, and geological landscape remarkably well. Finally, the implications of the often overlooked thermal and mechanical perturbations are used to develop a qualitative model of the evolution of characteristic signal of collisional orogens

(Chapter 6).

Chapter 2 : Numerical Methods

This study presents a series of fully coupled thermo-mechanical numerical models

that utilize simple boundary conditions to solve for simple stress–strain relationships, and

heat transfer (input files are included as Appendix A and In Pocket). The thermal

structure is used to define the rheological behavior of the model dynamically during the

model run period leading to more complete and realistic material behavior. These models

are developed to investigate the role of thermal and mechanical heterogeneities in the

development of the characteristic signals of deformation within an accretionary orogen,

such as southern Alaska. The theoretical aspects presented in Chapter 1.4 are further

expanded here within the framework of the numerical models.

2.1 Introduction

This study uses a 3D model geometry based upon observed natural geometry of

southern Alaska (Figure 2.1; Pavlis et al., 2004; Eberhart-Philips et al., 2006; Fuis et al.,

2008) on a large- (macroscale; 1000 km) and embedded small-scale (mesoscale; 100- km). The macroscale model encompasses a large portion of southern Alaska with a volume of 1600 km (X-axis) by 1200 km (Y-axis) by 50 km (Z-axis) and discretization of

40 km horizontal and 5 km vertical (Figure 2.2). This geometry consists of two separate blocks, a wedge shaped "Pacific" and "North America" block that are separated by a frictional interface (discussed further below). The embedded mesoscale model

Figure 2.1. Spatial location of numerical models.

Figure 2.1. Spatial location of numerical models. The white box outlines the location of the macroscale geometry and the red box outlines the location of the mesoscale model. The dashed white line shows the approximate location of the 2D geometry. Yellow shaded area depicts the location of the exposed Yakutat terrane and the gray shaded area depicts the extent of the subducted Yakutat terrane over the last ~30 Ma (after Eberhart- Phillips et al., 2006). Additional details are described in Figure 1.1 and Chapter 1.3.

Figure 2.2. Macroscale model geometry.

Figure 2.2. Macroscale model geometry. The North American block is shaded in blue and Pacific block is shaded in red. The yellow dashed line projected onto the surface indicates the approximate location of the Alaskan coastline. Cross sections A-A' and B- B' show the initial Vz and Vy conditions.

Figure 2.3. Mesoscale model geometry.

Figure 2.3. Mesoscale model geometry. This plot shows the topographic boundary conditions with a vertical exaggeration. The white arrows indicate the initial vectors applied to the model base over the area south of the white line (limit of the Yakutat- Pacific block. See Figure 2.1 for explanation of abbreviations.

encompasses the area of the St. Elias Range (640km x 890km x 20km) and discretized on

an approximately 5 km by 5 km horizontal and 2 km vertical grid (Figure 2.3). An

additional 2D geometry consists of a northeast-southwest (X-Z) cross-section of the

macroscale model (Figure 2.4; 1600 km (X-axis) by 50 km (Z-axis)) and is consistent with the geometry of the macroscale model. The general geometries used for this study have been standardized to allow for direct scaling and extrapolation of the results to other accretionary orogens.

Initial kinematics of the macroscale models are based upon realistic scaled values of measured relative plate velocities and assigned to the Pacific block (Figure 2.2;

DeMets et al, 1994; Fletcher and Freymueller, 1999; DeMets et al., 1999). Initial velocity conditions within the mesoscale models are imposed on the base of the model over an area corresponding to the spatial extent of the Yakutat terrane consistent with the

plate motions and macroscale models (Figure 2.3). All deformation within the models is driven by the initiated tectonic boundary conditions spatially defined by geological observations.

Two-dimensional models were developed using a commercial modeling environment (Comsol Multiphysics™) that solves the Fourier partial differential equation using simple boundary conditions on a triangular discretization of the user defined geometry (Comsol, 2006). Three-dimensional numerical models were completed using commercial explicit finite difference software (Fast Lagrangian Active Continuum;

FLAC3D) that simulates the mechanical behavior of materials undergoing deformation at

large strains (e.g. Cundall and Board, 1988; Koons et al., 2002; Upton et al., 2003;

Johnson et al., 2004) by transforming the model continuum into discrete nodes. FLAC3D

Figure 2.4. 2D Macroscale model geometry.

Figure 2.4. 2D Macroscale model geometry. Geometry of the 2D thermal model is defined similar to the Macroscale models, with an upper and lower crust and Pacific block. The Pacific block is given initial velocity conditions to facilitate thermal advection.

utilizes finite differences to approximate first-order space and time derivatives at the nodes within the three-dimensional mesh. To avoid errant mechanical responses, a mixed discretization meshing technique is utilized where coarser zones are superposed onto a finer tetrahedral discretization. This meshing system allows for deformation without volume change of individual tetrahedra, which would otherwise result in an over-stiff

response (Itasca, 2006).

Our modeling path first develops a Reference model to which later model

iterations are compared. The macroscale Reference model was developed with

mechanics decoupled from the thermal structure. Further models have fully coupled

thermal and mechanical solutions, as well as the application of an regional erosion model

(e.g. Hallet et al., 1996; Meigs and Sauber, 2000) to demonstrate the large-scale evolution

of southern Alaska and the role of temperature, development of rheological

heterogeneities, and erosion. The mesoscale reference Tectonic model was developed

without imposed surface boundary conditions. The mesoscale models then successively

vary the material properties, apply natural topography derived from a global 1-minute

DEM sampled at the model discretization (Smith and Sandwell, 1997), and apply a local

erosion scheme to demonstrate the evolution of the deformation patterns and the relative

importance of preexisting topography and climatic forcing.

2.2 Thermal Solutions

The initial stage of this project developed the thermal structure for the model

geometry using an isotropic conduction thermal constitutive model, which in later steps

forms the basis for assignment of mechanical constitutive models. The thermal structure

is developed by solving for the transient conductive-advective Fourier equation over the

2D and 3D model geometries at the discretization described above:

, 2.1

-1 -1 -3 where k is the thermal conductivity (Wm C ), ρ is the density (kgm ), Cv is the heat

capacity (Jkg-1C-1), is the kinematic velocity field (ms-1), and A is the radiogenic heat

source (µWm-3). The thermal model utilizes assigned initial velocities (2D initial thermal

structure models) or velocities derived from the mechanical solutions (3D fully coupled

thermo-mechanical models) to approximate thermal advection.

For all model geometries (both 2D and 3D), the surface temperature is maintained

at 0 ºC based upon abundant glacial ice, permafrost, and snow field cover within the

study area (Péwé, 1975), and fixed basal temperature of 780 ºC (50 km) over the North

America block, suggestive of previous heating following spreading ridge subduction

during the Eocene (e.g. Pavlis and Sisson, 2003 and references therein). This heating

event is well within the characteristic time for the decay of a thermal perturbation (~80

Ma) for a 50-km thick crust. The lateral boundaries of the North American block are no- flow boundaries, except the advective boundaries associated with the Pacific block.

3 The models are assigned constant values for heat capacity (Cv; 10 J/kgºC) and

density (ρ; 2700 kgm-3; Table 2.1). The North American block is further defined as two units: the upper 35 km of the model are assumed to be felsic composition and assigned a thermal conductivity (k; 2.6 Wm-1C-1) and radiogenic source term (A; 0.37 µWm-3)

Table 2.1. Material Properties.

Table 2.2. Material Properties. Material properties and constants assigned to the thermo-mechanical numerical models. Values of E, n, and A for diabase flow law after Mackwell et al (1998).

Constants Density 2700 kgm-3 Gas constant R 8.31e-3 JK-1mol-1

Mohr-Coulomb Model Bulk modulus K 105 MPa Shear modulus G 104.5 MPa Angle of friction φ 30 to 35 degrees Cohesion C 44 to100 MPa

Strain Softening Mohr-Coulomb Model Strain threshold ε 3 percent Angle of friction φ' 15 degrees Cohesion C' 0.1 MPa

Elastic Model Bulk modulus K 105 MPa Shear modulus G 104.5 MPa

Temperature-dependent Diabase Flow Law Activation Energy E 260 Jmol-1 Stress Exponent n-1 0.2941 A 2e-4 Pa-1s-1

Isotropic Thermal Model

-1 -1 Specific heat Cv 1000 Jkg C -1 -1 Conductivity ku 2.6 Wm C (upper 35 km) -1 -1 kl 1.0 Wm C (lower 15 km) Volumetric source A 0.37 µWm-3 (upper 35 km)

consistent with a granitic composition (Turcotte and Schubert, 2002). The lower 15 km

(35 km < z < 50 km), assumed to be mafic composition, are assigned a thermal

conductivity (1.0 Wm-1C-1) consistent with basaltic compositions and no radiogenic

source term. The assumption of a mafic lower crust is consistent with under-plating by the mafic oceanic plateau material of the Yakutat terrane (Plafker et al., 1994; Pavlis et al., 2004; Eberhart-Phillips et al., 2006; Fuis et al., 2008).

2.3 Mechanical Solutions

For the mechanical solutions, the indenting Pacific block is chosen to be an elastic material. The isotropic elastic model is the simplest representation of material behavior and is valid for a homogeneous, isotropic, continuous material that exhibits a linear stress-strain relationship. The Pacific block was used to apply the kinematic boundary condition and this study was not concerned with its internal deformation, therefore an elastic model was a valid choice. The isotropic elastic model stress increments are generated from strain increments according to the linear Hooke’s Law:

, 2.2

Where E is Young's modulus (MPa) related to the bulk (K ) and shear (G) moduli by

, 2.3

and,

, 2.4

in which ν is the Poisson's ratio. The shear and bulk moduli are assigned to the

numerical model at values characteristic of crustal rock (Table 2.1). The mechanical

solution utilizes K, G, and E during elastic deformation where all strain is recoverable. If

the stress state within the model dictates failure, deformation is permanent and is

calculated using the following Mohr-Coulomb or power law mechanical models. For the

numerical modeling presented here, only the Pacific block is defined as an elastic

material.

The rheological constitutive relationship for the North American block is

dynamically assigned based upon the resultant steady state advective-conductive thermal structure of the model, except for the macroscale Reference model. The mechanical behavior is linked to temperature by assuming the transition from frictional to plastic behavior occurs at a transient temperature and strain rate in a manner constrained by experimental analysis (Brace and Kohlstedt, 1980; Kohlstedt et al., 1995; Mackwell et al., 1998; Wijn et al., 2005), in our case the approximate onset of crystal plasticity in quartz at ~350 ºC (Hirth and Tullis, 1992; Gleason and Tullis, 1995).

The upper crust (<15 km or temperatures≤350 ºC) is assigned a pressure-

dependent Mohr-Coulomb constitutive model. The standard associated plasticity of the

Mohr-Coulomb behavior (macroscale Mohr-Coulomb models) is also altered in

subsequent models by allowing the model to weaken with accumulation of strain

(macroscale Strain-Softening models), simulating the formation of faults and shear zones

by the non-associated strain-dependent material weakening (e.g. Vermeer and de Borst,

1984; Hobbs et al., 1990; Buck and Poliakov, 1998; Montesi and Zuber, 2002). The shear yield function for the Mohr-Coulomb model is given by:

, 2.5

Where f s is the shear failure criterion, C is cohesion (MPa), and N is given by:

, 2.6 where φ is the internal angle of friction. The Mohr-Coulomb models had assigned values

(Table 2.1) for φ (30 to 35º), cohesion (C; 44 to 100 MPa), shear modulus (G; 104.5 MPa), and bulk modulus (K; 105 MPa) consistent with a broad range of geological materials.

The Strain-Softening models have assigned values of φ (15º) and C (0.1 MPa) dependent upon a hardening/softening criterion based on the second invariant of the plastic shear- strain increment tensor (0.03 to 0.05) consistent with faulted rock (Buck and Poliakov,

1998; Montesi and Zuber, 2002; Nagel and Buck, 2004).

The lower portion of the macroscale North American block, with temperatures greater than 350 ºC, was assigned a plastic yield behavior based upon the von Mises constitutive model. The shear yield function is given by:

, 2.7

where τ is tangential stress, kφ is a material property related to strength in shear (MPa).

The numerical model derives values for kφ through the experimentally derived flow law:

. 2.8

The values for A, E, and n are material constants based upon published experimental data

for diabase (Table 2.1; Mackwell et al., 1998), and R is the universal gas constant. The

temperature (K) is taken from the model thermal and mechanical solutions, respectively.

2.4 Model Megathrust

Previous 2D continuum numerical models developed to study the transfer of

stress across a megathrust used 10-km thick slip layers that included thermally-dependent rheologies based upon the age (and therefore temperature) of the oceanic lithosphere

(Wdowinski, 1992; Yanez and Cembrano, 2003). These studies found that the large-

scale, first-order deformation patterns are dependent upon the viscosity and/or

convergence rates, and suggest a "strength" for the slip layer of 30-90 MPa (equivalent to

k ; Yanez and Cembrano, 2003). This method assigned a constant viscosity to the slip

layer and therefore did not reproduce the variation of strength of the megathrust with slab

angle, depth, and temperature, as evidenced by seismicity patterns (Peacock and Wang,

1999; Gutscher et al., 2000a; Gutscher and Peacock, 2003).

This study departs slightly from the continuum mechanics approach by defining a

frictional interface separating the North America and Pacific blocks. This interface separates the two blocks and represents a plane on which sliding can occur. This sliding is characterized by Coulomb sliding criteria that relies on the angle of friction (φ=20º) and normal and shear stiffness (each 700 MPam-1) of the interface. The interface allows

for shear yield under the control of these material properties, which are assigned based upon properties of geological faults (e.g Buck and Poliakov, 1998; Nagel and Buck,

2004; Giger et al., 2008). The use of an interface allows for realistic transfer of strain from the Pacific to North American block similar to the expected transfer of strain across the Alaskan megathrust.

2.5 Sensitivity Analysis and Limitations

Observations from seismic analysis and geodetic studies have been employed to develop the model geometry, boundary kinematics, and an initial thermal state of the plate boundary (Pavlis et al., 2004; Eberhart-Philips et al., 2006; Abers et al., 2006; Fuis et al., 2008). The thermal structure of the model, and consequently the rheological structure, are strongly influenced by the plate boundary geometry and previous geological history (e.g. Byrne, 1979; Bradley et al., 2003). The advective thermal pattern developed in the initial thermal model is characteristic of long-standing subduction zones and agrees well with previous 2D numerical studies (e.g. Gutscher and Peacock, 2003). Additional verification and validation of mechanical and thermal solutions was completed utilizing a series of sensitivity analysis models (see Appendix B).

The sensitivity analysis models tested the influence of each mechanical and thermal variable by independently testing them across a broad range of values that encompassing geologically reasonable estimates. Values of φ, C, G, and K were varied for the pressure-dependent upper crustal Mohr-Coulomb rheology (see equations 2.1,

2.5, and 2.6). The model results were generally unaffected by changes to G and K, as the deformation was not recoverable. Resultant orogen shape (magnitude of width and vertical uplift) were sensitive to variation of φ and C, however, the location of the

orogens was consistent. Values of kφ are varied independently from the thermal structure

(uncoupled models) by calculating a strength profile using the flow law (Equation 2.8)

given static geothermal gradients of 12.5, 15.5, and 19.5 Cm-1. The 'normal' strength

profile (0.8 < kφ < 1000 MPa; given a geothermal gradient of 15.5 Cm-1) produces an

uplift pattern consisting of a developing orogen associated with the down-dip limit of the

Pacific block. This pattern is consistent for the slightly stronger and weaker lower crust

trials (0.5 < kφ 1500 MPa). Trials with very weak lower crust (kφ < 0.5 MPa) did not converge due to grid failure, suggesting a lower limit to the strength of the crust. Trails with very strong lower crust (kφ > 1000 MPa) produced a broad area of uplift.

Sensitivity to the thermal parameters was completed on the 2D thermal models.

These trials varied the values of k and A within the range of accepted values (Turcotte

and Schubert, 2002; Fowler, 2005). Additionally, boundary condition effects were also

tested by including a heat flux condition at base of the crust instead of a fixed

temperature. Varying the value of k causes the thermal perturbation of the down-going

slab to propagate much slower (lower k) or faster (higher k). The values of thermal

conductivity used for this study are approximate averages of published values for the

appropriate rock types (Fowler, 2005). Increasing the internal heat generation, A, and/or

basal heat flux will increase the thermal gradient of the crust.

Heating due to mechanical work along faults and shear zones has been suggested

as holding important implications to the thermal structure, and therefore strength of the

lithosphere (Hirose and Bystricky, 2007; Burg and Schmolhotz, 2008; Hartz and

Podladchikov, 2008; Regenauer-Lieb et al., 2008). Hartz and Podladchikov (2008)

suggest that shear heating can cause an order of magnitude decrease in the strength of the lower crust. Trials were completed with inclusion of mechanical heating (Am) utilizing:

2.9

-1 where τxz is the shear stress along the interface (MPa), v is the velocity (ms ), and w is the width of the fault zone (m; Ponko and Peacock, 1995). The sensitivity analysis models include a mechanical heat source term that varies in thickness (1 to 20 km) and magnitude (10-6 to 10-5 Wm-3) associated with the megathrust. At the values used, the inclusion of a mechanical source term had little effect on the thermal structure.

Geometric controls were tested by varying the Pacific block shape by changing the dip of the megathrust. These models indicate that the dip of the slab provides an important control on the location of the resulting orogen, with a relatively steeply dipping slab (~45º) producing a narrow, near trench (<200 km from the trench) orogen. The model used a relatively flat slab subduction consistent with the observed geometry of southern Alaska (Eberhart-Phillips et al., 2006; Fuis et al., 2008). Additionally, preexisting weaknesses in the forms of interfaces acting as large terrane bounding faults

(e.g. Denali, Boarder Ranges, and/or Chugach-St. Elias Faults) were also tested. These interfaces effectively partition strain onto localized slip planes and can drastically change the deformation patterns. However, they were not included within the models presented herein as they provide artificial control to the deformation patterns, over define the continuum model, do not allow thermal diffusion across the interface surfaces, and make the numerical solutions less stable.

Finally, the role of kinematics was tested by varying the angle and magnitude of

the initial vectors. Changes to the angle of convergence altered the location of maximum

deformation, with respect to the direction of maximum convergence. Altering the

magnitude of convergence caused higher strain rates and faster uplift. The kinematics used for the results presented in Chapters 3, 4, and 5 are based upon accepted plate vectors for the Pacific plate and Yakutat terrane (DeMets et al., 1994; Fletcher and

Freymueller, 1999; DeMets and Dixon, 1999).

The model technique that was used for this study precludes long-term (>2 Ma)

model runs due to the failure of the Lagrangian grid. Additionally, by focusing on the

deformation within the crust, the full lithosphere-asthenosphere system is not examined.

Static compositional estimates and the evolving thermal structure of the lithosphere are

utilized to define the mechanical behavior of the model. The subduction-driven

advective perturbation the model thermal structure results in a complex spatial

distribution of mineral stabilities and therefore rheological heterogeneities (e.g. Royden,

1996; Ellis et al., 1998; Liu and Yang, 2003; Hacker et al., 2003b). The rheological

models utilized in this study inadequately describe the petrologically-induced rheological

variations. Finally, due to overestimates in porosity, FLAC3D has a lower limit of 10%

porosity, the models do not include a fluid phase.

Chapter 3 : The Influence of Thermal Perturbations on the Characteristic

Signal of Terrane Accretion

3.1 Introduction

The general deformation patterns associated with terrane accretion have been partially captured within the mechanical force balance framework of critical wedges (e.g.

Chappel, 1978; Davis et al., 1983; Koons 1990; Willett et al., 1992; Whipple, 2009).

Intrinsic to the force balance description has been the assumption of lateral homogeneity.

This physically unrealistic assumption has been overcome in some investigations through employment of numerical solutions to investigate the influence of laterally varying material properties (e.g. material weakening related to strain, thermal, or fluid perturbations; Upton et al., 2003; Groome et al., 2006; Upton et al., 2009) but the assumptions of lateral material homogeneity have often been retained because of the lack of general predictability about the properties of local material caught up within an accretion zone. Here we argue that one major source of material heterogeneity during accretion is omnipresent due to thermal-mechanical non-linear positive feedbacks and that these feedbacks impose a characteristic shape on accretionary orogens.

The interaction of tectonic driving forces and variations of the rheological structure of the lithosphere influence strain localization and deformation patterns (Liu and Yang, 2003; Royden, 1996; Ellis et al., 1998). A hot, relatively weak lower crust favors the formation of large plateau where most of the strain is localized into large, crustal scale faults (i.e. Tibet and the central Andes; Babeyko and Sobolev, 2005;

Beaumont et al, 2001; Beaumont et al, 2004, Gerbault et al, 2005). Conversely a cold,

relatively strong lower crust decouples from mantle and forms a narrow thin-skinned

fold-and-thrust belt (i.e. Southern Alps; Ellis et al., 2004). The deformation patterns of

an evolving convergent orogen are therefore strongly influenced by advective

perturbation of the thermal structure by under-thrusting of relatively cold lithosphere.

In addition to the thermal dependence of the rheology of the crust, the transfer of

strain from the down-going to over-riding plate is controlled by strength of interface, which is also sensitive to temperature, sediment flux, and fluid production (Wdowinski,

1992; Reyners and Eberhart-Phillips, 2009). This transfer of strain to the overriding plate has long been recognized as the mechanism driving deformation within subduction zones.

Vertical thermal advection of cold material, herein referred to as tectonic refrigeration, resulting from shallow angle subduction of the relatively thick or young oceanic plate

causes the offset of the hot mantle wedge flow, development of a relatively wide

seismogenic zone, and creation of a cool lower crust within the fore-arc (Gutscher and

Peacock, 2003; Eberhart-Phillips et al., 2006; Rossi et al., 2006). The variation in

strength of the megathrust is a function of sediment flux, slab angle, depth, and

temperature, as evidenced by seismicity patterns (Peacock and Wang, 1999; Gutscher et

al., 2000a; Gutscher and Peacock, 2003). In southern Alaska, thermal advection driven

by the subduction of a cold, thick oceanic plateau forms a broad seismogenic zone that is

responsible for some of the largest strain releases in recorded history (Johnson et al.,

1996; Shennan et al., 2009).

Previous modeling of the thermal structure of subduction zones utilized two- dimensional stationary heat transfer models that included advective transport, conduction, mantle flow, and heat sources (Peacock and Wang, 1999; Hyndman et al., 1995; Hacker et al, 2003b). These models indicate thermal advection provides an important control on seismic activity of the megathrust through a series of metamorphic reactions (Hacker et al, 2003a,b).

It was also noted that subduction zones can be differentiated based upon their earthquake potential into seismic (e.g. Chile) and aseismic (e.g. Marianas) groups (Uyeda and Kanamori, 1979). This differentiation was based upon the generation of back-arc basin, which Uyeda and Kanamori (1979) suggested was due to stress generated due to the coupled (Chile, no basins developed) or uncoupled (Marianas, basin developed) nature of the contact zone between the two plates. The nature of this contact zone, the megathrust, is that the transfer of tectonic force (i.e. derived from motion of the down- going lithosphere) shears the overriding lithosphere (Wdowinski, 1992).

This study proposes that advective tectonic refrigeration caused by the flat-slab subduction of a relatively cold crustal block causes an overall strengthening of the fore- arc that influences a temporal and spatial structural reorganization of the regional tectonics. The feedback between the vertical advection of cold lithosphere and temperature-dependence of rheology creates a stronger fore-arc region taking the shape of a sliver of frictional material along the surface of the dipping slab. This frictional sliver, the model equivalent of the seismic megathrust, influencing the modeled spatial reorganization of deformation patterns of the overriding crust by controlling the transfer

of strain through shear coupling. The increased cooling associated with flat slab subduction results in a wider seismogenic zone and therefore broader zone of shear coupling. The development of the relatively strong frictional sliver provides an important control on the location and formation of diachronous secondary orogenic wedge, demonstrating the importance of the thermal structure on strain partitioning and indicating that the development of this local heterogeneity has profound effects on the regional scale evolution.

3.2 Numerical Methods

This study uses both 2D thermal and 3D thermo-mechanical numerical models to test the role of thermal perturbations in the development and evolution of deformation patterns within an accretionary orogen. The 2D thermal models use a commercial modeling environment that solves the Fourier partial differential equations (Equation 2.1) using simple boundary conditions (Comsol, 2006). The 3D models are completed using a commercial finite element code that solves for simple stress–strain relationships, applied forces or kinematic boundary conditions, and heat flow to simulate the mechanical and thermal behavior of Earth materials (Cundall and Board, 1988; Koons, 1990). The models employ a simple conductive thermal constitutive model coupled with the derived kinematics of the associated plastic mechanical model to simulate a conductive-advective thermal solution over the entire model geometry. The initial thermal solutions are then used to assign a rheological structure to coupled thermo-mechanical models that investigate the role of local thermal and mechanical heterogeneities in the development of

the regional characteristic signal of an accretionary orogen. See Chapter 2 for additional details of the numerical methods and model geometries.

The thermal models presented herein represent two end-members of thermal perturbation with southern Alaska representing an intermediate stage. The 2D modeling assumes the thermal structure evolves from initial conditions where no slab exists and tracks the evolution of the thermal structure over 10 million years. The 3D models assume an instantaneous emplacement of the shallow slab. Due to limitations of the models, the transition from normal subduction to flat slab subduction is not captured, however, the two end member cases provide constraints on the time frames.

3.3 Results and Discussion

3.3.1 Thermal Evolution

The subduction of the relatively cold incoming block alters the initial conductive thermal solution by causing the isotherms to become inverted in the fore-arc region due to advective tectonic refrigeration (e.g. Gutscher and Peacock, 2003). The extent of the inversion of the thermal structure is partially controlled by the kinematic, as well as the material properties (e.g. radiogenic heat source, thermal conductivity, etc) and geometry.

Long-standing subduction zones, such as that of southern Alaska (>30 Ma; Plafker et al.,

1994), should have well developed advective thermal signatures. The mechanical signature of tectonic refrigeration due to the subduction of a relatively cold plate is the formation of a thin zone of cooler material forming along the surface of interface between

the two plates. This sliver of material will be at temperatures where the dominant mode

of deformation is brittle fracturing and corresponds to the upper seismic zone (Gutscher

et al., 2000a; Gutscher and Peacock, 2003). The change in material behavior caused by

this perturbation of the thermal structure has profound influences on the pattern of

deformation expected and observed within subduction zones.

The 2D thermal and 3D fully coupled thermo-mechanical model models

reproduce the long term (10 Myr) and short term (500 ky) evolution of the thermal

structure associated with the subduction of oceanic lithosphere (Figures 3.1 and 3.2;

Animation 3.1). This model results in the inversion of the isothermal structure due to the

vertical motion of the model slab, the Pacific block, driving cooling of the lower crust

leading to inversion of the isotherms above the slab. The temporal evolution of this

thermal structure shows that the isotherms roll back away from the trench resulting in

overall cooling of the near trench, or forearc, region of the model. The constitutive

mechanical model and rheological properties are intimately linked to the thermal

structure through the temperature-dependent flow law (Equation 2.8). The inversion and

cooling of the material directly above the dipping Pacific block caused by of the

advective perturbation of the thermal structure develops the thin, cold, relatively strong

sliver that behaves as a frictional material (Figure 3.3). The lateral (Y-axis) variation in the thermal structure causes an additional thermal gradient to form along the eastern margin of the Pacific block, cooling and strengthening that portion of the lithosphere as well. These frictional slivers has important implications to the mechanics and seismicity within the evolving orogen.

Figure 3.1. 2D thermal results.

Figure 3.1. 2D thermal results. Time series of cross section plots (north-south) of the resulting thermal structure from the 2D thermal models (°C). The 350°C is highlighted for reference. The dipping black line denotes the interface between North American and Pacific blocks (See Figure 2.4). Vertical exaggeration - 2X.

Figure 3.2. 3D thermal results.

Figure 3.2. 3D thermal results. Time series of cross section plots (north-south @ 600 km) of the resulting thermal structure from the 3D thermal models (ºC). Note the advective signature due to convergence at X 300 to 400 km. Vertical exaggeration - 2X.

Figure 3.3. Model Rheology

Figure 3.3. Model rheology. A) Initial thermal structure (see Figure 3.2) is based upon the 3D thermal model. B) Reference and C) Fully-coupled model rheological structures. Differential stress vs. depth plot shown demonstrates the approximate crustal strength profile. Note the absence of the frictional sliver in the Reference model rheological structure.

3.3.2 Thermo-Mechanical Evolution

Coupling between the thermal and rheological structures of the lithosphere can be

seen within the observable, natural deformation patterns. The pattern of deformation

(Figures 1.1 and 2.1) formed in response to the ongoing terrane accretion in southern

Alaska consists of dual areas of uplift defined by the Alaska and St. Elias Ranges (Koons

et al., 2009). The fully coupled thermo-mechanical models produce similar deformation

patterns, including the formation of dual orogenic wedges. Material moves into the

deforming zone of the model from the south, forming the fore-arc Inlet Orogen. Material

exits the deforming zone near the northern limit of the down-going slab, forming the

Outlet Orogen.

The characteristic deformation patterns of accretionary tectonics are demonstrated

in the series of models developed using the simple conductive thermal structure, where

the mechanical constitutive models are not defined with respect to the thermal structure

(Reference model). The boundary conditions for these conductive solutions are the same

as those for the advective-conductive solution, only the development of advective

heterogeneities do not influence the development of rheological heterogeneities. The

boundary conditions of these models results in the formation of an orogenic wedge above

the leading edge of the down-going slab, the Outlet Orogen, with strain transferred across

the model by the lateral transform fault (Figure 3.4).

The advective mechanical strengthening the of fore-arc, discussed above, causes a spatial reorganization of deformation patterns associated with subduction within the

Fully-coupled models (Figure 3.5). These deformation patterns are characterized by the

Figure 3.4. Reference model mechanical results.

Figure 3.4. Reference model mechanical results. Time sequence (model time indicated at left) cross sections (north-south @ 600 km) of vertical displacement (scale bar at right; meters) and velocity vectors (see reference vector; myr-1). Location of the Outlet Orogen is indicated (see Figure 2.1).

Figure 3.5. Fully-coupled model mechanical results.

Figure 3.5. Fully-coupled model mechanical results. Time sequence (model time indicated at left) cross sections (north-south @ 600 km) of vertical displacement (scale bar at right; meters) and velocity vectors (see reference vector; myr-1). Locations of the Inlet and Outlet Orogens are indicated (see Figure 2.1). Compare patterns with those in Figure 3.4. Three paths (blue, black and red) are indicated on 500 ky time step are steady state flow paths through the orogen.

development of a second orogenic wedge, the Inlet Orogen, located spatially above the

down-dip limit of frictional sliver. Both orogenic wedges are actively deforming,

however, the bulk of the strain is accommodated by the Inlet Orogen, suggesting the two

orogens are teleconnected spatially. The location, width, and magnitude of this Inlet

Orogen orogenic wedge are related to the basal strength and slab dip. Only the effects of

the thermal structure on the development of a dual orogenic wedge system will be

considered here, however, sediment flux and fluid pressures also play important roles

(Upton et al., 2003; Reyners and Eberhart-Phillips, 2009). The basal boundary condition

and slab dip have similar effects on the location and geometry of the dual orogenic wedge

system. The coupled thermal-mechanic model diverges from the simple mechanical model due to the coupling between the down-going and overriding plates. This coupling is a function of the transfer of strain, which can be directly linked to the mechanical constitutive model. The pressure-dependent rheology allows for better coupling, which is responsible for the development of the Inlet Orogen. The formation of two teleconnected orogenic wedges is diagnostic of the influence of thermal and mechanical heterogeneities on the characteristic signal of accretionary tectonics.

3.4 Petrology and Strain Transfer

Metamorphism within the down-going basaltic composition slab is dominated by dehydration reactions (Figure 3.6; Bucher and Frey, 2002; Hacker et al., 2003b). These reactions have important implications for magmatism and seismicity within the active plate boundary. The following discussion will be framed in terms of comparing the

Figure 3.6. P-T Facies Diagram.

evolution of the down-going slab of consolidated mafic crust (Bucher and Frey, 2002;

Hacker et al., 2003b) with modest water content (5-6 wt %; Schmidt and Poli, 1998;

Hacker et al., 2003b) over three distinct particle paths; an average P/average T path

(average path; Figure 3.6 path A) characteristic of an average subduction angle of 45º

(Costa Rica; Protti et al, 1995), a low P/high T path (low pressure path; Figure 3.6 path

B) characteristic of flat slab subduction (Cascades; Hyndman and Wang, 1993) or

collisional tectonics (southern Alaska), and a high P/low T path (high pressure path;

Figure 3.6 Path C; NE Japan; Igarashi et al., 2001). Results from the numerical models will be discussed in terms of the petrologic evolution of the crust and implications for strain accumulation and seismicity of the megathrust.

The initial, relatively low pressure path reactions involve the breakdown of prehnite- or

pumpellyite-bearing subgreenschist facies rocks, which are stable under 300±30 ºC and

600 MPa, to form greenschist facies rocks. These reactions produce the first appearance

of the characteristic assemblage of actinolite + epidote + chlorite + albite ± quartz, at the

expense of prehnite at low pressures (<400 MPa; Path B) and pumpellyite at higher

pressures (>400 MPa; Path A). This transition from subgreenschist to greenschist facies

releases a modest amount of fluid (~2 wt. %).

In the high pressure path (Figure 3.6 path C), subgreenschist facies may transition

directly to glaucophane ± lawsonite ± chlorite-bearing blueschist facies rocks, which is

essentially a fluid conserving reaction (Schmidt and Poli, 1998; Hacker et al., 2003).

This transition involves the breakdown of the low grade chlorite + albite + actinolite

assemblage in favor of the relatively high-grade glaucophane + paragonite + epidote

assemblage. Remaining albite (if albite is more abundant than chlorite) will be converted

to jadeite + quartz with the transition to the blueschist facies, however, albite is usually

completely destroyed by the earlier reactions. The blueschist facies transitions at higher

temperature (400-550 ºC) and pressures (> 1.6 GPa) to eclogite facies as the assemblage

glaucophane + paragonite is replaced by omphacite + garnet. This reaction releases most

of the remaining fluid within the rocks. Hydrous phases lawsonite and paragonite are

progressively dehydrated and removed in the transition to eclogite facies and formation

of the eclogite kyanite + jadeite assemblage, respectively.

The low pressure and average paths can both transition directly from greenschist

to amphibolite facies conditions. This transition involves the progressive replacement of

actinolite + epidote + chlorite producing an assemblage of alkali- and Al-rich amphibole

+ anorthite ± garnet ± epidote or chlorite (depending upon relative abundances), and a

substantial release of fluids from initial concentrations of ~7 wt. % in the subgreenschist

facies to 2-3 wt. % in the amphibolite facies (Hacker et al., 2003b). The average path

may also record the transition from greenschist to epidote-bearing blueschist grade rocks.

The blueschist rocks may contain up to approximately 2 weight percent more water than their lower pressure amphibolite equivalents (e.g. Hacker et al., 2003b). The next major reaction to occur within the slab are those that consist of the transition from blueschist or amphibolite to amphibole-bearing eclogite (Figure 3.6; Hacker et al., 2003b).

Linkages between seismicity patterns and metamorphism are well established

(Peacock and Hyndman, 1999; Oleskevich et al., 1999; Gutscher and Peacock, 2003;

Hacker et al., 2003b). Previous studies have utilized projected P-T paths and the

metamorphism of the down-going plate to explain the pattern of seismic hazard (Gutscher

and Peacock, 2003). Our results are compatible and complimentary to these previous

studies in that the mechanics, metamorphism and P-T paths agree that the advective component provides the key influence on the transfer of strain. We, however, suggest that the heterogeneities in the thermal field also provide an important influence on the overall patterns of deformation within the overriding plate.

The upper (lower temperatures and pressures; Figure 3.6 path C) limit for seismicity is consistent with the transition from ductilely deforming clay minerals to illite and chlorite (Gutscher and Peacock, 2003). In warm subduction zones, such as Cascadia, the down-dip limit of seismicity is controlled by the interface temperature, ~350-450 ºC

(Peacock and Hyndman, 1999). However, within relatively cold subduction zones such as southern Alaska, the lower, down-dip limit of the seismogenic zone (higher

temperatures and pressures; Figure 3.6 path A) coincides approximately with the

intersection of the megathrust with the forearc mantle wedge (~40 km; Peacock and

Hyndman, 1999) where fluids migrating from the dehydrating slab alter the local

mineralogy to a serpentine + brucite assemblage within the mantle and talc-rich rocks at

the interface. All three of these phases promote aseismic behavior at relatively shallow

depths. Dehydration reactions involving serpentine and talc have been suggested to

cause observed intermediate depth earthquakes at the slab surface (Kirby, 1995; Hacker

et al., 2003b). The location of the Inlet Orogen and the extent of the frictional sliver

compare favorably with the observed maximum in seismic hazard for southern Alaska

(Figure 3.7).

Figure 3.7. Seismic hazard of south-central Alaska.

Figure 3.7. Seismic hazard of south-central Alaska. U.S. Geologic Survey seismic hazard map for Alaska with major fault systems noted (black lines; see Figure 1.1; Plafker et al., 1994; Pavlis et al., 2004). Scale for hazard map is peak ground acceleration (ms-2) with 10% probability of exceedance in 50 years (Petersen et al., 2008). Dashed red line indicates the down-dip extent of the frictional sliver within the Fully-coupled models and is derived directly from the evolved (advected) model results. Note that the western portion of the frictional sliver migrates to the south in response to the initial Vz conditions (see Chapter 2.1).

The transfer of strain across the frictional sliver is a function of the mechanical properties of the interface and rheology of the material on either side. The mechanical behavior of the megathrust with depth, with a transition from friction to plastic flow at down-dip limit of sliver, can be seen in the velocity vectors (Figure 3.8). The frictional zone, which is more efficient at transfer of strain, does not form a velocity gradient across the megathrust, and therefore only forms a large-scale shear zone encompassing the lower, temperature-dependent portion of the megathrust. This suggests that there is the potential for a larger accumulation of strain energy in the upper, low-temperature portion of the megathrust, which would result in potential for larger magnitude seismicity (Figure

3.8; Gutscher et al., 2000a).

A lithospheric scale shear zone forms along the entire down-dip extent of the megathrust indicating a more uniform transfer of strain across the entire interface within the Reference model (Figure 3.8). This indicates that the model interface is at temperatures above the down-dip limit for seismicity, consistent with less strain transfer across the megathrust. For the Fully-coupled model, a shear zone forms along the megathrust down-dip from the extent of the frictional sliver at ~30-40 km depth, approximately consistent with the transition from amphibolite to garnet amphibolite in path B of Figure 3.6 (Figure 3.8). The different strain patterns on the model interface demonstrate the influence of temperature on the transfer of strain and seismicity.

Figure 3.8. Megathrust strain patterns.

Figure 3.8. Megathrust strain patterns. Contour plots of vertical shear stress (εyz; See Appendix D) and displacement (Vz; m) for A) Reference and B) Fully-coupled models. The red and black lines on the displacement plot indicate the locations of the megathrust and large-scale thrust faults, respectively.

3.5 Petrological Implications

Hydrated oceanic crust contains between 5 and 6 wt% water hosted in lawsonite,

zoisite, chloritoid, talc (±phengite), chlorite, and serpentine. Complete dehydration, at

eclogite facies conditions, is achieved between 70 and >300 km depth and is a function of

individual slab dip and resulting geothermal gradients (Schmidt and Poli, 1998). The

flux of the expelled fluid into the mantle wedge is an important component in creating

partial melts (Peacock, 1990). It is generally accepted that the flux of this

metamorphically derived fluid into the hot mantle wedge (~1300 ºC; Schmidt and Poli,

1998) is at least partially responsible for subduction zone magmatism. Slabs with dips of

less than 25º do not have mantle wedge temperatures high enough to produce melts

(Philpotts, 1990; Peacock, 1990). This is consistent with the model results from this and

previous studies that indicate the observed break in volcanism in central Alaska is related

to the shallow dip of the subducting Yakutat terrane offsetting the hot mantle wedge flow

to the northwest (Munoz, 2005; Eberhart-Philips et al., 2006; Abers et al., 2006). Finally, the presence of adakitic volcanism associated with the Wrangell Mountains (Preece and

Hart, 2004) suggests that slab melt played a role in the early stages of flat-slab subduction. Gutscher et al. (2000b) suggest that during the initial stages of flat-slab

subduction the continued presence of the high-temperature mantle wedge allows for slab

melting. This adakitic component would continue to move spatially with the propagation

of the flat slab until the mantle wedge has been completely displaced, creating a gap in

volcanism commonly associated with flat-slab subduction (Gutscher et al., 2000b; Mori

et al., 2007).

It has long been noted that seismicity within continental scale transform faults has

been restricted to the pressure-dependent, brittle portion of the crust (e.g. Chen and

Molnar, 1983; Maggi et al., 2000; Handy et al., 2007), indicating a link between

accumulation and/or transfer of strain and metamorphic grade (Handy et al., 2007).

Previous numerical studies have relied upon "melt weakening" for the formation of large

shear zones, or zones of channel flow, to explain exhumation and deformation within the

Himalayan orogeny (Beaumont et al., 2004; Jamieson et al., 2002). The linkage between

crustal-scale deformation patterns, rheology and petrology relies in part upon pressure- temperature (P-T) paths, which can be easily derived from thermo-mechanical models, as

well as reaction kinetics (e.g. Thompson and Rubie, 1985; Baxter and DePaolo, 2000;

Koons et al., 2002). The advection of material through the orogen defined by the model

vectors can be linked to the model pressure and temperature conditions (Figure 3.9). This

produces counter-clockwise P-T paths that record maximum conditions within the

amphibolite (0.8 GPa and 600 ºC) range for material starting in the middle crust that

outlet in the Outlet Orogen. Rock exhumed in the Inlet Orogen would have experienced

at maximum subgreenschist facies metamorphic conditions (0.4 GPa and ~200 ºC). The

P-T paths for moderate- to high-grade lower crustal rocks that follow the subduction

Pacific block trajectory do not cross the dehydration melting solidus curve, indicating the

models have captured an additional plausible cause for the observed volcanic gap;

temperatures necessary for melting are not reached. This petrological model technique

does, however, assume that the metamorphic conditions are a result of the Yakutat

collision and not a previous tectonic event.

Figure 3.9. Derived P-T paths.

The pressure and temperature condition can be used to identify the 3D spatial

relationships of the simple metamorphic facies within the numerical model geometry

(Figure 3.10). This yields a 2D petrologic representation of the entire orogen that can

then be used to identify petrologic patterns associated with oblique terrane accretion,

tectonic corner structures, or flat slab subduction in ancient orogenies (Figure 3.9). In a

similar study, Hacker et al. (2003b) utilized simple physical properties of minerals

associated with subduction zones to construct theoretical phase diagrams and simple

material properties (e.g. water content and densities). These data were then compared

and superimposed on 2D P-T models of subduction zones yielding spatial distribution of

metamorphic facies.

The numerical model can be examined extensively, with the easy extraction of

cross sections, map view slices, and particle paths on a variety of spatial and temporal

scales that would allow for a complete three-dimensional description of the petrological

evolution of an orogen. This technique has an advantage over applied geobarometry and

geothermometry in identifying areas of ancient thermal perturbations or tectonic

aneurysms in that it allows for the 3D investigation of the entire study area. Additionally,

the thermodynamic calculation of P-T pseudo-sections (i.e. using PurpleX (Connelly,

2005), TWQ (Berman, 2007) or ThermoCalc (Powell et al., 1998)) based upon the identification and composition of mineral assemblages would be complimentary to the numerical model results.

Figure 3.10. Spatial distribution of metamorphic facies.

The use of coupling of thermo-mechanical modeling and petrology as the sole test

for the model results does require some knowledge of the petrology for the area of

interest, that is the pressure and temperature conditions and approximate particle

trajectories. However, this technique holds promise as a testing technique of the

tectonics of ancient orogens, such as the Acadian, through comparison of natural and

modeled petrologic evolution.

3.6 Conclusions

The numerical modeling results presented here demonstrate the important

influence exerted by perturbation to the thermal structure of the lithosphere during flat-

slab subduction. Deformation patterns produced in a Reference model, that does not

account for the effects of vertical advection, consist of the development of a single

orogen at the down-dip extent of the indenting Pacific block. This basic patterns are

altered within the Fully-coupled model where vertical advection of cold crustal rock

causes cooling and resultant strengthening of the model forearc through the creation of a

sliver of frictional material within the fully coupled thermo-mechanical model. This strengthening drives a temporal and spatial structural reorganization of the orogen, including a trench-ward jump in the deformation front. The modeled large-scale spatial

deformation pattern from the Fully-Coupled model is consistent with observed deformation within southern Alaska, as well as other locations such as the Laramide and

Acadian orogenies, indicating feedback between thermal perturbation and development of rheological heterogeneities has a profound impact on the characteristic signal of terrane accretion.

Chapter 4 : Macroscale Geodynamics of Southern Alaska

4.1 Introduction

There is a dynamic link between long-wavelength, 100 to 1000 km-scale

landscape development and characteristic spatial and temporal deformation patterns

within an evolving compressional orogen resulting from the accretion of an exotic

terrane. This chapter focuses on the thermal and mechanical evolution of the crust during the accretion of a micro-continent, the Yakutat terrane, within southern Alaska and

demonstrates the influence of the thermal structure and evolving rheology on the

observed geomorphology and characteristic signals of deformation. As such, southern

Alaska is framed within a single coherent three-dimensional, fully coupled, thermo- mechanical dynamic numerical model that studies the evolution of the crust during the early stages of convergence. The relationship between convergent tectonics, characteristic signal of deformation, and landscape development indicates a dynamic, non-linear relationship that explains some aspects of the observed geological evolution of southern Alaska.

4.2 Geodynamics of Southern Alaska

The tectonic geometry of the southern Alaskan active plate boundary can be

divided into zones based upon observed kinematics, geomorphology, and style of

deformation (Figure 4.1):

• Transform Boundary: The eastern margin of the Yakutat terrane is

bounded by the dextral Fairweather transform fault, while farther to the

east and continuing to the northwest is the dextral Denali fault. Strain may

be transferred from the Fairweather fault to the Denali fault through the

Totschunda fault, although that connection is not confirmed (e.g. Plafker

et al., 1977; Matmon et al., 2006; Kalbas et al., 2007). A narrow area of

uplift, the Fairweather range, occurring in a zone restricted to the

microplate boundary is most likely related to the oblique collision of the

Yakutat terrane and not to the low angle (~15º) westward bend within the

Fairweather fault (e.g. Plafker et al., 1994; Bruhn et al., 2004). The

dominant mode of deformation within this zone is related to the dextral

strike-slip motion of the Denali and Fairweather faults. Within this zone,

the <5 Ma volcanoes of the Wrangell Mountains consist of a complex

spatial distribution of calc-alkaline, adakitic and tholeiitic lavas (Preece

and Hart, 2003) derived from a mid-ocean ridge basaltic like mantle

wedge enriched by the addition of subduction components. These

volcanoes were constructed upon bedrock high potentially formed during

Figure 4.1. Geodynamics of southern Alaska.

Figure 4.1. Geodynamics of southern Alaska. Map showing location of important geological, geomorphological, and structural features of southern Alaska. The white box indicates the spatial extent of the numerical model geometry. See text for further explanation. Note the location of some noted features are used for spatial reference in the following plots.

the initial stages of Yakutat convergence. Earlier volcanic rocks deposited

from ~30 Ma to present have a pattern of successive deposition and

extinction from east to west (Plafker et al., 1994; Preece and Hart, 2003).

The relationship between the earlier and later volcanics and tectonics is

not clear.

• Transform-Compressional Transition: The transition from the northern

observed limit of the transform motion of the Fairweather and Denali

faults to convergent deformation associated with the St.Elias Range

comprises this boundary (e.g. Pavlis et al., 2004; Chapman et al., 2008).

This nexus of oblique strain is spatially associated with this tectonic

corner structure, and holds some very important implications. For

instance, within the eastern portion of the St. Elias Range, this zone has

high relief and is extensively glaciated, suggesting the area may be an

excellent location to study the interaction between tectonics and climate.

• Inlet Orogen: The St. Elias Range is composed of a broad area of uplift

along the coast, with the highest relief and topography in the eastern

portion (Mt. St. Elias – 5489m and Mt. Logan – 5959m) within the

vicinity of Yakutat Bay (Chapman et al., 2008). The topographic

expression of the orogen decreases to the west, with the deformation

associated with the convergence of the Yakutat terrane decreasing west of

Cordova (N60.54; W145.7). The zone of active deformation has been

reorganized progressively from Miocene to Pliocene E-W trending thrust

sheets, to Plio-Pleistocene offshore NE-trending foreland and offshore

(Pamplona zone) fold and trust belts (Figure 4.1; Chapman et al., 2008).

This orogen has been related to convergence of the Yakutat terrane.

• Outlet Orogen: To the northwest of the Inlet Orogen, the Alaska Range

and Wrangell Mountains define a broad, actively deforming highland

associated with the transpressional motion of the Denali Fault (Miller et

al., 2002; Nokleberg et al., 1994). The Alaska Range is composed of a

long standing highland that underwent initial uplift during the early

Paleocene (Ridgway et al., 2007). Ongoing observed deformation and

uplift is related to the St. Elias orogeny (Plafker and Berg, 1994).

• Subduction Boundary: The deformation patterns to the west of Cordova

are characteristic of the subduction of Pacific Plate. The vertical load

associated with the downward motion of the Pacific Plate creates

subduction basins within the Cook Inlet and Bering Sea (e.g. Bruhn and

Haeussler, 2006). The Cook Inlet basin continues to the east forming the

eastward shallowing Copper/Chitina River Basin, consistent with

shallowing of the slab dip to the east (Eberhart-Philips et al., 2006;

Gutscher and Peacock, 2003). Additionally, the uplift within the Alaska

and St. Elias Ranges creates a sediment flux to the Nenana and Gulf of

Alaska basins, respectively (Ridgway et al., 2007).

The relative buoyancy of the Yakutat terrane causes the dip of the subducting slab to decrease, or flatten (e.g Eberhart-Philips et al., 2006; Rossi et al., 2006; Fuis et al.,

2008). Flat subduction has important implications to the thermal structure: it effectively

cools the fore-arc by offsetting the hot mantle wedge flow, in this case toward the

northwest (e.g. Munoz, 2005). The relatively cool subduction zone alters the pressure-

temperature conditions of the lower crust, shutting off melt production and creating the

observed gap in volcanism between the Wrangells and Mt. Spurr (Preece and Hart, 2004;

Figure 4.1). Additionally, the refrigeration of the fore-arc creates a much wider seismogenic zone allowing for larger strain accumulation, leading to large mega-thrust

earthquakes (Munoz, 2005; Gutscher and Peacock, 2003; Eberhart-Philips et al., 2006;

Abers et al., 2006).

4.3 Numerical Modeling Methods

This study uses the macroscale 3D thermo-mechanical numerical model to

reproduce the evolution of deformation patterns within an accretionary orogen (Figures

2.1 and 4.1). A discussion of the corresponding mesoscale embedded models can be

found in Chapter 5. These models are completed using a commercial finite element code,

FLAC3D, solving for simple stress–strain relationships, applied forces or kinematic boundary conditions, and heat flow to simulate the mechanical and thermal behavior of

Earth materials (Cundall and Board, 1988; Koons et al., 2002). The initial thermal solutions from Chapter 3 are used to assign an initial rheological structure to coupled thermo-mechanical models to investigate the role of local thermal and mechanical heterogeneities in the development of the regional characteristic signal of an accretionary orogen (Figure 3.3).

The modeling path is to first develop a standard Reference model that has a

rheology defined independently from the thermal structure. The upper crust (<15 km) of

this model is defined as a pressure-dependent Mohr-Coulomb material where the standard associated plasticity retains strength as the model evolves. The lower crust has a von Mises which simulates a thermally-dependent flow law with a nonlinear decrease in strength with depth. This model is used as a base of comparison to later, fully coupled that include thermally-dependent rheologies for the lower crust (>15 km or 350 ºC). The upper crust of the fully coupled models also alter the Mohr-Coulomb behavior (Tectonic models) by allowing the model to weaken with accumulation of strain (Strain-Softening

models), simulating the formation of faults and shear zones by the non-associated strain-

dependent material weakening (e.g. Vermeer and de Borst, 1984; Hobbs et al., 1990;

Buck and Poliakov, 1998; Montesi and Zuber, 2002). The fully-coupled models are also successively conditioned with surficial regional erosional conditions based upon the approximate limits of major glacier within the St. Elias Range. See Chapter 2 for additional details of the numerical methods and model geometries.

The presence of episodic releases of strain (i.e. large earthquakes) associated with the Denali-Fairweather fault and the Alaskan megathrust is not captured within the models. The models assume the transfer of strain is efficient and that the faults are continuously locked. This may overstate the strain transferred, and therefore the deformation rate and magnitudes within the model as compared to the natural example.

The relative focusing of strain within the models can be related to the thermal and rheological evolution of the crust. The spatial and temporal evolution of active orogens can be related to the interaction of tectonics, rheology, and erosion/climate. The standard

Mohr-Coulomb mechanical model used in this study is a temporally homogeneous material where post-yield stress levels are maintained, and therefore does not account for the development of weak zones associated with accumulation of strain. The strain softening model progressively weakens with strain, allowing for further localization of deformation. The weakening models do not address the potential healing of strained rocks through time and are only altered by the described weakening criteria.

Additionally, metamorphism has been shown to affect strain patterns on a regional scale.

Mineralogical changes associated with metamorphism, liberation of metamorphic or igneous fluid, and accumulation of strain have been shown to create a relatively strong zone within the crust, affecting the overall deformation pattern (e.g. Groome et al., 2008;

Upton et al., 2003; Pavlis and Sisson, 2003; Horsman et al., 2008; Chapter 3).

Additionally, the large scale magmatism associated with the Wrangell volcanic field may weaken the crust regionally and alter how strain is transferred. These topics are beyond the scope of this project and will be addressed in the future elsewhere.

4.4 Results

4.4.1 Reference Model Results

The Reference model was developed with a rheological structure defined independently from the thermal structure to illustrate the effects of thermal perturbations

Figure 4.2. Reference model strain patterns.

Figure 4.2. Reference model strain patterns. Contour plots from the Reference model taken at ~6 km below sea level of the vorticity components A) Ωxy, B) Ωxz, and C) Ωyz. D) Plot of the shear strain increment. Note all horizontal contour plots are taken at 5 km below sea level with spatial reference location projected from the surface. See Figure 4.1 for comparison of spatial references.

on the rheological structure observed within the fully coupled models (Figures 3.3). The

rheologic structure of the Reference model is spatially and temporally homogeneous at

the scale of the geometry and behaves mechanical in a similar fashion to that of the fully

coupled models. The initial stages of deformation include the development of a dextral

transform fault along the eastern margin of the Pacific block, as can be seen in a vorticity

horizontal plane (ΩXY; Figure 4.2A). Vorticity, defined by velocity gradients in the

vertical and horizontal planes, is here used to approximate strain patterns and indicate

modes of deformation (i.e. transform, thrust, or normal faulting; See Appendix D). The

obliquity of the convergence increases northward due to the initial velocity conditions

(Figure 2.2), as can be seen by the increased importance of the vorticity in the vertical

plane perpendicular to the direction of convergence (ΩYZ; Figure 4.2B). A maximum in

ΩYZ at 400km along the Y-axis is a result of the variation in the initial vertical velocity

(VZ; Figures 2.2 and 4.2C). To the west, the mode of deformation is dominated by

compression, with the formation of an orogenic wedge above the down-dip limit of the

Pacific block (ΩXZ; Figure 4.2C). The strain pattern forms a right angle bend associated

with the northern Transition-Compressional Transition zone located within the eastern

Alaska Range (Figures 4.1 and 4.2D). This tectonic corner structure forms due to the

rotation of principle stress axes from transcurrent to compressional deformation (Figure

4.2D; Animation 4.1). Additionally, the overall highest rate and magnitude of uplift, as

well as largest thermal anomaly (50 ºC), occurs within this corner (Figures 4.3A, 4.3B, and 4.3C respectively). This deformation and thermal pattern can be explained by the pattern of convergence into the corner defined by the evolved kinematics and displacements (Figure 4.3D). The total displacement and velocity vector pattern (Figure

Figure 4.3. Uplift and thermal results for the Reference model.

Figure 4.3. Uplift and temperature results for the Reference model. Contour plots of A) vertical velocity (myr-1), B) vertical displacement (m) and kinematic boundaries (see Figure 4.1), and C) temperature (°C). D) Vector plot of the model kinematic (myr-1) results overlain on the total displacement (km) results.

Figure 4.4. Temporal uplift patterns in the Reference model.

Figure 4.4. Temporal uplift patterns in the Reference model. Contour plots of A) vertical velocity (myr-1) at A) 100 ky, B) 300 ky, C) 500 ky, and D) 700 ky.

4.3D) is indicate that the largest transcurrent offset is along the Transform Boundary with

increasing obliquity in convergence northward through the corner structure to the

transition to compressional strain within the Outlet Orogen.

The area of highest uplift shifts as the model evolves from initially spatially

consistent with the corner structure to an area further to the east corresponding to the

orogenic wedge (Figure 4.3B and 4.4; Animation 4.2). Temporally fragmented uplift

within the forearc region is on a minor scale relative to the corner structure suggesting

most of the strain is accommodated within the Outlet Orogen (Figure 4.4; Animation

4.2).

4.4.2 Tectonic Model Results

The initial thermal structure is indicative of long standing flat slab subduction

zones where vertical motion of the relatively cold down-going slab drags the isotherms

downward, inverting them at depth and effectively cooling the lower crust within the

fore-arc (Figure 4.5; Chapter 3). One of the more important aspects of this model is the

formation of a thin sliver (< 10 km thickness) of material along the down-dip surface of

the Pacific plate that behaves frictionally (Figures 3.3 and 4.5). During the initial stages

of the model, the frictional sliver is uniform in down-dip extent in the east-west (Y-axis)

direction. The velocity conditions of the Pacific block are variable along the Y-axis, with

a larger vertical velocity (VZ) along the western half due to the transition from subduction of the Pacific plate to subduction of the more buoyant Yakutat terrane. The down-dip

extent of the frictional sliver decreases along the western portion of the Pacific block as

the model evolves. The decrease in extent, or roll-back, of the frictional sliver can be

Figure 4.5. Spatial development of the frictional sliver.

Figure 4.5. Spatial development of the frictional sliver. Cross section plots of the rheological model at y-axis dimension A) 100 km, B) 300 km, C) 500 km, and D) 700 km showing the spatial development of the frictional sliver. The inset shows the initial vertical velocity conditions imposed upon the Pacific block.

Figure 4.6. Strain patterns associated with the megathrust.

Figure 4.6. Strain patterns associated with the megathrust. Cross section contour plots of the shear strain rates (s-1) at y-axis dimension A) 100 km, B) 300 km, C) 500 km, and D) 700 km. The location of modeled faults are show on the cross sections as black lines, projected onto the model surface as dashed black lines, and linked spatially by the gray shaded planes. Additionally, the kinematically defined boundaries from Figure 4.1 and the location of the megathrust (white outline) are noted.

. Figure 4.7. Tectonic model strain patterns.

Figure 4.7. Tectonic model strain patterns. Contour plots of the vorticity components A) Ωxy, B) Ωxz, and C) Ωyz. D) Plot of the shear strain increment.

seen in the resulting deformation patterns (Figure 4.5). A large scale megathrust shear

zone develops at the down-dip extent of the frictional sliver and continues to the base of

the crust (Figure 4.6).

The fully coupled thermo-mechanical Tectonic models incorporate simple

pressure-dependent upper crust (<15 km or 350 ºC) and temperature-dependent plastic

yield lower crust (>15 km or 350 ºC) that evolves as the model steps. The large-scale,

spatial strain patterns are captured by the vorticity components (Figures 4.1 and Figures

4.7). The maximum in vorticity in the horizontal plane defines a dextral transform fault

along the eastern margin of the Pacific block (ΩXY; Figures 4.7A). The increasing obliquity of the convergence northward can be seen by the increased importance of the vorticity in the vertical plane perpendicular to the direction of convergence (ΩYZ; Figure

4.7B). At the leading edge of the Pacific block (~0 km on the X-axis), north-south

vertical vorticity component shows that convergent strain is partitioned into an orogenic

wedge (ΩXZ; Figures 4.7C). At the down-dip limit of the frictional sliver (Figures 4.5

and 4.6), a relatively minor maximum in the north-south vertical vorticity component

defines the limits of a secondary orogenic wedge (ΩXZ; Figure 4.7C). The transition from the transcurrent motion of the Transform Boundary to the compressive deformation

within the dual orogenic wedges develops two tectonic corner structures (at ~400 km and

~0km along the X-axis; Figure 4.7D). Each of these corner structures is defined by the rotation of principle stress axes from transcurrent to compressional deformation

(Animation 4.3).

Figure

4.8. Uplift and temperature results for the Tectonic model.

Figure 4.8. Uplift and temperature results for the Tectonic model. Contour plots of A) vertical velocity (myr-1), B) vertical displacement (m) and kinematic boundaries (see Figure 4.1), and C) temperature (°C). D) Vector plot of the model kinematic (myr-1) results overlain on the total displacement (km) results.

The rate and magnitude of uplift define two sub-parallel orogens located spatially

consistent with the Inlet and Outlet Orogens (Figures 4.8A and 4.8B). Additionally, uplift within the Outlet Orogen develops a small magnitude (25 ºC) advective thermal anomaly within that location (Figure 4.8C). This deformation and thermal pattern can be explained by the pattern of convergence into the southern corner structure defined by the evolved kinematics and displacements pattern showing the maximum in velocity and displacement gradients associated with the Inlet Orogen (Figure 4.8D). The total displacement and velocity vector patterns (Figure 4.8D) indicate that the largest transcurrent offset occurs along the Transform Boundary with increasing obliquity in convergence northward through the corner structures to the transition to compressional strain within the Inlet and Outlet Orogens.

As the model evolves the area of highest uplift shifts from initially spatially

consistent with the corner structure westward to an area corresponding to the orogenic

wedge of the Outlet Orogen (Figure 4.8B and 4.9; Animation 4.4). Additionally, there is

a temporal shift from initial uplift within the Outlet Orogen to uplift dominantly

associated with the southern corner structure and Inlet Orogen (Figure 4.9B to 4.9C;

Animation 4.4).

4.4.3 Strain-Softening Model Results

The fully coupled thermo-mechanical Strain-Softening models incorporate simple

pressure-dependent upper crust (<15 km or 350 ºC) that weakens with accumulation of

strain. Similar to the previously discussed models, the maximum in vorticity in the

Figure 4.9. Temporal uplift patterns in the Tectonic model.

Figure 4.9. Temporal uplift patterns in the Tectonics model. Contour plots of A) vertical velocity (myr-1) at A) 100 ky, B) 300 ky, C) 500 ky, and D) 700 ky.

Figure 4.10. Strain-softening model strain patterns.

Figure 4.10. Strain-softening model strain patterns. Contour plots of the vorticity components A) Ωxy, B) Ωxz, and C) Ωyz. D) Plot of the shear strain increment.

horizontal plane defines a dextral transform fault along the eastern margin of the Pacific

block forming the Transform boundary (ΩXY; Figures 4.10A). The increasing obliquity

of the convergence northward, as well as the partitioning of deformation into the

secondary Inlet Orogen, can be seen by the pattern of the vorticity in the vertical plane

perpendicular to the direction of convergence (ΩYZ; Figure 4.10B). The northward extent of the oblique component defines the Transform-Compressional Transition associated with the Inlet Orogen. The north-south vertical vorticity component indicates that the bulk of the convergent strain is partitioned into the Inlet Orogen, with the Outlet Orogen forming a relatively minor structure (ΩXZ; Figures 4.10C). The transition from the

transform motion compressive deformation within the dual orogenic wedges develops

two tectonic corner structures, the dominant corner is located at the eastern terminus of

the Inlet Orogen and the relatively minor corner is located at the eastern terminus of the

Outlet Orogen (Figure 4.10D). These corner structures are defined by the rotation of

principle stress axes from transcurrent to compressional deformation (Animation 4.5).

The rate and magnitude of uplift define two sub-parallel orogens located spatially

consistent with the Inlet and Outlet Orogens (Figures 4.11A and 4.11B), with the

southern Inlet Orogen capturing most of the strain. A southward bend in the uplift

pattern west of the Inlet Orogen at 400km along the Y-axis evident in Figure 11B is consistent with the role-back of the frictional sliver (Figure 4.6) due to the lateral increase in VZ associated with subduction of the Pacific Plate to the west. Uplift within the two

orogens develop a small magnitude (25 ºC) advective thermal anomaly within the Outlet

Orogen and a larger magnitude (75 ºC) anomaly associated with the Inlet Orogen (Figure

4.11C). The pattern of deformation and thermal anomalies can be explained by

Figure 4.11. Uplift and temperature results for the Strain-softening model.

Figure 4.11. Uplift and temperature results for the Strain-Softening model. Contour plots of A) vertical velocity (myr-1), B) vertical displacement (m) and kinematic boundaries (see Figure 4.1), and C) temperature (°C). D) Vector plot of the model kinematic (myr-1) results overlain on the total displacement (km) results.

Figure 4.12. Temporal uplift patterns in the Strain-softening model.

Figure 4.12. Temporal uplift patterns in the Strain-Softening model. Contour plots of A) vertical velocity (myr-1) at A) 100 ky, B) 300 ky, C) 500 ky, and D) 700 ky.

partitioning of most of the convergence into the southern corner structure, as defined by

the evolved kinematics and displacements pattern showing the maximum in velocity and

displacement gradients associated with the Inlet Orogen (Figure 4.11D), with little strain

translated further northward. The total displacement and velocity vector patterns (Figure

4.11D) indicate that the largest transcurrent offset occurs along the Transform Boundary

with increasing obliquity in convergence northward through the corner structures to the

transition to compressional strain mainly within the Inlet Orogen.

As the model evolves the area of highest uplift shifts from initially spatially

consistent with the corner structures westward to an area corresponding to the orogenic

wedges of the Inlet and Outlet Orogens (Figure 4.11B and 4.12; Animation 4.6). The

temporal shift from initial uplift within the Outlet Orogen to uplift dominantly associated

with the southern corner structure and Inlet Orogen is much more evident in the Strain-

Softening model as the Inlet Orogen is better defined than in the Reference or Tectonic models (Figure 4.11B to 4.11C; Animation 4.6).

4.4.4 Erosional Tectonic Model Results

The utilization of a simple erosion model that assumes warm-based glaciers maintain near constant elevations provide a more realistic surface boundary condition

(e.g. Hallet et al., 1996; Meigs and Suber, 2000). The erosional model is applied to the approximate location of the Inlet Orogen and assumes the orogen is maintained at a constant elevation by the present glacial cover. The results of this series of models

Figure 4.13. Erosional Tectonic model strain and uplift patterns.

Figure 4.13. Erosional Tectonic model strain and uplift patterns. Contour plots of the vorticity components A) Ωxy, B) Ωxz, and C) Ωyz. D) Plot of the shear strain increment.

Figure 4.13. Continued. E) vertical velocity (myr-1), F) vertical displacement (m) and kinematic boundaries (see Figure 4.1), and G) temperature (°C). H) Vector plot of the model kinematic (myr-1) results overlain on the total displacement (km) results. The location of the defined erosional area is indicated on A) and E) by the black outline.

suggest that erosion can play an important role in partitioning of deformation within a large-scale (1000km) orogen and is not limited to relatively small-scale (10km) focusing of strain.

Inclusion of this simple erosion model in the Tectonic model (Figure 4.13) produces a strain pattern very similar to that produced within the non-erosive Tectonic model (Figures 4.7 through 4.9). This pattern is characterized with a lateral dextral transform fault (Figure 4.13A) that increases in obliquity to the north (Figure 4.13B) before bifurcating into the dual orogenic wedge system (Figure 4.13C). The inclusion of erosion causes an increase in magnitude of the large-scale strain (Figure 4.13D) within the tectonic corner structures, consistent with the rotation observed in the principle strain tensor (Animation 4.7). The displacement and velocity vectors (Figure 4.13E), uplift patterns (Figures 4.13F and 4.13G; Animation 4.8), and local thermal perturbations

(Figure 4.13H) also show focused deformation and uplift within the corner relative to the corresponding non-erosive models. The result of the modeled erosion is thinning and significantly weakening of the upper crust causing the focusing of strain within a narrow zone (Figure 4.14). Weakening within the erosive Tectonic model causes the Inlet

Orogen to capture proportionally more of the strain (relative to the non-erosive model) and further decreases the deformation within the Outlet Orogen.

4.4.5 Erosional Strain-Softening Model Results

The strain patterns of the erosional Strain-softening model (Figure 4.15) are also similar to those produced within the non-erosive version (Figures 4.10 through 4.12), with a lateral dextral transform fault (Figure 4.15A) that increases in obliquity to the

Figure 4.14. Effect of erosion on crustal strength.

Figure 4.14. Effect of Erosion on crustal strength. A) Schematic showing the effects of erosive and thermal thinning on the strength of the crust (after Koons et al., 2002). The plot on the right shows the initial strength (red line) prior to the model run. The gray shaded box indicates the zone of transition from a pressure- to a temperature-dependent rheology. With increasing crustal flow driving vertical advection that transition is offset upward, effectively thinning the crust. The addition of an erosional surface adds to this effect. The new, evolved strength (blue line) is weaker than the initial strength profile (dashed red line). B) A cross-section of the model (Y-axis = 600km) showing the two weakened areas: A) the Outlet Orogen is driven by thermal advection and B) the Inlet Orogen weakening is driven by the non-linear feedback between advection and erosion.

Figure 4.15. Erosional Strain-softening model strain and uplift results.

Figure 4.15. Erosional Strain-Softening model strain and uplift patterns. Contour plots of the vorticity components A) Ωxy, B) Ωxz, and C) Ωyz. D) Plot of the shear strain increment.

Figure 4.15. Continued. E) vertical velocity (myr-1), F) vertical displacement (m) and kinematic boundaries (see Figure 4.1), and G) temperature (°C). H) Vector plot of the model kinematic (myr-1) results overlain on the total displacement (km) results. The location of the defined erosional area is indicated on A) and E) by the black outline.

north (Figure 4.15B) before bifurcating into the dual orogenic wedge system dominated

by the Inlet Orogen (Figure 4.15C). The modeled erosion drives an increase in

magnitude of the large-scale strain (Figure 4.15D) within the tectonic corner structure to

the south at the eastern terminus of the Inlet Orogen. This pattern is also seen in the

rotation of the principle stress tensor (Animation 4.9). The resultant displacement and

velocity vectors (Figure 4.15E), uplift patterns (Figures 4.15F and 4.15G; Animation

4.10), and local thermal perturbations (Figure 4.15H) indicate that deformation and uplift

are focused within the corner more than the corresponding non-erosive model. The

combination of erosion and strain weakening cause the Inlet Orogen to capture most of

the strain (relative to the non-erosive model).

4.5 Discussion

4.5.1 Discussion of Model Results

The main location of deformation within the Reference model results includes the

formation of the Outlet Orogen (Figure 4.3B). Strain is transferred northward along the

Transform Boundary, which is spatially consistent with the Denali-Fairweather fault

system (Figures 4.1 and 4.7A) and the transition from transcurrent to convergent strain

defines the Transform-Compressional Transition located at the eastern termination of the

Outlet Orogen. However, this model does not form a well-developed Subduction

Boundary or Inlet Orogen (Figure 4.3B). This suggests the formation of the naturally

observed dual, teleconnected orogenic wedges located at down-dip extent of the Pacific

block (Outlet Orogen) and frictional sliver (Inlet Orogen) is related to the influence of the

development of the frictional sliver due to the advective thermal anomaly associated with

flat slab subduction.

There is a disparity between the timing of the model structural reorganization and

the natural geological evolution of southern Alaska. The initial thermal model used in

this series of model assumes an instantaneous perturbation of the thermal structure.

However, as seen in the 2D thermal model presented in Chapter 3 (Figure 3.1), the

advective perturbation due to the initiation of the flat slab subduction takes approximately

10 Myr to fully develop. This is consistent with the simple arithmetic that indicates that

the flat slab would have initiated at 10-14 Ma, assuming 500-600 km has subducted at a rate of 50 mmyr-1.

In the fully coupled models, strain is transferred between the Inlet to Outlet

Orogens along Denali-Fairweather fault system. The relative dominance of this fault

system is related to the degree of deformation occurring within the Inlet Orogen. In the

standard Tectonic model, deformation within the Outlet Orogen is dominant; the Denali-

Fairweather fault system is a large-scale, high-strain feature formed along the entire eastern margin of the Pacific block (Figure 4.7A and 4.10A). The major transition from transform to convergent deformation occurs with the tectonic corner associated with the

Outlet Orogen. Within continued evolution (Figure 4.9), the model develops the secondary Inlet Orogen. Both wedges are active simultaneously (Animation 4.4).

The deformation patterns are enhanced within the Strain-Softening model. The initial stages of the Tectonic and Strain-Softening models are similar, with a well defined

transitional fault system transferring strain to the Outlet Orogen (Figure 4.9 and 4.12).

However, dynamic weakening of model within the fore-arc causes the Inlet Orogen to

capture the majority of the deformation in the Strain-Softening model, the northern extent

of the Denali-Fairweather fault system becomes less important. The major transition from transform to convergent deformation discontinuously jumps to the eastern margin of the Inlet Orogen. While this structural reorganization does not stop all strain along the

Denali-Fairweather fault, it does greatly reduce the amount of transferred strain, and therefore the deformation within the Outlet Orogen (Figure 4.9).

As noted above, vertical advection driven by subduction of the Pacific block cools

the lower crust within the fore-arc region (Figure 4.5). This tectonic refrigeration drives

the observed model spatial and temporal structural reorganization of the deformation

patterns (Figures 4.9 and 4.12) by altering the rheological structure of the crust.

Additionally, evolution of the thermal structure within the orogens mirror the vertical

displacements, areas with the highest vertical uplift also have the most deflection of the

isotherms (Figures 4.8 and 4.11). The Outlet and Inlet Orogen have the largest perturbations, and in both cases the magnitude of the perturbation decreases from a maximum within the tectonic corner toward the west. These vertical thermal perturbations associated with the orogenic wedges create an upward material advection associated with crustal weakening facilitated by the vertical offset of the transition from frictional to plastic material behavior (Figures 4.8 and 4.11). These weakening zones created by vertical advection drive further localization of strain (e.g. Koons et al., 2002;

Koons et al., 2003).

The vertical motion of the Pacific block drives the creation of subduction basins.

The main basin forms above the western portion of both the Tectonic and Strain-

Softening models consistent with the Subduction Boundary. These basins shallow to the

east with the reduction of the VZ of the Pacific block to the east. The mechanics of the western portion of the model are dominated by the vertical velocity of the Pacific block, suggesting that the basins form in response to vertical loading at the base and not to sediment loading at the surface (Figures 4.7 and 4.9). In order to create thick sedimentary basins within tectonically active areas, flexure of the crust driven by tectonic deformation must first create a basin structure into which sediment can be accumulated.

Sediment loading can then contribute to the continued formation and evolution of these basins. The northern extents of these basins are truncated by the western margin the Inlet

Orogen and further north by the Outlet Orogen.

4.5.2 Comparison with Geological Observations

The development of long wavelength (macroscale; 100 km) topography is linked to the rheology, tectonic geometry, and kinematics of the orogeny. The numerical modeling presented here predicts the development of dual orogenic wedges that can be associated with large, crustal scale deformation patterns. The deformation patterns are coupled to perturbations of the rheological structure of the crust by thermal advection and accumulation of strain. This coupling suggests a fundamental link between the spatially and temporally evolving rheology and deformation patterns. Here the model results are explored in more detail with respect to geological observations of the timing of uplift

(Figure 4.16) and large scale strain patterns (Figure 4.17).

Figure 4.16. Geological uplift patterns.

Figure 4.16. Geological uplift patterns. Thermochronology of observed uplift from A) Denali (Fitzgerald et al., 1995) and B) Logan (O'Sullivan and Currie, 1996) and sedimentology from C) Tanana basin (Ridgway et al., 2007) and D) Gulf of Alaska (Plafker, 1987; Plafker et al., 1995).

Figure 4.17. Structural comparisons.

Figure 4.17. Structural comparisons. Schematic map showing the comparison of mapped faults in southern Alaska (Plafker, 1987; Plafker et al., 1994; Meriaux et al., 2009) and modeled areas of maximum strain.

Constraints on the deformation and uplift history are given by significant

thermochronological data within both the Alaska (Fitzgerald et al., 1995; Haeussler et al.,

2008; Haeussler, 2008) and St. Elias Ranges (Berger et al, 2008a; Berger et al, 2008b,

Berger and Spotilla, 2008; Enkelmann et al., 2008; Enkelmann et al., 2009). Apatite

fission track thermochronology indicates that exhumation within the Denali region

(vicinity of Mt. McKinley) began by the Late Miocene (~5-6 Ma; Figure 4.16 inset A),

with the amount of uplift decreasing to the south of the Denali fault (Fitzgerald et al.,

1995). Fitzgerald et al. (1995) suggested that the patterns and timing of uplift,

association with the Denali fault, and shift in plate motion at ~5.6 Ma (Engebretson et al.,

1985) are related to changing stress relationships between the Pacific and North

American plate. The authors suggested that the perturbation of the stress relationships

causes additional oblique compressional deformation within the McKinley strand of the

Denali fault, resulting in the observed uplift. The results of this study indicate that the

accretion of the Yakutat terrane at ~6 Ma played a significant role in altering this stress

relationship and drove uplift within the Alaska Range, consistent with the interpretation of Plafker et al. (1994). Additionally, O'Sullivan and Currie (1996) indicated two stages of uplift and exhumation within the Mt. Logan massif; one in the late Miocene (~16 Ma) and a younger Pliocene event (~4 Ma; Figure 4.16 inset B). The authors suggest if the

Miocene deformation event is restricted to the St. Elias Range, it could be related to subduction and initial under-plating of the cool Yakutat terrane resulting in crustal

thickening and uplift. Alternatively, if the event was regional, it could be related to rapid

cooling due to migration of the magmatic front associated with a ~16 Ma change in

relative plate motions (Engebretson et al., 1985; Skulski et al., 1992). Pliocene-

Pleistocene deformation (Campbell and Dodds, 1982), palynology studies indicating reduction of forest canopy after ~7 Ma due to uplift (O'Sullivan and Currie, 1996), and sedimentology of local Cenozoic marine glacial deposits indicate that the later event is associated with uplift of the St. Elias Range driven by accretion of the Yakutat terrane.

Additional low temperature thermochronology indicates extremely young exhumation

(<1 Ma) on the windward flank of the St. Elias Range driven by glacial erosion can be

linked to climatic changes within the Pleistocene (Berger et al., 2008a; Berger and

Spotilla, 2008). The observed disparity in timing between earlier uplift within the Alaska

Range and later, continuing uplift with the St. Elias Range is consistent with the modeled

temporal shift in deformation fronts. As noted in 4.3, the model time-scale is accelerated relative to the natural example due to the assumption that strain is efficiently transferred across a continuously locked Denali-Fairweather fault and Alaskan megathrust. The younger uplift history within the St. Elias Range will be further explored in Chapter 5.

The stratigraphic record provided by large basins within southern Alaska can provide a secondary record of uplift and exhumation within the study area. To the north of the Denali fault, the Tanana basin provides a record of uplift within the Alaska Range

(Figure 4.16 inset C). Provenance of the mid Miocene Suntana Formation (~15 Ma) indicates the Alaska Range was a developing highland at that time (Ridgway et al.,

2007). By the deposition of the Nenana gravels (6.7 Ma) the Alaska Range was a major orogenic source of coarse clastic sediments, corresponding favorably with the thermochronological uplift history. The Gulf of Alaska records a significantly younger sedimentary record (Figure 4.16 inset D). The synorogenic deposits within this basin are referred to as the Yakutaga Formation, and record a complex depositional environment

Figure 4.18. Seismic hazard map of southern Alaska.

Figure 4.18. Seismic hazard map of southern Alaska. Seismic hazard map (Petersen et al., 2008) overlain on map of southern Alaska including model faults patterns (black lines). Seismic moment tensor solutions are derived from the vorticity associated with the model faults (see Figure 4.2 for example). Scale for hazard map is peak ground acceleration (ms-2) with 10% probability of exceedance in 50 years (Petersen et al., 2008).

with multiple synorogenic unconformities and facies variations related to the relative dominance of glacio-marine sedimentation. This unit has poor paleontological age constraint (Plafker, 1987), however, recent work has suggested that the unit is entirely

Plio-Pleistocene, with an age of 4.5 Ma at the base of the unit (Eyles et al., 1991; Lagoe and Zellers, 1996). Again, the sedimentological record of uplift and exhumation corresponds favorably with the thermochronological data and supports the numerical modeling results.

The observed deformation and strain patterns within the model can be used to indicate the locations of modeled faults that compare favorably with the natural example

(Figure 4.17). The zone of highest strain and highest topography occurs within the tectonic corner structure. The corner structure associated with the St. Elias Range is located at the intersection of the Fairweather, Contact, and Boarder Ranges Faults, which comprise major terrane boundaries (Chapter 1). This corner zone is also spatially consistent with an area of high historical seismicity (Estabrook et al., 1992; Plafker and

Thatcher, 2008), present seismic hazard (Figure 4.18; Petersen et al., 2008), and high topography. The Alaska Range is characterized by a relatively narrow zone of high topography associated with the oblique motion on the Denali fault (Matmon et al., 2006) and subdued topography to the north associated with a fold and thrust belt (Meriaux et al., 2009). The model results suggest that the largest uplift should be associated with the oblique corner structure, with a relatively broad area of uplift across the region. The St.

Elias Range consists of three large, southward directed thrust sheets (Meigs et al., 2008).

Additionally, there is a complex pattern of overprinting minor faults that suggest an evolving stress state within the region consistent with a temporal change from dominantly

100

transform to oblique motion (Chapman et al., 2008), a similar patterns predicted by the

modeled jump in deformation front.

There are three hypotheses proposed to explain the motion of the Denali Fault and

neotectonics of southern Alaska. The first proposes counterclockwise rotation of the

Wrangell block (e.g. the major terrane immediately south of the Denali Fault; see Figure

1.1) about a vertical axis (Lahr and Plafker, 1980; Stout and Chase, 1980; Csejtey, 1990;

Page et al., 1995). This hypothesis implies the Denali fault is a major dextral transform fault with a constant slip rate along its length and also decouples the deformation within

the Alaska Range from the rotation of the Wrangell block. The second alternative

hypothesis suggests that slip is progressively partitioned from a northwest translating

Wrangell block into the fold and thrust structure of the Alaska Range (Bird et al., 1996;

Meriaux et al., 2009). This hypothesis relies on new cosmogenic dating of geomorphic

features as a means to measure fault slip rates along the strike of the Denali Fault

(Matmon et al., 2006; Meriaux et al., 2009). These studies present data that indicate a

decrease in slip rate from >14 mma-1 (145 W) to <7 mma-1 west of McKinley (Meriaux et al., 2009; 149 W). Meriaux et al. (2009) suggests that this decrease in slip is due to the increased partition of strain into the northward thrust faults associated with the foreland fold-and-thrust belt of the Alaska Range (Armijo et al., 1986; Bowman et al., 2003; Bird et al., 1996; Meriaux et al., 2009; Bemis and Wallace, 2007; Lesh and Ridgway, 2007).

Another important observation is the gradual widening of the Alaska Range east (30 km) to west (50 km) caused by crustal thickening. The crustal thickening is due to the increased partitioning of dextral transcurrent strain into the convergent fold-and-thrust

belt related to the kinematics of the Wrangell block and the strike of Denali Fault. The

101

neotectonic hypothesis of Meriaux et al. (2009) assumes no rotation of the Wrangell

block, which would decrease partitioning of dextral transcurrent slip to the fold-and- thrust belt and uplift rates within the Alaska Range. This is the preferred hypothesis and agrees well with the results of the models presented here. Finally, the third hypothesis utilizes lateral extrusion of southern Alaska to the west (Redfield et al., 2007), similar to the proposed gravitational collapse of the Tibetan Plateau (Tapponnier et al., 1982;

England and Houseman, 1989, Dewey, 1988). This proposed hypothesis does not seem to agree with the observed GPS plate vectors, deformation patterns, or slip partitioning along the major fault systems.

A tectonic model of the structural history of the accretion of the Yakutat Terrane in the St. Elias Range was presented by Chapman et al. (2008). During the early stages of convergence in the Miocene, the deformation was characterized by the stacked, thin- skinned fold-and-thrust belt and early exhumation and deposition of the Yakutaga

Formation. This early deformation was overprinted by the northeast trending Pliocene to early Pliestocene thrust faults in foreland. This structural reorganization has been linked to glaciation (Gulick et al., 2007; Worthington et al., 2008; Berger et al., 2008a) by erosion and mass redistribution driving the progradation of the shelf edge. However, the role of changes to the vertical normal and shear stresses by mass loading due to large ice sheets has as of yet been unexplored. Currently, the deformation front is located within the orogen and consists of a narrow band of en echelon northeast trending thrusts (Pavlis et al., 2004). Thermochronology (Meigs et al., 2008; Berger et al., 2008a; Berger et al.,

2008b) and metamorphic petrology (Dusel-Bacon et al., 1994) indicate that the highest exhumation occurs south of the Contact Fault and decreases from Mt. St. Elias east to

102

west. The basic patterns of deformation (fold-and-thrust belt) and exhumation of the model correspond favorably with those produced by the models. However, the local structural reorganization is not reproduced by the macroscale models. The structural evolution of the St. Elias Range will be addressed further in Chapter 5.

The transfer of strain associated with the Yakutat terrane accretion into the Alaska

Range is well documented (Plafker et al., 1994; Estabrook et al., 1992; Matmon et al.,

2006; Meriaux et al., 2009). This strain transfer is well represented within the models presented here and is the direct result of a early progressive shift in the deformation front in response to the initiation and evolution of flat slab subduction (Chapter 3). Additional far field effect are suggested for ongoing deformation within the Mackenzie and

Richardson Ranges of the Canadian Cordilleran (Mazzotti and Hyndman, 2002).

Mazzotti and Hyndman (2002) suggest strain is transferred along a relatively weak lower crustal decollement approximately 800 km to the northeast of the Yakutat orogeny. The initial rheological homogeneity, boundary conditions, and spatial dimensions of our model do not allow the development of these far field effects.

4.5.3 Influence of Thermal Advection, Rheological Weakening, and Erosion

Large-scale (100-1000 km) deformation patterns are related to the temperature structure driven by subduction related advection (Figures 4.8C and 4.11C). The flat slab segment of the model alters the observed deformation pattern by shifting the Inlet Orogen

away from the trench (to the northwest in the models). This is consistent with

observations of other flat slab subduction zones (Figures 4.5, 4.8 and 4.11; Dickinson et

al., 1988; Murphy et al., 1998; Gutscher et al., 2000a; Gutscher, 2001; Mori et al., 2007).

103

Furthermore, the Inlet Orogen location can be linked to the down-dip limit of the

frictional sliver, indicating the thermal control on the mechanical behavior of the crust

provides the first-order influence on the spatial and temporal evolution of regional, large-

scale deformation patterns observed within southern Alaska. The relative dominance of

one orogen over the other after the shift in deformation patterns can be linked to the strain

localization driven by the associated rheological weakening (Figures 4.10 and 4.11).

Small-scale (10-100 km) deformation patterns can be altered by vertical thermal advection, rheological weakening, development of topography, and erosion. The vertical advection associated with crustal flow patterns causes a vertical displacement of the thermal structure into the actively deforming areas. If the displacement is enough, >8 km

(Koons et al., 2002), it effectively thermally thins the strong upper crust, focusing deformation into these zones. The inclusion of the development of faults and shear zones by a strain-weakening rheology, and inclusion of an erosion surface boundary condition both enhance the localization of deformation. The effects of topography, erosion, and tectonics will be further explored in Chapter 5.

4.5.4 Petrological Implications

The nature of the numerical models presented in this study allows for the easy extraction of pressure-temperature-time (P-T-t) paths for the direct comparison and prediction of the petrological evolution of southern Alaska (Figure 4.19). The divergence of flow paths related to the structural reorganization can be seen in the predicted P-T-t

paths. The path for the Inlet orogen encounters a maximum of subgreenschist facies

conditions (<300 ºC and 0.4 GPa), while the path leading to the Outlet orogen encounter

104

Figure 4.19. Model P-T paths.

105

amphibolites conditions (600 ºC and 0.8 GPa), given enough time for exhumation (>10

Ma). Material associated with the slab, within 5-10 km of the top of the Pacific block,

will not be exhumed in either orogen. The maximum P-T conditions of material within the Inlet orogen are not favorable to the production of partial melting, suggesting the melt enhanced weakening does not play a significant role in the St. Elias orogen (e.g.

Beaumont et al., 2004).

4.5.5 Tectonics Implications

The simplest prediction of the modeling presented herein is the large-scale

reorganization of deformation resulting from the transition to flat slab subduction. The

temporal and spatial reorganization of the orogen with continuing convergence is well

supported by the structural geology and uplift history observations. The model results

can be scaled allowing for direct comparison to other areas suspected of being dominated

by flat slab subduction. This scaling and comparison relies on the relationship between

the patterns of deformation, rheology, and the tectonics.

Elsewhere, the combination of structural reorganization of deformation patterns,

development of thick sedimentary basins, and a gap in volcanism have also been

recognized in the Andes (Gutscher et al., 2000a), central Mexico (Mori et al., 2007),

Japan (Gutscher, 2001), and southern Alaska (Plafker and Berg, 1994). These features

have all been associated with shallow angle or flat slab subduction. The Paleozoic (415-

360 Ma) Acadian Orogeny in northern New England and eastern Canada was driven

oblique convergence between Laurentia and Gondwana (Figure 4.20A; Williams, 1979).

The temporal evolution of strain patterns indicates a progressive 600-km shift away from

106

Figure 4.20. Tectonics of terrane accretion.

Figure 4.20. Tectonics of terrane accretion. Comparison of schematic cross sections of the A) Acadian (after Murphy et al., 1998), B), Laramide (after Bird, 1998), and C) St. Elias (after Koons et al., in review) orogens showing approximate locations and timing of deformation fronts.

107

early deformation in the foreland westward (Bradley et al, 2001). Additionally, a

temporal gap in magmatism is apparent during the Early to Mid Devonian (395-380 Ma;

Murphy et al 1999). This pattern of shifting the deformation patterns and magmatic quiescence are characteristic of flat slab subduction.

The early stages of terrane accretion are captured in both the geology and numerical modeling of southern Alaska. However, these observations can be used to develop a more complete evolutionary model with the inclusion of additional data from other well-characterized flat slab subduction zones. The Laramide Orogeny (late

Cretaceous to Eocene; 75-40 Ma was the final phase of the Cordilleran that drove broad anticlinal uplift of Precambrian crystalline basement in Colorado and Wyoming (Figure

4.20B). The deformation front migrated eastward away from the fore-arc approximately

700 km during the initial 10 million years, much further to the east than previous Sevier belt deformation. Thick sequences of syn-orogenic sediments were accumulated within basins formed between these broad anticlinal uplifts (Dickinson et al., 1988).

Additionally, this phase of deformation is accompanied by a temporal and spatial break in volcanism. This has long been interpreted as the result of shallow subduction of the Kula and Farallon Plates due to abnormally high convergence rates or interaction with a mantle plume (Dickinson and Snyder, 1978; Murphy et al., 1998). The shallow angle of subduction creates a much broader thermal anomaly and wider seismogenic zone, allowing for strain to be transferred much further inboard than for a much steeper subduction angle. After 40 Ma, magmatism resumed within the fore-arc area, suggesting an end to the flat slab subduction.

108

4.6 Conclusions

Numerical models of the accretion of the Yakutat terrane onto southern Alaska produces deformation patterns that are characterized by temporal and spatial reorganization of deformation patterns. The location of the structural reorganization is linked to thermal and rheological perturbations associated with the accretion and partial shallow subduction of the buoyant Yakutat terrane. These model results capture the evolution the entire orogen remarkably well and are supported by independent geological observations. The deformation patterns provide a characteristic view of terrane accretion that allows for scaling to other suspect locations of flat slab subduction or terrane accretion. Addition derived information, such as the petrologic evolution of the model, can provide useful information allowing for interpretation of ancient patterns. Overall, the combination of the information from geological observations and numerical modeling from southern Alaska, as well as other recognized and well characterized locations, has been used to develop a qualitative model of the tectonics of terrane accretion suggesting an intimate link between tectonics, rheology, and climate.

109

Chapter 5 : Mesoscale Geodynamics of Southern Alaska

5.1 Introduction

An enduring question of landscape evolution regards the relative role of surface versus tectonic processes in influencing topographic and geological evolution of convergent orogens. Clearly both are involved at the orogen scale where strongly asymmetric orographic gradients impose major flux control on orogen evolution (e.g.

Koons, 1990; Willett et al., 1992; Whipple, 2009), as well as at catchment scale of local response to intense erosion (Zeitler et al., 2001; Stewart et al., 2008; Enkelmann et al.,

2009). In some instances where the rheological-erosional feedback persists, local tectonic aneurysm behavior can produce massive topographic relief associated with decompression melting, as in the two Himalayan syntaxes (Koons et al., 2002). The nature of the non-linear, positive feedbacks among erosion and tectonic processes that produce an aneurysm means that the initial stages of strain concentration are rarely observed. Consequently, the chicken and egg discussion of what controls exhumation

(e.g. England and Molnar, 1990) seldom has an opportunity of resolution in evolved aneurysm settings. In the eastern and western syntaxes of the India-Asia collision zone, the Indus and Tsangpo Rivers cut through the high elevation areas of the Nanga Parbat and Namche Barwa Massifs, respectively, producing local reliefs of >7000 m. This extreme local relief perturbs vertical normal and shear stress, allowing focusing of strain into the valleys. These large river systems efficiently transport material out of the

Himalayan orogen, amplifying the localization of strain and uplift (Zeitler et al., 2001;

Koons et al., 2002). Erosion acts to focus advective flow of material through the crust,

110

causing a zone of high thermal advection, which locally weakens the crust by thinning

the high strength upper crust. The nature of this feedback is well characterized in mature

settings, such as in the eastern and west syntaxes of the Himalayan orogen, however the

early stages have rarely been characterized in nature. In southern Alaska, where rapid,

glacially-driven exhumation is associated with anomalous thermochronological data

showing rapid cooling (Enkelmann et al., 2009), we have an opportunity to identify the

controlling parameter for initiation of aneurysm behavior in at least one orogen.

5.2 Numerical Modeling Methods

Presented here is the embedded portion of this macroscale model that focuses on

the influence of natural topography and erosion on the <5-km scale (mesoscale) thermal

and mechanical evolution of the St. Elias Range. The embedded model encompasses the

area of the St. Elias Range (640km x 890km x 20km) with discretization on an

approximately 5 by 5 km horizontal and 2 km vertical grid (Figure 2.3). The entire

model domain is defined with an isotropic conductive-advective thermal model and

Mohr-Coulomb (~upper 15 km) and thermally-defined plastic yield condition (~lower 5 km) mechanical models. For further description of the numerical methods please see

Chapter 2 and Appendix A.

Initial velocity conditions are imposed on the base of the model over an area corresponding to the spatial extent of the Yakutat terrane (south of the Chugach-St. Elias

Fault and west of the Fairweather Fault; Figure 5.1) consistent with the observed motions

111

of the Yakutat terrane and Pacific plate (Figure 2.3; ~45 mm/yr; De Mets et al., 1994;

Fletcher and Freymueller, 1999). All deformation within the model is driven by this imposed drag along the base (tectonic boundary conditions), which is spatially defined by geological observations and previous macroscale tectonic modeling (Chapter 4).

The mesoscale modeling path first develops a reference Tectonic model in the

absence of topography or erosion; then the natural topography (Topographic Model) and

an erosion scheme (Erosion Model) are successively applied as initial boundary

conditions. The present topography, including bathymetry, was derived from a global 1-

minute DEM sampled at the model discretization (Figure 5.1; Smith and Sandwell,

1997). To emphasize the influence of erosion, vigorous erosional conditions are applied

based upon the spatial extent of current glaciations (Figure 5.1) and are based on the

assumption that glacial erosion maintains a near constant elevation during exhumation

(Hallet et al, 1996; Meigs and Sauber, 2000).

5.3 Mesoscale Modeling Results

5.3.1 Mechanical Results

The development of the characteristic features of the deformation field within the

simple Tectonic reference model are driven entirely by the tectonic boundary conditions.

These modeled deformation patterns are consistent with the characteristic signal of corner

geometries found in accretionary tectonic settings where the margins of the accreting

terrane affect the regional strain patterns. A maximum in the horizontal

112

Figure 5.1. Mesoscale surface boundary conditions.

Figure 5.1. Mesoscale surface boundary conditions. A) Contour plot of the DEM used for the Topographic model. White arrows and dashed line indicate the vector and extent of the basal velocity conditions based upon defined northern and eastern limits of the Yakutat block (e.g. Plafker et al., 1995; Pavlis et al., 2004) and its measured velocity (Fletcher and Freymueller, 1999). Locations of important feature are used in later plots for spatial reference. B) Approximate locations of important glacial systems (shaded white) and Erosion model surface conditions (black boxes).

113

strain occurs along the eastern lateral boundary of the Yakutat terrane, consistent with the

large-scale, dextral Fairweather Fault system (εXY; Figure 5.2A). The vertical Cartesian

strain components (see Appendix D) demonstrate the transition from lateral dextral transform motion (εXZ; Figure 5.2B) to the development of a convergent orogenic wedge

(εYZ; Figure 5.2C) defined by oppositely dipping thrust faults. The total strain (shear

strain; Figure 5.2D) reaches a maximum within the tectonic corner structure, which is

defined by the kinematics as the relative rotation of velocity vectors (Figure 5.3A) and

principle stresses (Animation 5.1; please see Appendix B for captions for all animations)

from transform motion to convergence. The zone of highest vertical uplift is spatially

consistent with the transition from dextral transform strain along the Fairweather fault to

the convergent strain associated with the development of an orogenic wedge within the

St. Elias Range (Figure 5.3B and C). There is a 40-60 ºC vertical deflection of the

thermal structure of the model corresponding to the area of highest uplift (Figure 5.3C).

The Topographic model results produce a similar strain pattern as the Tectonic

model. This includes the development of a dextral transform fault spatially associated

with the Fairweather Fault (Figure 5.4A). As the lateral shear zone continues to the

northwest, the strain becomes more oblique (Figure 5.4B) and eventually bifurcates into a

two sided convergent orogenic wedge that is defined by oppositely dipping convergent

strain shear zones (Figure 5.4C) coinciding with the Chugach-St. Elias Range. The shear

zones define oppositely dipping thrust faults located south and north of the highest peaks,

the Mt. St. Elias and Mt. Logan, respectively. The maximum strain (Figure 5.4D),

114

Figure 5.2. Tectonic model strain patterns.

Figure 5.2. Tectonic model strain patterns. Contour plots of the Cartesian strain components A) εxy (horizontal strain, i.e. transform faults), B) εxz (north-south vertical strain, i.e. thrust faults), and C) εyz (east-west vertical strain, i.e. thrust faults), and D) shear strain increment. These and subsequent plots are horizontal planes taken at 5 km below sea level. Noted are the locations of the Alaskan coastline, Mt. St. Elias, Mt. Logan, Mt. Wrangell, and Mt. Fairweather (stars). See Figure 5.1 for location details.

115

Figure 5.3. Tectonic model uplift and thermal patterns.

Figure 5.3. Tectonic model uplift and thermal patterns. A) Velocity vector (myr-1) plot overlain on contours of displacement (km), contours of B) vertical velocity (mmyr-1; maximum of 5.5 myr-1), C) vertical displacement (m), and D) temperature (°C).

116

rotation of velocity vectors (Figure 5.5A) and principle stresses (Animation 5.2) again

occurs in the corner structure defined by the transition from transcurrent to convergent

deformation. This area of maximum deformation has uplift rates of ~5 mm/yr (Figure

5.5B), a total vertical displacement of 2933 meters (Figure 5.5C), and a 60-80 ºC vertical

deflection of the thermal structure, similar to the Tectonic model (Figure 5.5D).

However, the presence of topography causes the deformation patterns to diverge relative

to the basic characteristic signal of the Tectonic model by shifting the southward dipping

high strain zone toward the coast, and increasingly focused uplift in the Malaspina-

Seward glacier areas (Figure 5.4C). The perturbation of the deformation pattern can be

related to changes to the vertical normal (σzz) and shear stresses (σxz and σyz) related to

the increased load due to the presence of topography (Figure 5.6A and B). Topography

creates higher normal vertical stresses associated with the peaks and higher vertical shear

stress associated with the slopes and valleys.

The mechanical results from the Erosion model are spatially very similar to the

Tectonic model, developing a lateral transform shear zone (Figure 5.7A) that increase in

obliquity into the tectonic corner (Figure 5.7B) eventually bifurcating into and orogenic

wedge defined by oppositely dipping shear zones (5.7C). The zone of highest strain

(Figure 5.7D) is consistent with the rotation of velocity vectors (Figure 5.8A) and principle stress axes (Animation 5.3) due to the progressive change from transcurrent to convergent deformation. The Erosional model departs from the Tectonic model in rate

(Figure 5.8B) and magnitude (Figure 5.8C) of uplift and the advective thermal signal

117

Figure 5.4. Topographic model strain patterns.

Figure 5.4. Topographic model strain patterns. Contour plots of the Cartesian strain components A) εxy, B) εxz, and C) εyz, and D) shear strain increment. Compare with Figure 5.2.

118

Figure 5.5. Topographic model uplift and thermal patterns.

Figure 5.5. Topographic model uplift and thermal patterns. A) Velocity vector (myr-1) plot overlain on contours of displacement (km), contours of B) vertical velocity (mmyr-1; maximum of 5.0 myr-1), C) vertical displacement (m), and D) temperature (°C). of the model kinematic results overlain on the topographic results. Compare with Figure 5.3.

119

Figure 5.6. Comparison of vertical stresses.

Figure 5.6. Vertical stresses due to topography. Contour plots of the vertical stress index (VSI; see Appendix C for further explanation) for the Topographic model. VSI is a ratio of vertical shear to vertical normal stresses and can be related to effect of topographic load. Note the plot is taken at 1 km below sea level, the portion of the plot in the southeastern corner has a bathymetric depth below the cutting plane. See Appendix D for further information.

120

(Figure 5.8D). The presence of erosion within the Seward Glacier area, consistent with

the tectonic drives faster uplift rates (Figure 5.8B; ~0.007 myr-1). The deflection of the

isothermal structure mirrors the faster uplift rates, with a ~90 ºC increase in temperature

in the corner at 5 km depth (Figure 5.8D).

5.3.2 Petrologic Implications

Particle trajectories through the orogen are easily extracted from the model and

yield P-T data that can be directly compared to the petrologic and thermochronologic patterns observed in southern Alaska (Figure 5.9). The highest P-T conditions of the

mesoscale model correspond to the lower greenschist facies (~400 ºC and 675 MPa),

however, the particles that are exhumed within the actively deforming orogenic wedge

have not experienced conditions much past the transition from subgreenschist to

greenschist facies (~220 to 250 ºC at <270 MPa). Within the main portion of the

orogenic wedge, to the west of Mt. St. Elias and Mt. Logan, a pattern develops in the

maximum conditions the particles encountered while advecting through the model

(Figure 5.10). Material exhumed to the south of the main area of uplift does reach

conditions above 100 ºC and originated <5 km below sea level (<150 MPa). Material

that reached the maximum P-T conditions, upper subgreenschist to lower greenschist facies, is exhumed within the main area of deformation. The maximum conditions decrease again to the north. This pattern has implications to the observed thermochronologic record, which is discussed further below.

121

Figure 5.7. Erosion model strain patterns.

Figure 5.7. Erosion model strain patterns. Contour plots of the Cartesian strain components A) εxy, B) εxz, and C) εyz, and D) shear strain increment. Compare with Figures 5.2 and 5.4.

122

Figure 5.8. Erosion model uplift and thermal patterns.

Figure 5.8. Erosion model uplift and thermal patterns. A) Velocity vector (myr-1) plot overlain on contours of displacement (km), contours of B) vertical velocity (mmyr-1; maximum of 6.8 myr-1), C) vertical displacement (m), and D) temperature (°C). of the model kinematic results overlain on the topographic results. Compare with Figures 5.3 and 5.5.

123

Figure 5.9. Model P-T paths.

124

Decompression melting within the western syntaxis of the Himalaya has produced very young intrusions (<1 Ma) that are currently exposed at <4000 km elevation on

Nanga Parbat (e.g. Koons et al., 2002). These partial melts greatly reduce the strength of the crust and allow from further strain accumulation. The thermal structure in the

Topographic model is deflected by up to 60-80 ºC (~4 km vertical displacement) and the

Erosion model is deflected 80-100 ºC within the tectonic corner structure (Figures 5.6 and 5.8). This deflection is not enough to produce decompression melting within this narrow zone at the depth of the model geometry (20 km; Figure 2.2). At that depth and pressure (20km and ~700 MPa, respectively), the water saturated solidus of granite is

~625 ºC, about 225 ºC above the temperature at the base, given the model geothermal gradient. The current model run length is 0.5 Myr; continued convergence for another 2

Myr may produce a large enough advective signal to allow for decompression melting within the corner structure. Alternatively, the thickness of the crust may maintain crustal temperatures below those necessary for anatexis.

5.4 Discussion of Model Results

5.4.1 Role of Erosion

The tectonic aneurysm, as developed with respect to the eastern and western syntaxes of the Himalayan orogen, utilizes fast, efficient fluvial erosion to focus crustal flow and strain into the river valleys. The focused crustal flow drives thermal advection, which acts to thin the strength bearing upper crust. The Erosion model shows the early

125

Figure 5.10. Maximum modeled metamorphic conditions.

Figure 5.10. Maximum modeled metamorphic conditions. Contour plot of maximum P- T conditions encountered by material exhumed in the Topographic model. Note the inset of vertical displacement is taken from Figure 5.5B.

126

stages of enhanced focusing of strain, by approximately 50%, resulting in increased

magnitude of uplift (0.007 myr-1). However, the inclusion of erosion does not alter the spatial distribution of strain significantly, indicating that erosion does not provide a major influence over the deformation patterns. This indicates that the initial stages of a tectonic aneurysm are linked to the focusing of strain due to the tectonic geometry, and erosion enhances the development by further focusing crustal flow.

5.4.2 Comparison with Thermochronology

The modeled mechanical evolution of the St. Elias orogen indicates the presence of a tectonic corner, enhanced by high topography and efficient glacial erosion, influences focusing of strain and rock uplift within a narrow zone (Figures 5.3, 5.6, and

5.8). The areas with the highest strain rates are also the areas with the highest vertical velocities, resulting in localized thermal advection and compression of the isothermal structure. The location of this modeled thermal perturbation corresponds with the location

of young zircon FT ages derived from detrital deposits of the Seward Glacier (Figures

5.3, 5.6, and 5.8). Zircon fission track (FT) ages from detrital material transported by the

Seward-Malaspina glacier system revealed exceptional high cooling rates of 150–300

°C/Ma underneath the Seward Glacier (Enkelmann et al., 2009), suggesting the location

of a zone with intense rock exhumation and resultant thermal perturbation (Figure 5.11).

The modeled area of intense rock exhumation extends eastward from the Seward Glacier

into the area of Yakutat Bay. Unfortunately, thermochronologic data are sparse in this

area and no detrital data exist that give evidence about the cooling history of these ice

covered valleys. However, the 4.5 Ma zircon FT and young hornblende 40Ar/39Ar ages

127

(~19 Ma; ~500 °C closure temperature; Sisson et al., 2003) of a bedrock samples from the ridge north of Yakutat Bay (McAleer et al., 2009) might hint that the ice covered rocks of this area experience a similar fast cooling as the rocks located underneath the Seward

Glacier. The location of the modeled aneurysm, as defined by modeled strain and thermal patterns, is consistent with the thermochronological record.

Besides the Seward Glacier area, relative fast cooling of 100-120 °C/Myr is revealed in a narrow band along the Fairweather transform fault (O’Sullivan et al, 1995;

McAleer et al., 2009), and in the fold-and-thrust belt, west of the Malaspina Glacier (e.g.

Spotila et al., 2004; Berger et al., 2008a; Berger et al., 2008b; Berger and Spotila, 2008).

Bedrock apatite (U-Th)/He ages of <2 Ma occur along the windward flanks of the orogen that receive precipitation rates of >5 m/yr, whereas bedrock ages from the northern and much dryer Chugach terrane are significantly older (30-9 Ma; e.g. Spotila et al., 2004;

Berger et al., 2008a). This age pattern led to the conclusion that the Chugach terrane forms the deformational backstop and the Contact fault was suggested to have developed as a back thrust underneath the Bagley Ice field–Seward Glacier (Berger et al., 2008a;

Berger et al., 2008b; Berger and Spotila, 2008). This interpretation, however, conflicts with apatite and zircon FT results that revealed several phases of Miocene and Pliocene exhumation within the Chugach terrane with a general trend of younging towards the south, suggesting instead that the Wrangellia terrane forms the backstop, and the Bagley

Ice Field structure is within the deforming zone (O’Sullivan and Currie, 1996;

Enkelmann et al., 2008). The thermochronological record is consistent with this tectonic

128

Figure 5.11. Plot of thermochronological data.

Figure 5.11. Plot of thermochronological data. Colored contours are cooling rates after Berger and Spotilla (2008), shaded area indicates the location of Seward-Malaspina glacier system (Enkelmann et al., 2008; Enkelmann et al., 2009). Precipitation rate after Enkelmann et al., 2008). Thermochronological data overlain on vertical uplift rate (Vz; myr-1) from the Tectonic model. See text for additional details.

129

signal, recording uplift and cooling within the orogenic wedge (Figures 5.10 and 5.11).

The cooling rate increases to the east and culminates with a maximum in corner structure

in the area associated with the Seward Glacier (Enkelmann et al., 2009). Additionally,

the asymmetric orographic precipitation alters this pattern by increasing erosion on the

windward side of the St. Elias Range; producing the age discordance pattern with a signal that partially masks the more subtle signal on the drier, northern flank of the orogen.

The closure temperatures of the methods (apatite (U-Th)/He closure temperature

~70-90 °C and zircon FT closure temperature ~200-250 °C; Berger et al., 2008a; Berger

et al., 2008b; Enkelmann et al., 2008) of the thermochronology studies can be used to

compare the particle flow paths produced by the models with the natural geologic

example (Figure 5.10 and 5.11). The models produce a thermal pattern where rocks along the coast and to the north of the main area of deformation do not have reset apatite

(U-Th)/He ages indicating they did not reach the 70-90 °C closure temperature, whereas

Rocks exhumed within the central portion of the orogenic wedge have reset apatite (U-

Th)/He ages. This pattern is consistent with the observed pattern in the St. Elias and

Chugach Ranges.

5.4.3 Comparison with Structural Geology

Beginning during the Miocene, the impingement and attempted subduction of the

Yakutat Terrane onto southern Alaska has been driving uplift within the Chugach,

Fairweather, Kenai, and St. Elias Ranges in the Northern Cordillera (O’Sullivan and

Currie, 1996; Bruhn et al., 2004; Pavlis et al., 2004). The structural evolution of this

orogen varies along strike east to west, forming three distinct structural zones with all

130

deformation partitioned into major transform and thrust fault systems (Bruhn et al., 2004;

Chapman et al., 2008). The eastern margin of this orogen is bounded by the dextral

Fairweather Fault. Transcurrent motion associated with this fault, as well as minor thrust

faulting, is responsible for the uplift and deformation associated with the relatively

narrow Fairweather Range. The dominant mode of deformation gradually becomes more

oblique into the central portion of the orogen. The central zone is separated from the

eastern zone by the Pamplona zone (offshore) and the Malaspina Fault (onshore) thought

to be the eastern extension of the Aleutian subduction zone (Plafker et al., 1994).

Alternatively, this zone may represent a younger deformation front associated with a

structural reorganization during the Pliocene (Chapman et al., 2008).

Deformation within the central zone is characterized by dextral transpression with

central thrusting of the Yakutat terrane along the Chugach-St. Elias, Hope Creek,

Sullivan, and unnamed offshore thrust faults forming a thin skinned fold-and-thrust belt that absorbs at least 82 km (53%) shortening (Bruhn et al., 2004; Pavlis et al., 2004;

Chapman et al., 2008; Meigs et al, 2008). The later three thrust sheets are younger to south and splay from a decollement located within the basal part of the Yakutat cover rock sequence (Meigs et al., 2008). Localized dextral strike-slip motion is restricted to the hanging-walls of the thrust sheets or dissipated by later superimposed oblique thrusting (Bruhn et al., 1994; Pavlis et al., 1994). Early deformation within the fold-and- thrust belt is generally oriented orthogonal to margin of southern Alaska and oblique to motion of the Yakutat Terrane (Pavlis et al., 1994; Chapman et al., 2008). This earlier deformation pattern is overprinted by later northeast trending thrust faults (Chapman et al., 2008).

131

The basal unit of the Yakutat Terrane cover sequence is the Kulthieth Formation,

an early Oligocene to early Eocene (~55-28 Ma) sequence of a marine regressive sequence of arkosic sandstone with minor coal beds with recognizable non-marine alluvial, delta plain, and shallow marine facies (Plafker, 1987; Meigs et al., 2008). The

Kulthieth Formation is conformably overlain by the marine transgressive siltstone and sandstones and interbedded volcanic rocks of the late Eocene to early Miocene (~40-5.6

Ma) Poul Creek Formation (Plafker, 1987; Lagoe et al., 1993). The provenance of the

Poul Creek Formation has been interpreted as pre-orogenic strata of the Chugach

Terrane, likely from British Columbia, and has been suggested to have been deposited during the northward translation of the Yakutat Terrane (Plafker et al., 1994; Bruns,

1983; Perry et al., 2009). The sequence is capped by the synorogenic deposits of the

Yakutaga Formation deposited in response to Miocene orogenesis of the St. Elias Range

(<5.6 Ma to present; Plafker, 1987; Eyles et al., 1991; Lagoe et al., 1993; Zellers, 1993;

Zellers, 1995; Meigs et al., 2008). Timing of the development of these thrust seem to correspond with major tectonic events. Emplacement of the oldest thrust sheet, the Hope

Creek thrust, is difficult to determine, however, it had to be after the initial deposition of the Poul Creek Formation (<~40 Ma and >5.6 Ma; Meigs et al., 2008) and is likely related to the initial stages of convergence of the Yakutat Terrane. This was followed by the Sullivan thrust at ~6 Ma, consistent with foreland basin development and the deposition of the Yakutaga Formation. Finally, the unnamed offshore thrust consists of two fault cored anticlines that formed at 1.8 and 0.25 Ma, consistent with the timing of the reorganization of deformation associated with Pliocene to Pleistocene glaciations

(Gulick et al., 2007; Worthington et al., 2008; Berger et al., 2008a). The fold-and-thrust

132

belt is most deeply eroded on windward side of the orogen, south of the Contact Fault,

with less deformation and erosion on the leeward side, however this pattern cannot

distinguish between tectonic versus orographic effects (Meigs et al., 2008; Berger et al.,

2008a). Additionally, metamorphic grade increases to east peaking at amphibolites near

Mt. St. Elias (Dusel-Bacon et al.,1994) reflecting an along strike pattern to the exhumation (Sisson et al., 1989).

The western portion of the orogen, within the Cook Inlet region, consists of

complex dextral transpressive thrusting and fault-propagated anticlinal folding around

dipping hinge lines causing shortening and extension within the sedimentary cover

accreted onto overriding plate of the Aleutian subduction zone (Bruhn et al., 2004; Bruhn

and Haeussler, 2006). The area records the complex interaction between dextral shearing

and horizontal shortening driven by terrane accretion and compressive thrust-related

strain of the subduction zone. This leads to temporally variable stress fields that are on

the interval of the large seismic releases (Bruhn and Haeussler, 2006). Interplate

seismicity during 1987-1992 within the Pacific Plate reflects internal deformation due to

changes in stress state between the subduction zone to the west and convergent boundary

to the east (Pegler and Das, 1996).

The modeled fault and uplift patterns compare favorably with the natural mapped

fault patterns and geologic observations of the orogen. The location of stain maxima can

be used to predict fault patterns and potential for seismic hazards (Figure 5.12). All

models develop a similar pattern of a dextral transform fault bounding the eastern margin

of the Yakutat terrane. This fault bifurcates and the principle stresses rotate at the

133

tectonic corner, with a transition of the dominant model of deformation from transcurrent to compressional. An orogenic wedge, defined by oppositely dipping thrust faults, continues from the corner to the west. The transcurrent strain continues to the northwest, resulting in subdued uplift (<0.002 m/yr) within the Wrangell Mountains. This is consistent with the transfer of strain northward along the Totschunda Fault, which links the Fairweather to the Denali Faults and is potentially responsible for the 2002 magnitude

7.9 Denali Fault earthquake (e.g. Plafker et al., 1977; Matmon et al., 2006; Kalbas et al.,

2007).

5.5 Tectonic Aneurysm

Coupling between the faster uplift rates, erosive thinning, and vertical advection of hot material upward weakens the strong upper crust. This weakening initiates the non- linear feedback responsible for triggering a tectonic aneurysm. Deformed syn-collisional deposits in the fold-and-thrust belt revealed a localized fast exhuming rock source existed already 3 to 4 Ma ago (Enkelmann et al., 2008). Assuming that this sediment source was the same as today, we suggest that the observed tectonic ‘aneurysm’ at the Yakutat collision corner existed since the Pliocene and is thus not a response that developed due to Quaternary climate. Including the topography and glacial erosion, the mesoscale modeling suggests that a local perturbation of the thermal structure sufficient to produce a thermochronological and deformational record similar to the observations from southern Alaska can develop within 5 million years, or within the first ~50% of the

Yakutat convergence.

134

Figure 5.12. Comparison of interpreted and modeled faults.

Figure 5.12. Comparison of interpreted and modeled faults. Blue lines indicate interpreted faults formed within the Topographic model and red lines indicate mapped or interpreted fault traces. Moment tensor solutions are plotted to suggest mode of deformation observed within the model results.

135

5.6 Conclusions

The thermal-mechanical models presented here utilize simple boundary conditions and capture most of the variance in the signal of an oblique nonterminal collision. This signal includes the formation of a strain maxima in the tectonic corner structure spatially associated with the Seward Glacier area. Inclusion of heterogeneities in surface topography and erosion provide minor alteration to these tectonically developed strain patterns and captures remarkably well the evolution of local topography, observed fault zones, and cooling age patterns (Figures 5.10 through 5.12). In particular, the models reveal focused uplift that perturbs the thermal structure in the tectonic corner to the east of the present high topography. This pattern of focused strain demonstrates the first order control provided by the tectonic geometry on the focusing of strain and the secondary influence of topographic load and erosion. The models produce an incipient thermal perturbation that is consistent with a tectonic ‘aneurysm’, supporting the thermochronological evidence given by detrital samples for focused exhumation underneath the Seward Glacier.

136

Chapter 6 : Geodynamics of Terrane Accretion

The numerical modeling results presented herein provide a unique view of the

temporal and spatial evolution of southern Alaska that can be easily translated to other locations of suspect terrane accretion or flat slab subduction. This is accomplished

through the development of a qualitative model that relies on natural geological

observations (e.g. Pavlis et al., 2004; Enkelmann et al., 2008), numerical modeling results

(e.g. Koons et al., 2009), and related studies (e.g. Eberhart-Philips et al., 2006; Fuis et al.,

2008) to predict the observed associated features of the oblique collision and accretion, including spatial and temporal deformation and uplift patterns, petrological evolution, and multi-scale landscape development. Additionally, modeling results and information gleaned from studies of other orogens (Himalaya, Andes, Tibet, Southern Alps,

Laramide, and Acadian) were used to aid the development of this qualitative model of terrane accretion.

The early stages of deformation during terrane accretion are a function of the relationship between the rate of convergence, buoyancy of the down-going plate, and mechanical coupling between the plates. Each of these variables evolves with time and can vary spatially. For instance, within southern Alaska, the characteristic of the down- going slab varies from relatively ordinary oceanic lithosphere in the west to a thickened oceanic plateau in the east (Pavlis et al., 2004; Eberhart-Phillips et al., 2006). The

137

increasing thickness of the lithosphere causes a change in buoyancy that allows for a higher degree of mechanical coupling between the plates. With increasing convergence, the plate buoyancy eventually slows the rate of convergence and locks the subduction zone (e.g. Huang et al., 2000; Huang et al., 2006). The initial stages of terrane accretion in southern Alaska involve the impending collision of a micro-continent (i.e. the Yakutat terrane) with a large continent at a long standing subduction zone at an oblique angle of convergence (Figure 6.1; Plafker et al., 1994; Pavlis et al., 2004). The initial patterns of deformation are dominated by compression associated with subduction forming a relatively narrow zone of thrusting near the trench associated with formation of an accretionary wedge. Additionally, a narrow calc-alkaline volcanic arc forms inland above the down-going slab.

The pre-collisional history of the terranes can be important to the collisional deformation patterns. For instance, in the Himalayan orogen, subduction, magmatism, previous terrane accretion events, and crustal thickening over millions of years weakened the lower crust (e.g. Yin and Harrison, 2000; Tapponnier et al., 2001). This weak lower crust created a weak coupling between the Indian and Eurasian plates allowing for far field deformation and the formation of the Tibetan Plateau (Gerbault et al., 2005). A similar relationship between weak lower crust, weak coupling, and the formation of a broad plateau is seen in the Altaplano of the Central Andes (Gutscher et al., 2000a). In southern Alaska, in contrast, thermal advection associated with the long standing subduction zone has cooled and strengthened the lower crust prior to the accretion event creating a zone of strong coupling. The strong coupling favors the formation of a

138

Figure 6.1. Initial stages of convergence.

139

relatively narrow, thin-skinned fold and thrust belt associated with the accretionary prism

(Figure 6.1).

The angle of the down-going slab begins to shallow as the thickened oceanic crust

begins to be subducted (Figure 6.2). This shallow, or flat slab angle propagates away

from the thickened oceanic crust is subducted leading to the displacement of the hot

mantle wedge and ultimately the inversion of the slab at depth and displacement (Figure

6.3). The deformation front within the overriding plate will track the leading edge of the flat slab, gradually migrating away from the trench, similar to the observations with the

Acadian (Murphy et al., 1999) and Laramide orogens (Dickinson et al., 1988). The lateral margins of this deformation front may form large-scale transform faults as in the

eastern margin of the St. Elias Range, or may show a more gradual curvature of the

deformation patterns, similar to the western margin of the flat slab segment of the

Aleutian subduction zone. The final distal location of the deformation front is controlled

thermally by the spatial location of the transition to eclogite facies metamorphism, which

corresponds to a densification, loss of buoyancy, and eventual foundering of the slab (e.g.

Gutscher et al., 2000a; Kearney et al., 2009). In addition to the shift in deformation, the

initial calc-alkaline volcanism is replaced by adakitic or tholeiitic volcanism associated

the lateral margins of the slab (i.e. leaky transform faults), which also migrates away

from the trench (Gutscher et al., 2000b). With the final displacement of the mantle

wedge, all volcanism stops creating a volcanic gap as observed between Mount Spurr and

the Wrangell volcanics.

140

Figure 6.2. Initial flat slab subduction.

141

Figure 6.3. Reorganization of deformation patterns.

142

The propagation and establishment of the flat-slab subduction drives northwestward migration of Wrangell block, which is accommodated by dextral motion of the Denali fault and uplift and deformation within the Alaska Range fold-and-thrust

belt (Bruhn and Haeussler, 2006; Matmon et al., 2006; Meriaux et al., 2009). Additional

far field deformation within the Mackenzie and Richardson Mountains, Canadian Yukon,

is translated by a decollement in the hot, weak lower crust (Mazzotti et al., 2002). This

progressive development of far field deformation tracks the replacement of weak lower

crust capable of transferring strain over long distances with a relatively cold, strong lower

crust due to shallowing subduction and displacement of mantle wedge. The thermal

structure of the subduction zone is perturbed by offsetting the mantle wedge flow away

from the trench, resulting in an overall cooling. Cooling, or tectonic refrigeration, of the

lower crust alters the patterns of metamorphism associated with subduction resulting in a

widening of the seismogenic zone and shutting off the magmatism by either dehydrating

or sufficiently cooling the mantle wedge and lower crust. Alteration of the lower crust

strength and the development of the frictional sliver facilitate a structural reorganization

within the orogen.

The basement complex of the Yakutat terrane is subducting while the upper ~5-15

km thick sedimentary cover is incorporated within the evolving thin-skinned foreland

fold and thrust belt (e.g. Chapman et al., 2008). This deformation pattern is similar to

that observed in Taiwan: the formation of a fold-and-thrust belt outboard of and

incorporating the accretionary prism (Huang et al., 2000; Huang et al., 2006). The

evolving rheology of the lower crust, forming the strong frictional sliver, favors the

143

formation of a narrow zone of strain forming a localized fold and thrust belt (Royden,

1996; Ellis et al., 1998; Ellis et al., 2004; Babeyko and Sobolev, 2005). The continued

convergence and subduction of the Yakutat basement complex allows for excess strain to

be transferred along the lateral dextral transform fault system into the fold-and-thrust belt associated with the Alaska Range (Meriaux et al., 2009; Matmon et al., 2004; Meigs and

Sauber, 2000). Uplift and thrusting within both of these zones would thicken and flex the crust, creating foreland and back arc basins (Dickinson, 1974; Flemings and Jordan,

1990; Jordan and Watts, 2005). These basins, such as those in the Gulf of Alaska,

Copper River, and Tanana River, provide an excellent tracer for proximal uplift patterns and timing.

Evolution of material within the orogen, such as thermal- (Koons, 1987;

Beaumont et al., 1996; Batt and Braun, 1999; Willett, 1999) or strain-weakening (Behn et al., 2002; Montesi and Zuber, 2003; Nagel and Buck, 2004; Sobolev et al., 2005; Wijn et al., 2005) reactions, will allow for the development of dual orogenic systems where strain is accumulated into a trench proximal orogenic wedge. Additionally, metamorphic and melt-producing reactions can have drastic effects on the strength of the crust (e.g.

Beaumont et al., 2001; Beaumont et al., 2004). These changes to the rheological structure, as well as complex interactions with boundary conditions, influence the evolution of the deformation patterns (Tapponnier et al., 1982; Dewey, 1988; Hodges,

2000).

144

After the terrane is fully accreted, the subduction of normal (average) oceanic

lithosphere will resume at the outboard side of the new margin, causing the deformation

patterns and volcanism to eventually reorganize into the previous subduction related

patterns (Figure 6.4) with the resumption of subduction achieved by the foundering of the

oceanic crust outboard of the new continental margin. Silver et al (1983) noted two

thrust faults north of the Banda arc and interpreted these features as precursors to the

reinitiation of subduction following terrane accretion. The new magmatism may be

transitional from subduction related calc-alkaline and terrane accretion associated adakitic intraplate magmatism (e.g. Gutscher et al., 2000b; Cole et al., 2007), reflecting the transition of source areas and evolving forearc thermal structure.

145

Figure 6.4. Resumption of subduction.

146

Chapter 7 : Conclusions

This study has four general conclusions that have important implications for the growth of coupled orogenic wedges within terrane accretion events, development and evolution of 3D strain patterns, initiation and growth of tectonic aneurysms, and relationship between geodynamics and petrologic evolution of the crust.

1) The geometry of the plate interaction has profound effects on the

partitioning of strain and petrologic evolution of the crust

2) The thermal structure of the crust has an important first order

influence on the mechanical behavior of the lithosphere

3) Mechanical heterogeneities, both temporal and spatial, provide

important control on the evolution on how strain is partitioned at

both meso- and macroscales

4) Surficial processes, such as erosion and mass redistribution,

provide an influence on the mesoscale patterns of deformation.

7.1 Influence of the Thermal Structure

Perturbations to the thermal structure of the lithosphere in southern Alaska, specifically cooling and strengthening the fore-arc, provides an important primary influence on its mechanical behavior and deformation patterns. A long-standing subduction zone has defined the southern margin of Alaska since at least the end of the

147

Cretaceous (Plafker et al., 1994). Advection associated with the evolving flat-slab subduction zone conditioned the thermal structure of the lithosphere, influencing a mechanical strengthening of the fore-arc region over the last few million years. This strengthening provided a primary control on the associated deformation patterns within the overriding plate (Chapter 3) that resulted in a recent temporal shift from uplift within the Alaska Range during the early stages of convergence (>6 Ma) to uplift within the St.

Elias Range during later stages (<6 Ma; Koons et al., 2009).

The vertical advection of material into the actively deforming and uplifting zones within the overriding crust drives the formation of a local zone with relatively steep thermal gradients where the crust is hotter than surrounding areas (see Figures 4.xx and

5.xx). The role of temperature in the mechanical behavior of rocks is well characterized

(Brace and Kohlstedt, 1980; Hirth and Tullis, 1992; Gleason and Tullis, 1995; Kohlstedt et al., 1995; Mackwell et al., 1998; Wijn et al., 2005; Chapter 4). Advection of relatively hot material into an actively deforming region provides a non-linear feedback that further weakens the region by thinning the strong upper crust, leading to further focusing of deformation (Chapter 4). The initial stages of weakening and nonlinear feedback are due to the accumulation of strain in a tectonic corner structure, driving the initiation of a tectonic aneurysm in southern Alaska (Hooks et al., in review). Focused zones of strain provide a control on the overall strength of the crust. An efficient erosional regime, either fluvial or glacial (Chapter 5) would lead to further feedback weakening driven by thermally thinning the strong pressure-dependent upper crust by focusing crustal flow into these weakened zones.

148

In addition to altering the mechanical behavior, the zone of extreme uplift, strain accumulation, and advective perturbations to the thermal structure within the tectonic corner structure have important implications to the petrological evolution of the crust.

The models indicate the advective vertical deflection of the thermal structure is on the order of 80-120 ºC (5 to 6 km), with a 150 to 200 MPa reduction in the local pressure.

This deflection is not enough to drive adiabatic melting at depths of less than at least 50 km or pressures of ~1 GPa (e.g. Figures 4.19 and 6.1). Partially molten rocks or resultant intrusive bodies focus much more strain and alter the partitioning of deformation on a much larger scale.

7.2 Role of Heterogeneity at the Macroscale

The influence of temporal and spatial heterogeneities to the thermal structure are described in Chapter 3 and are further developed with the coupling with mechanical implication in Chapters 4 and 5. As noted in these previous chapters, the thermal and mechanical structures and boundary conditions of the models are based upon geological observations (e.g. Fletcher and Freymueller, 1999; Pavlis et al., 2004; Abers et al., 2006;

Eberhart-Philips et al., 2006; Fuis et al., 2008). The perturbations to the mechanical structure provide a primary influence on the spatial partitioning of strain within the model orogen. These thermo-mechanical heterogeneities are developed dynamically as the model steps, however, pre-existing macroscale geological spatial variations, such as earlier deformation patterns and relatively strong or weak rock units observed in nature,

149

are also important and were not addressed in the modeling completed for this study (e.g.

Yin and Harrison, 2000; Tapponnier et al., 2001; Upton et al., 2009).

There is an obvious scale-dependency on the mechanical heterogeneities in numerical modeling approaches that restricts the heterogeneities to sizes larger than the scale of discretization. The fold and thrust deformation of the Alleghenian foreland within Pennsylvania, Virginia, and Maryland has been suggested to be controlled by the formation of a decollement within the Lower Cambrian Waynesboro Shale (e.g. Faill and

Nickelsen, 1999). The thickness of this formation, ~1000' (Kauffman, 1999) would be less than the discretization of even the mesoscale model (5 km vertical). Even modest variations in strength, which are inherent in the heterogeneous rocks of the crust, can produce a range of structures at many scales (e.g. Faill and Nickelsen, 1999; Pavlis and

Sisson, 2003; Holyoke and Tullis, 2006; Horsman et al., 2008).

The long-term strength of the crust has long been a subject of debate (e.g.

Jackson, 2002; Afonso and Ranalli, 2004; Handy and Brun, 2004; Burov and Watts,

2006; Thatcher and Pollitz, 2008). Many food-oriented terms have been used to describe the crust, for instance, the "jelly sandwich" model denotes a strong upper crust and strong mantle "sandwiching" a weak lower crust. These interpretations favor a steady-state strength model and are based upon geophysical, experimental, and seismological data.

These models assume that the crustal strength profiles are not time- or rate-dependent.

Conversely, Thatcher and Pollitz (2008) present a transient strength estimate with the mechanical response varying as a function of the time scale of the perturbation (e.g. post- seismic relaxation, glacial loading, or tectonic isostatic adjustment). The transient

150

strength of the lower crust on a tectonic time scale (106-107 years) must be controlled by

the external load (i.e. tectonically derived stress) and internal material properties (i.e.

material 'strength'). This relationship is summed up in Newton's equation of viscosity:

, 6.1

where viscosity (η) is related to the ratio of stress (σ) and strain rate ( ). The strength of the crust, that is how it strains in response to a given stress, can be altered by metamorphic reactions and mechanical fracturing. Given sufficient time to achieve steady state conditions with respect to reactions and particle paths in an active orogen, the strength of the crust will essentially reach steady-state conditions spatially and temporally. Any additional changes to the system, such as flux of fluids through the crust, seismicity, or climatic effects, will cause the steady-state to be reset. These perturbations active over sufficiently long time periods (~107 years) fix the strength of

the crust forming an internally controlled self-ordered system (e.g. Thatcher and Pollitz,

2008).

7.3 Non-linear Feedbacks at the Mesoscale

The mesoscale interaction of tectonic geometry, partitioning of strain, rheological

evolution, and climate-erosion provides a well characterized non-linear feedback that

holds the potential for accumulation of strain, fast exhumation of mid crustal rock, large-

scale uplift, and creation of high relief topography. This feedback is most efficient at the

mesoscale. The initial stages of this feedback were described in Chapter 5 as the

151

accumulation of strain within the tectonic corner structure records the spatial transition from lateral transform motion to compressional deformation. The initiation of this feedback requires the accumulation of high strain and fast uplift within the tectonic corner structure. Erosion within this area leads to the continued weakening of the crust through focused strain, uplift, and thermal advection. Development of faults and shear zones through focused strain, as well as advective thermal perturbations leading to thermal weakening increase the further focusing of deformation and uplift within these regions. The high precipitation rates on the windward flank of the St. Elias Range (e.g.

Enkelmann et al., 2008; Berger et al., 2008a; Berger et al., 2008b) or the Himalayas

(Zeitler et al., 2001; Koons et al., 2002) promote high erosion rates. When the erosion is focused into a few narrow rivers or glaciers, the end result would be the thinning and weakening of the upper crust. The full development of tectonic aneurysms require non- linear feedback between the initial stages of strain focused into tectonic corner structures, thermal and mechanical weakening within the high strain zone, and fast erosion driven by high precipitation rates leading to thinning of the strong upper crust.

7.4 Suggestions for Future Work

The work presented add to the understanding of the geodynamics of the accretionary Yakutat orogen of southern Alaska. There are a few interesting and important topics that were beyond the scope of this study that should warrant further consideration:

152

• A more complete role of surface processes, erosion and deposition,

in the context of reorganization of the tectonic signature should

incorporate a more complete surface model. The vertical loading

of sediment within the Gulf of Alaska derived from the St. Elias

Range could provide an important perturbation to the stress state of

the crust.

• The role of climate-uplift interaction was not addressed within this

study. The time dependent nature of glacial erosion is an

important topic; the removal of a large glacial load would, again,

change the stress state of the crust. Also, the variable nature of

glacial erosion links directly with the climate signal. One other

important aspect of the climate signal is the variable glacial load

over the last ~1 Ma. Removal of the glacial load during the ~2000

years have led to extreme isostatically driven uplift rates (up to

0.032 myr-1) within the St. Elias Range and Glacier Bay regions

(Larsen et al., 2005) that are an order of magnitude greater than the

tectonic signal.

• A linkage between mechanics, rheology, and petrology was

explored in a cursory fashion. This coupling is of fundamental

importance to orogenesis, seismicity, economic geology, tectonics,

and petrology. A more complete linkage can be accomplished by

tying the rheology to petrology within the model. For instance, the

transition from greenschist to amphibolite facies roughly

153

corresponds to the transition from brittle to plastic deformation,

respectively. However, this transition, as well as the amphibolite

to granulate transition in the lower crust, can release considerable

fluid that may alter the mechanical behavior of the rocks, act as a

flux driving anatexis, drive hydrothermal alteration and mass flux

elsewhere in the crust, or simply act as an advective medium.

• The inclusion of a crustal scale fluid model would have important

impacts upon the stress state of the model. This topic was not

addressed due to the inaccuracy of the fluid model calculations in

the numerical method used.

• Finally, the inclusion of seismic data (G. Pavlis, personal

communication; Eberhart-Philips et al., 2006) would allow for the

development of a more realistic, site specific geometry and more

accurate material properties.

154

References Cited

Abers, G.A., P.E. van Keken, E.A. Kneller, A. Ferris, and J.C. Stachnik, 2006, The thermal structure of subduction zones constrained by seismic imaging: Implications for slab dehydration and wedge flow, Earth Planetary Science Letters, v. 241, pp. 387-397.

Afonso, J.C., and G. Ranalli, 2004, Crustal and mantle strengths in continental lithosphere: is the jelly sandwich model obsolete?, Tectonophysics, v. 394, pp. 221-232.

Armijo, R., P. Tapponnier, J. L. Mercier, and T. L. Han, 1986, Quaternary extension in southern Tibet – Field observations and Tectonic implications, Journal of Geophysical Research, v. 91, pp. 13803–13872.

Barker, A.D., 2007, 3D Mechanical Evolution of the Plate Boundary Corner in SE Alaska, unpublished M.S. thesis, University of Maine, Orono, ME, 145 pg.

Batt, G.E. and J. Braun, 1999, The tectonic evolution of the Southern Alps, New Zealand: insights from fully thermally coupled dynamical modelling, Geophysical Journal International, v. 136, pp. 403-420.

Baxter, E.F., and D.J. DePaolo, 2000, Field measurements of slow metamorphic reaction rates at temperatures of 500° to 600° C, Science, v. 288, pp. 1411-1414.

Beaumont, C., P.J.J. Kamp, J. Hamilton, and P. Fullsack, 1996, The continental collision zone, South Island, New Zealand: comparison of geodynamical models and observations, Journal of Geophysical Research, v. 101, pp. 3333-3359.

Beaumont, C., R. A. Jamieson, M. H. Nguyen, and B. Lee, 2001, Himalayan tectonics explained by extrusion of a low-viscosity channel coupled to focused surface denudation, Nature, 414, 738–742.

Beaumont, C., R. A. Jamieson, M. H. Nguyen, and S. Medvedev, 2004, Crustal channel flows: 1. Numerical models with applications to the tectonics of the Himalayan- Tibetan orogen, Journal of Geophysical Research, v. 109, B06406.

Behn, M.D., J. Lin, and M.T. Zuber, 2002, A continuum mechanics model for normal faulting using a strain-rate softening rheology: implications for thermal and rheological controls on continental and oceanic rifting, Earth and Planetary Science Letters, v. 202, pp. 725-740.

Bemis, S. P., and W. K. Wallace, 2007, Neotectonic framework of the north-central Alaska Range foothills, in Tectonic Growth of a Collisional Continental Margin: Crustal Evolution of Southern Alaska, edited by K.D. Ridgeway et al., Geological Society of America Special Paper, v. 431, pp. 549– 572.

155

Berger, A.L., and J.A. Spotila, 2008, Denudation and deformation in a glaciated orogenic wedge: The St. Elias orogen, Alaska: Geology, v. 36, p. 523-526.

Berger, A.J., J.A. Spotila, J. Chapman, T.L. Pavlis, E. Enkelmann, N.A. Ruppert, and J.T. Buscher, 2008a, Architecture, kinematics, and exhumation of a convergent orogenic wedge: A thermochronological investigation of tectonic-climatic interactions within the central St. Elias Orogen, Alaska: Earth and Planetary Science Letters, v. 270, pp. 13-24.

Berger, A.J., S.P.S. Gulick, J.A. Spotila, P. Upton, J.M. Jaeger, J.B. Chapman, L.A. Worthington, T.L. Pavlis, K.D. Ridgway, B.A. Willems, and R.J. McAleer, 2008b, Quaternary tectonic response to intensified glacial erosion in an orogenic wedge: Nature Geoscience, v. 1, p. 793-799.

Berman, R.G., 2007, winTWQ (version 2.3): a software package for performing internally-consistent thermobarometric calculations. Geological Survey of Canada Open File Report 5462, (ed. 2.32) 2007, 41 pages.

Billen, M.I., 2008, Modeling the Dynamics of Subducting Slabs, Annual Review of Earth and Planetary Sciences, v. 36, pp. 325-356.

Bird, P., 1996, Computer simulations of Alaskan neotectonics, Tectonics, v. 15, pp. 225– 256.

Bird, P., 1998, Kinematic history of the Laramide orogeny in latitudes 35º -49º N, western United States, Tectonics, v. 17, pp. 780-801.

Bowman, D., G. King, and P. Tapponnier 2003, Slip partitioning by elastoplastic propagation of oblique slip at depth, Science, 300, pp. 1121–1123.

Bradley, D. C., R. D. Tucker, D. R. Lux, A. G. Harris, and D. C. McGregor, 2000, Migration of the Acadian Orogen and Foreland Basin across the Northern Appalachians of Maine and adjacent areas, USGS Professional Paper 1624, 64 pp.

Bradley, D.C., T.M. Kusky, P.J. Haeussler, R.J. Goldfarb, J.L. Miller, J.A. Dumoulin, S.W. Nelson, and S.M. Karl, 2003, Geology of a transpressional orogen developed during ridge-trench interaction along the North Pacific margin, in Geology of a Transpressional Orogen Developed during Ridge-Trench Interaction along the North Pacific Margin, edited by V.B. Sisson, S.M. Roeske, and T.L. Pavlis, Geologic Society of America Special Paper, v. 371, pp. 19-49.

Brace, W.F., and D.L. Kohlstedt, 1980, Limits on lithospheric stress imposed by laboratory experiments, Journal of Geophysical Research, v. 85, pp. 6248–6252.

156

Bruhn, R.L., T.L. Pavlis, G. Plafker, and L. Serpa, 2004, Deformation during terrane accretion in the Saint Elias orogen, Alaska, Geological Society of America Bulletin, v. 116, p. 771–787.

Bruhn, R.L., and P.J. Haeussler, 2006, Deformation driven by subduction and microplate collision: Geodynamics of Cook Inlet basin, Alaska, Geological Society of America Bulletin, v. 118, pp. 289–303. Bruns, T.R., 1983, Model for the origin of the Yakutat Block, an accreted terrane in the northern Gulf of Alaska, Geology, v. 11, p. 718–721.

Bruns, T.R., 1985, Tectonics of the Yakutat Block, an allochthonous terrane in the northern Gulf of Alaska: U.S. Geological Survey Open-File Report 85–13, 112 p.

Bucher, K. and M. Frey, 2002, Petrogenesis of Metamorphic Rocks, 7th edition, Springer-Verlag, New York, 341 pg.

Buck, W.R., and Poliakov, A.N.B., 1998, Abyssal hills formed by stretching oceanic lithosphere, Nature, v. 392, p. 272-275.

Burg, J.-P., and S.M. Schmalholz, 2008, Viscous heating allows thrusting to overcome crust-scale buckling: Numerical investigation with application to Himalayan syntaxes, Earth and Planetary Science Letters, v. 274, pp. 189-203.

Burov, E.B., and A.B. Watts, 2006, The long-term strength of continental lithosphere: “jelly sandwich” or “crème brûlée”?: Geological society of America Today, v. 16, pp. 4–10.

Byrne, T., 1979, Late Paleocene demise of the Kula-Pacific spreading center, Geology, v. 7, pp. 341-344.

Campbell R.B. and C.J. Dodds, 1982, Geology, Mount St. Elias, Yukon Territory, Map Sheets 115B and C, Geological Survey of Canada Open File Map 830, 1: 125,000 scale.

Carver, G. and G. Plafker, 2008, Paleoseismicity and Neotectonics of the Aleutian Subduction Zone - An Overview, in Active Tectonics and Seismic Potential of Alaska, edited by J.T. Freymueller, P.J. Haeussler, R. Wesson, and G. Ekstrom, Geophysical Monograph 179, AGU, Washington, p. 43-63.

Chappel, W.M., 1978, Mechanics of thin-skinned fold and thrust belts, Geological Society of America Bulletin, v. 89, pp. 1189-1198.

157

Chapman, J. B., T. Pavlis, S. Gulick, A. Burger, L. Lowe, J. Spotila, R. Bruhn, M. Vorkink, P. Koons, A. Barker, K, Picornell, K. Ridgeway, B. Hallet, J. Jaeger, and J. McAlpin (2008), Neotectonics of the Yakutat collision: Changes in deformation driven by mass redistribution, in Active Tectonics and Seismic Potential of Alaska, edited by J.T. Freymueller, P.J. Haeussler, R. Wesson, and G. Ekstrom, Geophysical Monograph 179, AGU, Washington, p. 65-81.

Chen, W.-P., and P. Molnar, 1983, Focal depths of intracontinental and intraplate earthquakes and their implications for the thermal and mechanical properties of the lithosphere: Journal of Geophysical Research, v. 88, p. 4183–4214.

Cole, R.B., B. Hooks, C. Turner, and C. Slaugenhoup, 2000, Unroofing of Metamorphic Terranes and Subsequent Volcanism: Late Cretaceous(?)-Early Tertiary Basin Evolution along the McKinley Fault, McKinley 1x3 Quadrangle, Alaska: Geological Society of America, Abstract with Programs, v. 32, no. 1, p. A-11.

Cole, Ronald B., W.M. Brewer, and B.P. Hooks, 2002, Melange and Late Paleozoic- Triassic(?) Chert, Limestone, and Pillow Basalt Along a Terrane Suture Zone South of the Denali Fault, Reindeer Hills, Alaska, Geological Society of America Abstract with Programs, v. 34, no. 6, p. 44.

Comsol, 2006, Comsol Multiphysics User's Guide, Burlington, MA, 708 pgs.

Connolly, J.A.D., 2005, Computation of phase equilibria by linear programming: a tool for geodynamic modeling and its application to subduction zone decarbonation, Earth and Planetary Science Letters, v. 236, pp. 524-541.

Conrad, C.P., and C. Lithgow-Bertelloni, 2002, How Mantle Slabs Drive Plate Tectonics, Science, v. 298, pp. 207-209.

Conrad, C.P., S. Bilek, and C. Lithgow-Bertelloni, 2004, Great earthquakes and slab pull: interaction between seismic coupling and plate-slab coupling, Earth and Planetary Science Letters, v. 218, pp. 109-222.

Csejtey Jr., B., 1990, Discrepancies between geologic evidence and rotational models - Talkeetna Mountains and adjacent areas of South-Central Alaska, in Geological studies in Alaska by the United States Geological Survey, edited by Bradley, D.C., and A.B. Ford, United States Geological Survey Professional Paper 1999, pp. 71- 80.

Cundall, P. A., and J. Board, 1988, A microcomputer program for modeling large-strain plasticity problems, in Swododa, C., editor, Numerical Methods in Geomechanics, Proceedings 6th International Conference on Numerical Methods in Geomechanics, Innsbruck, Austria, April 11–15: Rotterdam, Balkema, p. 2101–2108.

158

Dahlen, F.A., 1984, Noncohesive critical Coulomb wedges; an exact solution: Journal of Geophysical Research, v. 89, p. 10,125–10,133.

Davis, D., J. Suppe, and F.A. Dahlen, 1983, Mechanics of fold-and-thrust belts and accretionary wedges, Journal of Geophysical Research, v. 88, p. 1153–1172.

DeMets, C., R.G. Gordon, D.F. Argus, and S. Stein, 1994, Effect of recent revisions of geomagnetic reversal time-scale on estimate of current plate motions: Geophysical Research Letters, v. 21, p. 2191–2194.

DeMets, C., and T.H. Dixon, 1999, New kinematic models for Pacific-North America motion from 3 Ma to present, I: Evidence for steady motion and biases in the NUVEL-1A model, Geophysical Research Letters, v. 26, pp. 1921-1924.

Dewey, J.F., 1988, Extensional collapse of orogens, Tectonics, v. 7, pp. 1123-1139.

Dickinson, W.R., 1974, Plate tectonics and sedimentation, in Tectonics and Sedimentation, edited by W.R. Dickinson, Soc. Econ. Pal. Mineral. Special publication, v. 22, pp. 1-27.

Dickinson, W.R. and W.S. Snyder, 1978, Plate tectonics of the Laramide Orogeny, Geological Society of America Memoir, v. 141, pp. 355-366

Dickinson, W.R., M.A. Klute, M. J. Hayes, S. U. Janecke, E. R. Lundin, M. A., McKittrick, and M. D. Olivares, 1988, Paleogeographic and paleotectonic setting of Laramide sedimentary basins in the central Rock Mountains, Geological Society of America Bulletin, v. 100, pp. 1023-1039.

Dusel-Bacon, C., B. Csejtey, Jr., H.L. Foster, E.O. Doyle, W.J. Nokleberg, and G. Plafker, 1993, Distribution, Facies, Ages, and Proposed Tectonic Associations of Regionally Metamorphosed Rocks in East- and South-Central Alaska, U.S.G.S. Professional Paper 1497-C, 73 p.

Eberhart-Phillips, D., Christensen, D.H., Brocher, T.M., Hansen, R., Ruppert, N.A., Haeussler, P.J., and Abers, G.A., 2006, Imaging the transition from Aleutian subduction to Yakutat collision in central Alaska, with local earthquakes and active source data: Journal of Geophysical Research, v. 111, p. B11303.

Ellis S., C. Beaumont, R.A. Jamieson, and G. Quinlan, 1998, continental collision including a weak zone: the vise model and its application to Newfoundland Appalachians, Canadian Journal of Earth Science, v. 35, pp. 1323-1346.

Ellis, S., G. Schreurs, and M. Panien, 2004, Comparisons between analogue and numerical models of thrust wedge development, Journal of Structural Geology, v. 26, pp. 1659-1675.

159

England, P. and G. Houseman, 1989, Extension during continental convergence, with application to the Tibetan Plateau, Journal of Geophysical Research, v. 94, pp. 17561-17579.

England, P., and P. Molnar, 1990, Surface uplift, uplift of rocks, and exhumation of rocks, Geology, v. 18, pp. 1173-1177.

Engebretson, D., A. Cox, and R.G. Gordon, 1985, Relative motions between oceanic and continental plates in the Pacific Basin, Geological Society of America Special Paper, v. 206, 59 pp.

Enkelmann, E., J.I. Garver, and T.L. Pavlis, 2008, Rapid exhumation of ice-covered rocks of the Chugach-St. Elias orogen, SE-Alaska: Geology, v. 36, p. 915-918.

Enkelmann, E., P.K. Zeitler, T.L. Pavlis, J.I. Garver, and K.D. Ridgway, 2009, Intense Localized Rock Uplift and Erosion in the St. Elias Orogen of Alaska, Nature Geoscience, v. 2, p. 360-363.

Enlow, R.L., and P.O. Koons, 1998, Critical wedges in three dimensions; analytical expressions from Mohr-Coulomb constrained perturbation analysis, Journal of Geophysical Research, v. 103, p. 4897–4914.

Estabrook, C.H., J.L. Nabelek, and Lerner-Lam, 1992, Tectonic model of the Pacific- North America plate boundary in the Gulf of Alaska from broadband analysis of the 1979 St. Elias, Alaska earthquake and its aftershocks, Journal of Geophysical Research, v. 97, p. 6587-6612.

Eyles, C.H., N. Eyles, and M.B. Lagoe, 1991, The Yakataga Formation; a late Miocene to Pleistocene record of temperate glacial marine sedimentation in the Gulf of Alaska, in Glacial-marine sedimentation; paleoclimatic significance, edited by J.B. Anderson and G.M. Ashley,: Geological Society of America Special Paper, v. 261, pp. 159-180.

Faill, R.T., and R.P. Nickelsen, 1999, Appalachian Mountain section of the Ridge and Valley province, in The Geology of Pennsylvania, edited by C.H. Schultz, Pennsylvania Geological Survey Special Publication No. 1, pp. 268-285.

Ferris, A., G.A. Abers, D.H. Christensen, and E. Veenstra, 2003, High resolution image of the subducted Pacific (?) plate beneath central Alaska, 50-150 km depth, Earth and Planetary Science Letters, v. 214, pp. 575-588.

Fitzgerald, P.G., R.B. Sorkhabi, T.F. Redfield, and E. Stump, 1995, Uplift and denudation of the central Alaska Range: A case study in the use of apatite fission- track thermochronology to determine absolute uplift parameters, Journal of Geophysical Research, v. 100, pp. 20175-20191.

160

Flemings, P.B., and T.E. Jordan, 1990, Stratigraphic modeling of foreland basins; interpreting thrust deformation and lithosphere rheology, Geology, v. 18, pp. 430- 434.

Fletcher, H. J., and J.T. Freymueller, 1999, New GPS constraints on the motion of the Yakutat block, Geophysical Research Letters, v. 19, pp. 3029–3032.

Fowler, C.M.R., 2005, The Solid Earth: an introduction to global geophysics, 2nd Edition. Cambridge University Press, Cambridge.

Fuis, G.S., Plafker, G., 1991. Evolution of deep structure along the Trans-Alaska Crustal Transect, Chugach Mountains and Copper River Basin, Southern Alaska. Journal Geophysical Research, v. 96, pp. 4229–4253.

Fuis, G.S., T.E. More, G. Plafker, T.M. Brocher, M.A. Fisher, W.D. Mooney, W.J. Nokleberg, R.A. Page, B.C. Beaudoin, N.I. Christensen, A.R. Levander, W.J. Lutter, R.W. Saltus, N.A. Ruppert, 2008, Trans-Alaska Crustal Transect and continental evolution involving subduction underplating and synchronous foreland thrusting, Geology, v. 36, pp. 267-270.

Funiciello F, C. Faccenna, D. Giardini, and K. Regenauer-Lieb, 2003, Dynamics of retreating slabs: 2. Insights from three-dimensional laboratory experiments, Journal of Geophysical Research, v. 108, 2207.

Gerbault, M. J. Martinod, and G. Herail, 2005, Possible orogeny-parallel lower crustal flow and thickening in the Central Andes, Tectonophysics, v. 399, pp. 59-72.

Giger, S.B, S.F. Cox, and E. Tenthory., 2008, Slip localization and fault weakening as a consequence of fault gouge strengthening; insights from laboratory experiments, Earth and Planetary Science Letters, vol. 276, no. 1-2, pp.73-84

Gleason, G.C., and J. Tullis, 1995, A flow law for dislocation creep of quartz aggregates determined with the molten salt cell: Tectonophysics, v. 247, p. 1–23.

Groome, W. G., S. E. Johnson, and P. O. Koons, 2006, The effects of porphyroblast growth on the effective viscosity of metapelitic rocks: Implications for the strength of the middle crust, Journal of Metamorphic Geology, v. 24, pp. 389-407.

Groome, W. G., P. O. Koons, and S. E. Johnson, 2008, Metamorphism, transient mid- crustal rheology, strain localization and the exhumation of high-grade metamorphic rocks, Tectonics, v. 27, TC1001.

Gulick, S.P.S., L.A. Lowe, T.L. Pavlis, J.V. Gardner, and L.A. Mayer, 2007, Geophysical insights into Transition fault debate: Propagating strike-slip in response to stalling Yakutat block subduction in the Gulf of Alaska, Geology, v. 35, pp. 763-766.

161

Gutscher, M.-A., W. Spakman, H. Bijwaard, and E. R. Engdahl, 2000a, Geodynamics of flat subduction: Seismicity and tomographic constraints from the Andean margin, Tectonics, v. 9, pp. 814–833.

Gutscher, M.-A., R. Maury, J.-P. Eissen and E. Bourdon, 2000b, Can slab melting be caused by flat subduction?, Geology, v. 28, pp. 535-538.

Gutscher, M.-A., 2001, An Andean model of interplate coupling and strain partitioning applied to the flat subduction of SW Japan (Nankai Trough), Tectonophysics, v. 333, pp. 95–109.

Gutscher, M.-A., and S. M. Peacock, 2003, Thermal models of flat subduction and the rupture zone of great subduction earthquakes, Journal of Geophysical Research, v. 108, 2009.

Hacker, B. R., G. A. Abers, and S. M. Peacock, 2003a, Subduction factory: 1. Theoretical mineralogy, densities, seismic wave speeds, and H2O contents, Journal of Geophysical Research, v. 108, 2029.

Hacker, B. R., S. M. Peacock, G. A. Abers, and S. D. Holloway (2003b), Subduction factory: 2. Are intermediate-depth earthquakes in subducting slabs linked to metamorphic dehydration reactions?, Journal of Geophysical Research, v. 108, pp. 2030.

Haeussler, P.J., P. O’Sullivan, A.L. Berger, and J.A. Spotila, 2008, Neogene exhumation of the Tordrillo Mountains, Alaska, and correlations with Denali (Mt. McKinley), in Active Tectonics and Seismic Potential of Alaska, edited by J.T. Freymueller, P.J. Haeussler, R. Wesson, and G. Ekstrom, Geophysical Monograph 179, AGU, Washington, p. 269-285.

Haeussler, P.J., 2008, An overview of the neotectonics of interior Alaska: far-field deformation from the Yakutat microplate collision: in Active Tectonics and Seismic Potential of Alaska, edited by J.T. Freymueller, P.J. Haeussler, R. Wesson, and G. Ekstrom, Geophysical Monograph 179, AGU, Washington, p. 83- 108.

Hall, J., F. Marillier, and S. Dehler, 1998, Geophysical studies of the structure of the Appalachian orogen in the Atlantic borderlands of Canada, Canadian Journal of Earth Science v. 35, pp. 1205-1221.

Hallet, B., L. Hunter, and J. Bogen, 1996, Rates of erosion and sediment evacuation by glaciers: A review of field data and their implications, Global and Planetary Change, v.12, p. 213-235.

Handy, M.R., and J.-P. Brun, 2004, Seismicity, structure, and strength of the continental lithosphere, Earth and Planetary Science Letters, v. 223, pp. 427-441.

162

Handy, M.R., G. Hirth, R. Burgmann, 2007, Continental Fault Structure and Rheology from the Frictional-to-Viscous Transition Downward, in Tectonic Faults: Agents of Change on a Dynamic Earth, edited by M. R. Handy, et al., pp. 139-181, MIT Press, Cambridge, MA.

Hartz, E.H., and Y.Y. Podladchikov, 2008, Toasting the jelly sandwich: the effect of shear heating on lithospheric geotherms and strength, Geology, v. 36, pp. 331- 334.

Hirose, T., and M. Bystricky, 2007, Extreme dynamic weakening of faults during dehydration by coseismic shear heating, Geophysical Research Letters, v. 34, L14311.

Hirth, G., and J. Tullis, 1992, Dislocation creep regimes in quartz aggregates, Journal of Structural Geology, v. 14, pp. 145–159.

Hobbs, B. E., H.-B Mühlhaus, and A. Ord, 1990, Instability, softening and localization of deformation, in Deformation Mechanism, Rheology and Tectonics, edited by R.J. Knipe and E.H. Rutter, Geological Society Special Paper, v. 54, pp. 143-165.

Hodges, K.V., 2000, Tectonics of the Himalaya and southern Tibet from two perspectives, Geologic Society of America Bulletin, v. 112, pp. 324-350.

Holyoke III, C.W., and J. Tullis, 2006, Mechanisms of weak phase interconnection and the effects of phase strength contrast on fabric development, Journal of Structural Geology, v. 28, pp. 621-640.

Hooks, B.P., E. Enkelmann, P.O. Koons, P. Upton, and T.L. Pavlis, in review, 3D Mechanical View of the Relative Importance of Surface and Tectonic Processes in the Initiation of the St Elias Tectonic Aneurysm, submitted to Geology.

Horsman, E., B. Tikoff, and D. Czeck, 2008, Rheological implications of heterogeneous deformation at multiple scales in the Late Cretaceous Sierra Nevada, California, Geologic Society of America Bulletin, v. 120, p. 238–255.

Huang, C.-Y., P.B. Yuan, C.-W. Lin, T.K. Wang, and C.-P. Chang, 2000, Geodynamic processes of Taiwan arc-continent collision and comparison with analogs in Timor, Papua New Guinea, Urals, and Corsica, Tectonophysics, v. 325, pp. 1-21.

Huang, C.-Y., P.B. Yuan, and S.-J. Tsao, 2006, Temporal and spatial records of active arc-continent collision in Taiwan: a synthesis, Bulletin of the Geological Society of America, v. 118, pp. 274-288.

163

Hyndman, R.D., and K. Wang, 1993, Thermal constraints on the zone of major thrust earthquake failure: The Cascadia subduction zone, Journal of Geophysical Research, v. 98, pp. 2039-2060.

Hyndman, R. D., M. Yamano, and K. Wang, 1995, Thermal constraints on the seismogenic portion of the southwestern Japan subduction thrust, Journal of Geophysical Research, v. 100, pp. 15373–15392.

Igarashi, T., T. Matsuzawa, N. Umino, and A. Hasegawa, 2001, Spatial distribution of focal mechanisms for interplate and intraplate earthquakes associated with the subducting Pacific plate beneath the northeastern Japan arc: A tripled-planed deep seismic zone, Journal of Geophysical Research, v. 106, 2177–2191.

Isacks, B.L., 1988, Uplift of the central Andean plateau and bending of the Bolivian orocline, Journal of Geophysical Research, v. 93, pp. 3211-3231.

ITASCA, 2006, FLAC3D (Fast Lagrangian Analysis of Continua in 3 Dimensions), Minneapolis, MN.

Jackson, J., 2002, Strength of continental lithosphere: time to abandon the jelly sandwich?, Geologic Society of America Today, v. 12 pp. 4-10.

Jaeger, J.M., C.A. Nittrouer, N.D. Scott, and J.D. Milliman, 1998, Sediment accumulation along a glacially impacted mountainous coastline: Northeast Gulf of Alaska, Basin Research, v. 10, p. 155– 173.

Jamieson, R. A., C. Beaumont, S. Medvedev, and M. H. Nguyen , 2004, Crustal channel flows: 2. Numerical models with implications for metamorphism in the Himalayan-Tibetan orogen, Journal of Geophysical Research, v. 109, B06407.

Jarrard, R.D., 1986, Relations among subduction parameters, Reviews in Geophysics, v. 24, pp. 217-284.

Johnson, J.M., K. Satake, S.R. Holdahl, and J. Sauber, 1996, The 1964 Prince William Sound earthquake; joint inversion of tsunami and geodetic data, Journal of Geophysical Research, v. 101, pp. 523-532.

Johnson, S. E., R. H. Vernon, and P. Upton, 2004, Foliation development and progressive strain rate partitioning in the crystallizing carapace of a tonalite pluton: Microstructural evidence and numerical modeling, Journal of Structural Geology, v. 26, pp. 1845 – 1865.

Jordan, T.A., and A.B. Watts, 2005, Gravity anomalies, flexure and the elastic thickness structure of the India-Eurasia collisional system, Earth and Planetary Science Letters, v. 236, pp. 732-750.

164

Kalbas, J.L., K.D. Ridgway, and G.E. Gehrels, 2007, Stratigraphy, depositional systems, and provenance of the Lower Cretaceous Kahiltna assemblage, western Alaska Range: Basin development in response to oblique collision, in Tectonic Growth of a Collisional Continental Margin: Crustal Evolution of Southern Alaska, edited by K.D. Ridgway, J.M. Trop, J.M.G. Glen, and J.M. O’Neill, Geologic Society of America Special Paper, v. 431, pp. 307-343.

Kauffman, M.E., 1999, Eocambrian, Cambrian, and transition to Ordovician, in The Geology of Pennsylvania, edited by C.H. Schultz, Pennsylvania Geological Survey Special Publication No. 1, pp. 58-73.

Kearney, P., K.A. Klepeis, and F.J. Vine, 2009, Global Tectonics, 3rd edition, Wiley- Blackwell, New Jersey, 482 pgs.

Kirby, S. H., 1995, Intraslab earthquakes and phase changes in subducting lithosphere, Reviews of Geophysics, v. 33 Suppl., pp. 287–297.

Koons, P.O., 1990, The two-sided wedge in orogeny; erosion and collision from the sand box to the Southern Alps, New Zealand: Geology, v. 18, p. 679–682.

Koons, P.O., 1994, Three-dimensional critical wedges : tectonics and topography in oblique collisional orogens: Journal of Geophysical Research, v. 99, p. 12,301– 12,315. Koons, P.O., 1995, Modeling the Topographic Evolution of Collisional Belts, Annual Review of Earth and Planetary Sciences, v. 23, pp. 375-408.

Koons, P.O., Zeitler, P.K., Chamberlain, C.P., Craw, D., and Melzer, A.S., 2002, Mechanical links between river erosion and metamorphism in Nanga Parbat, Pakistan Himalaya: American Journal of Science, v. 302, p. 749-773.

Koons, P.O., R.J. Norris, D. Craw, and A.F. Cooper, 2003, Influence of exhumation on the structural evolution of transpressional plate boundaries: An example from the Southern Alps, New Zealand, Geology, v. 31, pp. 306.

Koons, P.O., Hooks, B.P., Upton, P., Barker, A.D., and Pavlis, T.L., 2009, Three- dimensional mechanics of Yakutat convergence in the southern Alaskan plate corner: Tectonics, in review.

Kohlstedt D.L., and C. Goetze, 1974, Low-stress high-temperature creep in olivine single crystals, Journal of Geophysical Research, v. 79, pp. 2045-2051.

Kohlstedt, D.L., Evans, B., and Mackwell, S.J., 1995, Strength of the lithosphere— constraints imposed by laboratory experiments: Journal of Geophysical Research, v. 100, pp. 17,587–17,602.

165

Lahr, J., and G. Plafker, 1980, Holocene Pacific– North American plate interaction in southern Alaska: Implications for the Yakataga seismic gap, Geology, v. 8, pp. 483–486.

Lagoe, M. B., C.H. Eyles, N. Eyles, and C. Hale, 1993, Timing of late Cenozoic tidewater glaciation in the far North Pacific, Geologic Society of America Bulletin, v. 105, pp. 1542–1560.

Lagoe, M.B., and S.D. Zellers, 1996, Depositional and microfaunal response to Pliocene climate change and tectonics in the eastern Gulf of Alaska, Marine Micropaleontology, v. 27, pp. 121-140.

Larsen, C.F., R.J. Motyka, J.T. Freymueller, K.A. Echelmeyer, and E.R. Ivins, 2005, Rapid viscoelastic uplift in southeast Alaska caused by post-Little Ice Age glacial retreat, Earth and Planetary Science Letters, v. 237, pp. 548-560.

Lesh, M. E., and K. D. Ridgway, 2007, Geomorphic evidence of active transpressional deformation in the Tanana foreland basin, south-central Alaska, in Tectonic Growth of a Collisional Continental Margin: Crustal Evolution of Southern Alaska, edited by K. D. Ridgway et al., Geologic Society of America Special Paper, v. 431, pp. 573– 592.

Li, L., H. Long, P. Raterron, and D. Weidner, 2006, Plastic flow of pyrope at mantle pressure and temperature, American Mineralogist, v. 91, pp. 517-525.

Liu, M. and Y. Yang, 2003, Extensional collapse of the Tibetan Plateau: results of three- dimensional finite element modelling, Journal of Geophysical Research, v. 108, 2361.

Mackwell, S.J., M.E. Zimmerman, and D.E. Kohlstedt, 1998, High temperature deformation of dry diabase with application to tectonics on Venus, Journal of Geophysical Research, v. 103, pp. 975– 984.

Maggi, A., J.A. Jackson, D. McKenzie, and K. Priestly, 2000, Earthquake focal depths, effective elastic thickness, and the strength of the continental lithosphere, Geology, v. 28, p. 495–498.

Matmon, A., D.P. Schwartz, P.J. Haeussler, R. Finkel, J.J. Lienkaemper, H.D. Stenner, and T.E. Dawson, 2006, Denali fault slip rates and Holocene-late Pleistocene kinematics of central Alaska, Geology, v. 34, p. 645–648.

Mazzotti, S., and R.D. Hyndman, 2002, Yakutat collision and strain transfer across the northern Canadian Cordillera: Geology, v. 30, p. 495–498.

166

McAleer, R.J., J.A. Spotila, E. Enkelmann, and A.L Berger, 2009, Exhumation along the Fairweather Fault, Southeast Alaska, based on Low-Temperature Thermochronometry: Tectonics, v. 28, TC1007.

Meigs, A., and J. Sauber, 2000, Southern Alaska as an example of the long-term consequences of mountain building under the influence of glaciers, Quaternary Science Review, v. 19, pp. 1543-1562.

Meigs, A. S. Johnston, J. Garver, and J. Spotila, 2008, Crustal-scale structural architecture, shortening, and exhumation of an active, eroding orogenic wedge (Chugach/St. Elias Range, southern Alaska), Tectonics, v. 27, TC4003.

Meriaux, A.-S., K. Sieh, R.C. Finkel, C.M. Rubin, M.H. Taylor, A.J. Meltzner, and R.J. Ryerson, 2009, Kinematic behavior of southern Alaska constrained by westward decreasing postglacial slip rates on the Denali Fault, Alaska, Journal of Geophysical Research, v. 114, B03404.

Miller, M.L., D.C. Bradley, T.K. Bundtzen, and W. McClelland, 2002, Late Cretaceous through Cenozoic strike-slip tectonics of southwestern Alaska, The Journal of Geology, v. 110, pp. 247-270.

Montesi G. J. L., and M. T. Zuber, 2002, A unified description of localization for application to large-scale tectonics. Journal of Geophysical Research, v. 107, 2045.

Mori, L., A. Gomez-Tuena, Y. Cai, and S.L. Goldstein, 2007, Effects of prolonged flat subduction on the Miocene magmatic record of the central Trans-Mexican Volcanic Belt, Chemical Geology, v. 244, pp. 452–473.

Munoz, M., 2005, No flat Wadati-Benioff Zone in the central and southern central Andes, Tectonophysics, v. 395, pp. 41-65.

Murphy, J.B., C.R. van Staal, and J.D. Keppie, 1999, Middle to late Paleozoic Acadian orogeny in the northern Appalachians: A Laramide-style plume-modified orogeny?, Geology, v. 27; p. 653–656.

Nagel, T.J. and W.R. Buck, 2004, Symmetric alternative to asymmetric rifting models, Geology V. 32, pp. 937-940.

Naylor, M., H.D. Sinclair, S. Willett, and P.A. Cowie, 2005, A discrete element model for orogenesis and accretionary wedge growth, Journal of Geophysical Research, v. 110, B12403.

Nokleberg, W.J., J.N. Aleinikoff, and I.M. Lange, 1986, Cretaceous deformation and metamorphism in the northeastern Mount Hayes Quadrangle, eastern Alaska Range, U.S.G.S. Report C-0978.

167

Nokleberg, W.J., G. Plafker, and F.H. Wilson, 1994, Geology of south-central Alaska, in The Geology of North America, v. G-1, The Geology of Alaska, edited by G. Plafker and H.C. Berg, Geological Society of America, Boulder, CO., 311-366.

Oleskevich, D.A., R.D. Hyndman, and K. Wang, 1999, The updip and downdip limits to great subduction earthquakes: thermal and structural models of Cascadia, south Alaska, SW Japan, and Chile, Journal of Geophysical Research, v. 104, pp. 14965-14991.

O’Sullivan, P. B., Plafker, G., Murphy, J.M., 1995, Apatite Fission-Track Thermotectonic History of Crystalline Rocks in the Northern Saint Elias Mountains, Alaska: in Geological studies in Alaska by the United States Geological Survey, edited by Dumoulin, J.A., and J.E. Gray, United States Geological Survey Professional Paper 1574, p. 283-294.

O’Sullivan, P. B. and Currie, L.D., 1996, Thermotectonic history of Mt. Logan, Yukon Territory, Canada: implications of multiple episodes of middle to late Cenozoic denudation, Earth and Planetary Science Letters, v. 144, p. 251-261.

Page, R. A., G. Plafker, and H. Pulpan, 1995, Block rotation in east-central Alaska: A framework for evaluating earthquake potential?, Geology, v. 23, pp. 629– 632.

Pavlis, T.L., V.B. Sisson, H.L. Foster, W. J. Nokleberg, and G. Plafker, 1993, Mid- Cretaceous extensional tectonics of the Yukon-Tanana terrane, Trans-Alaska Crustal Transect (TACT), east-central Alaska, Tectonics, v. 12, pp. 103-122.

Pavlis, T.L., and V.B. Sisson, 1995, Structural history of the Chugach metamorphic complex in the Tana River region, eastern Alaska: A record of Eocene ridge subduction, Geologic Society of America Bulletin, v. 107, pp. 1333-1355.

Pavlis, T.L., and V.B. Sisson, 2003, Development of a subhorizontal decoupling horizon in a transpressional system, Chugach metamorphic complex, Alaska: Evidence for rheological stratification of the crust, in, Geology of a Transpressional Orogen Developed during Ridge-Trench Interaction along the North Pacific Margin, edited by V.B. Sisson, S.M. Roeske, and T.L. Pavlis, Geologic Society of America Special Paper, v. 371, pp. 1-18.

Pavlis, T. L., C. Picornell, L. Serpa, R. L. Bruhn, and G. Plafker, 2004, Tectonic processes during oblique collision: Insights from the St. Elias orogen, northern North American Cordillera, Tectonics, 23, TC3001.

Péwé, T.L., 1975, Quaternary geology of Alaska: U.S. Geological Survey Professional Paper 835, 139 p.

Peacock, S.M., 1990, Fluid Processes in Subduction Zones, Science, v. 248, pp. 329-337.

168

Peacock, S. M., and R. Hyndman, 1999, Hydrous minerals in the mantle wedge and the maximum depth of subduction thrust earthquakes, Geophysical Research Letters, v. 26, pp. 2517–2520.

Peacock, S. M., and K. Wang, 1999, Seismic consequences of warm versus cool subduction zone metamorphism: Examples from northeast and southwest Japan, Science, v. 286, pp. 937– 939.

Pegler, G., and S. Das, 1996, The 1987-1992 Gulf of Alaska earthquakes, Tectonophysics, v. 257, pp. 111-136.

Perry, S.E., J.I. Garver, and K.D. Ridgway, 2009, Transport of the Yakutat Terrane, Southern Alaska: Evidence from Sediment Petrology and Detrital Zircon Fission- Track, and U/Pb Double Dating, Journal of Geology, v. 117, pp. 156-173.

Petersen, M.D., A.D. Frankel, S.C. Harmsen, C.S. Mueller, K.M. Haller, R.L. Wheeler, R.L. Wesson, Y. Zeng, O.S. Boyd, D.M. Perkins, N. Luco, E.H. Field, C.J. Wills, and K.S. Rukstales, 2008, Documentation for the 2008 Update of the United States National Seismic Hazard Maps: U.S. Geological Survey Open-File Report 2008–1128, 61 p.

Philpotts, A.R., 1990, Principles of Igneous and Metamorphic Petrology, Prentice Hall, New Jersey, 498 pg.

Plafker, G., 1965, Tectonic deformation associated with the 1964 Alaska earthquake: Science, v. 148, pp. 1675-1687.

Plafker, G., and H.C. Berg, 1994, Overview of the geology and tectonic evolution of Alaska, in The Geology of North America, v. G-1, The Geology of Alaska, edited by G. Plafker and H.C. Berg, Geologic Society of America, Boulder CO., pp. 989–1021.

Plafker, G., T. Hudson, and D.H. Richter, 1977, Preliminary observations on late Cenozoic displacements along the Totschunda and Denali fault systems: U.S. Geological Survey Circular, Report C 0751-B, p. B67–B69.

Plafker, G., 1987, Regional geology and petroleum potential of the northern Gulf of Alaska continental margin: in Geology and resource potential of the continental margin of western North America and adjacent ocean basins, in Circum-Pacific Council for Energy and Mineral Resources, edited by D.W. Scholl, A. Grantz, and J.G. Vedder, Earth Science Series v. 6, Circum-Pacific Council for Energy and Mineral Resources, Houston, Texas, 229-268.

169

Plafker, G., W.J. Nokleberg, and J.S. Lull, 1989, Bedrock geology and tectonic evolution of the Wrangellia, Peninsular, and Chugach terranes along the Trans-Alaska Crustal Transect in the Chugach Mountains and Southern Copper River Basin, Alaska, Journal of Geophysical Research, v. 94, pp. 4255-4295.

Plafker, G., J. C. Moore, and G. R. Winkler, 1994, Geology of the southern Alaska margin, in The Geology of North America, vol. G-1, The Geology of Alaska, edited by G. Plafker and H. C. Berg, pp. 389– 449, Geologic Society of America, Boulder, Colo.

Plafker, G., and W. Thatcher, 2008, Geological and Geophysical Evaluation of the Mechanisms of the great 1899 Yakutat Bay Earthquakes, in Active Tectonics and Seismic Potential of Alaska, edited by J.T. Freymueller, P.J. Haeussler, R. Wesson, and G. Ekstrom, Geophysical Monograph 179, AGU, Washington, p. 215-236.

Poli, S., and M.W. Schmidt, 2002, Petrology of subducted slabs, Annual Review in Earth and Planetary Science, v. 30, pp. 207–235.

Ponko, S.C. and S.M. Peacock, 1995, Thermal modeling of the southern Alaska subduction zone: Insight into the petrology of the subducting slab and overlying mantle wedge, Journal of Geophysical Research, v. 100, pp. 22,117-22,128.

Powell, R., T.J.B. Holland, and B. Worley, 1998. Calculating phase diagrams involving solid solutions via non-linear equations, with examples using THERMOCALC Journal of Metamorphic Geology, v. 16, pp. 577-588.

Preece, S.J., and W.K. Hart, 2004, Geochemical variations in the <5 Ma Wrangell Volcanic Field, Alaska: implications for the magmatic and tectonic development of a complex continental arc system, Tectonophysics, v. 392, pp. 165-191.

Protti, M., F. Guendel, and K. McNally, Correlation between the age of the subducting Cocos Plate and the geometry of the Wadati-Benioff zone under Nicaragua and Costa Rica, in Geologic and Tectonic Development of the Caribbean Plate Boundary in Southern Central America, edited by P. Mann, Geologic Society of America Special Paper, v. 295, pp. 309– 326.

Redfield, T.F., D.W. Scholl, P.G. Fitzgerald, and M.E. Beck, Jr., Escape tectonics and the extrusion of Alaska: Past, present, and future, Geology, v. 35, pp. 1039-1042.

Regenauer-Lieb, K., G. Rosenbaum, and R.F. Weinberg, 2008, Strain localisation and weakening of the lithosphere during extension, Tectonophysics, v. 458, pp. 96- 104.

170

Reyners, M., and D. Eberhart-Phillips, 2009, Small earthquakes provide insight into plate coupling and fluid distribution in the Hikurangi subduction zone, Earth and Planetary Science Letters, v. 282, pp. 299-305.

Ridgway, K.D., E.E. Thoms, P.W. Layer, M.E. Lash, J.M. White, and S.V. Smith, 2007, Neogene transpressional foreland basin development on the north side of the central Alaska Range, Usibelli Group and Nenana Gravel, Tanana basin, in Tectonic Growth of a Collisional Continental Margin: Crustal Evolution of Southern Alaska, edited by K.D. Ridgway, J.M. Trop, J.M.G. Glen, and J.M. O’Neill, Geologic Society of America Special Paper, v. 431, pp. 507-547.

Rossi, G., G. A. Abers, S. Rondenay, and D. H. Christensen, 2006, Unusual mantle Poisson’s ratio, subduction, and crustal structure in central Alaska, Journal of Geophysical Research, v. 111, B09311.

Royden, L., 1996, Coupling and decoupling of the crust and mantle in convergent orogens: implications for strain partitioning in the crust, Journal of Geophysical Research, v. 101, pp. 17679-17705.

Schmidt, M., and S. Poli, 1998, Experimentally based water budgets for dehydrating slabs and consequences for arc magma generation, Earth and Planetary Science Letters, v. 163, pp. 361-379.

Sheaf, M.A., Serpa, L., and Pavlis, T.L., 2003, Exhumation rates in the St. Elias Mountains, Alaska: Tectonophysics, v. 367, p. 1–11.

Shennan, I., R. Bruhn, and G. Plafker, 2009, Multi-segment earthquakes and tsunami potential of the Aleutian megathrust, Quaternary Science Reviews, v. 28, pp. 7- 13.

Silver E.A., D. Reed, and R. McCaffery, 1983, Back-arc thrusting in the eastern Sunda arc, Indonesia: a consequence of arc-continental collision, Journal of Geophysical Research, v. 88, pp. 7429-7448.

Sisson, V.B., and L. Hollister, 1988, Low-pressure facies series metamorphism in an accretionary sedimentary prism, southern Alaska, Geology, v. 16, pp. 358-361.

Sisson, V. B., L.S. Hollister, and T.C. Onstott,1989, Petrologic and age constraints on the origin of a low-pressure/high-temperature metamorphic complex, southern Alaska, Journal of Geophysical Research, v. 94, pp. 4392–4410.

Sisson, V.B., and T.L. Pavlis, 1993, Geologic consequences of plate reorganization: An example from the Eocene southern Alaska fore arc, Geology, v. 21, pp. 913-916.

171

Sisson, V.B., A.R. Poole, N.R. Harris, H.C. Burner, T.L. Pavlis, P. Copeland, R.A. Donelick, W.C. McLelland, 2003, Geochemical and geochronological constraints for genesis of a tonalite-trondhjemite suite and associated mafic intrusive rocks in the eastern Chugach Mountains, Alaska; a record of ridge-transform subduction, Geology of a Transpressional Orogen Developed during Ridge-Trench Interaction along the North Pacific Margin, edited by V.B. Sisson, S.M. Roeske, and T.L. Pavlis, Geologic Society of America Special Paper, v. 371, pp. 293-326.

Skulski, T., D. Francis, and J. Ludden, 1992, Volcanism in an arc-transform transition zone: the stratigraphy of the St. Clare Creek volcanic filed, Wrangell volcanic belt, Yukon, Canada, Canadian Journal of Earth Sciences, v. 29, pp. 446-461.

Smith, W. H. F., and Sandwell, D. T., 1997, Global seafloor topography from satellite altimetry and ship depth soundings: Science, v. 277, p. 1956-1962.

Sobolev, S.V. and A.Y. Babeyko, 2005, What drives orogeny in the Andes?, Geology, v. 33, pp. 617-620.

Sobolev, S.V., A. Petruinin, Z. Garfunkel, and A.Y. Babeyko, 2005, Thermo-mechanical model of the Dead Sea transform, Earth and Planetary Science Letters, . 238, pp. 778-95.

Spotila, J.A., J.T. Buscher, A.J. Meigs, and P.W. Reiners, 2004, Long-term glacial erosion of active mountain belts: example of the Chugach–St. Elias Range, Alaska, Geology, v. 32, p. 501–504.

Stewart, R.J., Hallet, B., Zeitler, P.K., Malloy, M.A., Allen, C.M., and Trippett, D., 2008, Brahmaputra sediment flux dominated by highly localized rapid erosion from the easternmost Himalaya, Geology, v. 36, p. 711-714.

Stout, J., and C. Chase, 1980, Plate kinematics of the Denali fault system, Canadian Journal of Earth Science, v. 17, pp. 1527–1537.

Tapponnier, P., G. Peltzer, A.Y. Le Dain, R. Armijo, and P. Cobbold, 1982, Propagating extrusion tectonics in Asia: New insights from simple experiments with plasticine, Geology, v. 10, pp. 611-616.

Tapponnier, P., X. Zhiqin, F. Roger, B. Meyer, N. Arnaud, G. Wittlinger, and Y. Jinsui, 2001, Oblique stepwise rise and growth of the Tibet plateau, Science, v. 294, pp. 1671-1677.

Thatcher, W. and F.F. Pollitz, 2008, Temporal evolution of continental lithospheric strength in actively deforming regions, GSA Today: v. 18, pp. 4-11.

Thompson, A.B. and C.C. Rubie, 1985, Metamorphic reactions; kinetics, textures, and deformation, Springer-Verlag, New York.

172

Turcotte, D. L., and G. Schubert, 2002, Geodynamics, 2nd Edition, 456 pp., Cambridge University Press, Cambridge, UK.

Upton, P., P. O. Koons, and D. Eberhart-Phillips, 2003, Extension and strain-partitioning in an oblique subduction zone, New Zealand: Constraints from three-dimensional numerical modeling, Tectonics, 22, 1068.

Upton, P., P.O. Koons, D. Craw, C.M. Henderson, and R. Enlow, 2009, Along-strike differences in the Southern Alps of New Zealand: Consequences of inherited variation in rheology, Tectonics, v. 28, TC2007.

Uyeda, S. and H. Kanamori, 1979, Back-arc opening and the mode of subduction, Journal of Geophysical Research, v. 84, pp. 1049-1061.

Vermeer, P.A., and R. de Borst, 1984, Non-associated plasticity for soils, concrete, and rock, Heron, v. 29, pp. 3 – 64.

Wdowinski, S., 1992, Dynamically supported trench topography, Journal of Geophysical Research, v. 97, pp. 17651-17656.

Whipple, K.X., 2009, The influence of climate on the tectonic evolution of mountain belts, Nature Geoscience, v. 2, p. 97-104.

Wijns, C., R. Weinberg, K. Gessner, and L. Moresi, 2005, Mode of crustal extension determined by rheological layering, Earth and Planetary Science Letters, v. 236, pp. 120-134.

Willett, S.D., Beaumont, C., and Fullsack, P., 1993, Mechanical model for the tectonics of doubly vergent compressional orogens: Geology, v. 21, p. 371–374.

Williams, H., 1979, Appalachian orogen in Canada, Canadian Journal of Earth Sciences, v. 16, p. 792–807.

Worthington, L.A., S.P.S Gulick, and T.L. Pavlis, 2008, High-resolution datasets provide new constraints on active tectonics in the Kayak Island and Pamplona Zones, offshore Yakutat microplate, southern Alaska margin, in Active Tectonics and Seismic Potential of Alaska, edited by J.T. Freymueller, P.J. Haeussler, R. Wesson, and G. Ekstrom, Geophysical Monograph 179, AGU, Washington, p. 257-268.

Yanez, G., and J. Cembrano, 2004, Role of viscous plate coupling in the late Tertiary Andean tectonics, Journal of Geophysical Research, v. 109, B02407.

Yin, A. and T.M. Harrison, 2000, Geologic evolution of the Himalayan Tibetan Orogen, Annual Reviews of Earth and Planetary Science, v. 28, pp. 211-280.

173

Zeitler, P. K., Koons, P.O., Bishop, M. L., Chamberlain, C.P., Craw, D., Edwards, M.A., Hamidullah, S., Jan, M.Q., Khan, M.A., Khattak, M.U.K., Kidd, W.S.F., Mackie, R.L., Meltzer, A.S., Park, S.K., Pecher, A., Poage, M.A., Sarker, G., Schneider, D.A., Seeber, L., and Shroder, J., 2001, Crustal reworking at Nanga Parbat, Pakistan: Metamorphic consequences of thermal-mechanical coupling facilitated by erosion, Tectonics, v. 20, pp. 712-728.

Zellers, S. D., 1993, Controls on glacial-marine accumulation rates in the Yakataga Formation, Gulf of Alaska, in 14th Annual Research Conference on Rates of Geological Processes, Gulf Coast Section of Society for Sedimentary Geology, Tulsa, OK.

Zellers, S. D., 1995, Foraminiferal sequence biostratigraphy and seismic stratigraphy of a tectonically active margin: The Yakataga Formation, northeastern Gulf of Alaska, Marine Micropaleontology, v. 26, pp. 255 –271.

174

Appendix A: Numerical modeling Code

The Comsol™ and FLAC3D input code files are included within the attached CD.

The file structure for this Appendix is located within ":/Appendix A/":

Thermal/

2D Thermal Model/

Macroscale/

Reference Model/ Tectonic Model/ Strain-softening Model/ Erosional Tectonic Model/ Erosional Strain-softening Model/

Mesoscale/

Tectonic Model/ Topographic Model/ Erosion Model/

The thermal model input file is reported in HTML format and is viewable with

any web browser. The Mesoscale and Macroscale model input files are .txt and are

viewable with Microsoft Notepad or similar program.

The file "STEEP_Thermal_4.mph" file may be opened with COMSOL™.

To run the model codes in FLAC3D:

Macroscale model code - call "run_coupled_ramp.txt" Mesoscale model code - call "STE_coupled_topo.txt"

175

Appendix B: Numerical Model Sensitivity Analysis

All plots within this section are filled contours of vertical uplift (Dz; m), with each

variable tested as indicated. The initial conditions kinematics, except for the kinematics

trials are given such that Vy = ½Vx.

Figure B.1. Sensitivity analysis for pressure-dependent rheology.

Figure B.1. Sensitivity analysis for pressure-dependent rheology. Trials for A) 'strong' (φ = 45°, C = 1000 MPa) and B) 'weak' (φ = 25°, C = 10 MPa) upper crust. These and all following plots are contours of vertical displacement (Dz; m). Note the scales are constant for all plots.

176

Figure B.2. Sensitivity analysis for C and φ.

Figure B.2. Sensitivity analysis for C and φ. Sensitivity analysis for C for trials of A) C = 1,000 MPa and B) C = 10 MPa and φ for trials with C) φ = 45° and D) φ = 25°.

177

Figure B.3. Sensitivity analysis for thermally-dependent rheology.

Figure B.3. Sensitivity analysis for thermally-dependent rheology. A) variation of the thermal model used to define the thermally-dependent rheology for the lower crust. The "mid" value was used for the models presented in the main text. B) Strength profiles for wet granite (Kohlstedt et al., 1995), diabase (Mackwell et al., 1998), and olivine (Kohlstedt and Goetze, 1974; Li et al., 2006) based upon the "mid" geothermal gradient from Figure B.3.

178

Figure B.4. Sensitivity analysis for thermally-dependent rheological variations.

Figure B.4. Sensitivity analysis for thermally-dependent rheological variations. Results of sensitivity analysis to variations to the lower crust rheology using the A) olivine and B) diabase flow law with the low thermal gradient and C) olivine and D) diabase flow laws with the mid thermal gradient. Trials run with the high geothermal gradient and 'wet granite' flow law did not converge.

179

Figure B.5. Sensitivity analysis for initial kinematics.

Figure B.5. Sensitivity analysis for initial kinematics. Results of sensitivity analysis to variations to the value of Vy, where A) Vy = 0, B) Vy = ½Vx, and C) Vy = Vx.

180

Figure B.6. Sensitivity analysis for discretization and absence of interface.

Figure B.6 Sensitivity analysis for discretization and absence of interface. Results of sensitivity analysis to variations to the model grid spacing using a A) coarser and B) finer grid spacing. C) Results of sensitivity analysis completed without the inclusion of an interface.

181

Figure B.7. Sensitivity analysis for interface properties.

Figure B.7 Sensitivity analysis for interface properties. Results of sensitivity analysis to variations to the interface properties by varying A) φ = 10°, B) φ = 30°, and C) tension = 20°. Additionally, D) slip is turned off.

182

Appendix C: Animations of Model Results

Animations referred to within the text are included within the attached CD.

The file structure for this Appendix is located within ":/Appendix C - Animations/":

Chapter 3/

Animation 3.1: Thermal evolution in 2D. Animation showing the thermal evolution due to underthrusting by a cold block of material. This animation is meant to illustrate the effects of thermal advection within a subduction zone.

Chapter 4/

The keystone image of each of these movies shows the assigned mechanical constitutive models. Blue - elastic; Red - thermally-dependent plastic flow law; and Green - pressure-dependent Mohr-Coulomb.

Animation 4.1: Principle Stress axes for the Reference Model. Animation showing the principle stress tensor plotted on a horizontal plane (z = -5000 m). The principle stresses are σ1 - red; σ2 - green, and σ3 - blue. Plots are taken at the 500 ky time step. Velocity vectors are included for reference.

Animation 4.2: Uplift patterns for the Reference Model. Animation showing the uplift patterns (Dz; meters) within the Reference Model. See Figure 4.4 for more information.

Animation 4.3: Principle Stress axes for the Tectonic Model. . Animation showing the principle stress tensor plotted on a horizontal plane (z = -5000 m). The principle stresses are σ1 - red; σ2 - green, and σ3 - blue. Plots are taken at the 500 ky time step. Velocity vectors are included for reference.

Animation 4.4: Uplift patterns for the Tectonic Model. Animation showing the uplift patterns (Dz; meters) within the Reference Model. Figure 4.4 for more information. See Figure 4.4 for more information.

183

Animation 4.5: Principle Stress axes for the Strain-Softening Model. . Animation showing the principle stress tensor plotted on a horizontal plane (z = -5000 m). The principle stresses are σ1 - red; σ2 - green, and σ3 - blue. Plots are taken at the 500 ky time step. Velocity vectors are included for reference.

Animation 4.6: Uplift patterns for the Strain-Softening Model. Animation showing the uplift patterns (Dz; meters) within the Reference Model. Figure 4.4 for more information.

Animation 4.3: Principle Stress axes for the Erosional Tectonic Model. . Animation showing the principle stress tensor plotted on a horizontal plane (z = -5000 m). The principle stresses are σ1 - red; σ2 - green, and σ3 - blue. Plots are taken at the 500 ky time step. Velocity vectors are included for reference.

Animation 4.8: Uplift patterns for the Erosional Tectonic Model. Animation showing the uplift patterns (Dz; meters) within the Reference Model. Figure 4.4 for more information.

Animation 4.9: Principle Stress axes for the Erosional Strain-Softening Model. . Animation showing the principle stress tensor plotted on a horizontal plane (z = -5000 m). The principle stresses are σ1 - red; σ2 - green, and σ3 - blue. Plots are taken at the 500 ky time step. Velocity vectors are included for reference.

Animation 4.10: Uplift patterns for the Erosional Strain-Softening Model. Animation showing the uplift patterns (Dz; meters) within the Reference Model. Figure 4.4 for more information.

Chapter 5/

The keystone image for these movies is a contour plot of the evolved topography at 500 ky. Compare with Figure 5.1

Animation 5.1: Principle Stress axes for the Tectonic Model. Animation showing the principle stress tensor plotted on a horizontal plane (z = -5000 m). The principle stresses are σ1 - red; σ2 - green, and σ3 - blue. Plots are taken at the 500 ky time step. Velocity vectors are included for reference.

184

Animation 5.2: Principle Stress axes for the Topographic Model. Animation showing the principle stress tensor plotted on a horizontal plane (z = -5000 m). The principle stresses are σ1 - red; σ2 - green, and σ3 - blue. Plots are taken at the 500 ky time step. Velocity vectors are included for reference.

Animation 5.3: Principle Stress axes for the Erosion Model. Animation showing the principle stress tensor plotted on a horizontal plane (z = -5000 m). The principle stresses are σ1 - red; σ2 - green, and σ3 - blue. Plots are taken at the 500 ky time step. Velocity vectors are included for reference.

185

Appendix D: Explanation of Variables

Model strain results are plotted in one of two ways: 1) the Cartesian incremental strain

rate components:

D.1

D.2

D.3

or 2) the rotation or vorticity in horizontal (D.4) and vertical (D.5 and D.6) planes:

D.4

D.5

D.6

where Di is the displacement in the i direction. Instantaneous strain and rotation rates can

be calculated by substituting velocities for displacements. The rotation components are

useful as they show the model of deformation. The horizontal component (Ωxy) shows

strain maxima associated with transform faulting, while the vertical components (Ωxz and

Ωyz) show strain associated with normal and thrust faulting.

186

The effects of vertical stresses are plotted using the vertical stress index function (VSI):

D.7 which shows the proximity to failure due to a vertical loading by topography.

187

Biography of the Author

Benjamin Hooks was born in Butler, Pennsylvania on September 12, 1978. He was raised in Dayton, Pennsylvania and graduated from Shannock Valley High School in

1997. He attended Allegheny College, Meadville, Pennsylvania, and graduated in 2001

with a Bachelor’s degree in Geology with departmental honors. He entered the

Department of Earth Sciences graduate program at The University of Maine in the fall of

2001. He graduated in 2003 with a Master's degree in Earth Science with a focus in

Igneous Petrology and Geochemistry. Benjamin is a candidate for the Doctor of

Philosophy degree in Earth Sciences from The University of Maine in August, 2009.

188