This item is only available as the following downloads:
1 PORE PRESSURE AND FLUID FLOW BENEATH THE FRONTAL THRUST OF THE KUMANO BASIN TRANSECT, JAPAN: INFLUENCE ON DCOLLEMENT PROPAGATION By KATHERINE ROWE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTERS OF SCIENCE UNIVERSITY OF FLORIDA 2011
2 2011 Katherine Rowe
3 To my brothers, who encouraged me to pursue a career in science and to my mother, who has supported and encouraged me through my education
4 ACKNOWLEDGMENTS First and foremost, I would like to thank my adviser, Dr. Elizabeth Screaton, for her unwavering patience, help, and support in learning Abaqus and wr iting this thesis. I would not have been able to do this without her guidance for two and a half years. I would also like the thank Dr. John Jaeger and Dr. Jon Martin, for their comments and encouragement through this process. Additionally, I would like to thank Dr. Andre Hpers for sharing his data to help the thesis come together. This research used samples and data provided by the Integrated Ocean Drilling Program (IODP). Funding for this research was provided by the Consortium of Ocean Leadership U.S. S cientist Support Program post cruise grant to Screaton and National Science Foundation grants OCE 0623358, OCE 0727023 and OCE 0751497. I want to thank the Department of Geological Sciences for the learning experiences and great times. For edits and comme n ts, I woul d like to thank Pati Spellman. I woul d like to thank Stephen Rowe for being supportive and listening about my modeling issues. For encouragement, I wou ld like to thank Annette Farah. Lastly, I woul d like to thank Alex Hastings for the encourageme nt, multiple proof reads, edits, and peanut butter and jelly sandwiches.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF ABBREVIATIONS ................................ ................................ ............................. 9 ABSTRACT ................................ ................................ ................................ ................... 11 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 13 2 GEOLOGIC AL BACKGROUND ................................ ................................ .............. 16 Nankai Trough ................................ ................................ ................................ ........ 16 Kumano Basin Transect ................................ ................................ .......................... 17 3 EQUATION S OF FLUID FLOW AND DEFORMATION ................................ .......... 29 Fluid Flow ................................ ................................ ................................ ............... 29 Elastic and Poroelastic Deformation ................................ ................................ ....... 30 Plastic Deformation ................................ ................................ ................................ 33 Friction ................................ ................................ ................................ .................... 34 4 PREVIOUS NANKAI INVESTIGATIONS ................................ ................................ 36 Previous Nankai Trough Modeling ................................ ................................ .......... 38 Previous Fully Coupled Subduction Zone Fluid and Deformation Models .............. 39 5 MODEL DESIGN ................................ ................................ ................................ .... 41 Modeling Approach ................................ ................................ ................................ 41 Modeling Setup ................................ ................................ ................................ ....... 41 Modeling Procedure ................................ ................................ ......................... 42 Modeling Param eters ................................ ................................ ....................... 43 6 RESULTS ................................ ................................ ................................ ............... 50 Permeability Results ................................ ................................ ............................... 50 Modeling Results ................................ ................................ ................................ .... 50 Base Model Results ................................ ................................ ......................... 51 Sensitivity Results ................................ ................................ ............................ 52
6 7 DISCUSSION ................................ ................................ ................................ ......... 61 Pore Pressure and Fluid Flow ................................ ................................ ................. 61 Dco llement Propagation ................................ ................................ ........................ 63 8 CONCLUSIONS ................................ ................................ ................................ ..... 65 LIST OF REFERENCES ................................ ................................ ............................... 66 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 73
7 LIST OF TABLES Table page 4 1 Log linear permeability porosity relationships of the Shikoku Basin clay ........... 40 5 1 The parameters used within the base model simulations with references ......... 47 6 1 Maximum simulated excess pore pre maximum fluid velocity (m/s). ................................ ................................ ............. 54
8 LIST OF FIGURES Figure page 2 1 A map depicting the Ashizuri, Muroto, and Kumano Basin transects ................. 21 2 2 The seismic profile of the Kumano Basin transect ................................ .............. 21 2 3 The frontal thrust regi on of the Kumano Basin transect ................................ ..... 22 2 4 The stratigraphic profile of Site C0006 ................................ ............................... 2 3 2 5 The stratigraphic profile for Site C0007 ................................ .............................. 24 2 6 The stratigraphic profile for Site C0007 ................................ .............................. 25 2 7 The stratigraphic profile of Site C0011 ................................ ............................... 26 2 8 The stratigraphic profile of Site C0012 ................................ ............................... 27 2 9 The porosity depth profiles from the incoming sediments of the Kumano Basin and Muroto transects. ................................ ................................ ............... 28 3 1 A representation of consolidation curve and the swelling curve ......................... 35 5 1 An illustration of the model setup. ................................ ................................ ....... 48 5 2 The temperature distribution used for viscosity calculations ............................... 49 6 1 The grain size grouping of the permeability of Sites C0006 and C0007 samples plotted as a function of porosity. ................................ ........................... 55 6 2 The Shikoku Basin facies permeability porosity trends. ................................ ..... 55 6 3 Simulated excess pore pressure through time. ................................ .................. 56 6 4 The excess pore pressure resu lts ................................ ................................ ....... 57 6 5 Cumulative horizontal displacement simulated with the base model. ................. 58 6 6 The fluid flow results ................................ ................................ ........................... 59 6 7 The excess pore pressure ratio resul ts for the sensitivity analyses .................... 60
9 LIST OF ABBREVIATION S a Compressibility (ms 2 /kg) B C c Compression Index (Unitless) C s Swelling Index (Unitles s) Slope (Unitless) E e Void Ratio (Unitless) Strain (Unitless) g Gravity (m/s 2 ) h Head (m) K Hydraulic Conductivity (m/s) k Permeability (m 2 ) Log Bulk Modulus (Unitless) Plastic Log Bulk Modulus (Unitless) Pore Pressure Ratio (Unitless) Viscosity (kg/ms) n Porosity (Unitless) v P Pore Pressure (Pa) Density (kg/m 3 ) R m Fluid Source Term (kg/m 3 s) S Storage (1/m) Stress (Pa)
10 T t Time (s) V Volume (m 3 ) v f Fluid Velocity (m/s) Coefficient of Friction (Unitless) z Depth (m)
11 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Masters of Science PORE PRESSURE AND FLUID FLOW BENEATH THE FRONTAL THRUST OF THE KUMANO BASIN TRANSECT, JAPAN: INFLUENCE ON DCOLLEMENT PROPAGATION By Katherine R owe August 2011 Chair: Elizabeth Screaton Major: Geology Results of a coupled fluid flow deformation model illustrate the influence of excess pore pressure, friction, and stress on dcollement propagation in the Kumano Basin transect offshore Japan. Numerical modeling suggests that displacement along the shallow frontal thrust generates excess pore pressure of approximately one third of the overburde n. The maximum excess pore pressure is located ~0.6 km in depth below the frontal thrust and about 7 km landward of the trench. The fluid flow patterns within the simulation depict the fluid flowing horizontally through the permeable Lower Shikoku Basin sa ndstone and vertically through the permeable trench sediments. Modeling results suggest that two previously proposed mechanisms may both contribute to dcollement propagation At 7 km landward of the trench, a large excess pore pressure occurs with a minim um in the effective stress. This result is consistent with previous suggestions that the dcollement propagates along a horizon with high pore pressure and low effective stress. In contrast, at the trench, a small excess pore pressure occurs beneath the tr ench, with no observable minimum in effective stress. Instead, comparison of horizontal and vertical effective stresses indicates that horizontal
12 compression is occurring seaward of the frontal thrust in the upper portion of the underthrust sediments, whi le vertical effective stress exceeds horizontal effective stress within the lower footwall sediments. This stress rotation is consistent with dcollement propagation controlled by gradients in horizontal compaction, as has been proposed due to lateral comp ression transmitted across the frontal thrust.
13 CHAPTER 1 INTRODUCTION Sediment rich subduction zones, such as the Nankai Margin off shore of the coast of Japan, experience repeated earthquakes of magnitude 8 or g reater (Ruff and Kanamori, 1983 ). The Nankai Trough Seismogenic Zone Experiment (NanTroSEIZE) project is focused on understanding the mechanics and seismic activity within the Kumano B asin t ransect along the Nankai Margin. This project is actively studying the aseismic to seismic transition, earthquake and tsunami generation, and the hydrologic behavior of plate boundaries (Tobin and Kinoshita, 2006 a, b). The first stage of drilling as part of the Integrated Ocean Drilling Program (IODP) included examination of the frontal thrust at the toe of the accretionary prism investigating its evolution, the relationship between deformation, fluid behavior, and slip, and its function in large ea rthquakes (Tobin and Kinoshita, 2006 a, b). Understanding the hydraulic behavior of the system can allow insight into deformation. Pore pressure and permeability play an important role in deformation at subduction zones, and thus are an important componen t of NanTroSEIZE. Permeability influences the direction and rate of the fluid flow through the region. As a result, it controls the drainage of pore pressure and the increase in effective stress (Saffer and Bekins 2002). Effective stress can affect fault strength, deformation, and structural development (Saffer 2003), and can alter the shape of the accretionary co mplex (Saffer and Bekins 2002). Characterizing the pore pressure buildup within fault zones can allow for better predictions of fault movement (Tobin and Saffer 2009). In addition, pore pressures may also help to control the propagation of the dcollement zone. The dcollement zone is the structural and mechanical boundary between the accreted and
14 subducted sediments (Moore, 1989). At greater dep ths, this plate boundary becomes the seismogenic zone. The dcollement is thought to originate along a layer of low strength, underconsolidation, and excess pore pressure (Moore 1989). Le Pichon and Henry (1992) suggest that the formation of the protodcollement occurs in the level of least mechanical resistance, which is believed to contain high pore pressures. In contrast, Raimbourg et al. (2011) argue that the thrusting of the prism create s horizontal compression seaward o f the frontal thrust, which controls the propagation of the dcollement. Prior to NanTroSEIZE, Nankai margin drilling focused on two transects, Muroto and Ashizuri, located to the west of the Kumano Basin transect. The Muroto transect is unique in that it has high heat flow seaward of the deformation front (Moore et al., 2001; Moore et al., 2009; Saffer, 2010). In contrast, the Ashizuri and Kumano Basin transects are considerably cooler. The Kumano Basin frontal t hrust exhibits a shallow dip angle (< 10) and has been thrust over ~6 km seaward on top of the trench sed iments. In contrast the Muroto and Ashizuri frontal thrusts of the Nankai prism have steeper dip angles (~25 30) and are only displaced by ~1 2 km (M oore et al., 2009) While all three transects contain sandy trench sediment, both the Ashizuri and Kumano Basin transect have buried sandy turbidites while the Muroto transect does not Underwood et al. (2009) suggested underthrust turbidites may enhance d ewatering and focus fluid flow from depth. Recent drilling results suggest that the buried sand layer may allow geochemical signals of flow from depth to reach 20 km seaward of the deformation front (Underwood et al., 2009).
15 Previous modeling investigation s have allowed for better understanding of the hydrological behavior and formation of the dcollement including identifying variables that influence hydraulic behavior, estima t ing excess pore pressure, and examining hydrological influence on fault strengt h (Saffer, 2010; Skarbek and Saffer, 2009; Gamage and Screaton, 2006; Screaton et al., 2002). The majority of investigations have been focused on the Muroto transect due to more information available there (Skarbek and Saffer, 2009; Gamage and Screaton, 20 06). One exception is a study by Saffer (2010) who compared the lithological influence on pore pressure and prism shape of the Ashizuri and Muroto transects. Saffer (2010) found that buried sandy turbidites in the Ashizuri transect resulted in lower exces s pore pressures than the Muroto transect. Although previous hydrogeological models of Muroto and Ashizuri transects have helped to assess pore pressures and effective stress conditions there, differences in geometry, heat flow, and lithology complicate th e transfer of results to the Kumano Basin transect. In this paper, results of laboratory tests of permeability are integrated into a finite element model of plastic deformation and fluid flow in the Kumano Basin transect. Modeling is a key tool for investi gating the de pths below that reached by drilling First, the modeling examines whether deformation occurs rapidly enough relative to fluid escape to generate significant excess pore pressures, as have been inferred for the Muroto and Ashizuri transects, or whether the buried sand layers result in rapid dewatering. Second, the modeling will help to understand the relationship between fluid pressure, deformation, and the propagation of the dcollement zone. The use of a fully coupled deformation and fluid flo w model allows assessment of both hydrologic and mechanical conditions that might influence dcollement propagation.
16 CHAPTER 2 GEOLOGICAL BACKGROUND Nankai Trough The Nankai Margin is one of the most studied subduction zones with earthquake and tsunami records going back over 1300 years (Ando 1975). T he Nankai Trough is formed by the subduction of the Philippines Sea Plate beneath the Eurasian Plate at 4 0 65 km per million years (Seno et al. 1993; Miyazaki and Heki, 2001 ). Prior to NanTroSEIZE, drilling and other investigations along the Ashizuri and Muroto transects have helped to characterize the Nankai Margin (Figure 2 1). The Shikoku B asin facies is deposited on the incoming plate, and trench sediments are deposited near the deformation front. The predominantly hemipelagic mud of the Shikoku Basin is separated into the Upper Shikoku Basin and the Lower Shikoku Basin facies The boundary between the lower and upper Shikoku Basin facies is a diagenetic transition that is related to breakdow n of opal cements (Spinelli et al., 2007). Another distinction between the upper and lower Shikoku Basin (LSB) facies is the occurrence of sandy turbidites within the LSB (Moore et al., 2001). The LSB sandstones are generally deposited in basement lows and absent from basement highs (Ike et al., 2008). There are notable topographical, thermal, and mechanical differences between the Muroto and Ashizuri transects. The Muroto transect is located on top of a topographic high, which prevented the deposition of the LSB sandstones that are seen in the Ashizuri transect. Due to the Muroto transect being located near an extensional ridge that ceased spreading 15 Ma, it contains a higher heat flow and observed temperatures than the Ashizuri transect (Moore et al., 20 01). In addition, a rapid accretionary outgrowth of 20km/my during the past 2 My is observed in the slope of the Muroto. The
17 rapid growth is suggested to be caused by a seamount that was subducted ~2 Ma, creating a large embayment. The rapid growth reflect s the prism rapidly trying to recover the frontal accretion across over the embayment area (Moore et al., 2001). Such rapid growth is not observed in the Ashizuri transect. In both transects, the dcollement zone near the deformation front is localized i n the Lower Shikoku Basin facies within sediments with ages of ~5.9 7 .0 Ma (Moore et al., 2001). This lies above the LSB sandstones of the Ashizuri transect. The proto dcollement, the extension of the d collement beneath the frontal thrust has been esti mated to correlate to a depth of ~400mbsf in the incoming sediments (prior to dep osition of trench deposits) of the Lower Shikoku Basin facies, in both the Muroto and Ashizuri transects ( Shipboard Scientific Party 2000; Shipboard Scientific Part y 2001; Moore et al. 2005). Kumano Basin Transect The frontal thrust and incoming sediments of the Kumano Basin t ransect offshore the Kii Peninsula, Japan were drilled during IODP Expeditions 314 and 316 in 2007 2008 and Expedition 322 in 2009 (Figures 2 1 and 2 2) In this transect the frontal thrust of the accretionary prism has moved the prism more than6 km seaward on top of the trench sed iments at a shallow dip angle, < 10 (Moore et al. 2009). In contrast the Muroto and Ashizuri frontal thrusts of the Nankai prism have steeper dip angles (~25 30) and are only displaced by ~1 2 km. This unusual geometry of the Kumano Basin transect suggests that normal frontal imbrication was interrupted, perhaps due to a variation in incoming basement structure sediment thickness or lithology (Moore et al. 2009). Development of a protothrust zone (PTZ) seaward of the frontal thrust indicates that frontal imbrication may be resuming (Moore et al. 2009) (Figure 2 3) Screaton et
18 al. (2009) us ed seismic reflect ion profiles and data derived from IODP Expedition 316 cores to estimate that the earliest initiation of the frontal thrust fault occurred about 0.78 Ma 0.436 Ma. During IODP Expedition 316, Sites C0006 and C0007 were drilled to examine the main frontal t hrust region located at the toe of the accretionary prism and to investigate the relationship between slip, deformation, and fluid flow (Screaton et al., 2009) ( Figure s 2 2 and 2 3 ) Drilling during Expedition 316 reached ~ 603 mbsf for Site C0006 and ~484 mbsf for C0007 (Expedition 316 Scientist, 2009a and b). Greater depths ~885 m below seafloor, were reached at Site C0006 during Expedition 314, in which logging while drilling was conducted but no cores were collected (Expedition 314 Scientists, 2009). Te mperatures in the drilled section are predicted to be < 40C with a 27C/km gradient at Site C0006 and a 42C/km gradient at Site C0007, based on extrapolation of borehole temperature measurements (Figure 2 2) Using nannofossils the ages from Sites C0006 a nd C0007 ranged from less than 1 Ma near the surface to greater than 5 Ma with depth (Expedition 316 Scientists, 2009a and b). Drilling that occurred in Site C0006 and did not reach the underthrust sediments due to drilling conditions. Unlike Site C0006, d rilling at Site C0007 went through the slope, prism, and underthrust sediments. The general lithology between the two sites is similar and can be separated into four units. Unit I, the slope sediments, contain s fining upward successions of silty clay, sand, sand silt, and rare volcanic ash layers. Unit II, the accreted trench wedge is a succession of coarsening upward clay to sand and gravel. Unit III, the Shikoku Basin facies, is composed of mudstone with volcanic ash layers. Lastly, Unit IV, the underthrust trench wedge, is primarily composed of sand that
19 is presumably young in age, although poor recovery prevented dating (Expedition 316 Scientists 2009 a, b) (Figure s 2 4 2 5, and 2 6 ). Expedition 322 drilled into the incoming sediments including the basement, at Sites C0011 and C0012 (Figures 2 2, 2 7 and 2 8 ). The drilling at Site C0011 was planned to drill into the basalt, but due to complications and time shortage, the basement was not reached. The drilling started at ~340 mbsf, not coring the upper Shikoku Basin facies and the drilling stopped at 950 mbsf. Expedition 333, conducted in December 2010 through January 2011, returned to Sites C0011 and C0012 to complete the record at these sites and collect temperature data. The report for this expedition has not yet been released. Site C0012 is located on a basement high, known as the Kashinosaki Knoll (Figure 2 2) Because of the location, all the units within Site C0012 are thinner than those of Site C0011 and the basement was able t o be drilled at 537 mbsf. Unit I, the upper Shikoku Basin facies was not cored and the description is inferred from ODP Sites 808, 1173, 1174, and 1177. It is composed of hemipelagic mud with interbedded volcanic ash. Unit II, the middle Shikoku Basin faci es is Miocene in age and ranges from silty claystone with interbedded sandstone to silty claystone, volcaniclastic sandstone and siltstone. Unit III, the lower Shikoku is middle to late Miocene in age and is mainly composed of silty claystone. Unit IV, the lower Shikoku: turbidites are middle Miocene in age and are composed of silty claystone with interbedded clayey siltstone and siliciclastic sandstone. These LSB sandstones are similar to those described at the Ashizuri transect. Due to its location Site C0011 contains a greater number of sand layers from gravity flows and turbidity currents than Site C0012. Unit V, the
20 volcaniclastic rich facies ranges in the middle Miocene and is primarily compose d of silty claystone and tuff wit h sandy s iltstone Unit VI, which was only recovered in Site C0012, is a pelagic clay facies from the early Miocene in age (Underwood et al. 2009). Pore f luid from samples obtained from Site C0011 greater than 600 mbsf, which is in the LSB, had chloride ( Cl ) concentrations that were 7% lower than seawater concentration. Underwood et al. (2009) suggested this fluid found at Site C0011 could originate from clay dehydration from depth and the fluid arrived a t Site C0011 through permeable layer in the Lower S hikoku Basin sandstones Also, hydrocarbon concentrations, such as propane, were found to i ncrease with depth at Site C0011 (Figure 2 2) Unlike Site C0011, the Cl concentration at Site C0012 increases by 12% with no indications of freshening, and propane was absent throughout the core. The lack of propane and Cl freshening signal suggests the influence of fluids from depth did not reach Site C0012 (Underwood et al., 2009). Low chloride concentrations have been observed within the LSB at Site 1177 seaward o f the deformation front along the Ashizuri transect. P orosity depth profile s are available from the drilling results of both Site s C0011 and C0012. Porosity is an important indication of the compaction state of sediments. The incoming sediments contain a decreasing trend of porosity with depth, which is typical of sediment consolidation (Underwood et al., 2009). When comparing the porosity depth profile of the lower Shikoku Basin facies from the Kumano Basin transect to the Muroto transect the profiles contain a similar trend in consolidation (Figure 2 9).
21 Figure 2 1. A map depicting the Ashizuri, Muroto, and Kumano Basin transects Figure 2 2. The seismic profile of the Kumano Basin transect ( modified from Underwood et al. 2009 and based on the interpretations of Moore et al., 2009 ). Sites C0006, C0007, C0011, and C0012 are the primary focus within this investigation
22 Figure 2 3. The frontal thrust region of the Kumano Basin transect, with the inferred protodcollement (in red) and the locatio ns of profiles at the trench and 7 km from the trench (in yellow) (m odified from Moore et al., 2009)
23 Figure 2 4. The stratigraphic profile of Site C0006 (Expedition 316 Scientists, 2009a).
24 Figure 2 5. The stratigraphic profile for Site C0007 (Expedition 316 Scientists, 2009b).
25 Figure 2 6. The stratigraphic profile for Site C0007 (Expedition 316 Scientists, 2009b).
26 Figure 2 7. The stratigraphic profile of Site C0011 (Underwood et al., 2009)
27 Figure 2 8. The stratigraphic profile of Site C0 012 (Underwood et al., 2009).
28 Figure 2 9. The porosity depth profile s from the incoming sediments of the Kumano Basin (Expedition 322) and Muroto (Leg 190) transects. 300 400 500 600 700 800 900 1000 0 0.2 0.4 0.6 0.8 1 Depth below seafloor (m) Porosity POROSITY DEPTH PROFILE Expedition 322 Site C0011 Leg 190 Site 1173
29 CHAPTER 3 EQUATIONS OF FLUID FLOW AND DEFORMATION Fluid Flow Fluid flow is gove fluid and sediment properties, hydraulic head, and fluid velocity: v f = ( 3 1) where v f (m/s) is the velocity of fluid flow, k (m 2 f (kg/m 3 ) is fluid density, g (m/s 2 ) is gravity, n e the fluid, and h (m) is hydraulic head ( Domenico and Schwartz, 1998 ). Hydraulic head can be expressed in terms of pore pressure (P): h= ( 3 2) where z (m) is the elevation head and P (Pa) is the pore pressure The pressures with the conservation of mass equation (Ingebritsen et al., 2006; modified by Screaton, 2010): ( 3 3) where t (s) is time, n is porosity, and R m (kg/m 3 s) is a fluid source term.
30 Permeability describes the ease of fluid flow through a porous material and can vary by orders of magnitude for different sediment types A relationship between the logarithm of permeabi lity and porosity of clays was developed by Bryant et al. (1975) and Neuzil (1994): ( 3 4) Where log (k 0 is the slope of the porosity permeability relationship. Hydraulic conductivity (K) (m/s) combines the fluid properties and the permeability: ( 3 5) Temperature can affect fluid flow by altering fluid viscosity (Voss and Provost, 2010): ( 3 6) Elastic and Poroelastic Deformation In a deforming porous material, fluid pressures and flow are closely related to strain and stress. A linear elastic strain is proportional to the applied stress and will
31 revert back to the original state once the stress is relieved T he relationship between stress and strain for two dimensions can be expressed as: ( 3 7) ( 3 8) xx zz (Pa) is the stress in the Modulus. E is the incompressibility or stiffness of the rock, while is the ratio of compression and thickening (Ingebritsen et al., 200 6). In porous materials, pore pressure cannot be ignored Pore pressure is related to stress through the effective stress principal in the poroelasticity theory. The change in related to the change in pore pressure and the change in th e effective e ): ( 3 9) where a s is the compressibility of the mineral grains (ms 2 /kg) and a b is the bulk compressibility of the porous medium: ( 3 1 0 )
32 ( 3 1 1 ) where V is the total volume and V s is the volume of the solids (Ingebritsen et al., 2006; Screaton, 2010). Bulk compressibility (a b Ratio are related through a linear relationship (Wang, 2000): ( 3 12) If the bulk compressibility of the porous medium is significantly greater than the bulk compressibility of the grains (a b > > a s ), Equation 9 simplifies to: ( 3 1 3 ) To tie everything together, fluid flow in terms of pore pressures is coupled with def ormation in terms of effective stress : ( 3 1 4 ) where S s3 (1/m) is the specific storage in three Coef ficient: ( 3 1 5 )
33 ( 3 1 6 ) where n (dimensionless) is the porosity and a f (ms 2 /kg) is the compressibility of the fluid (Screaton 2010). Plastic Deformation For large strains, most sediment s are plastic rather than elastic, and will not return to their original undeformed shape once the applied stress is removed. To characterize the stress and strain behaviors of soils, Roscoe and Burland (1968) and Schofield and Wroth (1968) developed the Cam Clay and Modified Cam Clay models. The Modified Cam Clay model is based on the ability to predict volume changes of a material due to loading, or the plasticity theory. The Modified Cam Clay model is designed to simulate the behavior of soils by predicting the pressure dependent soil strength and the change in volume and compression caused by applying a shear stress ( s ) or stress tangential or parallel to the surface The m odified Cam Clay model allows calculat ion of the deformation or compaction of a solid during loading (Helwany, 2007). To find the Cam Clay parameters for a material a triaxial test or a one dimensional consolidation test can be performed on a sample and results plotted along an e v ) plot (Fig ure 3 1) For one dimensional consolidation, v ) is the log of the effective vertical stress and e is the void ratio (e= )). The break in the slope on Figure 3 1 is the effective stress above which the material begins to deform plastically. The consolidation curves are used to find the compression and swelling index from the slope. The compression index (Cc) is the slope of the loading curve,
34 while the swelling index (Cs) is the slope of t he unloading curve that reflects the elastic rebound when the effective stress is reduced (Figure 3 1 ). As an alternative to laboratory experiments, field data can be used to relate observed void ratio and effective stresses to find the compression index. The Cam Clay parameters, are related to Cc and C s (Helwany, 2007): ( 3 17 ) ( 3 1 8) Friction Slip along a fault occurs when the shear stress (the stress that is parallel o r tangential to a surface ) is greater than the frictional strength. Coulomb failure relationship describes the failure under stress conditions, and describes the slip tendency: ( 3 19) where the c is the tendency for slip, n is the c is positive, there is an c is negative, there is a decrease in the tendency for the fault to slip (King et al., 1994; Masterlark and Wang, 2000; Hughes et al., 2010 ).
35 Figure 3 1 A representation of consolidation curve and the swelling curve ( modified from Helwany, 2007).
36 CHAPTER 4 PREVIOUS NANKAI INVESTIGATIONS Based on data and samples from the Muroto transect excess po re pressures have been estimated (Screaton et al., 2002; Saffer, 2003) and the permeability porosity relationship of the Shikoku Basin Clay calculated (Saffer and Bekins, 1998; Gamage and Screaton, 2006; Skarbek and Saffer, 2009). Screaton et al. (2002) us ed the measured porosity along the Muroto transect to estimate the excess pore pressure within the proto underthrust sediments (Site 1174) and underthrust sediments (Site 808). Assuming hydrostatic pore pressure (P L e ) within Site 1173, they created a po rosity versus effective stress relationship. Using this relationship as a reference, they compared it to the porosity versus the excess lithostatic stress at Sites 808 and 1174, to calculate the pore pressures at these two sites. Screaton et al. (2002 ) cal culated the as the excess pore pressure (P*) divided by the excess lithostatic pressure (P L *) : ( 4 1 ) T he lithostatic pressure (P L b z) can be described as the pressure of the overlying body, and P H is the hydrostatic pressure (Screaton et al., 2002). For Site 1174, Screaton et al (2002) These results suggest underconsolidation of the sediments, potentially due to rapid loading of the low permeability sediments.
37 Saffer (2003) used a similar method but included laboratory consolidation experiments, logging while drilling (LWD), and void ratio reduction to estimate pore pressure development within underthrust sediments at S ites 1173 and 808. His results were similar to but slightly lower than the results of Screaton et al. (2002), yielding a Site 1174 of ~0.34 and Site 808 of ~0.44. The permeability porosity relationship (Equation 16) has been estimated based on inverse numerical modeling (Saffer and Bekins, 1998) and laboratory tests on Muroto and Ashizuri core samples (Gamage and Screaton, 2006; Skarbek and Saffer, 2009; Saffer 2010 ). Saffer and Bekins (1998) calculated the first permeability porosity relationship for the Muroto transect ba sed on inverse modeling (Table 4 1 ). Gamage and Screaton (2006) used a constant flow method to calculate the vertical permeability and create a p ermeability porosity relationship. The Gamage and Screaton (2006) relationship contains the highest permeability intercept. Skarbek and Saffer (2009) and Saffer (2010) created a log linear relationship between the porosity and permeability measured during constant rate of strain uniaxial consolidation testing (CRS).Saffer (2010) and Skarbek and Saffer (2009) found their relationship was steeper but consistent with Gamage and Screaton (2006) (Table 4 1). General patterns of subduction zone permeability poros ity relationships w ere investigated f urther by Gamage et al. (201 1 ) by c ompiling the available permeability porosity data from s iliciclastic samples from Barbados, Nankai, Peru, and Costa Rica; they found that grain size distribution is the most important control on the permeability porosity trend. Their results illustrated a general trend of increase in permeability with a
38 decrease in clay size particles at any given porosity, despite different locations and thermal conditions. Previous Nankai Trough Model ing The evolution of pore pressure within a subduction zone and the influence of other parameters, such as morphology of the prism, potential heat transport, chemical transport, fault strength, and sediment strength have been investigated on several conver gent margins (Saffer and Bekins, 1998; Stauffer and Bekins, 2001; Saffer and Bekins, 2002; Saffer, 2003; Saffer and Bekins, 2006; Saffer, 2007). Specifically, modeling of the pore pressure and fluid flow within the Nankai margin has been performed by multi ple scientists (Gamage and Screaton, 2006; Skarbek and Saffer, 2009; Saffer, 2010). Gamage and Screaton (2006) used laboratory permeability data to investigate porosity and excess pore pressure in the Muroto Transect They incorporat e one dimensional mode ling of loading into a fluid flow model by increasing the pore pressure by the added weight of effective overburden. Gamage and Screaton (2006) simulated the previous find ings of Screaton et al (2002) and Saffer (2003). If excess pore pressures are as high as Screaton et al. (2002) and Saffer (2003) observed beneath the frontal thrust then the permeabilities must be lower due to a low permeability barrier above the dcoll ement, a lower bulk permeability, or both (Gamage and Screaton 2006). Skarbek and Saffer (2009) used laboratory permeability measurements and one dimensional model of loading to investigate the evolution of pore pressure within the underthrust sediments o f the Muroto transect from shallow to great depth. Their
39 approach is similar to the approach of Gamage and Screaton (2006), except Skarbek an average pore pressure ratio, 0.60 along the dcollement up to 20 km landward of the trench. Saffer (2010) used a two dimensional model to compare and understand the fluid flow of the Muroto and Ashizuri transect. He simulated compaction by computing the fluid released from compaction and dehydration for a reference frame fixed at the deformation front. He found that the lithostratigraphy influences the mechanical behavior of subduction zones by affecting the distri bution and magnitude of excess pore pressure. He calculated the average basal pore pressure ratio for 30 km. The modeling yielded a of ~ 0 59 for the Muroto transect and a of~0.38 for the Ashizuri transect (estimated from Figure 8 in Saffer, 2010) T he underthrust sediments were found to dewater in the Ashizuri transect through the LSB sandstone This dewatering resulted in lower excess pore pressure ratios, by ~15%, in the Ashizuri transect as compare d to the Muroto transect, which lacks the buried L SB sandstone Previous Fully Coupled Subduction Zone Fluid and Deformation Models Due to the data needs and difficulty, fully coupled deformation fluid flow models are less commonly applied than the partially coupled models discussed above. Finite Element Models (FEMs) have been used to understand the influence of fully coupled deformation and fluid flow (Hughes et al., 2010; Masterlark and Hughes, 2008). Hughes et al. (2010) illustrated the importance of pore pressures and elastic behavior within the subdu ction zone off the coast of Sumatra using a FEM Abaqus Stauffer and Bekins (2001) investigated mechanisms that prevented consolidation and maintained high permeability along the dcollement of the northern Barbados accretionary complex.
40 Using a coupled f luid flow/consolidation model that included the modified Cam Clay relationship, they found high pore pressures migrated seaward of the defor mation front, consistent with the underconsolidation inferred from high porosities in sediment cores. The simulation s depicted minor swelling in the incoming sediments up to 3 km before subduction, which could explain shear and faulting proximal to the protodcollement (Stauffer and Bekins, 2001 ). Table 4 1. Log linear permeability porosity relationships of the Shikoku Basin clay estimated from core samples from the Muroto transect or inverse modeling. Reference Relationship Saffer (2010) log (k)= 20.450.5+6.93n Skarbek and Saffer (2009) log (k)= 20.45+6.93n Gamage and Screaton (2006) log (k)= 19.82+5.39n Saffer and Bekins (1998) log (k)= 20+5.5n
41 CHAPTER 5 MODEL DESIGN M odeling Approach We use the FEM package Abaqus ( www.simulia.com ) which includes coupled fluid flow and deformation for soil mechanics. With Abaqus, the design of a model can be in two or three dimensions, with no restrictions on the overall shape. Displacement of the nodes allows the calculation of stress and strain, which affects the other variables within the element (Abaqus 6.7 2007 ). We include the influence of lithos tratigraphy of the region by using permeability porosity data from samples of Expeditions 316 and 322 and literature, consolidation parameters from previous investigations of the Muroto transect and the observed porosity data of the incoming sediments, and a thermal profile to simulate the effects of temperature dependent viscosity on hydraulic conductivity. Modeling does not incorporate the effects of dehydration reactions, such as the smectite illite transition This was excluded due to the findings of Sa ffer and Bekins (1998), who found minimal influence the dehydration on fluid pressure for the Muroto transect and Bekins et al. (1995) who found that the dehydration driven fluid sources are small when compared to the compaction driven sources for simulati on of the Barbados subduction zone Modeling Setup Our model is composed of the prism, the trench sediments, the Shikoku Basin Clay, the Lower Shikoku Basin (LSB) sandstone, and basalt layer. The entire modeled region is 73 km long, 23 km landward from the trench (prior to displacement) and 50 km seaward f rom the trench. The prism is ~5 km thick at the most landward location. The incoming Shikoku Basin facies is ~1.4 km thick (after compaction), and includes the
42 LSB sandstone that is simulated as ~0.4 km thick. The trench sediment contains a maximum thickness of 0.5 km extending from ~7km beneath the prism to 10 km seaward of the prism. Due to the difficulties of the coupled modeling, some simplifications were made to the geometry. In reality, the prism does not start on top of the trench sediments. F urthermore, the trench sediments are deposited as the subduction is occurring. Lastly, the trench sediments are about half the observed thickness of 1 km. T he upper 3 km of basalt is included in the model as its strength was found to be important to the de f ormation of the region (Figure 5 1). The mesh is composed of 4218 quadrilateral elements and 4722 nodes. The frontal thrust is represented as an interaction between two parts, the prism and the footwall (including the trench turbidites, the Shikoku Basin clay, and the LSB sandstone), and is assigned a coefficient of friction. Hydrostatic pore pressure is applied to the top of the model to represent the bottom of the ocean. We also apply a hydrostatic boundary along the right hand side of the model. The bo ttom of the model is constrained to have no vertical movement, while the left and right side of the footwall are constrained to have no horizontal movement. To test the sensitivity of the hydrostatic boundary on the seaward side of the footwall, we removed the boundary condition creating a no flow boundary. A simulation with the seaward footwall able to move horizontally was created to see the sensitivity of the no movement boundary. Each resulted in minimal influence (~1%) on simulated excess pore pressure s. Modeling Procedure Due to temperature influencing the fluid flow viscosity, an initial model simulation of thermal conduction was run to develop a distribution of temperatures throughout the
43 model region. The top of the model is set to 2C, based on the temperature data of seawater from Expedition Scientists (2009a, b). With the input of heat flow, based on measured temperatures and inferences from bottom simulating reflectors (Kinoshita et al., 2008), and average thermal conductivity for each of the mod el lithologies (Kinoshita et al., 2008; Expedition 316 Scientists, 2009a and b), Abaqus calculated the temperatures based on thermal conduction (Figure 5 2). Prior to movement of the frontal thrust, gravity was applied to each part of the model as a body f orce, based on the acceleration due to gravity (g=9.81 m/s) and the average effective density of the rock (bulk density of the rock minus the density of salt water) in each of the model lithologies. The body force was applied gradually over 4 million years to create starting conditions for frontal thrust movement. Following the initial equilibration with the body force, we displaced the left side of the accretionary prism 7.2 km to generate a toe movement of about 6 km after prism shortening. The displaceme nt was applied for the duration of 0.61My, which is the middle of the range of age estimates for frontal thrust initiation (Screaton et al., 2009). Because the actual displacement history is not known, we assume a constant frontal thrust slip velocity thro ughout the model. As the prism moves seaward, a pressure is applied to the exposed region of the footwall to compensate for the removed body of the prism Modeling Parameters The five specific regions of the model are assigned mechanical and fluid flow par ameters. Each part of the model has a unique permeability porosity relationship. Basalt permeability varies with depth and ranges vastly. In previous studies at other locations, the regional scale permeability for the upper basement aquifer, which is compo sed of sheet flows and pillow basalt, has been constrained by pressure transient
44 borehole test s and the thermal state of the basalt to range from 1 x 10 19 to 1 x 10 10 m 2 (Becker and Davis, 2004; Davis et al., 2000; Davis and Becker, 2002; Fisher 2005 ). Kummer and Spinelli (2009) note variations of permeability with depth and the lack of knowledge concerning how lateral permeability changes during subduction. Underwood et al. (2009) found evidence for circulation in upper basement at C0012 but note that i t is hydrologically isolated from the sediments. To prevent high rates of drainage of the sediments into the basalt, contrary to the Site C0012 observations, we assign a permeability value of 1 x 10 17 m 2 The permeability porosity relationships of the pr ism and the Shikoku Basin clays are taken from the data results of Sites C0006 and C0007 and previous permeability results presented below. To obtain permeability porosity data for the Kumano Basin transect, we measured the vertical permeability and porosi ty of nine core samples from Sites C0006 and C0007 using the flow through system described by Gamage and Screaton (2003). Detailed methods and results of the laboratory tests are provide by Rowe et al. (in review). F luid f low was driven by generating a kno wn difference in pressure at the top and bottom of the sample and flow rate was measured Using then converted into permeability based on fluid properties at laboratory tempe rature Using porosity data from Expedition 316, the change in porosity was calculated at each effective stress, which is based on the change in fluid volume in the cell. We use the permeability porosity relationship of the LSB sandstone measured by Hper s et al., (in review) log (k) = 6.6108 n 17.71 The trench sediments are simulated to be a combination of sand, gravel, and clay. To develop a permeability porosity
45 relationship for this layer, we combined the known clay relationship, with generic sand permeability (Schwartz and Zhang, 2003). Heterogeneous vertical hydraulic between two lithological layers is calculated by dividing the total thickness by the sum of t he each thickness layer divided by the hydraulic conductivity: ( 5 1 ) Due to the trench sediments being sandy to clayey sandy, w e estimated half th e trench to be clay and half to be sand, and developed a permeability porosity relationship for the trench sediment yielding log (k) = 19.444 + 6.8851 n. The dcollement and frontal thrust are not assign ed a specific permeability because of its location a nd the shallow design of our model. The frontal thrust abuts the sand rich trench sediments This sand rich layer provides a channel in which fluid can escape from beneath the prism. Thus assigning high permeability to the fault zone is unnecessary for our shallow region. or stiffness and the based on data c ompiled by Turcotte and Schubert (1982) w ho ranged from 0.1 0.4, while Christensen (1996) depicted the range of bulk representation of crust to range from 0.25 he model. The prism is treated as elastically based on the conclusion of Moore et al. (2009) who suggested that the prism has come up from depth. Thus the applied effective stresses are expected to be less than those previously experienced through burial, and the deformation will be elastic. In addition, the focus in our simulations is on the footwall
46 rather than the prism because the excess pore pressure primarily occurs within the footwall in previous investigations (Screaton et al., 2002 and Saffer, 2003) We used E=5 x 10 8 Pa for the prism. In Abaqus, during the displacement of the prism, if the prism is too stiff, the model fails. can be applied without the model failing. This stiffness is a little high for clay but is representative of sandy gravel based on the compressibility found in Domenico and Schwartz (1998 ) The basalt is treated as elastic. Based on the literature values of compressibility given in Domenico and Schwartz, (1998), we apply an E of 5 x 10 11 Pa for the basalt. We apply the Cam Clay parameters to the Shikoku Basin clay, LSB sandstone, and trenc h sediments. The compression tests performed on Muroto transect samples (Bellew, 2004). Results from samples of the Ku mano Basin transect are not yet available. Despite the thermal differences between the Muroto and Kumano Basin transects, the similarities between the porosity depth profiles of the incoming sediments suggest that they share a similar consolidation behavio r (Figure 2 9). The appropriateness of the laboratory data was checked by comparing the simulated porosities of the incoming sediments to those measured at Site C0011. The value derived from the laboratory test underestimated compaction. To better match th e field data, the value was increased to 0.3. Cam Clay parameters of th e sand rich sediments are not available. For trench sediments, we applied the Cam clay parameters of the Shikoku Basin c lay to obtain the maximum influence of compaction of interbedded clay layers. For the LSB sandstone,
47 we used the deeper LSB clay parameters from Muroto samples (Hpers et al., 2010 ) to obtain maximum compaction results (Table 5 1 ). A is necessary for the contact between the accretionary hangi ng and footwall of the frontal thrust. Ikari et al. (2007) studied the effects of hydration on frictional properties of montmorillonite quartz fault gouge material. They found their coefficient of friction values were in agreement with Saffer and Marone (2 003) values for 100% and 50% montmorillonite. Dependent on the amount of stress, montmorillonite and 5 MPa) to 0.12 (with 100% montmorillonite and 100 MPa). We used region and was judged to be unrealistic. Table 5 1. The parameters used within the base model simulations with references Part E Reference v Reference Corrected Reference Prism 5 x 10 8 0.3 Masterlark (2003) Trench 0.3 Masterlark (2003) 0.117 0.3 0.014 Bellew (2004); Hpers et al. (2010) Shikoku Basin Clay 0.3 Masterlark (2003) 0.117 0.3 0.014 Bellew (2004); Hpers et al. (2010) LSB Sandstone 0.3 Masterlark (2003) 0.78 0.02 Bellew (2004); Hpers et al. (2010) Basalt 1 x 10 11 Domenico and Schwartz (1998) 0.3 Masterlark (2003)
4 8 Figure 5 1. An illustration of the model setup A) The left side represents landward and the frontal thrust fault is highlighted in red. B) The mechanical boundary co nditions applied to the model C) The hydrostatic and no flow boun dary conditions for the model D) The m odel after it has been meshed
49 Figure 5 2. The temperature distribution used for viscosity calculations 5 km Temperature
50 CHAPTER 6 RESULTS Permeability Results Permeability tests were run on a total of nine samples from the toe region. All were from shallow depth (< 0.6 km beneath the seafloor). The porosities of Sites C0006 and C0007 samples ranged from ~0.2 to 0.4 and ~0.3 to 0.5, respectively. The measured pe rmeabilities at Site C0006 span from 3 x 10 19 to 4 x 10 17 m 2 and Site C0007 permeabilities ranged from 2 x 10 18 to 9 x 10 17 m 2 (Figure 6 1). The lithology of Sites C0006 and C0007 varies with depth, ranging from sandy slope sediments to hemipelagic mud. Grain size analysis performed by Kopf et al. (in review) allows the quantitative separation of the samples. The grouping is based on the grouping of Bryant (2002) and also used by Gamage et al (201 1 ). Group 1 includes sediments with greater than 80% clay size particles. Group 2 includes sediments with 60 80% clay size material. Group 3 includes the sediments with less than 60% clay sized material and less than 5% sand. Group 4 includes the sediments that are less than 60% clay sized material and great er than 5% sand (Gamage et al., 2011) Three measured samples were from the Shikoku Basin clay. The permeability porosity results of the Shikoku Basin clay samples from Kumano appear consistent with the permeability porosity trends of the Shikoku Basin cla y based on samples fr om the Muroto transect (Figure 6 2 ). The relationship of Skarbek and Saffer (2009) was used for the base model simulation Modeling Results Initial model runs were used to determine the appropriate Cam Clay parameters for the Shikoku B asin clay by comparing simulated porosity depth profiles 20 km
51 seaward of the trench to the observed profile from Site C0011. A of 0.3 was determined to provide a reasonable match to the observed data. Base Model Results As the hanging wall is displaced over the footwall, excess pore pressures were generated beneath the toe of the accretionary prism and migrat ed horizontally with the prism. A plot of excess pore pressure through time within the Shikoku Basin clay in the footwall, illustrates the gradual increase in pore pressure due t o the overriding prism (Figure 6 3 ). Superimposed on the gradual increase, slight fluctuations with lower and higher pressure occur during the disp lacement of the prism (Figure 6 4 ). Though a constant velocity is applied to t he frontal thrust slip, the coefficient of friction influences the ease of slip. With a constant coefficient of friction, variations in pore pressures can alter the effective stress and cause intermittent slip of the frontal thrust. Looking at the final m odel results, which represent the present time, the base model shows that excess pore pressure forms a horizontally elliptical shape beneath the frontal thrust within the Shikoku Basin clay and ab ove the LSB sandstone (Figure 6 4 ). The model results in an of 0.34.The maximum pore pressures are generated landward of the trench and do not appear seaward of the trench. Excess pore pressure and effective stress with depth profile were created at the trench and 7 km landward of the trench. At the trench, the maximum excess pore pressure occurs around ~0.7 5 km in depth from the top of the footwall. The effective stress at the trench continually increases with depth (Figure 6 4 ). At 7 km landward of the trench, the maximum EPP is located ~0. 6 5 km below from the top of the footwall.
52 The stress profile reflects a minimum in effective stress ~0. 6 km below the top of the footwall (Figure 6 4 ). At the trench, the effective horizontal stress is greater than th e effective vertical str ess until 0.8 km depth (Figure 6 4 ). At 7km landward of the trench, the vertical effective stress is consistently larger than the effective horizontal stress. There is a localized zone of horizontal compression located near the tren ch. The horizontal displacement illustrates that the thrust of the prism along the frontal thrust is felt within the footwall se diments at the trench (Figure 6 5 ). In general, the fluid flow in the Shikoku Basin clay beneath the thrust fault is diverted up ward towards the trench sediment or downwards towa rds the LSB sandstone (Figure 6 6 ). The highest flow rates occur horizontally along the buried LSB sandstone. The maximum fluid flow velocity in the base model was 2.3 x 10 10 m/s within the LSB sandstone. Sensitivity Results With permeability being a key parameter in the investigation, the prism, trench sediment, LSB sandstone, and Shikoku Clay permeabilities were altered in the sensitivity runs. To represent the multiple faults within the prism, we increas ed the permeability of the prism sediments. Increasing permeability within the prism had very minor e ffect on the pore pressure ratio and fluid flow within the footwall. To test sensitivity to the permeability of the LSB sandstone, we decreased the permeab ility an order of a magnitude. This resulted in a of 0.3 8 (Figure 6 7 ) The velocity along the LSB sandstone decreased to 1.5 x 10 1 0 m/s from the base model ( 2.3 x 10 1 0 m/s) (Table 6 1 ).In a separate run, the permeability of the trench sediments was decreased by an order of magnitude. The simulation yielded 6 (Figure 6 7 ) and
53 the fluid velocity within the LSB sandstone increased to 2.7 x 10 1 0 m/s. The EPP pattern in both models was very similar to the base model, with the highest EPP within the Shikoku Basin clay, about ~7 km seaward of the trench and about ~0.6 km in depth from the top of the footwall. The sensitivity to the Shikoku Basin clay permeability porosity relationship was examined by using the relationship found by Gamage and Scre aton (2006) rather than that of Skarbek and Saffer (2009) used in the base model. The resulting excess pore pressure ratio, was 0 19 which is significantly lower than the base run ( =0.34 ) ( Figure 6 7 ). The maximum pore pressure location was similar t o the base model, as well as the fluid flow pattern. The maximum fluid flow rate, 2. 3 x 10 1 0 m/s, appeared in the LSB sand stone (Table 6 1 ). We tested by Bell e w (2004) that yielded a =0.117. Decreasing the compaction parameter resulted in a significant reduction of to 0.00 5 (Figure 6 7 ) The velocity of the fluid flow within the LSB sandstone is decreased to 6.9 x 10 1 1 m/s (Table 6 1 ) The pattern of EPP and fluid flow are similar to the base model results. With a higher coefficient of friction (0.35) applied to the frontal thrust, the pore of 0.40 (Figure 6 7 ) With a lower coefficient of friction (0.1) a decrease in pore pressure occurred, with a maximum of 0.20 (Fig ure 6 7 ) The fluid patterns are similar to the base model yielding a maximum fluid velocity within the LSB sandstone of 2.5 x 10 10 m/s and 1.9 x 10 10 m/s, for a 0.35 and 0.1 coefficient of friction run, respectively (Table 6 1 )
54 Altering the fault slip rate generated varying excess pore pressure ratios. Decreasing the amount of time to 0.436 My, the lower end of the range given by Screaton et al (2009a) resulted in a of 0.40 while increasing the time to 0.78 My resulted in a lo wer of 0.29 (Figure 6 7 ) The velocity of fluid flow within the LSB sandstone was 2.9 x 10 1 0 m/s and 1.8 x 10 1 0 m/s for the slip time of 0.436 My and 0.78 My, respectiv ely (Table 6 1 ). The EPP locations and the fluid flow patterns are similar to the results of the base model. Table 6 1 Maximum simulated excess pore pressures (Pa), the calculated and the maximum fluid velocity (m/s). Run EPP (Pa) Velocity (m/s) Base 8.37E+06 0.37 2. 26 E 1 0 Decrease Slip Rate 6.72E+06 0.30 1.79 E 1 0 Increase Slip Rate 1.02E+07 0.44 2.93 E 10 Decrease k in Trench 1.14E+07 0.56 2. 74 E 10 Decrease k in LSB s andstone 8.42E+06 0.38 1.50 E 10 Gamage and Screaton (2006) k 3.98E+06 0.19 2. 72 E 1 0 High Coefficient of Friction 9.82E+06 0.39 2.48 E 10 Low Coefficient of Friction 5.67E+06 0.23 1.90E 10 Cam Clay 7.71E+04 0.00 6.87 E 11
55 Figure 6 1 The grain size grouping of the permeability of Sites C0006 and C0007 samples plotted as a function of porosity. Samples are grouped by the grain size classification of Bryant et al. ( 2002 ). Figure 6 2 The Shikoku Basin facies permeability porosity trends from the Muroto transect and the Shikoku Basin clay results from the Kumano Basin transect (in red). 19 18.5 18 17.5 17 16.5 16 15.5 0.00 0.10 0.20 0.30 0.40 0.50 0.60 Log Permeability (m 2 ) Porosity Group 3 Group 4 20 19.5 19 18.5 18 17.5 17 16.5 16 15.5 15 0.10 0.20 0.30 0.40 0.50 0.60 0.70 Log Permeability (m 2 ) Porosity Rowe and Screaton Skarbek and Saffer (2009) Gamage and Screaton (2006) Saffer and Bekins (1998)
56 Figure 6 3 Simulated excess pore pressure through time. 6.00E+06 6.50E+06 7.00E+06 7.50E+06 8.00E+06 8.50E+06 1.41E+14 1.46E+14 Excess Pore Pressure (Pa) Time (s) Time Vs Pore Pressure Time Vs Pore Pressure
57 Figure 6 4. The excess pore pressure results A) The excess pore press ure results of the base model. B) The trench profile of the excess pore pressures (noted as PORE) at depth in both the base and maximum model. The top depth represents the top of the fo otwall. C) The effective stresses and excess pore pressures occur 7 km landward of the trench.
58 Figure 6 5. Cumulative horizontal displacement simulated with the base model. Displacement (m)
59 Figure 6 6. The fluid flow results. A) The (velocity*porosity) of the base model B) A schematic illustrating flow direction in the trench sediments and LSB sandstone
60 Figure 6 7 The excess pore pressure ratio results for the sensitivity analyses. The base model is the blue column on the far left. 0.00 0.10 0.20 0.30 0.40 0.50 0.60 Base Decrease Slip Rate Increase Slip Rate Decrease Trench Permeability Decrease LSB Sandstone Permeability Increase COF (0.35) Decrease COF (0.10) Gamage and Screaton (2006) k Cam Clay
61 CHAPTER 7 DISCUSSION Pore Pressure and Fluid Flow Overall, the model simulated high excess pore pressures within the Shikoku Basin clay. Sensitivity runs confirmed the high pore pressure generatio n except for the simulation using the Gamage and Screaton (2006) porosity permeability relationship and the simulation with a Cam Clay parameter .The excess pore pressures reach a maximum beneath the frontal thrust ~7 km landward of the trench, but significant excess pore pressures do not extend seaward of the trench. The simulated excess pore pressure ratios of the base model are not as high as th at estimated for the Muroto transect (Screaton et al 2001; Saffer, 2003), likely due to fluid escape along the LSB sandstones, which do not occur in the Muroto transect. The model results are very sensitive to the permeability porosity relationship and consolidation parameters of the Shikoku Basin clay. The Shikoku Basin clay samples tested in this study exhibit a very similar permeability porosity trend to the previous relationships of the Muroto transect (Figure 6 2 ) despite very different temperature condition along the Muroto and Kumano Basin transects. This similarity supports the general results of Gamage et al. (201 1 ) who found that the permeability porosity trends of shallow sediments are less dependent on temperature or region than on the grain size within the samples. However, the seemingly small change in the relationships between the Skarbek and Saffer (2009) and the Gamage and Screaton (2006) drops Basin clay samples tested leaves room to suggest that either permeability porosity
62 relationship could be applied. Additional results will become available from laboratory tests on samples from Expedition 322 and 333 and may help refine the model results. The Cam Cla magnitude of excess pore pressure production The model resulted in minimal excess pore pressures =0.005, within the Shikoku Basin clay, ~3 km landward of the trench, and at a deeper depth, ~900 m from the top of the footwall. is considered unrealistically low because it results in simulated porosities higher than measured at Site C0011. Because of their high permeability relative to the Shikoku Basin clays, the LSB sandstone and trench sediments are zones of high flow velocity. At their highest, the velocities of the fluid flow within LSB sandstone are in the same order of magnitude of th e frontal thrust slip rate. As the fluid flows seaward, the velocity decreases by a n order of magnitude >20 km from the trench. This fluid velocity within our model is not consistent with the freshening and potentially deeper regional chemical signals obse rved at Site C0011 that suggest that fluid has traveled through the sand rich units from deeper within the subduction zone. For geochemical signals to reach Site C0011 the fluid velocity must have exceed ed the movement of the prism toe One reason that th e simulated flow rates are underestimates could be the assumption of a steady frontal thrust slip rate. Larger, but short lived, slip could cause transient bursts of fluid flow or small units of high permeability Saffer and Bekins (1998) illustrated a sim ilar idea by simulating a sudden increase in fault zone permeability allowing the fluid to escape to achieve the velocity and flow of a chemical signal.
63 Another reason for lower simulated velocities could be the exclusion of fluid migrating from landward of our model boundary due to our model focusing on shallow processes. If the sand units remain an open conduit with greater depth, the water may be escaping with an increased velocity that is not modeled. Dcollement Propagation The PTZ observations from the seismic reflection data locate the protodcollement ~1 km below the seafloor at the trenc h (Moore et al., 2009) (Figure 2 3 ). Le Pichon and Henry (1992) suggest that the protodcollement location is controlled by a maximum excess pore pressure and a m inimum in effective stress caused by rapid loading whereas Raimbourg et al. (2011) suggest that the location is controlled by horizontal stress created by movement of the prism. At 7km landward of the trench, the excess pore pressure profile illustrates a maximum in EPP and an effective stress minimum ~0.6 km from the top of the footwall. These results coincide with the inferred PTZ location (Figure 2 dcollement propagation. Looking at the eff ective horizontal and vertical stresses at this location, we can see the vertical stress is consistently greater than the horizontal stress with depth. There is no rotation of stress or suggestions of horizontal stress formation of the protodcollement as suggested by Raimbourg et al., 2001 (Figure 6 5 ). Results are significantly different at the trench, where the excess pore pressure is small. There is a n increase in excess pore pressure around ~ 0.8km of depth from the top of the footwall which is consist ent with the PTZ location (Figure 2 3). However, t he effective stress continuously increases with depth which disagrees with the mechanism for initiation of the proto dcollement suggested by Le Pichon and Henry (1992).It is possible that the loading by tr ench sediment deposition, which is not simulated in our
64 model, could increase pore pressures and decrease effective stress. O ur model results show a stress rotation below the trench. From 0 0. 7 km of depth from the top of the footwall, the horizontal stres s is greater than the vertical stress, while below 0.8 km the vertical stress is greater tha n the horizontal (Figure 6 5 ). From our model, the zone (0.7 and 0.8 km depth from footwall ) of horizontal and vertical stress transition appears consistent with po tential dcollement formation. The horizontal and vertical stress profile illustrates the potential for the protodcollement localization due to horizontal stress mechanism suggested by Raimbourg et al (2011). The mechanisms of protodcollement initiatio n by Le Pichon and Henry (1992) were observed landward of the trench, while the mechanisms by Raimbourg et al. (2011) were observed at the trench. The modeling results suggest both mechanisms contribute to creating and propagating the protodcollement.
65 CHAPTER 8 CONCLUSIONS Our modeling results suggest that the Kumano Basin transect could contain significant excess pore pressure s, with a simulated of 0.3 7, located 7 km landward of the trench beneath the frontal thrust ~0.7 km in depth from the top of the footwall and within the Shikoku Basin clay The resulting excess pore pressures are small at the trench and continually decrease seaward of the trench. The flow patterns of the simulations illustrate horizontal flow within the LSB sandstone and vertical flow in the trench sediments. Our fluid flow velocity is consistently less than the thrust fault slip, contrary to geochemical observations. Geochemical signals could require rapid but short lived thrust movements that can generate a pulse of fluid velocity or additional flow from greater depths than simulated in our model. At 7 km landward of the trench, the protodcollement propagation is dependent on the depth of maximum excess pore pressures and a minimum effec tive stress as discussed by Le Pichon and Henry (1992). At the trench, location of the protodcollement is dependent on the effective horizontal stresses that are creat ed from the prism displacement, as discu ssed in Raimbourg et al. (2011). The model resul ts illustrate both proposed mechanisms for protodcollement propagation, which suggests both could contribute to the formation and propagation of the protodcollement.
66 LIST OF REFERENCES Abaqus 6.7 (2007), Abaqus/CAE user's manual. Dassault Systme s Simulia Corp Ando, M. (1975), Source mechanisms and tectonic significance of historical earthquakes along the Nankai Trough, Japan, Tectonophysics, 27, 119 140. Becker, K. and E. E. Davis (2004), In situ determinations of the permeability of the igneous oceanic crust, in Hydrogeology of the Oceanic Lithosphere, edited by E. E. Davis and H. Elderfield, pp. 189 224, Cambridge Univ. Press, United Kingdom. Bekins, B. A., A. M. McCaffrey, and S. J. Dreiss (1995), Episodic and constant flow models for the origin of low chloride waters in a modern accretionary complex Water Resour. Res., 31 (12), 3205 3215. Bellew, G. (2004), Consolidation properties stress history, and modeling pore pressures for deep sea sediments at the Nankai Trough, M. S. thesis, University of Missouri Columbia. Bryant, W. (2002), Permeability of clays, silty clays and clayey silts, Transactions Gulf Coast Association of Geological Societies, 52, 1069 1077. Christensen, N. I. (1996), Poisson's ratio and crustal seismology, Journal of geophysical research, 101(b2), 3139 3156. Davis, E. E. and K. Becker (2002), Observations of natural state fluid pressures and temperatures in young oceanic crust and infe rences regarding hydrothermal circulation, Earth Planet. Sci. Lett., 204(1 2), 231 248, doi : 10.1016/S0012 821X(02)00982 2. Davis, E. E., K. Wang, K. Becker, and R. E. Thomson (2000), Formation scale hydraulic and mechanical properties of oceanic crust inferred from pore pressure response to periodic seafloor loading, Journal of geophysical research, 105(b6), 13423 13. Domenico P. A. and F. W. Schwartz (1998 ), Physical and Chemical Hydrogeology, 2nd Edition ed., John Wiley & Sons Inc., New Yo rk. Expedition 314 Scientists (2009), Expedition 314 Site C0006. In Kinoshita, M., H. Tobin J. Ashi G. Kimura S. Lallemant E. J. Screaton D. Curewitz, D., H. Masago K. T. Moe and the Expedition 314/315/316 Scientists, Proc. IODP, 314/315/316: Washi ngton, DC (Integrated Ocean Drilling Program Management International, Inc.), doi:10.2204/iodp.proc.314315316.118.2009
67 Expedition 316 Scientists (2009a), Expedition 316 Site C0006. In Kinoshita, M., H. Tobin, J. Ashi G. Kimura S. Lallem ant E. J. Screato n D. Curewitz H. Masago K. T. Moe and the Expedition 314/315/316 Scientists, Proc. IODP, 314/315/316: Washington, DC (Integrated Ocean Drilling Program Management International, Inc.), doi:10.2204/iodp.proc.314315316.134.2009 Expedition 316 Scientists (2009b), Expedition 316 Site C0007. In Kinoshita, M., H. Tobin J. Ashi G. Kimura S. Lallemant E. J. Screaton D. Curewitz H. Masago K. T. Moe and the Expedition 314/315/316 Scientists, Proc. IODP, 314/315/316: Washington, D C (Integrated Ocean Drilling Program Management International, Inc.), doi:10.2204/iodp.proc.314315316.135.2009 Fisher, A. T. (2005), Marine hydrogeology: recent accomplishments and future opportunities, Hydrogeology Journal, 13. Gamage, K., and E. Screat o n (2003), Data report: Permeabilities of Nankai accretionary prism sediments. In Mikada, H., Moore, G.F., Taira, A., Becker, K., Moore, J.C., and Klaus, A. (Eds.), Proc. ODP, Sci. Results, 190/196, 1 22 [Online]. Available from World Wide Web: < http:// www dp.tamu.edu/publications/190196SR/VOLUME/CHAPTERS/213.PDF >. Gamage, K. and E. Screaton (2006), Characterization of excess pore pressures at the toe of the Nankai accretionary complex, Ocean Drilling Program Sites 1173, 1174, and 808; results of one dimens ional modeling, Journal of geophysical research, 111(b4), doi: 10.1029/2004JB003572. Gamage, K., E. Screaton, B. Bekins, and I. Aiello (2011), Permeability porosity relationships of subduction zone sediments, Mar. Geol., 279(1 4), 19 36, doi: DOI: 10.1016 /j.margeo.2010.10.010. Helwany, S. (2007), Applied Soil Mechanics : With ABAQUS Applications, John Wiley & Sons. Hughes, K. L. H., T. Masterlark, and W. D. Mooney (2010), Poroelastic stress triggering of the 2005 M8.7 Nias earthquake by the 2004 M9.2 Sumatra Andaman earthquake, Earth Planet. Sci. Lett., 293(3 4), 289 299, doi: DOI: 10.1016/j.epsl.2010.02.043. Hpers, A., S. Kreiter, and A. J. Kopf (2010), Ramifications of high in situ temperatures for laboratory testing an d inferred stress states of unlithified sediments A case study from the Nankai margin, Mar. Pet. Geol., 27(7), 1565 1571, doi: DOI: 10.1016/j.marpetgeo.2010.03.014. Hpers, A (in prep.): Data Report: Permeabilities and consolidation properties of sandy sediments from the Nankai Trough, IODP Sites C0011 and C0012, Proceeding of the Integrated Ocean Drilling Program
68 Ikari, M. J., D. M. Saffer, and C. Marone (2007), Effect of hydration state on the frictional properties of montmorillonite based fault gouge Journal of geophysical research, 112(b6), B06423. Ike, T., G. F. Moore, S. Kuramoto, J. Park, Y. Kaneda, and A. Taira (2008), Variations in sediment thickness and type along the northern Philippine Sea Plate at the Nankai Trough, Island Arc, 17(3), 342 357, doi: 10.1111/j.1440 1738.2008.00624.x. Ingebritsen, S. (2006), Groundwater in Geologic Processes, 2nd Edition ed., Cambridge University Press, Cambridge. King, G. C. P., R. S. Stein, and J. Lin (1994), Static stress changes and the triggering of ear thquakes, Bulletin of the Seismological Society of America, 84(3), 935 953. Kinoshita, M., T. Kanamatsu, K. Kawamura, T. Shibata, H. Hamamoto, and K. Fujino (2008), Heat flow distribution on the floor of Nankai Trough off Kumano and implications for the g eothermal regime of subducting sediments, JAMSTEC report of research and development, 8, 13 28. Kopf, A. J., M. Strasser, N. Monsees, M.B. Underwood, and J. Guo (in p ress ), Data Report: Particle size analysis of sediments recovered during IODP Expe dition 315 and 316, Sites C0001 C00008 Nankai Trough forearc, off Japan Proceedings of the integrated Ocean Drilling Program Kummer, T. and G. A. Spinelli (2009), Thermal effects of fluid circulation in the basement aquifer of subducting ocean crust, Journal of geophysical research, 114(b3), B03104, doi: doi: 10.1029/2008JB006197. Le Pichon, X. and P. Henry (1992), Erosion and accretion along subduction zones: a model of evolution, Eos (Washington, D.C.), 73(14), 294. Masterlark, T. (2003), Finite element mode l predictions of static deformation from dislocation sources in a subduction zone: Sensitivities to homogeneous, isotropic, Poisson solid, and half space assumptions, Journal of geophysical research, 108(b11). Masterlark, T. and K. L. Hughes (2008), Next generation of deformation models for the 2004 M9 Sumatra Andaman earthquake, Geophys. Res. Lett., 35(19), L19310. Masterlark, T. and H. F. Wang (2000), Poroela stic coupling between the 1992 Landers and Big Bear Earthquakes, Geophys. Res. Lett., 27(22), 3647 3650.
69 Miyazaki, S. and K. Heki (2001), Crustal velocity field of southwest Japan: Subduction and arc arc collision, Journal of geophysical research, 106(b3), 4305 4326. Moore, G.F., H. Mikada, J.C. Moore, K. Becker, and A. Taira (2005), Legs 190 and 196 synthesis: deformation and fluid flow processes n the Nankai Trough accretionary prism. In Mikada, H., G. F. Moore, A. Taira, K. Becker, J.C. Moore, A. Klaus (Eds.), Proc.ODP, Sci. Results, 190/196, 1 26 [Online]. Availab le from World Wide Web: http://www odp.tamu.edu/publications/190196SR/VOLUME/SYNTH/SYNTH.PDF Moore, G. F., J. O. Park, N.L. Bangs, S.P. Gulick, H.J. Tobin, Y. Nakamura, S. Sato, T. Tsuji, T. Yoro, H. Tanaka, S. Uraki, Y. Kido, Y. Sanada, S. Kuramoto, A. T aira (2009), Structural and seismic stratigraphic framework of the NanTroSEIZE Stage 1 transect. In Kinoshita, M., H. Tobin, J. Ashi, G. Kimura, S. Lallement, E.J. Screaton, D. Curewitz, H. Masago, K.T. Moe, the Expedition 314/315/316 Scientists, Proc. IOD P, 314/315/316: Washington, DC (Integrated Ocean Drilling Program Management International, Inc.). doi:10.2204/iodp.proc.314315316.102.2009 Moore, G. F., A. Taira, A. Klaus, L. Becker, B. Boecker, B. A. Cragg, A. Dean, C. L. Fergusson, P. Henry, S. Hirano, T. Hisamitsu, S. Hunze, M. Kastner, A. J. Maltman, J. K. Morgan, Y. Murakami, D. M. Saffer, M. Sanchez Gomez, E. J. Screaton, D. C. Smith, A. J. Spivack, J. Steurer, H. J. Tobin, K. Ujiie, M. B. Underwood, and M. Wilson (2001), New insights into deformation and fluid flow processes in the Nankai Trough accretionary prism: Results of Ocean Drilling Program Leg 190, Geochem. Geophys. Geosyst., 2, 10.129/2001GC000166, 2001. Moore, J. C. (1989), Tectonics and hydrogeology of accretionary prisms: role of the dcollement zone, J. Struct. Geo l., 11(1 2), 95 106. Raimbourg, Hugues, K.Ujiie, A. Kopf., T. Hisamitsu, Y. Hamano, S. Saito, and M. Kinoshita (2011), The role of compaction contrasts in sediments in dcollement initiation in an accretionary prism, Marine Geology, doi:10.1016/j.margeo.2011.02.011 Roscoe, K.H., and J.B. Burland (1968), On the generalised stress strain behaviour of wet clay. In Engineering plasticity, Edited by J. Heyman and F.P. Leckie. Cambridge University Press, New York Rowe, K., E. Screaton, J. Guo, and M. B. Underwood (in review), Data Report: Permeabilities of sediments from the Kumano Basin transect offshore the Kii Peninsula, Japan, Proceedings of the integrated Ocean Drilling Program Ruff, L a nd H Kanamori (1983), S eismic coupling and unc oupling at subduction zones, T ectonophysics, 99(2 4), 99 117.
70 Saffer, D. (2003), Pore pressure development and progressive dewatering in underthrust sediments at the Costa Rican subduction margin: Comparison with northern Barbados and Nankai, Journal of g eophysical research, 108(b5), 2261. Saffer, D. M. (2007), Pore pressure within underthrust sediment in subduction zones, In Dixon, T. H. and J. C. Moore (eds.), The seismogenic zone of subduction thrust faults, Columbia University Press, New York, 171 205 S a ffer D. (2010), Hydrostratigraphy as a control on subduction zone mechanics through its effects on drainage: an example from the Nankai Margin, SW Japan, Geofluids, 10(1 2), 114 131, doi: 10.1111/j.1468 8123.2009.00276.x. Saffer, D. and B. A. Bekins (1998), Episodic fluid flow in the Nankai accretionary complex: Timescale, geochemistry, flow rates, and fluid budget, Journal of geophysical research, 103(b12), 30351 30370. Saffer, D. M. and B. A. Bekins (2002), Hydrologic co ntrols on the morphology and mechanics of accretionary wedges, Geology (Boulder), 30(3), 271 274. Saffer, D. M. and B. A. Bekins (2006), An evaluation of factors influencing pore pressure in accretionary complexes: Implications for taper angle and wedge m echanics, Journal of geophysical research, 111(b4), B04101. Saffer, D. M. and C. Marone (2003), Comparison of smectite and illite rich gouge frictional properties: application to the updip limit of the seismogenic zone along subduction megathrusts, Earth Planet. Sci. Lett., 215(1 2), 219 235, doi: DOI: 10.1016/S0012 821X(03)00424 2. Schofield, A., and P. Wroth (1968), Critical state soil mechanics, McGraw Hill, London Schwartz, F. W., and H. Zhang (2003), Fundamentals of groundwater, Wiley, New York Scre aton, E. (2010), Recent advances in subseafloor hydrogeology: focus on basement sediment interactions, subduction zones, and continental slopes, Hydrogeology Journal, 18, 1547 1570. Screaton, E., G. Kimura, D. Curewitz, and et al. (2009), Interactions bet ween deformation and fluids in the frontal thrust region of the NanTroSEIZE transect offshore the Kii Peninsula, Japan: Results from IODP Expedition 316 Sites C0006 and C0007, Geochemistry, geophysics, geosystems : G3, 10, Q0AD01, doi: 10.1029/2009GC002713
71 Screaton, E., D. Saffer, P. Henry, S. Hunze, and Leg 190 Shipboard Scientific Party (2002), Porosity loss within the underthrust sediments of the Nankai accretionary complex: Implications for overpressures, Geology (Boulder), 30(1), 19 22. S e no T., S. Stein, and A. E. Gripp (1993), A model for the motion of the P hilippine S ea plate consistent with nuvel 1 and geological data, Journal of geophysical research, 98(b10), 17941 17948. Shipboard Scientific Party (2000), Leg 190 Preliminary Report: Deformati on and fluid flow processes in the Nankai Trough accretionary prism. ODP Prelim. Rpt., 190 [Online]. Available from World Wide Web: < http://www odp.tamu.edu/publications/prelim/190_prel/190PREL.PDF >. Shipboard Scientific Party (2001), Leg 190 summary. In M oore, G.F., Taira, A., Klaus, A., et al., Proc. ODP, Init. Repts., 190: College Station TX (Ocean Drilling Program), 1 87. Skarbek, R. M. and D. M. Saffer (2009), Pore pressure development beneath the dcollement at the Nankai subduction zone: Implications for plate boundary fault strength and sediment dewatering, Journal of geophysical research, 114(b7), B07401. Spinelli, G. A., P. S. Mozley, H. J. Tobin, M. B. Underwood, N. W. Hoffman, and G. M. Bellew (2007), Diagenesis, sediment strength, and pore coll apse in sediment approaching the Nankai Trough subduction zone, Geological Society of America bulletin, 119(3 4), 377 390. Stauffer, P. and B. A. Bekins (2001), Modeling consolidation and dewatering near the toe of the northern Barbados accretionary compl ex, Journal of geophysical research, 106(b4), 6369 6383. Tobin, H.J., and M. Kinoshita (2006a), Investigations of seismogenesis at the Nankai Trough, Japan, IODP Sci. Prosp., NanTroSEIZE Stage 1, doi:10.2204/iodp.sp.nantroseize1.2006 Tobin, H.J., and M. K inoshita (2006b), NanTroSEIZE: the IODP Nankai Trough Seismogenic Zone Experiment. Sci. Drill., 2:23 27, doi:10.2204/iodp.sd.2.06.2006 Tobin, H. J. and D. M. Saffer (2009), Elevated fluid pressure and extreme mechanical weakness of a plate boundary thrust, Nankai Trough subduction zone, Geology (Boulder), 37(8), 679 682. Turcotte, D. L., and G. J. Schubert (1982), Geodynamics: Applications of Continuum Physics to Geological Problems, 450 pp., John Wiley, Hoboken, N. J.
72 Underwood, M.B., Saito, S., Kubo, Y., and the Expedition 322 Scientists (2009), NanTroSEIZE Stage 2: subduction inputs. IODP Prel. Rept., 322, doi:10.2204/iodp.pr.322.2009 Voss, C. I. and A. M. Provost, (2010), SUTRA: A model for saturated unsaturated, variable density ground water flow with solute or energy transport, USGS, http://water.usgs.gov/nrp/gwsoftware Wang, H. F. (2000), Theory of linear poroelasticity with applications to geomechanics and hydrogeology, Princeton University Press, New Jersey
73 BIOGRAPHICAL SKETCH Katherine Rowe was born in Augusta, Georgia. She attended New Smyrna Beach High School in New Smyrna Beach, Florida. She attended the University of Florida and graduated with a Bachelor of Science in g eology in 2008. She received her Masters of Science degree in g eology from the University of Florida in the summer of 2011.