Demian M. Saffer

Updip limit of the seismogenic zone beneath the accretionary prism of southwest Japan: An effect of diagenetic to low-grade metamorphic processes and increasing effective stress

Off southwest Japan the seaward limit of coseismic displacement (or updip limit of the seismogenic zone) of the 1946 Mw 8.3 thrust earthquake reaches to 4 km depth and ∼40 km landward of the trench. This limit coincides with the estimated location of the 150 °C isotherm, and has been linked to changes in physical properties associated with the smectite to illite clay-mineral transition. Here we show that this limit correlates with a suite of diagenetic to low-grade metamorphic processes characterized by (1) declining fluid production and decreasing fluid pressure ratio (λ*) and (2) active clay, carbonate, and zeolite cementation and the transition to pressure solution and quartz cementation. These diagenetic to low-grade metamorphic changes cause the onset of velocity weakening during thrust faulting, an increase in effective stress, and strengthening of the hanging wall, which together combine to produce recordable earthquakes.

Comparison of smectite- and illite-rich gouge frictional properties: application to the updip limit of the seismogenic zone along subduction megathrusts

Along plate boundary subduction thrusts, the transformation of smectite to illite within fault gouge at temperatures of ∼150°C is one of the key mineralogical changes thought to control the updip limit of seismicity. If correct, this hypothesis requires illite-rich gouges to exhibit frictionally unstable (velocity-weakening) behavior. Here, we report on laboratory experiments designed to investigate the frictional behavior of natural and synthetic clay-rich gouges. We sheared 5-mm-thick layers of commercially obtained pure Ca-smectite, a suite of smectite–quartz mixtures, and natural illite shale (grain size ranging from 2 to 500 μm) in the double-direct shear geometry to shear strains of ∼7–30 at room humidity and temperature. XRD analyses show that the illite shale contains dominantly clay minerals and quartz; within the clay-sized fraction ( 40 MPa). Our data, specifically the velocity-strengthening behavior of illite shale under a wide range of conditions, do not support the hypothesis that the smectite–illite transition is responsible for the seismic–aseismic transition in subduction zones. We suggest that other depth- and temperature-dependent processes, such as cementation, consolidation, and slip localization with increased shearing, may play an important role in changing the frictional properties of subduction zone faults, and that these processes, in addition to clay mineralogy, should be the focus of future investigation.

New insights into deformation and fluid flow processes in the Nankai Trough accretionary prism: Results of Ocean Drilling Program Leg 190

Moore, G. F., Taira, A., Klaus, A., Becker, L., Boeckel, B., Cragg, B. A., Dean, A., Fergusson, C. L., Henry, P., Hirano, S., Hisamitsu, T. et al. (2001). New insights into deformation and fluid flow processes in the Nankai Trough accretionary prism: Results of Ocean Drilling Program Leg 190. Geochemistry, Geophysics, Geosystems, 2, Article No: 2001GC000166.

Episodic fluid flow in the Nankai accretionary complex: Timescale, geochemistry, flow rates, and fluid budget

Down-hole geochemical anomalies encountered in active accretionary systems can be used to constrain the timing, rates, and localization of fluid flow. Here we combine a coupled flow and solute transport model with a kinetic model for smectite dehydration to better understand and quantify fluid flow in the Nankai accretionary complex offshore of Japan. Compaction of sediments and clay dehydration provide fluid sources which drive the model flow system. We explicitly include the consolidation rate of underthrust sediments in our calculations to evaluate the impact that variations in this unknown quantity have on pressure and chloride distribution. Sensitivity analysis of steady state pressure solutions constrains bulk and flow conduit permeabilities. Steady state simulations with 30% smectite in the incoming sedimentary sequence result in minimum chloride concentrations at site 808 of 550 mM, but measured chlorinity is as low as 447 mM. We simulate the transient effects of hydrofracture or a strain event by assuming an instantaneous permeability increase of 3–4 orders of magnitude along a flow conduit (in this case the decollement), using steady state results as initial conditions. Transient results with an increase in decollement permeability from 10−16 m2 to 10−13 m2 and 20% smectite reproduce the observed chloride profile at site 808 after 80–160 kyr. Modeled chloride concentrations are highly sensitive to the consolidation rate of underthrust sediments, such that rapid compaction of underthrust material leads to increased freshening. Pressures within the decollement during transient simulations rise rapidly to a significant fraction of lithostatic and remain high for at least 160 kyr, providing a mechanism for maintaining high permeability. Flow rates at the deformation front for transient simulations are in good agreement with direct measurements, but steady state flow rates are 2–3 orders of magnitude smaller than observed. Fluid budget calculations indicate that nearly 71% of the incoming water in the sediments leaves the accretionary wedge via diffuse flow out the seafloor, 0–5% escapes by focused flow along the decollement, and roughly 1% is subducted.

On the relation between fault strength and frictional stability

A fundamental problem in fault mechanics is whether slip instability associated with earthquake nucleation depends on absolute fault strength. We present laboratory experimental evidence for a systematic relationship between frictional strength and friction rate dependence, one of the key parameters controlling stability, for a wide range of constituent minerals relevant to natural faults. All of the frictionally weak gouges (coefficient of sliding friction, μ < 0.5) are composed of phyllosilicate minerals and exhibit increased friction with slip velocity, known as velocity-strengthening behavior, which suppresses frictional instability. In contrast, fault gouges with higher frictional strength exhibit both velocity-weakening and velocity-strengthening frictional behavior. These materials are dominantly quartzofeldspathic in composition, but in some cases include certain phyllosilicate-rich gouges with high friction coefficients. We also find that frictional velocity dependence evolves systematically with shear strain, such that a critical shear strain is required to allow slip instability. As applied to tectonic faults, our results suggest that seismic behavior and the mode of fault slip may evolve predictably as a function of accumulated offset.

Elevated fluid pressure and extreme mechanical weakness of a plate boundary thrust, Nankai Trough subduction zone

Pore fluid pressure is a key parameter governing both the shear strength of faults and the conditions for seismic and tsunamigenic slip on subduction zone megathrusts. However, quantitative constraints on this parameter based on in situ data in these, or any, faults have proven elusive. Using seismic interval velocities derived from a three-dimensional seismic reflection survey, we compute porosity, fluid pressure, and effective stress at the plate interface in the Nankai Trough subduction zone (offshore southwestern Japan) to 20 km down the fault dip using well-constrained, locally calibrated empirical relationships between velocity, porosity, and effective stress. We show that the fault and immediately subjacent footwall are nearly undrained, implying that the subduction megathrust slips under a shear stress of

Critically pressured free-gas reservoirs below gas-hydrate provinces

Palaeoceanographic data have been used to suggest that methane hydrates play a significant role in global climate change. The mechanism by which methane is released during periods of global warming is, however, poorly understood. In particular, the size and role of the free-gas zone below gas-hydrate provinces remain relatively unconstrained, largely because the base of the free-gas zone is not a phase boundary and has thus defied systematic description. Here we evaluate the possibility that the maximum thickness of an interconnected free-gas zone is mechanically regulated by valving caused by fault slip in overlying sediments. Our results suggest that a critical gas column exists below most hydrate provinces in basin settings, implying that these provinces are poised for mechanical failure and are therefore highly sensitive to changes in ambient conditions. We estimate that the global free-gas reservoir may contain from one-sixth to two-thirds of the total methane trapped in hydrate. If gas accumulations are critically thick along passive continental slopes, we calculate that a 5 °C temperature increase at the sea floor could result in a release of ∼2,000 Gt of methane from the free-gas zone, offering a mechanism for rapid methane release during global warming events.

Porosity loss within the underthrust sediments of the Nankai accretionary complex: Implications for overpressures

Subduction complexes provide an opportunity to examine the interactions of deformation and fluid flow in an active setting. Ocean Drilling Program Leg 190 investigated the relationship between deformation, physical properties, and fluid flow in the toe of the Nankai Trough accretionary complex. With three sites (two from Leg 190, one from a previous leg) penetrating the decollement zone at various stages of development along the same transect, it is now possible to examine the change in porosity during rapid loading by trench turbidites and subsequent underthrusting. Results indicate inhibited dewatering and probable overpressure development seaward of the frontal thrust. Comparison of a reference site porosity versus depth curve to data from a site located within the protothrust zone indicates an overpressure ratio, λ * , of ∼0.42, where λ * = [(pore pressure - hydrostatic pressure)/(lithostatic pressure - hydrostatic pressure)]. These overpressures suggest that the hemipelagic sediments have insufficient permeability for fluid escape to keep pace with the rapid loading by turbidite deposition within the trench. At a site 1.75 km farther arcward, an excess pore pressure ratio of λ * = ∼0.47 was estimated, reflecting the additional loading due to recent thickening by the frontal thrust.

Hydrologic controls on the morphology and mechanics of accretionary wedges

At many subduction zones, accretionary complexes form as sediments are offscraped from the subducting plate. Existing mechanical models that treat accretionary complexes as critically tapered wedges of sediment demonstrate that pore pressure controls their taper angle by modifying rock strength. We combine a model of groundwater flow with critical-taper theory to show that permeability and plate-convergence rate are important controls on accretionary wedge geometry through their influence on pore pressure. Low permeability and rapid convergence sustain nearly undrained conditions and shallowly tapered geometry, whereas high permeability and slow convergence result in steep geometry. Our results are generally in good agreement with data from active accretionary complexes, but also illustrate the importance of other factors, such as incoming sediment thickness and stratigraphy. One key implication is that strain rate and hydrologic properties may strongly influence the strength of the crust in a variety of geologic settings.

Laboratory results indicating complex and potentially unstable frictional behavior of smectite clay

A central problem in explaining the apparent weakness of the San Andreas and other plate boundary faults has been identifying candidate fault zone materials that are both weak and capable of hosting earthquake-like unstable rupture. Our results demonstrate that smectite clay can be both weak and velocity weakening at low normal stress (<30 MPa). Our data are consistent with previous work, which has focused on higher normal stress conditions (50 MPa and greater) and found only velocity strengthening. If natural fault zones contain significant smectite, one key implication of our results is that localized zones of high pore pressure, which reduce effective normal stress, could be important in controlling potential sites of earthquake nucleation. Our experiments indicate that friction of smectite is complex, and depends upon both sliding velocity and normal stress. This complexity highlights the need for detailed experiments that reflect in-situ conditions for fault gouges.