Compositional and Fluid Pressure Controls on the State of Stress on the Nankai Subduction Thrust

We present the initial results of an ongoing experimental study that examines the role that clay minerals play in controlling the properties and state of stress on subduction plate boundary faults. We focus here on the Nankai Trough, SW Japan, because it is representative of a major class of sediment-rich subduction zones (including Alaska, Chile, and Cascadia) that are capable of generating very large earthquakes (M ~8 and greater) and are almost fully coupled in their interseismic periods. The plate boundary fault also seems to be extremely weak because the maximum principal stress currently has a trench parallel orientation in the backstop region. For this to occur the maximum shear stress on the subduction thrust cannot exceed approx. 18 MPa even at the down-dip end of the seismogenic zone (recent numerical simulations by Wang and He). Our preliminary results suggest that both fault mineralogy and regional excess fluid pressure contribute to the low resolved shear stresses on the subduction zone plate boundary. Ring and direct shear tests show that saturated clay phases in the fault possess intrinsically low residual friction coefficients (RFC) at stress levels between 1 and 30 MPa. The RFC values for smectite are 0.14 ñ 0.02, for illite 0.25 ñ 0.01, and for chlorite 0.26 ñ 0.02. We find that the relatively weak illite phase is mechanically dominant in the incoming Muroto section and smectite contributes to even lower RFC values (as low as 0.18) in the incoming strata off the Ashizuri peninsula. Off Muroto, the overall magnitude of the shear stress we predict in the frontal 30 km remains below 2 MPa and only rises towards and above 4 MPa at 50 km from the toe of the wedge because of the overpressuring (lambda* values of between 0.6 to 0.8) and weak clays (RFC values of aprox.0.32 for 60 % Illite + 40 % Qtz mudstone). This is consistent with the low critical wedge taper (4.1degrees) in this region and values are well below the maximum stress estimates derived by Wang and He. Off the Ashizuri region, where the critical taper is higher (7.9 degrees), the basal lambda* value for the decollement could be as low as 0.0 to 0.4 for a decollement lying in a clay-rich portion of the deeper underthrust section, which may have a RFC values as low as 0.24-0.26 even after clay diagenesis has removed the smectite. Even with the lower overpressures we estimate that the basal shear stresses should remain below approx. 12 to 18 MPa, consistent with the Wang and He maximum basal shear stress estimates, further suggesting there is no low stress paradox at this subduction zone. Overall, our data do not, support the proposal that the smectite to illite reaction is directly responsible for the onset of seismogenic behavior throughout the Nankai system because both smectite and illite have low RFC values, tend to velocity strengthen, and there is already a preexisting mechanical dominance of in much of the incoming section (particularly off Muroto).