Continental Fault Structure and Rheology from the Frictional-to-Viscous Transition Downward

Faulting is an expression of the interaction between rock rheology, kinematic boundary conditions, and associated stress fields. The structure and rheology of faults vary with depth, such that pressure-dependent frictional behavior predominating in the upper, brittle part of the crust is transitional to strongly temperature- and rate-dependent behavior in the lower part of the crust and mantle. This frictional-to-viscous transition (FVT) is characterized by changes in rock structure, rheology, and fluid activity that are closely tied to the earthquake cycle. As such, the FVT is a first-order decoupling zone, whose depth and lateral extent vary in time. Brittle, sometimes seismic, instabilities perturb the ambient stress field within the lithosphere on timescales ranging from seconds to years. These instabilities are measurable as transient motions of the Earth's surface and are manifest both at, and below, the FVT by the development of structural anisotropies (fractures, foliations). Surface motion studies of plate-boundary strike-slip faults indicate that shearing below the FVT is more localized in the lower crust than in the upper mantle. Structural investigations of exhumed shear zones reveal that this localization involves the nucleation of fractures at the FVT, as well as the buckling and rotation of existing foliations below the FVT. In some cases, rotation of these surfaces can initiate transient deformation, transferring stress upward and potentially triggering earthquakes. The networking of shear zones on several length scales allows them to function as decoupling horizons that partition three-dimensional strain within the lithosphere. The simplification of fault geometry with progressive strain lends justification to the use of laboratory-derived flow laws to estimate the bulk rock rheology on length scales at which strain is homogeneous. In general, the longer the timeand length scales of faulting considered, the greater the potential influence of the kinematic and thermal history on the rheology of the fault system. Taken together, studies suggest that future fault modeling must include parameters that quantify the thermal and structural aspects of rock history, as well as the fluid activity in and around faults.