On the emergence of fault afterslip during laboratory seismic cycles

Abstract

Spatial fault heterogeneity is often invoked to explain the occurrence of fault afterslip following seismic ruptures. In this study, we tested this hypothesis in the laboratory by performing triaxial experiments on both homogeneous and heterogeneous faults, under confining pressures of 30, 60, and 90 MPa. The faults were composed of granite, prone to seismic behaviour, and marble, prone to aseismic behaviour. Unlike homogeneous granite faults, which display a nucleation stage followed by regular seismic events, heterogeneous faults can contain the co-seismic dynamic event within the experimental fault length. During this phase, the aseismic areas adjacent to the dynamic event undergo a stress increase, which is then released by fault afterslip over an extended post-seismic phase. The magnitude and duration of this post-seismic phase increase with confining pressure and with the proportion of aseismic-prone areas. We infer that the enhancement of post-seismic afterslip originates from the increase in the frictional stability of the aseismic-prone area, and of the normal stress acting on the fault. In addition, the observed increase in initial strain rate with normal stress is well explained by the rate-and-state framework. At the scale of our experiments, fault frictional heterogeneities play a primary role in the emergence of fault afterslip.