Effects of fault roughness on estimating critical slip-weakening distance from fault slip history: A laboratory study

Earthquakes are the dynamic rupture of faults governed by fault weakening processes. Critical slip-weakening distance (Dc) is a crucial source parameter of earthquakes, and the determination of Dc is of great concern to semiologists. However, determining Dc for natural earthquakes is challenging due to the trade-off in inversed source models. To solve this problem, Fukuyama and his coworkers proposed a simple method (denoted as the F&M method) to estimate Dc directly from slip-velocity functions. According to the F&M method, the fault slip at the peak slip velocity (Dc') can be used as an approximation of Dc. However, the feasibility of this method has not been completely resolved. Here, we performed laboratory earthquake rupture experiments to examine the validity of the F&M method. Experiments were conducted on faults with different roughness and stress level. We studied the effect of fault roughness on the validity of the F&M method. Our results show that the increase in fault roughness could complicate the fault weakening process, producing repeated weakening and strengthening phase, which leads to a prominent deviation between Dc' and Dc. Furthermore, we also observed a correlation between the critical slip-weakening distance Dc and the final fault slip D. Such correlation implies the scaledependent nature of Dc, which is consistent with seismic observations in the field.