Response of laboratory faults to transient stressing at seismic frequencies: stressing-rate and duration-dependent triggering regimes.
Friday, 19 December 2014: 11:50 AM
Earthquakes are sometimes triggered by transient stresses carried by seismic waves. There are several features of triggering that require explanation, including 1) the often-observed delay between the triggering waves and the triggered earthquakes, and 2) the effectiveness of transient stressing in the seismic frequency band. Previous theoretical and laboratory work has suggested that seismic transients should have little impact on faults if the duration of the transient is smaller than the timescale for nucleation of slip. We examine stress triggering during laboratory stick-slip sliding of granite and focus on several observations that pertain to earthquake triggering in nature. 1) Delayed triggering (clock advance) occurs when slip accumulates during the transient but remains below a critical value dc. 2) Delayed and instantaneous triggering both occur much earlier in the stick-slip cycle than expected for a simple Coulomb threshold. 3) Shorter-period (higher stressing-rate) pulses are more effective at triggering than longer-period pulses of the same stress amplitude. We use numerical simulations to show that rate-state friction can explain each of the observed features, but not all three simultaneously. Only the Ruina slip law can reproduce the observations of stressing-rate dependence and early triggering onset. The Dieterich aging law predicts only duration-dependent triggering with relatively late triggering onset. The observations show that moderate-amplitude, short-period transients can have a strong triggering effect even at durations far below the nucleation timescale. Transient oscillations at seismic frequencies may therefore be more effective at earthquake triggering than previously recognized.