Effect of postseismic creep on earthquake triggering

Studies of small repeating earthquake sequences are important in characterizing fault frictional properties. Some of the repeating sequences located close to each other appear to interact, such as the LA and SF sequences on the creeping section of the San Andreas Fault in California, which are close to the SAFOD drilling site. In this study, our goal is to quantify the interaction of repeating earthquakes in the framework of rate-and-state fault models. Such models can reproduce the behavior of isolated repeating earthquake sequences. Our model consists of repeating earthquakes occurring on two small velocity-weakening (VW) patches embedded into a larger velocity-strengthening (VS) fault area. Our simulations show several types of stress transfer that occur between the VW patches which host the repeating sequences. When an earthquake occurs on one patch, it causes a direct static stress increase on the other patch, a direct static stress increase on the VS area surrounding the other patch that leads to a higher stressing rate on that patch, and an additional evolving stress increase due to the postseismic creep that is largest in the areas immediately next to the earthquake and then travels through the VS region. We find that this last contribution, that of the traveling postseismic slip, is the dominating factor in triggering rupture on the other VW patch. That findings is based on the comparison of the time advance of the triggered earthquake in our simulation and in the Coulomb failure model which only accounts for the direct stress increase on the patch due to an earthquake on the other patch. The difference between the time advance in the two models can be as large as 75% of the original recurrence time interval. These results show the great importance of the evolving stress increases on earthquake triggering, which are primarily due to the postseismic creep in our simulations. However, similar effects should occur due to aseismic transients that have been observed on a number of faults. In our model, the interaction is established early in the simulation. However, it requires a certain period of time for the interaction between the two sequences to approach the steady state. Our preliminary results show that the shorter the distance between the two repeating sequences, the longer is the time needed to reach the steady state. We will report on our detailed analysis of the postseismic creep as the dominant triggering factor, including its role in establishing the interactive sequence, how its features evolve in space and time, and the comparison of various stress transfer effects using the rate-and-state-based model of time advance of Dieterich (1994). We will also report on our current studies on the effect of heterogeneity on this interaction, to reproduce the well-documented complex interaction behavior observed between the LA and SF repeating sequences.