Project Abstract
|
Faults are usually not planar and a single earthquake often involves multiple faults (1992 Landers, 2001 Kunlun, 2002 Denali etc.). But how and why does the earthquake rupture jump between discontinuous segments remain mostly unknown. A few studies have address this problem before. Previous work put a lot of efforts on quantifying the key geometrical parameters such as step-over distances and overlapping distances with which rupture might jump from one fault to another (Harris et al, 1991; Harris and Day, 1993; Magistrale and Day, 1999). This project had 3 objectives: 1) Calculate multiple earthquake scenarios on a multi-segment fault to understand how the absolute initial stress conditions influence the rupture behavior and the resulting ground motion. 2) Investigate the geometrical conditions (degree of overlapping and step over between adjacent segments) under which a cascading earthquake can occur. 3) Study the velocity strengthening layer, its depth and its implementation. We use the finite element code MAFEM (Ma and Liu, 2006) to study the planar, rectangular faults in different kinds of tectonic settings. Although in our model the "velocity strengthening" only exists in the first couples of kilometers on the fault from the free surface, the maximum slip rates are only influenced in the shallow part, but the total slip tends to be adjusted throughout the fault. We ran a number of earthquake scenarios and investigated the nucleation by dynamic triggering on a multi-segment normal fault. We calculated the development of Coulomb stress on sub-parallel planes due to the dynamic rupture process on the main fault plane. We developed a simple proxy curve to characterize the complex temporal and spatial distribution of the Coulomb stress change field. We successfully applied the proxy TA-curve (threshold area curve) with theoretical formulas for critical length of nucleation patch to predict the initial stress level for the rupture to be triggered on adjacent fault. (Liu et al., 2011) |
