The Effect of Earthquake Source Complexities on the Variability of Tsunami Inundation Parameters

Abstract

To estimate earthquake induced tsunami hazard and risk posed to communities and coastal infrastructure, it is important to characterize well the variability of this physical process. For this purpose, we dissect the tsunami wave generation process into more manageable pieces, by isolating the variability which stems solely from earthquake source rupture complexity. To accomplish this, we use a simplified digital elevation model to simulate tsunami wave propagation and inundation using a nonlinear shallow water equations, from which we extract different tsunami intensity measures (TIMs) at the coast. To characterize the earthquake source, we implement a novel and flexible method to simulate synthetic slip across faults, by prescribing directly the slip correlation between each subfault pair. In particular, we assume that the correlation between subfault pairs decreases with increasing euclidean distance between them. From several thousand stochastic realizations of slip, we select realizations that comply with observed earthquake physics.First we compute mean and variability of the potential energy, which increase with increasing spatial slip correlations. In the near-field, the mean of all simulated TIMs remain constant with changing spatial correlations, slightly higher than the TIMs generated by a homogeneous slip earthquake model. In contrast, TIM variability increase with increasing spatial correlations, a feature that does apply to homogenous slip models, which do not produce TIM variability. Finally, larger spatial slip correlations produce higher exceedence probability of near field TIMs, which highlight the need to reduce the uncertainty of earthquake spatial slip correlation parameters.