Time-Dependent Probabilistic Tsunami Hazard Analysis by Coupling a Space-Time Earthquake Interaction Model with Stochastic Sources

Abstract:

This study develops a novel framework of time-dependent probabilistic tsunami hazard analysis (PTHA) by coupling a space-time earthquake occurrence model and a probabilistic tsunami hazard model. We apply the framework to Western Sumatra, Indonesia, where future tsunamigenic events are anticipated from the Mentawai segment of the Sunda subduction zone. The earthquake occurrence model characterizes spatiotemporal interaction of major tsunamigenic events as a correlated, multivariate Bernoulli process of neighboring segments, each following Brownian passage time recurrences (Ceferino et al., 2020). We calibrate the earthquake occurrence model on historically and instrumentally recorded ruptures of the Mentawai thrust. A total of 6,300 stochastic source models are simulated for six magnitude scenarios between M_w 7.75 and M_w 9.0. The stochastic slip models are employed to run Monte Carlo tsunami simulations and conduct probabilistic tsunami inundation analysis (Muhammad et al., 2016). Moreover, the space-time earthquake interaction model is coupled with the stochastic tsunami simulation method to obtain time-dependent PTHA results. The PTHA results in terms of tsunami hazard curves and spatial maps in Padang, West of Sumatra, are produced for two different durations (10 and 50 years). We then compare the classical (time-independent) PTHA to the time-dependent PTHA model. The results indicate that the new time-dependent model produces many earthquakes within a period of 50 years (which is a concerning period of regional tsunami risk management) and, hence, leads to higher tsunami hazard estimates than the classical approach. The tsunami hazard maps at 10% and 2% probability of exceedance in 50 years (i.e. 475-year and 2475-year return periods) also show that the classical approach significantly underestimates the hazard areas (i.e. 50% smaller than the new time-dependent approach) leading to an insufficient tsunami risk mitigation plan.