Author(s): Yong-Sik Cho; Hyun-Han Kwon
Linked Author(s):
Keywords: No Keywords
Abstract: A probabilistic tsunami hazard analysis (PTHA) has been largely adopted over the last decade to jointly consider the likelihood of all possible tsunami sources and the distribution of tsunami heights in both the near and far fields. To begin with, this study explored the changes in probability distribution associated with tsunami run-up heights along the Korean Peninsula, resulting from various factors related to tsunami hazard such as the location, magnitude, fault parameters and slip distributions. In these regards, this study employed numerical simulations of the wave propagation with respect to virtual tsunamis for obtaining the tsunami run-up heights along the eastern coast of the Korean Peninsula. We examined a number of probability distributions for the tsunami run -up heights, and Weibull distribution was finally selected for a given BIC (Bayesian information criterion). It was found that there are significant distribution changes in tsunami run -up heights relying on the location and magnitude of earthquakes. This study further investigated an integrated PTHA analysis to jointly estimate tsunami run-up heights from multiple earthquake events, based on a generalized linear model (GLM) within a fully Bayesian framework. More specifically, we estimated a set of parameters and their credible intervals relying on earthquake characteristics in a probability distribution. A fur ther discussion on the role of the distribution changes for tsunami risk is provided.
Year: 2018