Comparison of source-base estimate of peak ground acceleration (Amax) in Zagros by Bayesian method with non-source approach

Abstract

Regarding Bayesian probability theory is an appropriate and useful method to estimate parameters in seismic hazard analysis. The analysis in Bayesian approaches is based on a posterior belief, also their special ability is to take into account the uncertainty of parameters in probabilistic relations and a priori knowledge. In this study, we used the program for seismic hazard Bayesian estimate which was elaborated by Alexey Lyubushin. Our study is the next in the sequence of applications of this software to seismic hazard assessment in different regions of the world. In this study, Bayesian approach has been used to obtain estimated seismic parameters. In order to reach this aim, 30 different source regions in Zagros seismotectonic province have been considered. The main assumptions are Poissonian character of the seismic events flow and properties of the Gutenberg-Richter distribution law. The a posteriori probability distribution functions of Mmax(T) and the tail probabilities P(Mmax(T) > M), that will occur in future time intervals of 10, 20, 50, 100 and 475 years are illustrated for source regions. The map of peak ground acceleration (PGA) zonation by probability level of 90% (in g) in rock bed for average return period of 50, 100 and 475 years is presented. According to the results, the maximum acceleration is estimated for the cities of Kermanshah, Ilam, Khorram Abad and Bandar Abbas which are related to NWZ1, NWZ2, NWZ3, NWZ7, NWZ8, SZ3, SH1, PG1 and PG2 sources. Finally, the results of this study are compared with obtained results of non-source approach.