Effect of Declustering Algorithm in Probabilistic Seismic Hazard Analysis for Kathmandu Valley

Abstract

An earthquake is a random event with uncertainty on time of occurrence, size of events, and location of rupture. So earthquake phenomenon is a model based on Poisson’s distribution. Thus, all the dependent event like foreshock and aftershock has to be removed before performing any Probabilistic Seismic Hazard Analysis (PSHA). Among the number of methodologies proposed for declustering methods, the process proposed by Gardener and Knopoff (1974) and Reasenberg (1985) was mostly used due to the availability of source code and reliability of declustering work. In this study, an attempt has been made to compare the effect of the declustering process at hazard level for the scenario of Kathmandu Valley. Probabilistic seismic hazard analysis of Kathmandu valley is performed using RCRISIS software with different seismicity parameters calculated for without declustering, declustering using Gardener and Knopoff (1974), and declustering using Reasenberg (1985). Calculated PGA value at bedrock level for 760 years return period for three declustering models compared with the measured square root of the sum of square (SRSS) PGA by 2015 Gorkha earthquake at Kathmandu valley at KTP station. Obtained results highlighted the suitability of the Gardener and Knopoff source code for the study area.