Seismic magnitude conversion and its effect on seismic hazard analysis

Abstract

The aim of this study is to demonstrate the bias created in the seismic hazard studies due to the choice of magnitude scaling equations without any statistical basis. The earthquake catalogue of Tripura, India, has been used for the purpose of this study. The catalogue was homogenized using the various scaling equations suitable for the region. Then, the bias created on parameters, like the magnitude of completeness (Mc), a and b values of the Gutenberg–Richter recurrence relation, maximum magnitude (Mmax), and peak ground acceleration, was demonstrated. The standard deviations of Mc, a, and b parameters were observed to be 0.23, 0.27, and 0.037 respectively. The maximum variations in the Mmax and ground motion estimates were found to be 0.7 magnitude units and 0.2 g respectively. Then, the robustness of the regional rupture characters in overcoming the observed variations has been demonstrated. The trend of the rupture behavior of the seismic sources seems to be unaffected by the change in the magnitude scaling equations. The Mmax calculated from the rupture-based procedure was observed to be higher than that calculated from the probabilistic method. This variation in Mmax estimation has been utilized to critically assess the suitability of the magnitude scaling equations for the particular study area.