An Artificial Statistical Method to Estimate Seismicity Parameter from Incomplete Earthquake Catalogs, a Case Study in Metropolitan Tehran, Iran

Abstract

Uncertainties in earthquake catalogs, earthquake recurrence parameters, and in the variation of ground motion parameters are often considered in the evaluation of seismic hazard analysis. The purpose of this study is to develop an artificial statistical procedure based on Bayes’ formulation and weighted bootstrap sampling to estimate seismicity parameter (b-value of the Gutenberg-Richter law) from both historical and instrumental data in a given region. The procedure allows for uncertainty in the period of completeness, and assigns different weights to historical seismicity as compared to instrumental seismicity. Variation of seismicity within seismic sources is allowed with this procedure. This variation generalizes the condition of spatially homogeneous seismicity within seismic sources and permits accurate representation of historical seismicity. As a case study, the earthquake catalog of the greater Tehran, Iran is considered to estimate seismicity parameters as well as probabilistic seismic hazard analysis (PSHA) using the proposed procedure, and then the results are compared with those obtained from a conventional PSHA method. This comparison confirms the applicability of the procedure used in this study.