Effect of Earthquake Catalog Declustering on Seismic Hazard Assessment

Abstract

The paper investigates the effect of different methods for preprocessing earthquake catalogs (declustering, i.e., removal of dependent events from them, and selection of the magnitude of completeness) on seismic hazard assessment. Seismic catalogs of the Kamchatka and Caucasus regions have been used for the analysis, because synthetic catalogs do not always reflect the real features of regional seismicity. Test sites were selected for these regions. Three declustering methods are considered that leave different numbers of events in the catalogs. The plotted seismic hazard curves indicate a complex interaction of catalog declustering and selection of the magnitude of completeness. Since both methods affect the b-value, it cannot be predicted in advance, which will lead to an increase or decrease in the estimate for b. After it is applied, the declustering method leaves the largest number of events in the catalog, but does not always give the highest seismic hazard rating. Therefore, it is necessary to be extremely careful and attentive when declustering a catalog. At the very least, it should be borne in mind that the end result may be unpredictable.