Characterization of strong‐motion recording sites in the Himalayas and adjoining areas

Abstract

In this paper, an attempt has been made to arrive at sufficiently realistic estimates of the average shear-wave velocity in the top 30 m of the ground (V S30 ) for 200 strong motion recording sites spread over the Himalayas, adjoining Indo-Gangetic plains and northeast Indian regions. This is achieved by extending the application of the horizontal to vertical spectral ratio (HVSR) method for the strong motion recording sites with a limited number of records. The knowledge of surface geology at the recording sites and the V S30 estimates based on the topographic slope method has been utilized as supplementary information to arrive at a good first-order approximation of V S30 for the sites with a single record also. A total of 669 three-component acceleration records have been obtained at the 200 sites considered from 158 different earthquakes in the magnitude range of 2.3-7.9 during 1986-2015. It had not been possible to use this database fruitfully due to lack of the required site characterization. Compared to the costly and time-taking field studies required for the estimation of exact site-specific values, the proposed scheme can be considered to provide a convenient methodology for obtaining sufficiently reliable approximations of V S30 from the recorded strong motion data. The V S30 values obtained in this study are found to be in very good agreement with the site-specific experimental values available for 29 of the sites, establishing the validity of the proposed scheme. An alternative site characterization is also provided in terms of the site soil and the local geological conditions, both defined in terms of three categories specified by indicator variables with values of 0, 1 and 2. Lee and Trifunac have illustrated that this characterization has stronger correlation with the ground motion amplitudes than the V S30 as well as the NEHRP site classes based on broad intervals of V S30 . The site characterization of the 200 strong motion recording sites in the present study will find useful applications in developing region-specific empirical ground motion prediction equations with the site amplification effects modeled realistically and in selecting suitable equations from the published literature using score-based data-driven methods.