Seismic Hazard Assessment of Kashmir Region Using Logic Tree Approach: Focus on Sensitivity of PSHA Results Towards Declustering Procedures and GMPEs

Abstract

Probabilistic seismic hazard assessment (PSHA) for the Kashmir region located in the northwestern Himalayas has been performed to quantitatively estimate the probability of exceedance of various ground-shaking levels. An updated earthquake catalog composed of 7826 events was prepared by combining historical events (from 1250 BC) and instrumental events (1900–2020). Kijko's maximum likelihood technique yielded seismicity parameter b-value 0.92–1.05 and mmax ~ 7.98 for the entire Kashmir region. A comparison of three seismicity delustering methods has been presented based on the resulting seismicity parameters. PSHA computations were conducted using RCRISIS software based on a logic tree approach to account for the model uncertainties in attenuation models and epistemic uncertainties due to declustering methods. Seismic hazard maps at bedrock for four return periods of 475, 950, 2475, and 4950 years were prepared using peak ground accelerations (PGA) as well as short- (0.2 s) and long-period (1 s) spectral accelerations (Sa). Sensitivity analysis of the computed hazard revealed the substantial effect of attenuation relationships as well as declustering methods on the outcomes. Furthermore, hazard curves and uniform hazard response spectra at each of the 1228 grid points were developed. The southwestern, northwestern, and northern regions of the valley including Pulwama, Shopian, Kulgam, and Budgam were found to have the highest hazard as opposed to the central regions like Ganderbal and parts of Baramula. Kashmir region was divided into five zones (ZA-ZE) of high to low seismicity with mean PGA values of 0.175, 0.258, 0.379, 0.456, and 0.514 g, respectively, for the 2475-year return period. © 2023, The Author(s), under exclusive licence to Springer Nature Switzerland AG.