Ground-Motion Prediction Equations for the Geometric mean of Horizontal Component of Peak Ground Acceleration, and 5%-Damped Pseudo-Spectral Acceleration at Spectral Periods between 0.01 s and 8.0 s for Indo-Gangetic Plains, India

Indo-Gangetic Plains (IGP) and the adjacent Siwalik Himalaya are the seismically most vulnerable regions due to their proximity to the Himalayan seismic zone. In the absence of any authentic Ground Motion Prediction Equation (GMPE) for this region, it is not possible to assess the seismic hazard risk. In this study, an attempt has been made to derive an empirical attenuation relationship by applying nonlinear regression analysis, which would predict Peak Ground Acceleration (PGA) and Pseudo-Spectral Acceleration (PSA) on IGP. For this study, strong motion accelerograph database of PESMOS (Program for excellence in strong motion studies) was utilized to compile 37 earthquakes of magnitude (5.0 - 7.8) recorded at 211 seismic stations on IGP and its surrounding region with epicentral distance R_epi < 200 km. For PSA, 5%-Damped Elastic Pseudo-Acceleration Response Spectra for periods 0.01 s to 8.0 s of the recorded accelerograms were calculated. The functional form of a non-linear model was trained using an iterative estimation algorithm to determine the coefficients. The model includes site class, style of faulting, magnitude and closest distance to rupture surface as the explanatory variables and the natural logarithm of the geometric mean of the two peak horizontal components of the PGA/PSA as the response variable. For the fitted model, PGA/PSA values were calibrated with past earthquakes on three different sites (alluvium, soft rock, and hard rock), and the result shows that about 50% of the variation in PGA/PSA is explained by the explanatory variables. It can be further tested through simulation of ground motion on IGP to increase datasets and develop GMPE with recorded and simulated database combined that would predict ground motion parameters with great accuracy. GMPEs will be helpful to predict ground motion parameters to compute the seismic hazard risk on the IGP and the adjacent Siwalik region.