Stability Assessment and Quantitative Evaluation of H/V Spectral Ratios for Site Response Studies in Kumaon Himalaya, India Using Ambient Noise Recorded by a Broadband Seismograph Network

The Kumaon Himalaya region in India has accumulated considerable seismic risk potential from moderate to high seismicity due to ongoing tectonic evolutionary processes. To assess conditions of high seismic risk arising from local site effects at populated locations, we applied the empirical horizontal to vertical (H/V) spectral amplitude ratio method (Nakamura in Quarterly Reports of the Railway Technical Research Institute Tokyo, 30:25-33, 1989) using ambient seismic noise recorded by a network of 32 digital broadband seismographs from June 2005 to June 2008. The data and the estimated parameters were subjected to stability tests to assess the effect of seasonal variations. Seasonal variations in the ambient noise data seemed to show a lesser effect on fundamental frequency estimates and a slightly greater effect on the peak H/V amplitudes. Validation of the ambient noise results was done by complementary tests using H/V ratios of local and regional earthquakes. The 'peak' corresponding to the fundamental resonance frequency is prominently present in both the ambient noise and the earthquake datasets. The study locations showed distinct H/V curve topologies, corresponding to the effects of both velocity contrast at well-defined frequencies and characteristic topographic effect around a certain frequency range. The clearly indicated 'peaks' in the H/V curves possibly correspond to velocity contrasts created by weathered sediments overlying hard rock basements in rocky hills. Our study indicates high site responses at many populated locations near the surface trace of the seismically active Main Central Thrust (MCT) and other fault boundaries. The fundamental resonance frequencies estimated from the site response studies at the 32 locations could be useful in preliminary site characterization, ground motion prediction and seismic hazard analysis.