Site Effects of Linear Method and Modified Horizontal-to-Vertical Spectral Ratio

In order to further study the influence of site effects from engineering ground motion, a comparative analysis of site effects from ground motion in the Sichuan strong earthquake area was carried out based on two empirical methods, namely, horizontal-to-vertical spectral ratio (HVSR) and linear inversion. Firstly, 233 sets of strong motion waveforms captured by 17 strong motion stations during the aftershocks of the Wenchuan Earthquake were selected based on the magnitude, station, distance from the epicenter, and other factors. Second, the linear inversion method was used to estimate the quality factor Q_s of the S-wave based on the hinged three-stage attenuation model in the Longmenshan area. Meanwhile, two methods, rotational HVSR and multidirectional HVSR, were applied to explain the mechanism of the site effect of azimuthal angle on HVSR in seismic wave propagation. Finally, the possible reasons for the variability of the site effects calculated by the linear inversion and HVSR methods were analyzed, based on which an improved HVSR method was proposed to improve the accuracy of the site effect assessment. The results show that the quality factor of the S-wave in the area is frequency-dependent and approximately equals to 199.2f ^0.8 in the range of 0.4–20.0 Hz. The results of HVSR may have a certain dominant direction, and the site effect may increase suddenly at a certain angle with the change of the azimuthal angle, which may be related to the anisotropy of the soil layer at the site. The amplification of vertical ground motion, i.e., the vertical site effect formed by the coupling of soil layer at the site, engineered bedrock, or middle and deep hard rock layers, is the main reason for the difference between HVSR and linear inversion methods.