Regional spectral characteristics, quality factor and site responses in western-central Sichuan, China (I): Application of parametric generalized inversion technique

To study the source, site amplification and seismic wave attenuation characteristics in the Sichuan area, 1242 ground motion recordings corresponding to 62 earthquake events with Ms 4.3–6.4 recorded by 55 observation stations in the study area were selected. The source spectra, site effects and frequency-related quality factors in the western Sichuan area were inverted by the generalized inversion technique. Then, the source parameters, such as corner frequency, seismic moment and stress drop, were estimated by a grid-searching tool. In the generalized inversion method, a bedrock station was used to remove the trade-off between the site and source components. The inverted quality factor of 0.25–20 Hz frequency band is characterized as Q(f) = 219.4f0.7383. The results indicate that the inverted source spectrum is well matched with the ω2 source model, in which the stress parameter values range between 0.2 MPa and 3.5 MPa, with an average value of 1.2 MPa and source radii between 0.3 and 4 km. In addition, a comparison between the inversion results obtained from the horizontal-to-vertical spectral ratio and the generalized inversion methods indicates that these two approaches actually describe the same level of site effect, while the averaged site effect obtained by the generalized inversion method is always greater than the result obtained by the horizontal-to-vertical spectral ratio method. Finally, a residual defined as a function of hypocentral distance and moment magnitude was used to investigate whether there was any bias in the inversion results. The results indicate that the geometric spreading and source model used in this study resulted in an underestimation of the Fourier amplitude level in the far field and for large earthquakes. The results of this study can be applied in ground motion simulation and prediction to help the study of seismic disasters and risk assessment in the study area.