Surface waves carry valuable information about subsurface structures but they are often considered noise in shallow seismic reflection (SSR) surveys. One reason for limiting the application of surface waves in these surveys is the deployment of high-frequency geophones, which can attenuate the surface-wave signals. However, in SSR surveys, multiple shot gathers sample the same region, providing opportunities to enhance surface waves using appropriate techniques. In this study, we apply a moving spatial window method to select seismic records from the SSR data and extract spectrograms from the selected data by the frequency-Bessel transform (F-J) method. Then, the spectrograms extracted from different shot gathers with the same spatial window are stacked together to improve the signal-to-noise ratio (S/N). Multimodal dispersion curves of Rayleigh waves are picked from the stacked spectrograms and used to construct a 2D S-wave velocity (VS) model, which is validated against borehole data. In addition, we compare the dispersion curves and the inverted VS model with those obtained from a surface-wave survey that overlaps with the SSR survey. Although the surface-wave survey uses low-frequency geophones, which are more sensitive to surface waves, the seismic reflection data yield better dispersion curves and a higher-resolution 2D VS model. We attribute this to the multiple seismic sources in the seismic reflection survey, which allows us to enhance the S/N by stacking spectrograms. Furthermore, we compare the spectrograms extracted by different methods, indicating that the F-J method is particularly effective for extracting dispersion curves from the SSR data. Our study determines that the resolution of the VS model derived from the SSR survey, using the F-J method and moving spatial window method, is even higher than that obtained from the seismic surface-wave survey.

You do not have access to this content, please speak to your institutional administrator if you feel you should have access.