Imaging a near‐surface feature using cross‐correlation analysis of multi‐channel surface wave data

In this study, we demonstrate that cross‐correlation gathers of multi‐channel and multi‐shot configurations provide accurate estimations of shear‐wave velocity (Vs) perturbations from Rayleigh wave data for the reconstruction of two‐dimensional (2D), high‐resolution velocity distributions without requiring the systematic calculation of surface wave dispersion, as in the spectral analysis of surface waves (SASW). Data acquisition for cross‐correlation analysis is similar to that for a 2D seismic common midpoint reflection survey. The data processing involved is similar to the coda wave interferometry used for seismological data but differs in the sense that the cross‐correlation of the original waveform is calculated for active source seismic data. Data processing in cross‐correlation analysis consists in the following three steps: first, cross‐correlations are calculated for every trace in each shot gather with the same offset trace of a reference shot to flatten the linearly sloping events of surface waves. A common receiver location stack section can then be obtained, which allows for the assessment of lateral variations in the elastic properties of the medium. Second, the maxima of the time‐shifted cross‐correlation gathers and the maxima of the shot gathers’ envelope traces are picked for different frequencies and trace values having the same receiver location are averaged to calculate Vs perturbations, as described in the theory of coda wave interferometry. Finally, a 2D Vs profile is reconstructed by applying the lateral Vs perturbation to a homogeneous Vs velocity profile obtained by inversion of the surface wave dispersion of the reference shot gather. Analyses of waveform data from numerical modelling and field observations indicate that this new method is valid and greatly improves the accuracy and resolution of near‐surface imagery using surface waves and reconstructed subsurface velocity distributions compared with a conventional SASW or multi‐channel application.