Automatic and adaptive picking of surface-wave dispersion curves for near-surface application

Abstract

Multichannel analysis of surface waves (MASW) is a widely used non-invasive method to reveal the S-velocity structure in near-surface applications. Automatic and adaptive picking of surface wave dispersion curves is crucial for the later surface wave analysis. The traditional way to acquire the dispersion curves is to identify the local extremum energy in phase-velocity images and manually pick the dispersion curve by following peaks at different frequencies. However, the large number of interactive works in extensive surveys could be more efficient, time-consuming, and prone to human error. In this work, we developed an automatic adaptive picking of dispersion curves (AAPDC) strategy, which automatically searches the surface-wave dispersion curve based on the position of the local energy maxima in the dispersion spectrum. This method is straightforward and efficient and requires only a few or no additional parameters to obtain a stable result. The typical synthetic model tests show that the automatic picking agrees with the theoretical dispersion curves. In addition, the proposed AAPDC method is used to land geophone data and DAS data. The final S-velocity tomograms have a good agreement with the geological and other geophysics results, which demonstrate that the AAPDC method could accurately and adaptive extract fundamental dispersion curves from the complex dispersion spectrum.