Improving horizontal resolution of high-frequency surface-wave methods using travel-time tomography

Abstract

In surface-wave methods, horizontal resolution can be defined as the ability to distinguish anomalous objects that are laterally displaced from each other. The horizontal length of a recognizable geological anomalous body is measured by the lateral variation of shear (S)-wave velocity. Multichannel analysis of surface waves (MASW) is an efficient tool to determine near-surface S-wave velocities. The acquisition of the MASW method involves the same source-receiver configuration moved successively by a fixed distance interval (a few to several stations) along a linear survey line, which is called a roll-along acquisition geometry. A pseudo-2D S-wave velocity section is constructed by aligning 1D models, and each inverted 1D S-wave velocity model reflects the vertical S-wave velocity variation at the midpoint of each geophone spread. Although the MASW method can improve the horizontal resolution of S-wave velocity sections to some degree, the amount of fieldwork is increased by the roll-along acquisition geometry. We propose surface-wave tomography method to investigate horizontal resolution of surface-wave exploration. Phase-velocity dispersion curves are calculated by a pair of traces within a multichannel record through cross-correlation combined with a phase-shift scanning method. Then with the utilization of travel-time tomography, we can obtain high resolution pure-path dispersion curves with diverse sizes of grids at different frequencies. Finally, the pseudo-2D S-wave velocity structure is reconstructed by inverting the pure-path dispersion curves. Travel-time tomography of surface waves can extract accurate dispersion curves from a record with a short receiver spacing, and it can effectively enhance the ability of random noise immunity. Synthetic tests and a real-world example have indicated that travel-time tomography has a great potential for improving the horizontal resolution of surface waves using multi-channel analysis.