Quantification of the resolution of dispersion image in active MASW survey and automated extraction of dispersion curve

Abstract

Multichannel analysis of surface waves is a non-invasive method in the field of geophysics which is used for identifying shear wave velocity profile of the subsurface along its depth. The reliability and accuracy of the obtained profile strictly depends on the resolution of the dispersion image obtained from an active survey. The dispersion image is a representation of the distribution of the energy of propagating waves on a frequency-phase velocity space. Based on the relative energy accumulation in the dispersion image, the fundamental (M0) and higher modes of dispersion can be identified. As there are several parameters that can affect the resolution of dispersion images, it is important to determine the optimum magnitude of each parameter that would contribute in generating dispersion images of higher resolution. Until date, the resolution of dispersion image is described only qualitatively, wherein fundamental mode dispersion curve was extracted based on the best possible visual discretion of the user. Erroneous selections of dispersion points often led to the misidentification of the dispersion mode, leading to incorrect shear wave velocity profile. This paper reports an image-processing based methodology to quantify the resolution of dispersion images and automated extraction of M0 dispersion curve through threshold energy filtering of the dispersion image. Such methodology removes the user discretion-induced subjectivity associated with the manual extraction of dispersion curve, thereby resulting in a robust extraction technique.