Study on linearized inversion of surface wave dispersion for estimating near-surface shear wave velocity

Abstract

Abstract Near-surface shear wave velocity is very important for geophysical and geotechnical engineering. The use of surface wave for characterization of the near-surface subsurface involves three steps: acquisition, extracting dispersion curve, and inversion method to generate near-surface shear wave velocity. In this study, we explored a linear inversion technique to see the effect of some parameters such as the initial model and the damping factor on the inversion results. We used the Levenberg-Marquardt algorithm and written in Matlab. Synthetic data used three-layer earth model with assumed homogeneous, isotropic, flat, and half-space. Synthetic examples demonstrated that inverted result depends on the initial model. Based on our results, we have ranged from 30% to 120% of the true model in providing the initial model. It is found that there is an asymmetry between the range of the initial model which is larger and smaller than the true model. This is due to the non-linearity of the differential equation and the solution of the Jacobian matrix using numerical approach. There are several suggestions for solving this problem: (1) using an analytical approach, (2) higher order deductions when using numerical approaches. In term damping factor, our result show range between 10−2 to 1 give a stable inversion.