Shear wave velocity of soils by the spectral analysis of surface waves (SASW) method.

Abstract

Spectral analysis of surface waves (SASW) method is an in-situ seismic method used for determining the thickness and elastic properties of soil and pavement. The SASW method is fast and economical to perform since no boreholes are required. The method is suitable for sites where the use of large equipment is difficult or where sublayer conditions make it difficult to perform other seismic tests. The SASW method is also ideal for preliminary field investigations to be conducted prior to more detailed site investigation, and for quality control and monitoring of ground improvement. The purpose of this research was to improve the SASW method by incorporating multi-mode propagation in the backcalculation procedure. In order to facilitate the investigation carried out in this study, two computer programs were developed to simulate SASW tests (and also Steady-State surface wave tests) and to calculate theoretical dispersion curves. The program for calculating theoretical dispersion curves was based on the root-searching procedure used in existing backcalculation methods. The computer programs developed in this study were used in a case study to demonstrate difficulties encountered by existing methods in dealing with multi-mode situations. It was shown that: (i) wavelength filtering criteria used by existing methods yield inconsistent (i.e. erroneous) dispersion curves when more than one propagation mode participate in the wave field, and (ii) backcalculation procedures based on root-searching cannot identify predominant propagation modes and hence fail to yield accurate results in the case of multi-mode propagation. Two developments were made in the present study to overcome the above difficulties. First, a new wavelength filtering criterion was adopted. In this criterion, the dispersion data point for a particular frequency is rejected (i.e. filtered out) if the values of phase velocity obtained from two different receiver-to-receiver spacings are not in close agreement. In this manner, inconsistencies that might result in the dispersion due to multi-mode propagation are avoided. Second, a new procedure to calculate the theoretical dispersion curve was developed. This procedure is based on the maximum vertical flexibility coefficient (at each frequency) of the theoretical layered model. Unlike root-searching methods, the maximum vertical flexibility coefficient method easily identifies predominant propagation modes. A computer program was developed in this study for backcalculation of SASW data based on the flexibility coefficient method. Least-squares optimization using the down-hill simplex method was also implemented in this program to automate the backcalculation process. The accuracy of the above proposed procedures was demonstrated using SASW field tests. The shear wave velocity profiles obtained using the procedures developed in this study are in good agreement with those obtained from other in-situ seismic tests. (Abstract shortened by UMI.)