COMPARISON OF PHASE VELOCITIES AND SHEAR-WAVE VELOCITY INVERSION RESULTS OF SASW METHOD OBTAINED BY UNIFORM RECEIVER SPACING ANALYZED BY SurfSeis PACKAGE SOFTWARE WITH NON-UNIFORM RECEIVER SPACING ANALYZED BY THE GENETIC ALGORITHM INVERSION SCHEME

Conformity of inverted shear-wave velocity profile with the corresponding real profile of the earth has been a place for debate in seismic survey methods such as multi-channel spectral analysis of surface waves (MASW) due to deficiencies such as the non-uniqueness associated with the inversion process. The non-uniqueness problem can be eliminated by introducing additional information and constraints such as the optimum range of shear-wave velocity and introduction of higher modes. For careful implementation of higher modes, the receiver array length should be long enough to avoid artifacts in the resulting phase velocity dispersion data which will result in using a large number of receivers and an increased deployment, handling, and analysis efforts. In this study, we used a non-uniform spacing to form a longer array length, and we used an electrical vibrator generating sinusoidal wave as the source. The ensuing phase velocity dispersion data of such application is compared with the conventional MASW method with uniform receiver spacing and sledge hammer source. The phase velocity dispersion data for the conventional MASW method is obtained using the widely-used SurfSeis program which provides a basis for comparison. The corresponding dispersion data for non-uniform receivers spacing is obtained using the modified f-k method. Inclusion of higher modes in the inversion process demands a reliable forward method and an inversion algorithm. We used the Computer Programs in Seismology developed by professor Herrmann (St. Louis University) as a forward method to obtain multi-mode dispersion curve. We also used a Genetic Algorithm (GA) approach for the inversion process. Two study cases indicate that the experimental phase velocity dispersions are identical for both the uniform and the non-uniform receiver spacing. However, the SurfSeis program used to construct the experimental dispersion curves (uniform spacing receiver array and sledge hammer) is time consuming because user has to select the dispersion data for each shot separately. The results of both MASW applications are compared with the result of the downhole seismic survey, and it is observed that inverted soil profile using a GA can be more reliable to that of SurfSeis program considering the artifacts in inversion procedure and in higher modes calculations.