Numerical Investigation of Geophysical Measurements for Liquefaction Triggering Evaluation in Soils Exhibiting Natural Spatial Variability

One of the first steps in assessment of liquefaction hazards consists of evaluating the potential for liquefaction triggering at a site. Procedures based on the standard penetration test (SPT) and cone penetration test (CPT) are often used. However, geophysical measurements of shear wave velocity (V-S) have also been used to evaluate liquefaction triggering since VS is a proxy for soil stiffness. Measurements of VS for these triggering evaluations are often acquired using downhole methods (e.g., seismic CPT or P-S suspension logging) or cross-hole methods. SPT, CPT, and borehole-based geophysical measurements provide data in the localized region surrounding the boreholes. This presents challenges when attempting to evaluate liquefaction triggering in natural soils that exhibit spatial stiffness variability. Recently, surface wave measurements have grown in popularity for geotechnical investigations and have been used to evaluate V-S profiles for liquefaction triggering analysis. However, surface wave methods typically apply wavefield transformations to evaluate dispersion characteristics. This essentially introduces a form of spatial averaging that can lead to uncertainty in V-S measurements when soils exhibit appreciable spatial variability along the survey line. One promising development in surface wave testing involves the use of full waveform inversion (FWI). FWI attempts to match the entirety of each recorded signal rather than a dispersion curve derived from the corresponding wavefield transformation. Therefore, FWI has the potential to offer broad spatial coverage while avoiding the inherent limitations of a dispersion-based surface wave approach. In this study, numerical modeling was performed to simulate wave propagation in soils exhibiting natural spatial variability. Comparisons were made regarding the extent to which different geophysical approaches could reliably estimate liquefaction triggering. The results demonstrated that FWI can outperform downhole/cross-hole measurements and a dispersion-based surface wave approach when implementing V-S-based liquefaction triggering procedures in spatially variable soil conditions.