Multivariate geophysical index-based prediction of the compression index of fine-grained soil through nonlinear regression

Abstract

The determination of consolidation settlement is highly significant in fine-grained soil. However, evaluating compression index (Cc) is time-consuming and cumbersome. For the approximate estimation of Cc, several correlations based on various geotechnical properties have been established, involving boreholes and sample extraction. In such circumstances, non-invasive geophysical techniques can provide a quick and cost-effective alternative in predicting the compression index. In this study, two geophysical techniques, namely electrical resistivity tomography (ERT) and multichannel analysis of surface waves (MASW), were coupled together to predict the compression index of fine-grained soil at Silchar, Assam, India. Electrical resistivity (R) and shear wave velocity (Vs) were determined from the ERT and MASW test respectively, and a nonlinear regression technique was employed to estimate the Cc. The R2 and RMSE values observed from the regression analysis are 0.874 and 0.036, respectively with R and Vs as independent variables. The multiple-geophysical index-based technique resulted in a 20% improvement in the R2 value of the nonlinear fit compared to that obtained using individual geophysical index-based predictions. The computation of statistical indices validates the regression results. Further, the geophysical index-based predictions of Cc were compared with the conventional method of determining Cc using the geotechnical parameters including liquid limit, water content, and void ratio. The concept and the proposed empirical equation can be implemented for different locations, thereby benefiting the researchers and the construction engineers.