Estimated relative permittivity of contaminated laterite soil: An empirical model for GPR waves

Abstract

Estimated relative permittivity performed on soil is essential for forecasting the performance of Ground Penetrating Radar (GPR) in an in-depth manner. This study investigated and verified the empirical relationship model between relative permittivity and volumetric water content in soil to predict the relative permittivity of contaminated laterite soil. In this study, a 24-hour measurement involving 800 MHz shielded antenna GPR was carried out in a concrete simulation field tank filled with Terap Red soil (1.5 m x 2.6 m x 1.5 m) at UiTM Perlis, Malaysia. Embedded moisture content probe was simultaneously measured to monitor the response of volumetric water content in contaminated soil in order to formulate an empirical relationship between relative permittivity and moisture content. The GPR data were pre-processed and filtered with Reflexw 7.5, while regression analysis was performed to evaluate the empirical relationship model. The model outcomes were retrieved from a number of cross-validation schemes, including correlation analysis (R2), root mean square error (RMSE), and calibrated Agilent Technologies Automated Vector Analyser (VNA). A third-order polynomial for analysis of variance (ANOVA) best fitted the model with positively strong correlation (R2=0.989, N=24, P < 0.01) and RMSE 0.003< RMSEpredicted < 0.19. Verification of the proposed model using calibrated VNA displayed exceptional agreement between 0.06% comparisons.