Petrophysical relationship moisture-dielectric of lateritic soils infected by landfill wastewater using ground penetrating radar

Ground Penetrating Radar (GPR) is an effective non-destructive technique for assessing subsurface conditions, particularly in wastewater-contaminated soils at landfill sites. However, accurate detection remains challenging due to the limitations of existing empirical models linking dielectric and moisture content. This study investigates the dielectric behavior of wastewater migration in lateritic soils and develops a predictive petrophysical model to enhance GPR interpretation accuracy. Controlled laboratory and field experiments were conducted using Cherang Hangus, Terap Red, and Sandy soils with a 250 MHz GPR antenna. Standard geotechnical tests were performed to determine soil properties, while a Vector Network Analyzer (VNA) was used to validate GPRderived dielectric data. The developed third-order polynomial model demonstrated strong correlations between dielectric and moisture content, with R² values of 0.9883 for Cherang Hangus, 0.9489 for Terap Red, and 0.8763 for Sandy soil. Validation against the Topp, Malicki, and Roth models, as well as VNA measurements, yielded low RMSE values of 0.015 (Cherang Hangus), 0.031 (Terap Red), and 0.006 (Sandy soil). Field verification at Padang Cina and Rimba Mas further confirmed its reliability, with RMSE values ranging from 0.2613 to 0.4937. The proposed model establishes a more accurate dielectric moisture relationship for lateritic soils, enhancing GPR’s capability in detecting landfill wastewater and supporting sustainable landfill monitoring and groundwater protection in tropical environments.