Estimation of buried pipe depth in an artificial soil tank using ground-penetrating radar and moisture sensor

A method is required for accurate determination of the location (horizontal position and depth) of subsurface utilities with minimal digging. Ground-penetrating radars (GPRs) have been used at many sites as a nondestructive method for detecting underground structures. Previous studies have examined the accuracy and resolution of GPRs based on comparisons of exploration images and factors such as soil, pipe material, and environmental conditions. Although numerous studies have explored the influence of relative permittivity and water content on GPR surveys, only a limited number have conducted direct measurements of water content using a moisture sensor within a soil tank and applied this data to enhance the precision of depth estimation. In this study, the effects of pipe material, relative permittivity, and antenna frequency on the depth estimation accuracy were quantitatively investigated using an artificial soil tank made of silica sand with known physical properties. In addition to the GPR survey, relative permittivity measurements were conducted using time-domain reflectometry (TDR) sensors located at different water levels. The results show that the antenna frequency and pipe material affect the resolution but not the depth estimation accuracy. Furthermore, the GPR survey using the relative permittivity from the TDR sensors estimated the pipe depths with a relative error of < 7.33% relative to the actual burial depth under the following conditions: a volumetric water content of 2.6-38.2%, the relative permittivity of 2.9-23.2, and buried pipes of 0.1 m in diameter and 1.5 m in depth. The study findings show that the selection of an appropriate frequency and the estimation of soil permittivity using a moisture sensor can improve the depth estimation accuracy of buried pipes by GPR survey.