A non-destructive detection of soil bulk density in reclaimed land of opencast coal mines using ground penetrating radar method

Abstract

Land reclamation of opencast coal mines is of critical for environmental restoration and sustainable land use. Soil bulk density (SBD) is an important index to evaluate the impact of land reclamation on soil quality in mining areas. However, the traditional soil sampling method is destructive and time-consuming. Therefore, developing a non-destructive, repeatable, high-precision, and efficient method for detecting the SBD of reclaimed soil in opencast mining areas is required. This study evaluated the feasibility of using ground penetrating radar (GPR) to estimate SBD in reclaimed mining areas. The south dump of the Antaibao opencast coal mine in Pinglu District of Shuozhou City, Shanxi Province, China, was selected for the study. The random Hough transform algorithm was used to identify automatically the hyperbolic reflection in radar images and obtain the soil characteristic information from different locations. Inverse distance weighted interpolation was used to identify and characterize SBD and the soil dielectric constant (SDC) in 3 different soil layers and 2 measuring points. The method provided the SDC with high precision. The Pearson correlation coefficient (r) between the estimated and measured SDC was the highest at sampling point S1 (0.935), and the root mean square error (RMSE) was the lowest (0.272, from 0.272 to 0.542), indicating the feasibility of using the SDC to characterize SBD. The r for SBD and SDC ranged from 0.689 to 0.857 at sampling point S1, and from 0.724 to 0.747 at sampling point S2. The estimated and measured SBD had different distributions at different soil depths. A numerical model describing the relationship between SBD and SDC was used for the non-destructive identification of SBD with high precision. The proposed method expands the application potential of GPR to detect soil properties and provides a theoretical basis and technical support for the non-destructive detection of soil physical properties using GPR. This study contributes advancing non-destructive soil assessment techniques and provides a practical tool for assessing and optimizing reclaimed soil properties in opencast coal mine rehabilitation projects.