Detection of soil thickness in forest land based on electrical resistivity tomographic scanning technology

Abstract: Soil thickness is the key index for the soil quality evaluation and has great effect on the productivity and properties of the forested soils. Soil thickness is commonly defined as the distance from the soil surface to the bedrock. The direct estimates of bedrock depth through drilling steel or from destructive boreholes and pits are too expensive and laborious in large areas, for the study area is often far away from the road and large tools are not accessible. Geophysical method seems to be a possible non-invasive and efficient alternative. Among available techniques, ground penetrating radar, seismic waves or sound waves often provide confusing or even wrong information, which prohibits the signal to go deeper from the shallower rocks. However, electrical resistivity tomography (ERT) offers interesting assets. The electrical resistivity depends on a number of soil textural and structural characteristics and can be regarded as a proxy of many soil properties. The electrical resistivity of the soil is found to be sensitive to the soil inclusion such as the tunnel or rocks. The ERT provides the electrical image of subsurface soil, and the result provided is easy to be explained and straightforward. It has been proved that the approach is suitable to investigate heterogeneous soils having components with contrasted electrical resistivity. This is often the case in soil between the soil matrix and the bedrock. In this research, the ERT was applied to the stony forested soils to study the soil thickness, and the applicability of the ERT was evaluated. Electrical resistivity of the soil samples overlying the bedrock was measured in the lab with four-electrode method and characterized as 48-640 Ωm. Electrical resistivity of the bedrock was obtained by data extraction from the ERT surveys and measured on site using four-electrode method, and the results showed that the electrical resistivity of the bedrock could be characterized as 1 200-1 600 Ωm. The two-dimensional ERT was carried on site with gradient electrode array and the results were inversed by RES2DINV; finally the inversed resistivity section were combined with the characteristic electrical resistivity of the bedrock and its overlying soils to estimate the soil thickness. The estimated results by the ERT were compared to the real soil thickness obtained by digging pits. The results indicated that the soil thickness varied between 0 and 2 m in the Mt-ventoux hilly area located in the south of France. The estimated results fitted well with the real soil thickness measured by digging pits with root mean square error (RMSE) of 0.2678 and R2 of 0.8223. Thus the excellent applicability of the ERT to the soil thickness evaluation was proved on the forested soils. In this research, the transition between the bedrock and soil surface was found in the field, and the electrical properties of the transition were studied. And still, the applicability of the ERT was verified on account of the transition between the bedrock and soil surface, which is not common in the literature available. The method proposed showed higher accuracy and indicated that the applicability of the ERT for soil thickness distribution evaluation. In the future research, the method will be tried to apply to the soil dry layer which is popular in the Loess plateau and to evaluate the applicability. Soil moisture and temperature has great effect on the soil electrical resistivity, so the soil moisture and temperature is suggested to be measured in the field in order that the results more fit the reality. The research will provide an important tool for the problems related to soil and will be an important aid for the soil quality evaluation and land utilization.