The Role of Water Content in Electrical Resistivity Tomography of a Vertisol

Electrical resistivity tomography (ERT) has been suggested as a noninvasive approach for measuring cracks in Vertisols; however, its use has been limited, in part due to a lack of knowledge of the interactions between soil water content and electrical resistivity in cracking soils. In this study, we examined the relationship between soil water content and electrical resistivity for a Vertisol under cracked and noncracked conditions. Electrical resistivity was measured on soil clods, where desiccation does not result in cracking. Results showed that soil matrix electrical resistivity could be predicted from soil water content using an empirical model (RMSE = 8.3 Ωm and r² = 0.84). On a 4.5- by 5-m area of a Vertisol, in situ measurements of bulk soil electrical resistivity and soil water content were made using three-dimensional ERT and neutron soil moisture meter, respectively. Using the empirical model from soil clod measurements and field measured soil water content, we predicted the electrical resistivity of the in situ soil. When cracks were absent, measured soil electrical resistivity fell within the prediction intervals of the empirical model. When cracks were present, average measured electrical resistivity between the 0.2- and 0.8-m depth exceeded predicted electrical resistivity by 15,200 Ωm. The increased electrical resistivity measured on cracked soil cannot solely be attributed to lower soil water content. We, therefore, concluded that cracking directly increased the bulk soil electrical resistivity. Additionally, these data showed that for soils with and without cracks, in situ measurements of soil electrical resistivity could not predict soil water content within the accuracy of the neutron soil moisture meter. Observed differences between in situ electrical resistivity of the soil matrix and bulk soil indicate that ERT might be utilized to nondestructively map soil cracking.