Three-dimensional spatial and temporal monitoring of soil water content using electrical resistivity tomography

In this paper, we propose a noninvasive method for monitoring three-dimensional (3-D) spatial and temporal variations of soil water content in the field, soil moisture tomography. The basic idea of the method originates from Archie's relationship between soil resistivity and water content. Initially, 88 electrodes were densely buried within a 3.5 m x 3.5 m square area, and potentials at the electrodes were measured by pole-pole and Wenner array methods at given time intervals. An inversion calculation of the 3-D soil resistivity was then conducted based on these potential data. Next, 46 soil samples were taken at representative positions in the square, and the parameters in the Archie's relationship were measured in the laboratory. Then, the 3-D distributions of the parameters were obtained by a distance weight interpolation method. Finally, based on Archie's relationship and the 3-D distribution of the soil resistivity and the related parameters, 3-D distributions of soil water content were calculated. To evaluate the obtained water content, the calculated water contents were compared with those measured by heat-probe-type soil moisture sensors, and a comparison between the spatial distribution patterns of calculated water content and soil bulk dry density was conducted. The 3-D variations of the calculated water content during a rainfall event were also analyzed. The results show that there are +/-0.10 cm(3)/cm(3) errors in the calculated water content, but between the calculated and the measured water content there exists a good linear relationship. It is possible to use the calculated water content to analyze the very general 3-D distribution characteristics of the soil moisture and investigate the 3-D rainfall infiltration process, the redistribution of soil water after rain, and other hydrological processes in the field. The proposed method is preferred for porous media where the water resistivity is relatively stable.