Development of a method for estimating soil moisture using obliquely reflected light in two-dimensional soil experiments

Slope failure, groundwater behavior, and other geotechnical and hydrological phenomena are commonly investigated through two-dimensional (2D) physical model experiments. A method using reflected light from the front (MRF) has recently been proposed as a low-cost and noninvasive method for the volumetric water content and/or degree of saturation in 2D soil experiments. However, the applicability of the MRF to various soils—particularly when dye enhancement is absent—remains uncertain. Herein, the relationship between the average gray value (G-value) obtained from the MRF and the volumetric water content in 2D models of four soils with different colors and grain sizes was investigated. The results indicated that the MRF was less sensitive when applied to dark soils and in low-moisture environments, limiting its general applicability. Therefore, we proposed a new method—a method using reflected light from oblique angles (MRO)—quantifying the water content from the dark pixel proportion (black value) in images. The black value was found to be clearly and monotonically related to the volumetric water content across all soil types, including black soils, demonstrating the robustness and versatility of the method. Although the MRO should be further optimized for soils with extremely high or low moisture levels, it guarantees the practical visualization and quantification of the soil moisture in 2D model experiments.