Influence of Wetted-Phase Structure on the Electrical Conductivity-Moisture Content Relationship

Knowledge of the relationship between the soil bulk electrical conductivity (EC) and moisture content is required to quantify subsurface moisture variations using electrical resistance tomography (ERT). Repacked field samples are typically used to establish the EC-moisture content relationship, because of difficulty removing and analyzing intact samples. However internal sample structure can influence EC measurements, as complicated wetted-phase structures can evolve during drainage, and repacking errors can lead to increased uncertainty in subsurface moisture contents estimated with ERT. We use a numerical approach to investigate the impact of wetted-phase structure on EC-moisture content relationships in layered and homogenized sandy materials. We simulate laboratory drainage in samples using a macroscopic invasion percolation model that includes capillary and gravity forces. Cell moisture contents are then mapped to cell-specific EC values using the constitutive model of Mualem and Friedman (1991). Resulting heterogeneous EC fields are then used in a finite-difference model to simulate sample-scale EC measurements. We consider EC measurements on heterogeneous layered samples oriented both perpendicular and parallel to layering and samples with no internal spatial correlation. Sample EC values are tend to be highest in samples oriented parallel to layering and lowest in samples oriented perpendicular to layering, while statistically homogeneous samples have moderate EC values. EC differences are greatest at intermediate saturation where they can exceed an order of magnitude.