Soil Water Sensor Response to Bulk Electrical Conductivity.

Soil water monitoring using electromagnetic (EM) sensors can facilitate observations of water content at high temporal and spatial resolutions. These sensors measure soil dielectric permittivity (K_a) which is largely a function of volumetric water content. However, bulk electrical conductivity (σ_a) can dominate the low frequency loss spectrum in soils, masking changes in the real permittivity and thus water content. All EM sensors exhibit some degree of sensitivity to bulk electrical conductivity. The objectives for this talk are to review the origin of the K_a dependence on σ_a for broadband and capacitance technologies, present study results evaluating sensitivities of selected sensors in two contrasting soils, and discuss implications for sensor calibrations for use in soil water management. Study results demonstrated that time-domain reflectometry (TDR) measures K_a at an effective frequency of 160 MHz for a clay loam and 520 MHz for a sand. Accordingly, the slope of the permittivity – bulk electrical conductivity relationship (ΔK_a/Δσ_a) for TDR averaged 3.5 m dS^-1 in clay loam, similar to a capacitance sensor running at 70 MHz. However, in sand, TDR sensitivity to σ_a was negligible. Permittivity measured using a digital time-domain transmission probe exhibited little or no sensitivity to σ_a (< 0.32 m dS^-1) in both clay loam and sand. Including σ_a in the calibration equation can improve water content estimation. However, such calibrations can be impractical for sensors that exhibit nonlinear K_a responses to σ_a or measure permittivity and bulk electrical conductivity in differing or disproportionate sampling volumes.