Response of Three Soil Water Sensors to Variable Solution Electrical Conductivity in Different Soils

Abstract

Commercial dielectric soil water sensors may improve irrigation management by providing continuous field soil water information. The accuracy of these sensors may be compromised, however, when low quality, high electrical conductivity irrigation water is used. The magnitude of this effect is expected to vary among sensors depending on the sensor measurement frequency (f), decreasing with increasing f, thus providing a potential criterion for sensor selection. We examined the effect of varying solution electrical conductivity (σw) on the sensor‐measured dielectric permittivity (εs) of the following three soil water sensors: the WET (f = 20 MHz), 5TE (f = 70 MHz), and ML2 (f = 100 MHz). We also evaluated the efficacy of using a simple, cost effective, two‐point calibration technique (CAL) to adjust for differences among soils. We found that εs varied dramatically, in some cases more than 15 dielectric units, among sensors in different soils under uniform soil water and σw conditions. The relative εs response of the WET and ML2 sensors was consistent with their respective f, and the 5TE response was not. Thus, the WET εs was almost always greater than the ML2, while the 5TE was generally less than the ML2. In all cases, the ML2 εs could be described with a single, commonly used calibration equation. We also found that varying σw from 1.2 to 6 dS m−1 had little effect on εs, indicating that a single calibration may work for that range of irrigation water quality. Finally, we found that the CAL procedure, as evaluated with the RMSE, was effective for all soils and sensors.