Comparison of Laboratory and Field Calibration of a Soil-Moisture Capacitance Probe for Various Soils

AbstractThroughout the American west, irrigated agriculture has been targeted to increase water-use efficiency because of increased urban demands. Soil-moisture sensors offer a method to achieve efficiency improvements, but have found limited use primarily because of high cost and lack of soil-specific calibration equations. In this paper, the Decagon EC-20 soil-moisture sensor (a low-cost capacitance sensor) has been examined and a unique laboratory-calibration method has been developed. Field-and laboratory-calibration equations were developed for six soil types (sand, sandy loam, silt loam, loam, clay loam, and clay) in the Middle Rio Grande Valley for alfalfa and grass hay fields. The average absolute error in volumetric water content for field calibration was 0.430 m3/m3, and 0.012 m3/m3 for the laboratory calibration. The factory-calibration equation for the EC-20 was also evaluated and found to yield an average absolute error of 0.049 m3/m3. In this study, it was found that the EC-20 is a reliable, cost-effective, and accurate sensor, and it is recommended that the laboratory-calibration method presented in this paper be used to obtain maximum accuracy. It is also recommended that the field calibration of the EC-20 soil-moisture sensor be foregone, as this type of calibration exhibits large error rates that are associated with colocation of samples, voids, organic residues, and root densities. Additionally, it was found that the field-calibration method was time-consuming, covered a small range of moisture content values, and was destructive to the area around installed sensors, which could lead to additional measurement errors.