Evaluation of newly developed capacitance probes for continuous soil water measurement

Abstract

Precise and accurate measurement of soil water for monitoring crop water uptake, irrigation scheduling and management can be provided through continuous soil water monitoring, which is made possible with the aid of automated capacitance probes such as HydraScout. This study was carried out to develop HydraScout capacitance probe calibration equations for selected soil types and to evaluate the accuracy levels of the calibration equations for continuous measurement of soil water. The selected soil textural classes were sand, loamy fine sand, sandy loam, sandy clay loam, clay loam, sandy clay and clay. The volumetric water contents were plotted against the sensors readings to develop calibration equations. The best fit of the regression function was used on the basis of highest adjusted coefficient of determination (R2). Results indicated that calibration functions developed for the soils were linear (clay, clay loam and sandy soils) and polynomial (loamy fine sand, sandy loam, sandy clay loam, sandy clay soils). Volumetric water content was well predicted with the text-specific equation for clay loam (RMSE of 0.001 m3 m−3 and D-index of 0.99) and sandy loam (RMSE of 0.001 m3 m−3 and D-index of 0.99) soil textural classes. Prediction of volumetric water content for sandy clay produced the highest RMSE (0.009 m3 m−3) and lowest D-index (0.68) for all the soil textural classes. It was recommended that the text-specific laboratory calibration equations developed during this study should be used especially, for sandy loam soil type, in a situation where high numbers of the probes are in use. In the case where few of the probes are in use, each sensor must be calibrated.