Experimental Test of a Model of Water Uptake by Soybean1

Abstract

AbstractMechanistic models of soil water uptake have differed in the form of the water uptake term of Richard's equation. One of the most widely used uptake terms is that proposed by Hillel et al., although a detailed study comparing predicted soil water contents with those observed experimentally has been lacking. The purpose of this work was to develop a model that would test the accuracy of this water uptake term on a seasonal basis at various levels of water stress. Simulated soil water contents were compared with water contents observed biweekly in a 3‐yr field experiment that included three moisture treatments: irrigated, normal rainfall, and plastic covered. When a longitudinal resistance was included, 69 to 88% of the predicted soil water contents fell within the 95% confidence limits of the observed values from the irrigated and normal rainfall treatments in 3 yr. In the plastic‐covered treatment, however, only 17 to 49% of the predicted water contents were within the confidence limits. We conclude that the uptake term is accurate under normal and well watered conditions, but not under severe water‐stress conditions. Sensitivity analysis showed that the model was sensitive to net radiation, LAI, and rainfall and insensitive to hydraulic conductivity and root density.