Soil moisture characteristic curve (SMC) is a fundamental soil property and its direct measurement is tedious and time consuming.Therefore,various indirect methods have been developed to predict SMC from particle-size distribution (PSD).However,the majority of these methods often yield intermittent SMC data because they involve estimating individual SMC points.The objectives of this study were 1) to develop a procedure to predict continuous SMC from a limited number of experimental PSD data points and 2) to evaluate model predictions through comparisons with measured values.In this study,an approach that allowed predicting SMC from the knowledge of PSD,parameterized by means of the closed-form van Genuchten model (VG),was used.Through using Moharnmadi and Vanclooster (MV) model,the parameters obtained from fitting of VG to PSD data were applied to predict SMC curves.Since the residual water content (θr) could not be obtained through fitting of VG-MV integrated model to PSD data,we also examined and compared four different methods estimating θr.Results showed that the proposed equation (MV-VG integrated model) provided an excellent fit to all the PSD data and the model could adequately predict SMC as measured in forty-two soils sampled from different regions of Iran.For all soils,the method in which θr was obtained through parameter optimization procedure provided the best overall predictions of SMC.The two methods estimating θr with Campbell and Shiozawa (CS) model resulted in less accuracy than the optimization procedure.Furthermore,the proposed model underestimated the moisture content in the dry range of SMC when the value of θr was assumed to equal zero.θr could be attributed to the incomplete desorption of water coated on soil particles and the accurate estimation of θr was critical in prediction of SMC,especially for fine-textured soils at high suction heads.It could be concluded that the advantages of our approach were the continuity,robustness,and independency of model performance on soil type,allowing to improve predictions of SMC from PSD at the field and watershed scales.
展开▼
