Fitting Different Soil Water Characteristic Curve Models on the Experimental Data of Various Textural Classes of Guilan Province Soils

In addition to soil water content، soil matric potential is also important. Soil water characteristic curve (SWCC) describes soil water content and soil matric potential relationship. Many models have been developed to describe the SWCCs and their fitting to the experimental data. Their fitting accuracies in different soils، however، have been rarely investigated. In this study، 75 disturbed and undisturbed soil samples were taken from topsoils and subsoils of Guilan province and were partitioned into textural groups after textural measurements. Water contents were measured at 12 matric potentials. Thirteen well known and frequnetly applied SWCC models were fitted to the measured data. The most accurate models of total soil samples، each textural group and each textural class were determined using root mean square error، coefficient of determination and mean error. The Brooks and Corey، Kosugi and Gardner models had the least accuracies among all of the models. The results showed that Fredlund-Xing and Dexter models had better fitting accuracies than the van Genuchten model for total soil samples، textural groups and some of textural classes and showed superiority for the soils of Guilan province. Since، there was no significant difference between the fitting accuracies of the Fredlund-Xing and Dexter models on the empirical data، so Dexter model would be superior to the other models for the soils of Guilan province because of its simplicity.