Search for a command to run...
Soil-Water Characteristic Curves (SWCC) are fundamental in understanding the hydrological behavior of soils, particularly in field conditions where environmental factors significantly influence soil moisture and suction. This study aims to evaluate the performance of various SWCC models in fitting field-instrumented data collected from a compacted clay cover at a shallow depth. Moisture sensors and tensiometers were used to measure volumetric moisture contents and matric suction at shallow depths (0.3 m from the ground surface). The models considered include van Genuchten (vG), Fredlund and Xing (FX), Brooks and Corey (BC), Haverkamp (H), Kosugi (K), Gardner (G), and Campbell (C), among others, and their performance was assessed using metrics such as Index of Agreement (IA), Coefficient of Determination (R2), and Root Mean Square Error (RMSE). The study compared the metrics to identify the model that best represents the field-observed data. Results highlighted the similarity of different SWCC models in specific metrics; however, they demonstrated some degree of contrast when comparing the metrics. Among the models tested, the vG model was the best predictor for field observation. Its combination of high R2 (0.9482), low RMSE (0.01299), and good IA (0.9686) makes it the most reliable choice for capturing the soil-water retention behavior. Haverkamp (H) and Fredlund and Xing (FX) are close competitors, with FX excelling in RMSE (0.01294) and Haverkamp showing the best IA (0.9834). Kosugi (K), on the other hand, was the least effective model for this dataset.