Performance evaluation of methods for the estimation of soil hydraulic parameters and their suitability in a hydrologic model

Direct and indirect methods are available to estimate soil hydraulic properties, which are prerequisites to solve water movement equations. Direct methods are usually time consuming and require complex measuring devices, and a skilled operator compared to indirect methods. Indirect methods, e.g., pedotransfer functions (PTFs) estimate parameters based on easily obtained soil texture data, have lately been widely used. However, the reliability of these parameters should be scrutinized, because PTFs are based on general data sets, and verification with true field data is often lacking. In this study, we investigate how well direct and indirect methods estimate soil hydraulic parameters that are used as input in hydrological models to simulate water balance variables, such as soil water-content results, are compared to experimental data for a practical problem. The data sets are collected from different sources which are classified as aggregated and plot-specific soil and soil water data. Three different methods: (1) regression analysis to obtain PTFs, (2) artificial neural networks to obtain PTFs, and (3) parameter optimization to fit retention and hydraulic conductivity data were executed to estimate parameter sets from these data sets. The parameter sets from these estimation methods were used in MIKE SHE, a physically based hydrologic model, to simulate water content for 3 years at an agricultural site in the Central Valley, California. Simulated water content from MIKE SHE was similar to measured soil-water content only if the soil hydraulic parameters were representative of the site. Parameter sets from the plot-specific data with direct estimation, and the aggregated soil texture with average hydraulic parameter performed better than any other methods in simulating soil-water content.