Estimation of Streamflow with Incomplete Soil Dataset in Krasioa Basin Using Soil-Landscape Evaluation Approach and SWAT Model

Abstract

Data on soil properties are indispensable for process-based hydrological modeling. Soil information of Thailand is primarily provided by the Land Development Department (LDD), nevertheless soil property data are available only in arable land whose slope is less than 35%. The steep-slope land was generally labeled as Slope Complex (SC), there is no information available. This paper demonstrated the application of soil-landscape evaluation approach for predicting the missing properties of soil which resulted on enhancement of model performance in streamflow estimation in Krasioa Basin by the Soil and Water Assessment Tool (SWAT) model. The physical properties of soil-soil thickness, fraction of soil particles (clay, sand, organic matter) were predicted using the Soil-Landscape Estimation and Evaluation Program (SLEEP). The additional properties of soil including bulk density, hydraulic conductivity, and available water content were estimated using the pedo-transfer functions (ROSETTA). It was found that SLEEP model could provide consistent information on physical properties of soil. The SWAT model performance in streamflow simulation at the Krasioa Reservoir was improved using the proposed approach. Appropriate model inputs can generate reasonable output. Model performance can further be improved by calibration.