Modelling and Assessment of Streamflow and Sediment Yield using SWAT under Varying Accuracy and Resolution of Input Data
Abstract
Digital Elevation Models (DEM) and Land Use/Land Cover (LULC) are critical inputs for the SWAT model, providing a spatial framework for hydrological simulations. However, the combined influence of their sources and resolutions on hydrological outputs such as streamflow and sediment yield remain not widely explored. This study systematically evaluated the impacts of seven DEMs and four LULCs, varying in spatial resolution and source, through 28 SWAT model configurations. These models underwent monthly calibration and validation to assess performance and the influence of input variables. Results highlighted that DEM resolution is crucial for watershed delineation, while source has minimal impact. Finer DEMs consistently delineated larger areas, with SRTM yielding slightly broader regions. Maximum altitude correlated directly with DEM resolution, whereas minimum altitude showed an inverse relationship. Finer DEMs generated more Hydrological Response Units (HRUs), with Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) yielding the highest HRU count. LULC resolution significantly influenced HRU numbers based on the number of classes. Streamflow analysis revealed no clear linear relationship between DEM and LULC resolutions; the most accurate predictions did not always stem from the highest-resolution data. For sediment yield, finer DEMs generally produced higher outputs, with ALOS-30 m and SRTM 90 m yielding the highest sediment levels, 0.4% and 0.2%, respectively. However, no consistent trend emerged for LULC’s impact on sediment yield. These findings underscore the importance of carefully selecting input resolutions for SWAT modelling and provide crucial insights to bridge existing knowledge gaps in hydrological modelling.