Comparative Assessment of SWAT Model Performance in two Distinct Catchments under Various DEM Scenarios of Varying Resolution, Sources and Resampling Methods

The aim of this study is to evaluate the performance of hydrological model - Soil & Water Assessment Tool (SWAT) in two distinct catchments, under various Digital Elevation Model (DEM) scenarios of varying resolution (from 30 m to 300 m), sources (SRTM, ASTER and CartoDEM) and resampling methods (nearest neighbour, bilinear interpolation, majority and cubic convolution) available with ArcGIS software package. A comparison was made in between model response of highly elevated Himalayan Upper Teesta catchment and peninsular monsoon dominated Upper Narmada catchment in India. Model performance was assessed & subsequently compared based on statistical measures such as coefficient of determination (R-2) and Nash-Sutcliffe Efficiency (NSE), for monthly runoff and sediment yield. The sensitivity of monthly model outputs of runoff, sediment yield, Total Nitrogen (TN) & Total Phosphorous (TP) towards DEM scenarios was studied based on Relative Difference (RD). The key findings of this study are: 1) topographic characteristics of Upper Teesta catchment were found to be more sensitive towards various DEM scenarios compared with Upper Narmada catchment; 2) model performance in simulating monthly runoff was found to be unaffected for both catchments due to changes in DEM resolution & resampling method; 3) in simulating monthly sediment yield, model performance was affected due to all DEM scenarios for Upper Narmada catchment, while scenarios of changing DEM resolution and resampling method have affected model performance for Upper Teesta catchment.