Assessment of the impacts of different Digital Elevation Models on runoff and ‎suspended sediment estimations using SWAT model (Case Study: Galikesh ‎watershed, Golestan Province)‎

Simulation of runoff and sediment in watersheds require different modelling approaches, each provided for certain condition. Semi-distributed hydrological model (SWAT model) is one of the most widely used modelling approaches in this context. Among the most important spatial information needed by SWAT model, is the map of digital elevation model (DEM) which plays an important role in the model results for the exploration of hydrographic properties of watersheds, estimation of the spatial distribution of runoff and sediment load and its accuracy. The aim of this study was to evaluate the effect of different digital elevation models on runoff and suspended sediment in Galikesh watershed, Golestan Province. For this purpose, three types of DEM with the spatial resolutions of 30, 90 and 1000 meters were selected and for a period of 27 years, SWAT model was implemented. SWAT-CUP software and SUFI2 method were used for the model calibration and validation. Nash-Sutcliffe (NS) criterion for discharge and in calibration phase (1990-2007) for the resolutions of 30, 90 and 1,000 meters, was obtained respectively, 0.63, 0.63 and 0.62 and for suspended was obtained respectively 0.69, 0.68 and 0.67 which is considered satisfactory, given the presumed numerical ranges acquired from the previous studies. The amounts of watershed’s annual runoff for the resolutions of 30, 90 and 1000 meters, were respectively, 11.23, 11.30 and 11.39. The results that the use of different of digital elevation models would result in essentially similar results in runoff and sediment estimations, although changes in suspended sediment was slightly overestimated. This is attributable to the inverse modelling logic and lack of considering the parameters which arise directly from the DEM map. , due to data limitations in the Swat model calibration, the digital elevation model used to simulate the runoff did not have much impact on runoff and sediment estimations.