Investigation of SWAT Model Efficiency in Spatial Analysis of Runoff in Watersheds without Local Soil Data (Case Study: Damghanrood Watershed)

In semi-arid area the dangerous floods are usually generation by runoff. Therefore, spatial analysis of runoff is necessary for crisis management. Hydrological models are useful tools for simulation of water cycle-related processes and predict future events. Among the hydrological models, SWAT is one of the most popular of these models which the soil map and information are one the most important input data in SWAT model. On the other hand, the qualitative and quantitative soil maps are not available in most of the watershed in Iran. The main purpose of this study was to determine the spatial analysis of runoff generation using global soil data and using SWAT model in Damghanrood watershed in Semnan province. For this purpose, climatic data from 2008 to 2018 were used to simulate runoff. The data from 2010 to 2014 was used for model calibration and data from 2015 to 2018 was used for model validation. Sensitivity analysis, calibration, validation and model uncertainty were performed in SWAT-CUP software using SUFI-2 algorithm. The model was evaluated with coefficients of determination (R2) and Nash Sutcliffe (NS) statistics. The parameters of curve number, mean slope length, manning channel roughness coefficient and saturated soil hydraulic conductivity were determined as more sensitive parameters. The values of the R2 and NS statistics were 0.48, 0.47 for the model calibration stage and 0.46 and 0.45 for the validation stage, respectively. The results showed that the performance of the model in runoff simulation with global soil data was acceptable. Spatial analysis of runoff was performed using the average weight of runoff per unit area. The results of spatial analysis of runoff showed that the most important sub-basins in runoff generation was sub-basin number 5 and the least important sub-basins were number 1 and 11. Results indicated that sub-basin location and land use were most effective variables in runoff prioritization. SWAT model divide the watershed into the homogeneous units, therefore, it is possible to determine the critical areas of runoff generation.