Comparison and Assessment of Two Lumped AWBM and Semi-Distributed SWAT Models in Monthly Runoff Simulation of Gharah-Sou River in Kermanashah Province, Iran

In recent years determination of different components of water balance has been the focus of many hydrological modeling techniques. In most basins, it is not feasible to measure all the different quantities for a detailed and enhanced hydrological modeling. Therefore, it is necessary to select a model capable of simulating hydrological events with the least number of variables; while being simple to use. Hence, in this paper the monthly runoff of Ravansar Sanjabi basin, Kermanshah, Iran was simulated through AWBM and SWAT models. AWBM is a lumped model simulating the runoff in basins using rainfall and evaporation variables. On the other hand, SWAT model is a continuous and semi-distributive model, which can simulate the hydrological processes in basins through a wide range of information such as physical information of basins (soil, land use, slope) as well as weather data such as precipitation, temperature, wind, relative humidity, and solar radiation. Simulation results during the calibration and validation periods were evaluated through two statistical indices: Nash–Sutcliffe efficiency (NSE) and coefficient of determination, R2. Comparison of calculated statistical coefficients showed that SWAT model has better results in simulating monthly runoff in calibration and validation periods so that the calculated NSEcoefficient was equal to 0.7 and 0.81 respectively and 0.63 and 0.5 for AWBM model respectively.