The aims of this project was to assess the capability of SWAT model and SWAT-CUP software in hydrological simulation and to evaluate the uncertainty of SWAT model in estimating runoff in arid and semi-arid watersheds. Model calibration and uncertainty analysis were performed using the Sequential Uncertainty Fitting (SUFI2) algorithm. In the stage of calibration and validation of water flow, performance of SWAT model was evaluated using R², bR² and NS coefficients between observed and simulated records. Based on the results, R², bR² and NS were estimated to be respectively 0.75, 0.59 and 0.67 in the calibration stage, and 0.46, 0.24 and 0.42 in the validation stage. The results showed that the performance of the model is weak during calibration. Also output of the model is acceptable but performance is not significantly high in general. The most important reasons behind the poor accuracy are lack of data on water diversion for agriculture, irrigation use, and cropping pattern within the watershed.