Abstarct: Abstract Due to the impact of climate change and human activities, water shortages have become increasingly serious and water problems have become an obstacle for sustainable socio-economic development. Water is a key factor in the formation and development of civilizations. The requirement for the implementation of integrated management of water resources is to understand the complex processes within the watershed. The processes of the watershed finally end in runoff, and simulation of runoff can be a suitable solution to deal with problems such as floods and droughts. Accurate estimation of runoff is considered a central operation in the field of water resources management, and its prediction plays an important role for optimal exploitation of a hydrological system. In this research, after evaluating the IMERG satellite product of the Final Run type in the period from 2008 to 2017 at the location of five rain gauge stations and one synoptic station in and around the Taleghan watershed on a daily and monthly time scale, the monthly runoff of the basin was simulated using from the water and soil assessment model (SWAT) and the conceptual model of the water balance of Torrent White. Runoff simulation was carried out in the ten-year statistical period from 2008 to 2017. SWAT model was calibrated by SUFI-2 algorithm in SWAT-CUP software package. Also, RMSE, R2, KGE and NSE coefficients were used to calibrate the Thornthwaite conceptual model. The evaluation of the accuracy of satellite precipitation products has shown a better performance of the results on a monthly scale than on a daily basis. The results of the SWAT model also showed that the model baxsed on the satellite precipitation product performed better than the ground-baxsed model in simulating runoff; So that the Nash-Satcliff coefficient of the model baxsed on the satellite product in calibration and validation was 0.85 and 0.86, respectively. Also, the simulation results of the Thornthwaite conceptual model showed that the model baxsed on the satellite product has a better performance than the ground data; So that the Kling Gupta efficiency function for the simulation baxsed on satellite and terrestrial product was obtained as 0.38 and 0.23, respectively. In total, the results obtained from both models indicate the better performance of the SWAT model.