Spatial variability of input data for distributed and physically based models has a significant impact on the results of model simulations. The objective of this study was to compare the methods of precipitation spatial interpolation, namely, the method of Thiessen polygons (THI), Inverse Distance Weighting (IDW), and Universal Kriging (UNK) using the physically based and spatially distributed hydrological WetSpa model. The paper also illustrates how these methods allow more accurate estimation of river discharge simulated with the WetSpa model for the Gharesou basin in Iran. Daily data of 22 precipitation gaging stations were used in this study. In order to simulate hydrological processes for the basin, especially river runoff, we used the precipitation data interpolated with THI, IDW, and UNK as well as digital information on topography, the land use and soil type of the basin. For comparing three interpolation methods, they were applied to the simulation of river discharge using the WetSpa model. Simulated daily values of water discharge in the basin were compared with observational data. The quality of data simulated using these three methods was very good or excellent. According to the efficiency criterion of Aggregated Measure (0.879), the application of THI method provided more accurate forecast of river discharge than that of IDW and UNK methods.