An accurate estimation of evapotranspirationis a key issue in running climate models, especially for the calculation of surface fluxes. In recent years, by development of global and regional climate models,long-term predictions of weather parameters affecting evapotranpiration have been done more easily.The ability of the RegCM regional climate model (Version 3) in simulating the potential evapotranspiration over Mashhad in Northeastern Iran was evaluated during the base period of 1961 to 1985 and the future period of 2021-2035. Due to lack of a measured amount of evapotranspiration, the Penman–Montith (P-M) equation was chosen to estimate actual values of ET in the base period. For initialization of the RegCM3 model, boundary conditions from EH5OM General Circulation Model (GCM) were used as initial and boundary conditions to feed the RegCM3 regional model. Future predictions of evapotranspiration were done​under A1B emission scenario. A horizontal resolution of 50 km was considered for the model. Based on the assumed conditions in running the model, the results showed that the non-post-processed RegCM model outputs cannot be used for the estimation of potential evapotranspiration.However, after post processing using multivariate regression, results were closer to those calculated by P-M equation. Temperature and radiation parameters were considered as independent variables in the multiple regression models to perform post-processing. Our results showed that the mean annual evapotranspiration in the future period (1075 mm) will be increased by 16.34%, compared to the base period (924 mm). On the whole, the average potential evapotranspiration will be increased in April, May, June, August, September and October, while it will be decreased in January, February, July, November and December, compared to the base period. The outcomes of the present study reveal that the maximum evapotranspiration in the future period will be in June, while in the base period the maximum amount of evapotranspiration occurs in July.