Climate change is the most important problem in water resources limiting crops. Currently, prediction of climate change is the most important part of strategic plans. This study was conducted to assess effects of climate changes on surface runoff in Maharlu Lake basin, Iran. We processed climate data during the future period in two general circulation models (MICROC5 and HadGEM2-ES) under radiative forcing scenarios (RCP8.5 and RCP4.5), and considered changes in temperature and precipitation to the base period (1980–2013) during P1 (2021–2040), P2 (2041–2060), and P3 (2061–2080). We used statistical indices (interpretation and Nash-Sutcliffe efficiency coefficients) for assessing the model accuracy. The results showed the model successful simulation of daily runoff. The statistical and SWAT model results at three stations had good accuracy for calibration (NES = 0.72 and R2 = 0.75) and validation (NES = 0.85 and R2 = 0.86 ~ 0.92). The basin annual maximum mean temperature during the future period to the base period, under RCP4.5 scenario, increased between 1.45 and 3.94 °C, and 1.34 and 5.55 °C under RCP8.5 scenario. Annual minimum mean temperatures during future period increased, relative to the base period. This increase under RCP4.5 scenario is 1.20–3.34 °C, while 1.28–5.51 °C under RCP8.5 scenario. Changes in precipitation vary. The change in the rainy seasons (November to April) is greater than the arid seasons (May to October). Finally, soil and water assessment tool model simulated runoff response during the future period under two different scenarios. In spring, runoff will increase. Climate change is useful for cropping and presenting managerial solutions by managers and policymakers of water resources.