Understanding future trends in climate change and evaluating the performance of climate models in predicting temperature and precipitation is a fundamental step in water resources planning and management. In this study, precipitation, minimum and maximum temperatures, and evapotranspiration were simulated at the Qazvin synoptic station. For this purpose, monthly data of minimum temperature(℃), maximum temperature (℃), evapotranspiration (mm), and precipitation (mm) from the CanESM5, GFDL-ESM4, and HadGEM3 climate models were compared with the Qazvin synoptic station data in the base period 1986-2014, individually and in groups. The results showed that the group implementation of three general atmospheric circulation models using the random forest method for the study area has led to a reduction in the simulation error and, as a result, an increase in the accuracy of the predictions. So that in the minimum temperature simulation, the individual models have RMSE of about 2.6–2.4 (℃), while the combined model has 1 and 1.83 (℃) in the training and test data, respectively. The Spearman correlation is high in both cases (0.85–0.97). The trend of changes in climate variables was examined using the Kendall test and the slope method, and the results showed that in the future periods under the SSP5-8.5 and SSP2-4.5 scenarios, the precipitation trend is significantly decreasing, while the minimum and maximum temperatures and evapotranspiration are significantly increasing. In general, the results showed that the study area will face warmer temperatures and reduced annual precipitation in the coming decades under the influence of climate change scenarios; an issue that highlights the need for adaptive planning in the water and agriculture sectors.