sajad ghasemzadeh

Attention to the phenomenon of climate change and its impact on water resources is of great importance. The aim of this study is to evaluate the effect of different methods of weighting the output of AOGCM models and its effect on runoff in Qarahsu Basin during the period 2041-2021. The best climatic models (HADGEM2-ES, MICRO IPSL-CM5A-LR, NOERESM1-M, ESM2M-GFDEL) were selected from 14 general circulation models and weighed with three weighting methods including the same method, Bayesian averaging method and REA method. The results showed that the highest increase of maximum temperature in summer was 1.58 ° C by the same weighting method and the lowest increase of minimum temperature in winter with REA method was 0.96 ° C. The highest percentage of precipitation changes was in August with the same method and the lowest percentage of precipitation changes was in February with REA method. Evaluation of SWAT model simulation results for calibration period using R2 and NS statistical indices in calibration stage is equal to 0.74 and 0.79 and in validation stage is 0.68 and 0.72, respectively, which indicates the accuracy of the model in a runoff simulation. Prediction of runoff changes with the approach of ensemble the output of climate models, shows reduction in runoff in the future period. The lowest and highest percentages of runoff changes will be corresponded to Bayesian method which is -13 and -34.4%, respectively. Overall, the results indicate a change in the temporal distribution of flow in the Qarahsu Basin in the coming period, which will cause significant changes in the quality and quantity of water resources.

Given that global warming is a threat to human life on Earth, predicting future climate change is essential. One of the objectives of this study is to predict daily temperature and precipitation in the near future (2021-2040), using general atmosphere-ocean circulation (AOGCM) models and reduce their uncertainties. Therefore, in this study, first, among the studied models, 5 suitable models including HADGEM2-ES, MICRO IPSL-CM5A-LR, NOERESM1-M ESM2M-GFDEL were selected and downscale by LARS-WG method. Then, using two approaches of Raisanen weighting and observational mean temperature and precipitation (MOTP) were used to weight and ensemble multiple models. The results show that the maximum temperature increases in all months, with the highest temperature increase in the Riesen model in February and the lowest increase in October. In the MOTP weighting model, the highest temperature increase is in January and the lowest temperature increase is in August. The percentage of rainfall changes in June, July and August decreased significantly due to slight observational rainfall. In this study, the MOTP weighting method was chosen to predict climate variables in future periods due to having a higher R2 and the lowest RMSE as a better method in predicting climate data. Therefore, according to the results, it is expected that significant feedback in the water balance of the region will be due to the effect of increasing temperature and increased evaporation, as well as the potential for changes in atmospheric events and hydrology of the basin.