Investigating the role of downscaling and reference evapotranspiration estimation method in analysis of the impact of climate change on water resources

Climate change could largely affect the surface water and groundwater sources. This impact should be simulated using appropriate models. In the present study, using LARS-WG and SDSM statistical models, the output of HADCM3 general circulation model under four emission scenarios was downscaled in the Hablehroud Basin in the period of 2018-2047. The SWAT model was calibrated in the basin and utilized to simulate the discharge, groundwater recharge, and soil water content in the mentioned period. Three different methods including Hargreaves, Penman-Monteith and Priestley-Taylor in the SWAT model were used to estimate reference evapotranspiration. Different combinations of the factors effecting the uncertainty, including downscaling model, emission scenario, and evapotranspiration estimation method were used. The results showed that the method used in downscaling the output of the general circulation model is the most effective factor affecting the uncertainty of the output of the SWAT model. It was also observed that the different combinations produce more outliers in simulating groundwater recharge, in comparison with simulating the discharge and soil water content. The median of the annual discharges simulated using all combinations was calculated to be 13.32 cms. The results showed that the combinations of downscaling model, emission scenario, and evapotranspiration estimation method that simulate values less than 13.32 cms have less uncertainty than other combinations. In the case of groundwater recharge (with a median of 2.07 mm/ year) and soil water content (with a median of 112.4), same results were observed.