High‐resolution climate change simulations for the Jordan River area

For the estimation of future climate conditions in the Jordan River region, the National Center for Atmospheric Research–Penn State University meteorology model in the versions 3.5 and 3.7 driven with boundary data from the Max‐Planck‐Institute for Meteorology and Hadley Centre global circulation models and the Special Report on Emission Scenarios A1B emission scenario has been used. The spatial resolution of the nested dynamic downscaling approach was 18.6 km, and the transient runs were performed for the period 1960–2099. The investigated statistics include mean precipitation, frequency and intensity of wet days and strong precipitation events, as well as mean temperature and heat wave duration index. The results show that the models satisfactorily reproduce the mean temperature and precipitation patterns. The comparison with the observational reference for the period 1961–1990 reveals a bias in the annual mean precipitation ranging from −20% to +17%, with an ensemble mean of −3%. The models show limitations in reproducing the precipitation seasonality. All models underestimate the wet day frequency and show differences in the strong precipitation events. The simulations of the future climate signal indicate an ensemble mean increase of the annual mean temperature of approximately 2.1 K in the period 2031–2060 and 3.7 K for the period 2070–2099 related to the 1961–1991 mean. In the same periods, the annual mean precipitation is simulated to decrease by approximately −11.5% and −20%, respectively, which means a reduction of expected water availability in the Jordan River region. All models show an increase of the heat wave duration index. A significant elevation dependence is present in the simulated future climate signal on both temperature and precipitation. The simulations show an increased coefficient of variation in annual precipitation, indicating that larger interannual precipitation variability can be expected in the future.