Regional climate change over Europe in COSMO-CLM: Influence of emission scenario and driving global model

The variability of climate change over Europe is analyzed in a new ensemble of eight high-resolution scenario simulations with the dynamically nested regional climate model COSMO-CLM. The simulations are a contribution of the CLM-Community to EURO-CORDEX - the European branch of the international CORDEX initiative. The study documents the influence of two emission scenarios (RCP4.5 and RCP8.5) and four driving global climate models (MPI-ESM-LR, HadGEM2-ES, CNRM-CM5, EC-EARTH) on the climate change signal of seasonal near-surface air temperature, seasonal precipitation amount and daily precipitation intensity in eight different regions of Europe. Both temperature and precipitation changes show comparable spatial patterns across the ensemble. The warming magnitude, however, strongly depends on the emission scenario and the driving GCM. The analysis demonstrates that the influence of the driving GCM on the temperature change signal can be of comparable magnitude as the influence of the emission scenario. For seasonal precipitation changes the influence of the driving GCM is even more important. All ensemble members show a consistent increase in the frequency of days with intense precipitation while the total number of precipitation days decreases. A further aspect is the comparison between the climate change signals in the regional and the driving global simulations. A critical intensity class is used to quantify the transition from decreasing to increasing precipitation frequencies. The analysis reveals that the variability of frequency changes is smaller in the regional than in the global model ensemble, which can be interpreted as an improved robustness and a possible added value of the regional projections. It is also shown that the regional simulations tend to reduce the seasonal mean temperature change of the driving global simulation. In some seasons and regions the damping of the warming signal can amount to more than 2K. Overall, the analysis reveals a strong influence of comparable magnitude of the choice of the emission scenario and the driving global model on regional climate change signals, and stresses the necessity of a large regional model ensemble for a reliable and robust assessment of regional climate change.