Regional climate modeling on European scales : A joint standard evaluation of the EURO-CORDEX RCM ensemble
Résumé
EURO-CORDEX, the European branch of the CORDEX initiative, is providing regional climate simulations at spatial resolutions of 12 and 50 km. An important aspect for both model developers and end-users of regional climate projections is the assessment of model performance and the quantification of bias ranges. Here we present an evaluation of the ERA-Interim-driven EURO-CORDEX ensemble for the period 1989-2008, focusing on the two standard variables near-surface air temperature and precipitation and using the E-OBS dataset as an observational reference. In total, nine 12 km and eight 50 km simulations are considered conducted with seven different models. Several performance metrics, mostly relying on monthly and seasonal mean values, are used to assess model performance on a European scale. Results are compared to those for the ERA40-driven ENSEMBLES simulations. The evaluation reveals a satisfying model performance in many cases, but also considerable model biases for selected metrics and selected sub-domains of the continent. Seasonally and regionally averaged temperature biases are mostly smaller than 1.5 °C, while precipitation biases are typically located in the - 40 to +40 % range. For regional-scale averages, no clear benefit of an increased spatial resolution (12 km compared to 50 km) can be identified. The bias ranges of the EURO-CORDEX ensemble often correspond to those of the ENSEMBLES simulations, but several improvements in model performance can be identified, such as a less pronounced southern European warm summer bias. Bias ranges found for different configurations of one individual model can be of a similar magnitude as bias ranges across different models, demonstrating a strong sensitivity of the chosen combination of physics on model performance. The present overview study will be complemented by further evaluation efforts of the EURO-CORDEX community, addressing selected aspects of model performance such as the representation of extreme events in more detail.