The added value of high-resolution EURO-CORDEX simulations to describe daily wind speed over Europe

In the context of the CORDEX project, an ensemble of regional climate simulations (RCMs) of high resolution on a 0.11 degrees grid has been generated for Europe with the objective of improving the representation of regional to local-scale atmospheric phenomena. However, such simulations are computationally expensive and do not always reveal added value. Here, a recently proposed metric (the distribution added value [DAV]) is used to determine the added value of all available EURO-CORDEX high-resolution simulations at 0.11 degrees for daily mean wind speed compared to their respective coarser-gridded 0.44 degrees counterparts and their driving fields, hindcast and historical experiments. The analysis consists in comparing the degree of similarity between normalized wind probability density function (PDF) of simulations and observations. In addition, the use of a normalized PDF allows for a direct spatial comparison among the different regions and time periods. Results show that RCMs add value to their reanalysis or forcing global model, but the nature and magnitude of the improvement on the representation of wind speed may vary depending on the model, region and season. We found most RCMS at 0.11 degrees to outperform models at the 0.44 degrees resolution in terms of their quality in capturing the measured wind speed PDF. When looking at the upper tail of the wind speed PDF, the benefits of downscaling are generally larger. At the regional scale, added value is obtained for 0.11 degrees with respect to 0.44 degrees resolution for all subdomains studied, particularly over the Mediterranean, the Iberian Peninsula and the Alps. When analysing the added value dependence on altitude, runs at 0.11 degrees models represent better the locations below 50 m and above 350 m of altitude, and the 0.44 degrees increasingly under-perform for higher altitudes. Overall, DAV are larger at 0.11 degrees than at 0.44 degrees resolution, due to a better performance of local-scale feedbacks at high resolution.