Evaluating a state of the art, internationally coordinated pan-Arctic regional climate model ensemble

Within the Horizon 2020 project PolarRES, a new ensemble of regional climate simulations has been developed using the latest generation of regional climate models (RCMs) for the Arctic. These state-of-the-art RCMs downscale the ERA5 reanalysis over the period 2001-2020, covering the entire Arctic region at a grid spacings of approximately 12km. Furthermore, all simulations follow the Polar CORDEX protocol for the next generation of regional climate projections of the polar regions. This new ensemble of high-resolution climate simulations offers considerable opportunities to advance our understanding of the present-day climate of the Arctic. However, a first step to realising this potential is to evaluate the performance of the regional climate models, highlighting their strengths and limitations. This is also necessary for understanding and interpreting the future projections that will be generated by these RCMs using a novel storylines approach to downscale CMIP6 models.

The work presented here will focus on the simulations of the present-day climate driven by the ERA5 reanalysis. As part of the evaluation process, a clustering technique is applied to reanalysis data to identify regions with similar annual and seasonal characteristics of surface temperature and precipitation. This approach allows for a better understanding of the regional climates of the Arctic, provides a more physically consistent basis for model evaluation, and eases the investigation of model deficiencies in simulating regional scale forcings. This work will focus on the regionalisation of the Arctic for model evaluation and present preliminary results of the application of this regionalisation to the aforementioned Arctic climate simulations.