ClimateEU: A high-resolution database of historical and future climate for Europe developed with deep neural networks

This study contributes an accessible, comprehensive database of interpolated climate data for Europe that includes monthly, annual, decadal, and 30-year normal climate data for the last approximately 120 years (1901 to present) as well as multi-model CMIP6 climate change projections for the 21st century. The database includes variables relevant for ecological research and infrastructure planning, and comprises more than 25,000 climate grids that can be queried with a provided ClimateEU software package to extract time series for lists of sample locations, or custom grids for specific study areas at any resolution and projection. In addition, continent-wide 1km resolution gridded data are available for download (http://tinyurl.com/ClimateEU). The climate grids were developed with a three-step approach, using thin-plate spline interpolations of weather station data as a first approximation (replacing otherwise needed lengthy pre-training of the neural network). Subsequently, a novel deep learning approach is used to model orographic precipitation, rain shadows, lake and coastal effects at moderate resolution (2.5 arcmin). Lastly, lapse-rate based downscaling is applied to generate high-resolution grids (up to a useful resolution of 250 m in mountenous terrain). The climate estimates were optimized and cross-validated with a checkerboard approach to ensure that training data was spatially distanced from validation data. We conclude with a discussion of applications and limitations of this database.