Summary

EuRadCA brings together researchers from Germany and the Netherlands to improve radar precipitation determination across Europe. The project will develop new methods for combining radar data with information from rain gauges and private weather stations, using advanced techniques such as deep learning and geostatistics.

EuRadCA will focus on four key areas. Firstly, new quality control and bias correction techniques will be developed to work directly in the radar's polar coordinate system, significantly increasing the accuracy of radar data. Second, the project will define pixel-level quality indicators to track uncertainties in each processing step and enable probabilistic precipitation estimation. Thirdly, EuRadCA relies on deep learning models that are trained with both simulated and real radar data. The target is to overcome the previous limitations of conventional methods and create more accurate precipitation products from radar data. Fourthly, new approaches are being developed to reliably identify and filter out implausible readings from private weather stations, particularly with a view to improving data quality in the event of extreme weather events.

The project team consists of experts from the Karlsruhe Institute of Technology (KIT), the University of Stuttgart and Delft University of Technology. Collaboration with the national weather services DWD (Germany) and KNMI (Netherlands) ensures that the solutions developed are relevant in practice and that easy access to data and computing resources is guaranteed.