Preprints  

Numerical weather prediction depends essentially on high-quality upper-air wind observations to constrain the initial conditions. This study by Martin et al. investigates the impact of spaceborne Doppler wind lidar measurements from the Aeolus mission on forecast quality of the operational global forecasting system with ICON at Deutscher Wetterdienst. An observing system experiment shows an overall beneficial impact, and the authors go one important step further and present illustrative examples how events with strong forecast quality improvements can be related to specific dynamical phenomena such as ENSO and the extratropical transition of tropical cyclones.

Understanding the specific dynamical processes leading to extreme floods is an important but challenging task. Ruff and Pfahl used an innovative approach that allowed them to go beyond single case studies. They used operational ensemble forecasts from the ECMWF during the period 2003-2019 and focused on five major river catchments in Central Europe. Comparing extreme events (with a return period of 100 years) with more moderate events revealed important differences between the catchments. For some catchments the main factors that distinguish 100-year events were the intensity of the upper-level cutoff and surface cyclone, whereas in other catchments the main factor was an increased low-tropospheric moisture supply. The original results clearly illustrate that regional variability is substantial and no single atmospheric process can be claimed responsible for the distinction between extreme and moderate flood events.