Modelling Wind Induced Impacts for Application in a New Weather Warning System

The enhancement in user communication is one of the key aspects in the RainBoW (Risk-based, Application-oriented and INdividualizaBle delivery of Optimized Weather warnings) programme, which encompasses the development of the new weather warning system for the German Meteorological Service (Deutscher Wetterdienst). Within RainBoW, a probabilistic weather warning system is developed that supplies both information to users with special requirements but also generates standardised warnings for the general public. It addresses a number of weather elements, such as wind, precipitation, and thunderstorm. Hence, a good understanding of the issued weather warnings by the recipients plays a crucial role in order to enable both professional forces but also individuals to take appropriate action in affected regions. In RainBoW, one of the means to achieve a better perception is to move impact information into the focus for future weather warnings provided by the German Meteorological Service.

In this contribution, we present a study on hazards of wind gusts, which is among the weather elements that show the largest impacts on infrastructure and society. In order to shed light on the severeness of impacting events, we discuss the generation of an impact proxy for Germany, which we model by evaluating historical wind gust data by means of extreme value statistics. Despite the available several years of detailed reanalysis data (e.g. from the COSMO REA6 dataset), one of the main challenges is the sparsely available extreme weather data. One reason for this lies in the very profound nature of extreme value statistics, but wind gusts are also not very well represented in the model data. This is further complicated by the rather short period of sufficiently detailed and consistent data, which is governed by few more or less local extreme weather situations that occured during roughly two decades. We discuss the potential of this approach and motivate its implementation into the future warning system (RainBoW) of the German Meteorological Service. Moreover, we complement this with user generated crowd sourcing data and discuss its suitability for an impact proxy of wind.