During the last decades weather forecasts have improved continuously. However, in addition to the pure weather forecast the interest in the forecast of weather impacts is increasing. The understanding of risks and impacts is of particular importance in case of extreme weather events to provide useful warnings and recommendations and to execute protective measures. Studies show that the communication of information about both the meteorological event and its impacts can lead to more appropriate reactions of recipients (e. g. Weyrich et al. 2018).
The impacts of weather events can be analysed by combining meteorological data and impact information with the help of statistical methods. Within the framework of WEXICOM we develop statistical impact models for three types of impacts: Wind storm damages to residential buildings (Pardowitz et al. 2016), the impact of convective events on fire brigade operations (Pardowitz et al. 2017, 2018), as well as weather impacts on traffic and car accidents. By using these impact models, probabilistic risk indices are developed. These indices are implemented in real-time forecasting systems and tested by potential users. Impact models as developed within WEXICOM can help developing risk-based warning systems.
Furthermore, methods for statistical calibration are applied to a network of internet-of-things-based measurement devices used in a citizen science context.