Winter storms are one of the riskiest natural hazards in Europe. They are often assessed either from a climatology perspective or the exposed land use type. However, for accurate storm protection management, analyzing compatible wind speed and land use data at the highest spatiotemporally accuracy is critical. Thus, the study goals were (1) modeling wind speed for the 27 most extreme winter storm events from 1993 to 2022 at high spatial resolution (250 m × 250 m) and (2) defining and developing new and representative storm hazard and storm exposure indices for forested area, built-up area, and population. The winter storm event-related wind speed was modeled using a least-squares boosting approach based on ERA5 reanalysis data and 1730 wind speed measurement sites from the integrated surface global hourly meteorological dataset. In combination with temporally variable population density, built-up area, and forested area grids, the storm exposure indices for the European NUTS3 regions were calculated. The coefficient of determination of the median of the modeled wind speed fields is 0.73 for an independent subset of measurement sites. The wind speed maps reveal the uniqueness of the winter storm events. Each storm only affects parts of Europe. The study also emphasizes the relevance of distinguishing between storm hazard and storm exposure. The exposure of built-up areas to winter storms has increased considerably in recent years. This example underlines that storm exposure is not temporally constant but depends on the development of wind speed and land use. Decision-makers can apply the results of this study to evaluate and improve the effectiveness of storm protection measures. The developed storm exposures indices enable a sector- and location-specific evaluation with high precision.
Keywords: GIS; Storm Kyrill; Storm Lothar; Storm exposure; Storm hazard.
Copyright © 2024 The Authors. Published by Elsevier B.V. All rights reserved.