Hail frequency in Germany: regional trends over the last 20 years

Although loss-intensive hail events occur relatively rarely, these events account for a large proportion of total insured losses (SwissRe, 2024). There are individual events such as the hailstorm in Reutlingen on July 28, 2013, in which hail of up to 10 cm caused damage of over one billion euros (Kunz, 2017; SwissRe, 2015). However, other hailstorms with smaller hailstones also repeatedly cause major damage, including to house facades and roofs, photovoltaic systems, cars, and agricultural land. In recent years, the total amount of insured damage caused by hail events has increased both worldwide and in Europe (SwissRe, 2024). However, this cannot necessarily be attributed to an intensification of hail events in recent years, as not only meteorological causes but also higher insured values can influence the trend.

Due to the high amounts of damage caused by individual hail events, it is important to investigate the effects of climate change on hail events. However, a statistical evaluation of the temporal development of direct hail observations in Germany (and also in other countries worldwide) is not possible due to the lack of a homogeneous data set. This is because hail is formed in thunderstorms, which enable hail formation in the first place due to their strong updrafts and the supercooled water present, but usually have a horizontal extent of only a few kilometers. This makes it difficult to detect hail with conventional meteorological observation networks. In the HailClim project of the BMBF-funded joint project ClimXtreme (Phase 2), potential hail tracks are therefore derived with a tracking algorithm from the 3D radar data of the German Weather Service (DWD) as a data basis. In this algorithm, regions with high reflectivity in the radar signal are searched for in the first step and then put together into hail tracks (Handwerker, 2002; Schmidberger, 2018).

Considering all the events determined in this way over the last 20 summer half-years (April to September, 2005 – 2024), it is clear that, in principle, all regions of Germany can be affected by hail. However, there are clearly hotspot regions in which hail events are more likely than elsewhere (Fig. 1). In southern Germany there are the highest values in hail frequency: due to flow around and overflow effects from the Black Forest and the resulting convergences on the Swabian Jura and in the Neckar valley, there is an increased formation of deep convection with hail (Siegmann, 2022). The Bavarian Alpine foothills and parts of Hesse also show an increased frequency of hail.

In the period from 2005 to 2024, the total number of annual hailstorms in Germany did not change significantly (Fig. 2). There is an interannual variability visible, however, no statistically significant decrease or increase can be observed over the 20 years. This applies to hail tracks of all sizes and longer tracks. However, a different picture emerges from a spatial breakdown of the time series. For 25 km x 25 km grid cells over Germany, regional trends emerge in Figure 3: negative trends can be observed over large parts of northern and central Germany, which are also significant over a larger area, particularly in the western part. Only in southern Germany, primarily on the border between Baden-Württemberg and Bavaria, has there been a significant increase in hail events. These opposing regional trends balance each other out at the overall German level in Figure 2. The same spatial distribution of trends can also be seen when looking at the potential hail tracks with a minimum length of 50 km (Fig. 3b).

The results of a decreasing hail frequency over large parts of Germany are quite surprising. According to the Clausius-Clapeyron equation, a warmer atmosphere can hold more water vapor. This means that more energy is available for the formation of thunderstorms and hail. In addition to these thermodynamic processes, dynamic processes and mechanisms in the atmosphere (e.g. changes in the large-scale flow) also play a role. Initial results show that teleconnection patterns (processes with large-scale climatic interactions) provide first explanations for a change in hail frequency, including its annual variability, by changing the environmental conditions that favor hail.

References:

Handwerker, J. (2002): Cell tracking with TRACE3D – A new algorithm. Atmos. Res. 31, 15–34, doi: https://doi.org/10.1016/S0169-8095(01)00100-4.

Kunz, M., U. Blahak, J. Handwerker, M. Schmidberger, H.J. Punge, S. Mohr, E. Fluck, K.M. Bedka (2017): The severe hailstorm in southwest Germany on 28 July 2013: characteristics, impacts and meteorological conditions. Quart. J. Roy. Meteor. Soc. 144, 231–250, doi: https://doi.org/10.1002/qj.3197.

Schmidberger, M., 2018: Hagelgefährdung und Hagelrisiko in Deutschland basierend auf einer Kombination von Radardaten und Versicherungsdaten. Wissenschaftliche Berichte des Instituts für Meteorologie und Klimaforschung des Karlsruher Instituts für Technologie (KIT), 78, KIT Scientific Publishing, Karlsruhe, Germany, doi: https://doi.org/10.5445/KSP/1000086012.

Siegmann, F., 2022: Spatial and temporal variability of low-level convergence zones triggering deep moist convection in south-western Germany. Masterarbeit, Instituts für Meteorologie und Klimaforschung, Karlsruher Instituts für Technologie Karlsruhe, Germany: https://www.imk-tro.kit.edu/5734_11619.php

SwissRe, 2015: Sigma – Natural catastrophes and man-made disasters in 2014: convective and winter storms generate most losses. Technical report, Swiss Reinsurance Company Ltd, Zurich, Switzerland. https://www.swissre.com/institute/research/sigma-research/sigma-2015-02.html.

SwissRe, 2024: Sigma – Natural catastrophes in 2023: gearing up for today’s and tomorrow’s weather risks. Technical report, Swiss Reinsurance Company Ltd, Zurich, Switzerland https://www.swissre.com/institute/research/sigma-research/sigma-2024-01.html.

Working group: Atmospheric risks
Author: Mathis Tonn (January 2025)