Trends in Severe Convective Storm Activity over Europe (1983–2024)

Severe convective storms are among the most damaging natural hazards worldwide, with insured losses reaching tens of billions of US dollars annually. All severe convective storms originate from deep convective clouds (DCCs), making DCC occurrence a suitable proxy for assessing long-term changes in severe storm activity. However, robust observational evidence of DCC trends over Europe remains limited.

This study investigates long-term trends in DCC frequency over Europe during 1983–2024. We use observations from the Meteosat satellite series, combining data from the first-generation Meteosat Visible and Infrared Imager (MVIRI) and the second-generation Spinning Enhanced Visible and Infrared Imager (SEVIRI). The analysis is based on two spectral channels: the water vapour absorption channel centered near 6.5 µm and the infrared window channel centered near 11 µm. Satellite observations are complemented with atmospheric fields from the ERA5 reanalysis.

To ensure temporal homogeneity between sensors, spectral band adjustments were applied using correction functions derived from Infrared Atmospheric Sounding Interferometer observations. Parallax correction was performed using a cloud-top height estimation method based on infrared brightness temperatures combined with ERA5 temperature data. A Meteosat pixel was classified as a DCC when the brightness temperature difference between the water vapour and infrared window channels exceeded 2.5 K, a threshold established through validation with CloudSat–CALIPSO and Moderate Resolution Imaging Spectroradiometer observations. Additionally, convective available potential energy (CAPE) from ERA5 was required to exceed 500 J/kg.

The results reveal two distinct regional patterns of DCC frequency trends across Europe. Central and Western Europe exhibit positive trends, reaching up to 0.001 per decade in the annual mean, with the strongest increases observed over northern Italy and eastern Austria. The increase is most pronounced during boreal summer (June–August), with trends up to 0.004 per decade, while no significant trends are detected during other months. In contrast, negative trends occur over western France, the Iberian Peninsula, and the Mediterranean Sea, with annual mean decreases reaching −0.004 per decade. In these regions, the sign of the trend varies substantially between individual months.

Due to the relatively short time series and the low frequency of DCC occurrence, only the strongest trends are statistically significant (p < 0.05). Nevertheless, although the absolute trend magnitudes appear small, DCCs are rare phenomena, and the observed changes correspond to relative increases of approximately 10–25% in DCC frequency in parts of Europe. These findings indicate a potentially meaningful increase in severe convective storm risk under ongoing climate change.

This research was funded by the National Science Centre of Poland, grant no. UMO-2020/39/B/ST10/00850.  We gratefully acknowledge Polish high-performance computing infrastructure PLGrid (HPC Centers: ACK Cyfronet AGH) for providing computer facilities and support within computational grant no. PLG/2025/018115