Assessing Severe Storm Frequency and Risk Using Multi-Decadal Geostationary Infrared Satellite Data Records

The effects of severe convective storms (SCS) are experienced by much of the world, but whether communities can recover from the impacts depends upon their resilience – the capacity to prepare and mitigate. Hail is the costliest SCS hazard, accounting for 70% of losses. Swiss Re found that SCS accounted for “record high” global loss in 2023 ($64 billion), mostly originating from the U.S., but growing fastest in Europe. In the U.S. in 2023 (2024) there were 28 (27) “Billion Dollar Disasters”. 19 (17) of were caused by SCS. The (re)insurance industry relies on hazard models to statistically estimate risk. These models simulate risk based on long-term data records of weather hazard occurrences. Over much of the world, there are uncertainties with SCS risk assessment, primarily due to a lack of human spotter reports and/or insufficient weather radar climate data records.SCS are maintained by strong updrafts that create unique patterns at their cloud tops: overshooting tops and above anvil cirrus plumes. Though overshooting tops do not guarantee that extremely severe weather will occur, most severe storms are found to generate them. The distinct benefits of geostationary (GEO) satellites are their long-term observational data records, hemispheric spatial coverage, and high spatio-temporal frequency sampling. Reliable methods for overshooting top detection have been demonstrated, and long-term detection datasets have been previously used for SCS risk assessment. GEO observations can provide unique, long-term insights into SCS intensity, frequency, spatio-temporal distribution, and trends that support development of SCS hazard models, complementing alternative methods derived from reanalyses, weather radar, and human reporting.In this NASA Applied Sciences Disasters Program project, we employ an automated convective storm and overshooting top detection method, based on GEO infrared imagery and reanalysis model inputs. This method supports the development of SCS hazard models through creation of multi-decadal data records over Europe and North America. In partnership with Willis Towers Watson (WTW), Gallagher Re, the Karlsruhe Institute of Technology (KIT), and the European Severe Storm Laboratory, we seek to better inform the public and industry about SCS exposure and risk. This presentation will highlight analyses of a recently developed 21-year duration convective storm and overshooting top data record over Europe derived from 15-min MSG SEVIRI infrared imagery on a 4 km grid. It will also highlight comparisons between the SEVIRI data record and the European Severe Weather Database to evaluate regional severe storm detection performance and satellite-based storm intensity estimation.