Guy Carpenter launched its new leading-edge probabilistic hail model for Europe. The G-CAT® Hail Model features a unique approach to producing historic hail tracks using a lightning detection system developed for Guy Carpenter by atmospheric research firm nowcast GmbH.
The hazard module of the new hail model is based on the historic hail event set developed using nowcast GmbH’s patented global lightening detection system, LINET. The accuracy of the algorithm that uses certain parameters of lightning data to reconstruct hail occurrence in a storm track has been verified using policy claims from past events, hail reports from the European Severe Weather Database and radar data from the German Space Agency, DLR. The hazard module’s stochastic event set has been developed by analysis of the physical attributes of the historic hail events using a proprietary statistical model to generate events that capture the erratic nature of severe convective storms.
Commenting on the use of lightning data, Mark Weatherhead, Head of Catastrophe Model Development at Guy Carpenter, said: “There is no direct way of measuring hail intensity over broad scales and so the challenge is to find a good proxy measure of hail occurrence that can be used on a consistent basis over the entire area of interest. To this end, we have formed a partnership with nowcast GmbH who are using leading-edge technology to track the occurrence of those lightning parameters that best reconstruct the possible occurrence of hail in a storm track. We are very excited to be part of this collaboration, especially as it is producing very interesting data to use as a basis for the hazard component of our model.”
The launch of the model comes at a time when recent insured losses from hail events in Europe have reached record levels. Max Strasser, Project Leader for the G-CAT Hail Model, said: “While much of the focus is on the larger perils such as windstorms and floods, hail is a major European peril responsible for high attritional losses in many countries. In a bad year hail can be the largest source of loss in some places. Hailstorm Andreas which occurred in Germany in July 2013 generated an industry loss in excess of USD3 billion, but we don’t know if this is at the extreme tail-end of the range or something we could expect on a more regular basis. Experience-based approaches can only take us so far, and so we have invested in this new fully probabilistic model to allow us to properly explore the possibilities and provide a more comprehensive view for our clients.”
Massimo Reina, CEO of Continental Europe and MENA, added: “Understandably our clients are very interested in managing the hail peril and so our new model allows them to take a view on hail across their entire European portfolio.”
The European hail model will serve as a key tool in the identification of accumulations of insured risks exposed to hail, from the production of reliable estimates of hail losses to property and motor lines at a country level, to an entire insured portfolio across the Continental European region. The model is currently available for Austria, Germany, Italy, Poland and Slovenia, and will be extended to a number of other European territories in 2015.