logo image

In December I wrote about Lothar and Daria, a cluster of windstorms that emphasized the significance of ‘location’ when assessing windstorm risk. This month we have the 25th anniversary of the most damaging cluster of European windstorms on record—Daria, Herta, Wiebke, and Vivan.

This cluster of storms highlighted the need for better understanding the potential impact of clustering for insurance industry.

At the time of the events the industry was poorly prepared to deal with the cluster of four extreme windstorms that struck in rapid succession over a very short timeframe. However, since then we have not seen such a clustering again of such significance, so how important is this phenomena really over the long term?

There has been plenty of discourse over what makes a cluster of storms significant, the definition of clustering and how clustering should be modeled in recent years.

Today the industry accepts the need to consider the impact of clustering on the risk, and assess its importance when making decisions on underwriting and capital management. However, identifying and modeling a simple process to describe cyclone clustering is still proving to be a challenge for the modeling community due to the complexity and variety of mechanisms that govern fronts and cyclones.

What is a cluster of storms?

Broadly, a cluster can be defined as a group of cyclones that occur close in time.

But the insurance industry is mostly concerned with severity of the storms. Thus, how do we define a severe cluster? Are we talking about severe storms, such as those in 1990 and 1999, which had very extended and strong wind footprints. Or is it storms like those in the winter 2013/2014 season, that were not extremely windy but instead very wet and generated flooding in the U.K.? There are actually multiple descriptions of storm clustering, in terms of storm severity or spatial hazard variability.

Without a clearly identified precedence of these features, defining a unique modeled view for clustering has been complicated and brings uncertainty in the modelled results. This issue also exists in other aspects of wind catastrophe modeling, but in the case of clustering, the limited amount of calibration data available makes the problem particularly challenging.

Moreover, the frequency of storms is impacted by climate variability and as a result there are different valid assumptions that could be applied for modeling, depending on the activity time frame replicated in the model. For example, the 1980s and 1990s were more active than the most recent decade. A model that is calibrated against an active period will produce higher losses than one calibrated against a period of lower activity.

Due to the underlying uncertainty in the model impact, the industry should be cautious of only assessing either a clustered or non-clustered view of risk until future research has demonstrated that one view of clustering is superior to others.

How does RMS help?

RMS offers clustering as an optional view that reflects well-defined and transparent assumptions. By having different views of risk model available to them, users can better deepen their understanding of how clustering will impact a particular book of business, and explore the impact of the uncertainty around this topic, helping them make more informed decisions.

This transparent approach to modeling is very important in the context of Solvency II and helping (re)insurers better understand their tail risk.

Right now there are still many unknowns surrounding clustering but ongoing investigation, both in academia and industry, will help modelers to better understand the clustering mechanisms and dynamics, and the impacts on model components to further reduce the prevalent uncertainty that surrounds windstorm hazard in Europe.

You May Also Like
pigs
February 21, 2019
Addressing the Challenges in Assessing Livestock Insurance Risks
November 16, 2015
Are (Re)insurers Really Able To Plan For That Rainy Day?

Many (re)insurers may be taken aback by the level of claims arising from floods in the French Riviera on October 3, 2015. The reason? A large proportion of the affected homes and businesses they insure in the area are nowhere near a river or floodplain, so many models failed to identify the possibility of their inundation by rainfall and flash floods. Effective flood modeling must begin with precipitation (rain/snowfall), since river-gauge-based modeling of inland flood risk lacks the ability to cope with extreme peaks of precipitation intensity. Further, a credible flood model must incorporate risk factors as well as the hazard: the nature of the ground, such as its saturation level due to antecedent conditions, and the extent of flood defenses. Failing to provide such critical factor can cause risk to be dramatically miscalculated. A not so sunny Côte d’Azur This was clearly apparent to the RMS event reconnaissance team who visited the affected areas of southern France immediately after the floods. “High-water marks for fluvial flooding from the rivers Brague and Riou de l’Argentiere were at levels over two meters, but flash floodwaters reached heights in excess of one meter in areas well away from the rivers and their floodplains,” reported the team. This caused significant damage to many more ground-floor properties than would have been expected, including structural damage to foundations and scouring caused by fast-floating debris. Damage to vehicles parked in underground carparks was extensive, as many filled with rainwater. Vehicles struck by more than 0.5 meters of water were written off, all as a result of an event that was not modeled by many insurers. The Nice floods show clearly how European flood modeling must be taken to a new level. It is essential that modelers capture the entire temporal precipitation process that leads to floods. Antecedent conditions—primarily the capacity of the soil to absorb water must be considered, since a little additional rainfall may trigger saturation, causing “saturation excess overland flow” (or runoff). This in turn can lead to losses such as those assessed by our event reconnaissance team in Nice. Our modeling team believes that to achieve this new level of understanding, models must be based on continuous hydrological simulations, with a fine time-step discretization; the models must simulate the intensity of rainfall over time and place, at a high level of granularity. We’ve been able to see that models that are not based on continuous precipitation modeling could miss up to 50% of losses that would occur off flood plains, leading to serious underestimation of technical pricing for primary and reinsurance contracts. What’s in a model? When building a flood model, starting from precipitation is fundamental to the reproduction, and therefore the modeling, of realistic spatial correlation patterns between river basins, cities, and other areas of concentrated risks, which are driven by positive relationships between precipitation fields. Such modeling of rainfall may also identify the potential for damage from fluvial events. But credible defenses must also be included in the model. The small, poorly defended river Brague burst its banks due to rainfall, demolishing small structures in the town of Biot. Only a rainfall-based model that considers established defenses can capture this type of damage. Simulated precipitation forms the foundation of RMS inland flood models, which enables representation of both fluvial and pluvial flood risk. Since flood losses are often driven by events outside major river flood plains, such an approach, coupled with an advanced defense model, is the only way to garner a satisfactory view of risk. Visits by our event reconnaissance teams further allow RMS to integrate the latest flood data into models, for example as point validation for hazard and vulnerability. Sluggish growth in European insurance markets presents a challenge for many (re)insurers. Broad underwriting of flood risk presents an opportunity, but demands appropriate modeling solutions. RMS flood products provide just that, by ensuring that the potential for significant loss is well understood, and managed appropriately.…

Laurent Marescot
Laurent Marescot
Senior Director, Model Product Strategy, RMS

Based in Zurich, Laurent initially joined RMS in 2008 as part of the Zurich account management team, servicing the European (re)insurance and ILS market. He then moved to the model product management group, leading the technical product management team for European climatic perils, such as windstorm, severe convective storm and flood. Since 2014, he has joined the model product strategy group for Europe model product line.

Prior to RMS, Laurent worked 3 years at the Swiss Federal Institute of Technology Zurich (ETHZ) as a Research Associate and Lecturer, managing multidisciplinary natural hazard research projects. Laurent still lectures regularly on geophysics and catastrophe modeling at universities, and gives seminars and invited talks in international meetings. He is a Lecturer and Scientific Collaborator at the University of Fribourg (Switzerland). Laurent co-authored numerous industry publications, reviewed scientific articles and proceeding papers. He holds an MSc in Geology from the University of Lausanne and a PhD in Geophysics from the University of Lausanne and the University of Nantes.

cta image

Need Help Managing Your Portfolio?

close button
Video Title

Thank You

You’ll be contacted by an RMS specialist shortly.

RMS.com uses cookies to improve your experience and analyze site usage. Read Cookie Policy or click I understand.

close