Flood modelling is challenging but vital
The recent hurricanes have turned the spotlight on to the importance of flood modelling, which has come a long way in a short amount of time, Pete Dailey, vice president and head of global flood models at RMS, told PCI Today.
“We began evaluating flood risk and the potential for modelling these risks nearly a decade ago, when we started development of our first flood models for the UK and Europe. Those models have been expanded to cover other parts of the world,” he said.
According to Dailey, tackling the US—a very large area with many flood catchments, a multitude of complex river networks, as well as risks from coastal flooding, aka hurricane-induced storm surge—is a major undertaking. Multiple sources of flood need to be accounted for to capture the risk in a comprehensive way.
Dailey admitted that it’s been a long trek for RMS in terms of developing the RMS US Inland Flood Model. The company started about six years ago and over time it has seen development of the hazards and vulnerabilities.
The hazard accounts for precipitation leading to inland flooding and the amount of coastal storm surge that leads to coastal flooding. RMS has also developed the vulnerabilities piece—the relationship of the flood hazard to the engineering vulnerabilities of individual buildings. It accounts for various construction types, occupancies, different ages and heights and building classes—all unique in terms of how they translate the flood depth into a damage and cost of repair.
“The final step is accumulating all that knowledge into an economic loss and then ultimately into an insured loss risk for use by the insurance industry,” Dailey said. “It is a conventional approach for modelling natural catastrophes, but what makes modelling flood different is that it’s a ‘high gradient’ hazard.
“What we mean is that within a short distance, the hazard can change dramatically, owing to local land use and soil conditions, distance to nearby rivers and local defences, proximity to low-lying areas in urban environments, etc. All these make the local estimation of flood risk a major challenge.”
RMS has a model going through the final stages of calibration which it expects to release in the spring of 2018. It will include the latest research on heavy precipitation events, such as Houston experienced with Harvey, and will also draw on lessons from Irma.
Get the latest re/insurance news sent to your inbox every day - Sign up to our free email newsletters
More players are drawn to mortgage reinsurance by its growth, innovation, returns
Casualty remains tough but pockets of opportunity will emerge
Rates will now increase as reinsurers refill their depleted cat reserves
Keep your discipline when innovating
PCI looks to boost awareness of rises in auto fatality rates
HIM losses an earnings event for reinsurers
Divergence in risk helps prove role of data
AM Best questions insurers’ data modelling ability on cyber
Sleepwalking into unexpected cyber loss
Hurricane Nate will not generate material losses, says RMS
The fear factor will determine rate hikes
Cat losses prove uncertain nature of industry
Regulators need education on big data
