BI’s Article search uses Boolean search capabilities. If you are not familiar with these principles, here are some quick tips.

To search specifically for more than one word, put the search term in quotation marks. For example, “workers compensation”. This will limit your search to that combination of words.

To search for a combination of terms, use quotations and the & symbol. For example, “hurricane” & “loss”.

Login Register Subscribe

Flood risk models grow more sophisticated

Flood risk models grow more sophisticated

The havoc wreaked by Hurricane Harvey and other storms in recent years has taught valuable lessons about the changing nature of flood risk in the United States and the need to rethink old assumptions about the vulnerability of inland properties to flooding and the impact of weaker storms that instigate significant precipitation.

Evolving catastrophe risk maps and probabilistic models are factoring into this calculus — taking advantage of technological and data advances to enable risk managers and insurers to get a better handle on flood risk.

“Flood is a peril,” said John Burkholder, risk and insurance director for Tampa International Airport and chair of the Risk & Insurance Management Society Inc.’s external affairs committee. “You have to have underwriting data. Models aren’t perfect, but they’re certainly getting much better because we have better technology and better data. As we develop those models, it should improve the underwriting and should improve the ability to take care of the risk.”

Private mapping and modeling tools are allowing risk managers to simulate flood scenarios and more strategically deploy resources to mitigate flood risk based on which facilities are most vulnerable, particularly because they can determine the probability of a high loss, even in areas that Federal Emergency Management Agency maps designate as low-hazard flood zones, experts say.

“This is absolutely essential to the success of the flood insurance industry, and changes are happening fast and furiously,” said John Dickson, president of NFS Edge Insurance Agency Inc., an affiliate of Aon National Flood Services, in Kalispell, Montana.

In 2017, FM Global published its interactive Global Flood Map, which provides a global view of high- and moderate-risk flood zones and won a Business Insurance Innovation Award.

“The models are just now starting to catch up with the demand and are still probably a little bit behind,” said Louis Gritzo, vice president of research at FM Global’s Center for Property Risk Solutions in Norwood, Massachusetts. “The demand is driven by increased flood losses due to development into more value at risk. This is one of those areas where there is pretty sound science that says if the air temperature warms, there’s going to be more extreme precipitation.”

Advances in technology, including in digital elevation modeling that creates a digital representation of ground topography and terrain, and the availability of extensive data have enhanced the ability to evaluate individual properties for flood risk, experts say. Improved satellite technology, planes mounted with LIDAR — Light Detection and Ranging remote sensing to measure distances to Earth — and evolving GPS technology are contributing to the improved quality and quantity of data collected.

“Data accuracy is very important,” said Jayanta Guin, chief research officer at Bostonbased catastrophe modeling firm AIR Worldwide, a unit of Verisk Analytics Inc. “It’s not a perfect world yet, but we are well on our way to getting closer to perfection.”

Critical information such as ground elevation can be provided at resolutions of 10 meters compared with the previous norm of 90 meters.

“More than anything, flood is a high-resolution peril,” said Joshua Woodbury, flood specialist with Swiss Re Ltd. in Armonk, New York. “The difference between being flooded or not can be measured in meters. Resolution is important.”

Lessons from Harvey

The 2017 hurricane season demonstrated the importance of modeling not only flooding caused by storm surges during hurricanes, but inland flooding caused by major precipitation events, experts say.

Harvey made landfall in Texas as a Category 4 storm and lost strength relatively quickly, but lingered as a precipitation event with catastrophic consequences.

Harvey represented a “paradigm shift in how people think of hurricanes” because “the big story with Harvey was inland flood,” said Peter Dailey, vice president for global flood model with Newark, Californiabased risk modeler Risk Management Solutions Inc.

In September, Boston-based catastrophe modeling firm Karen Clark & Co. estimated Harvey insured losses at $15.4 billion, with $12 billion caused by inland flooding. “What we’re seeing more and more, particularly with weaker or stalled storms like Harvey, there’s a lot of rainfall associated with a hurricane,” said Chief Executive Officer Karen Clark.

More precipitation events unrelated to hurricanes are causing significant losses, and risk managers would be wise to plan for and mitigate against the risk, experts say.

“That is all consistent within the backdrop of climate change,” Mr. Guin said. “From a pure science point of view, we know that as the Earth is warming, the atmosphere can hold more and more moisture. From a physics point of view, the data is showing that the biggest impacts of climate change are in precipitation and extreme temperatures.”

For example, severe storms and flooding in Louisiana in August 2016 led to more than $2.5 billion in National Flood Insurance Program payouts, according to FEMA data.

“A lot of people mistakenly assume that flood is only an issue for communities on the coasts and next to some of the major rivers. But if you think about the 2016 rainstorms in the Baton Rouge area, that was an $8 billion event, and it affected over 55,000 homes,” said Nancy Watkins, a San Francisco-based principal and consulting actuary for Milliman Inc., which is designing a new nationwide rating plan for NFIP policies. “Most of those people were not insured for flood. Any place in the country where it can rain a lot in a small amount of time can have a flood. This is a 50-state problem.”





Read Next

  • Probabilistic flood models gain a following

    Flood maps provided by the Federal Emergency Management Agency had traditionally been the go-to tool for analyzing flood risk, but even the agency has acknowledged the value of and shift toward probabilistic models developed by the private sector.