If a new study led by the University of Bristol is to bebelieved, 41 million Americans are at risk from flooding rivers,according to phys.org. That's more than three times thanthe current estimate of 13 million people.

The Bristol study is based on a new high-resolution model thatmaps flood risk across the entire continental U.S., whereas theexisting regulatory flood maps produced by the Federal EmergencyManagement Agency (FEMA) cover about 60% of the continentalU.S.

The increase in numbers of those at risk is a result of theexpanded coverage of the map combined with its ability to estimateflooding on small streams — something that wasn't adequatelycaptured in previous flood-risk models, according to the study'sresearchers.

Related: Tackling the U.S. flood insurancechallenge

The 41-million estimate does not include the millions ofadditional Americans that are at risk of coastal flooding — and thestudy predicts that more than 60 million Americans may bevulnerable to a 100-year flood by 2050.

While it does make a terrific NU headline to state thatU.S. flood risk is far greater than FEMA would suggest, is thesituation as really as dire as it sounds? Should the findings of aresearch team at a U.K. academic institution to be given enoughcredence to change the way insurers — and the buying public —perceive flood exposure in the U.S.?

PropertyCasualty360.com reached out to Peter Bingenheimer, Senior Vice President ofthe Consulting and Clients Service Group at AIR Worldwide, a leadingprovider of catastrophe risk modeling software and consultingservices. (Bingenheimer works with AIR's U.S. insurer, insurancebroker, and reinsurance broker clients.) He commented on thefindings of the study but not the validity of the study itself,since AIR has not performed a thorough review of the underlyingmodel or methodology.

We asked, what might the Bristol report mean for the way theP&C industry should view flood exposures in the U.S.? AreFEMA's previous models really that reliable?

Historically, Bingenheimer explained, FEMA maps are the standardby which flood exposure is measured in the U.S.: By and largethey're of very good quality, and an excellent benchmark. FEMAflood maps, he noted, have many strengths: They take into account agreat deal of granular local data, but they do have theirlimitations in terms of “variation in their vintage” (a turn ofphrase I now plan on employing whenever possible) and consistencyin how they're developed. Geographically speaking, he added, theydo possess some gaps.

“We're now at an inflection point where science and data andtechnology can create better models than ever before, morecomplete, consistent views of risk at scale,” Bingenheimerstated.

That said, he added the Bristol study's findings aredirectionally consistent with AIR's own research, that “there issubstantial risk beyond the FEMA flood maps, broadly speaking,consistent with what we observe.”

While the Bristol report won't necessarily spur more majorstudies by U.S. modelers in the near future, he said, “it'sindicative of what you can expect to see more of” in years tocome.

In other words, just because U.S. cat modelers use FEMA as abenchmark doesn't mean they believe the level of flood risk is whatFEMA says it is. And the Bristol report does much to illustrate howmuch further research needs to be done on this side of thepond.

The biggest takeaway from all this for insurers? “The findingsof this report underscore a major opportunity that's in front ofthe industry, because it highlights how much property isuninsured,” Bingenheimer added.

For more information about the Bristol study, checkout EnvironmentalResearch Letters, where it was firstpublished.