GISWeekly examines select top news each week, picks out worthwhile reading from around the web, and special interest items you might not find elsewhere. This issue will feature Industry News, Top News of the Week, Acquisitions/Agreements/Alliances, Announcements, New Products, Around the Web and Events Calendar.
GISWeekly welcomes letters and feedback from readers, so let us know what you think. Send your comments to me at Email Contact
Susan Smith, Managing Editor
Announcing the U.S. Flood Catastrophe Model
by Susan Smith
Flood risk models have been in the news lately, most likely in preparation for flood and hurricane seasons ahead. Last we covered a new Flood Risk Model for Great Britain. Each flood risk model looks to incorporate more data into its database, as the more pertinent data that resides in the database, the more accurate an assessment of flood risk can be.
Just announced is U.S. Flood Catastrophe Model and advanced flood risk solutions proprietary software from the First American Flood Data solutions division, which is part of First American Spatial Solutions group based in Austin, Texas. First American’s Flood Model has been validated against hundreds of thousands of claims, and has assembled data from digitizing federal flood maps, proven hydraulic science and actual historical flooding events.
The Flood Model, a key product in the company’s new suite of Advanced Flood Risk Solutions, includes geospatial analytics, basic flood data, flood risk scoring and near real-time monitoring of actual flood conditions. It is a. sophisticated risk management solution designed for insurance companies to measure their financial risk with regard to floods on several fronts: property, portfolio and policy.
Click to Enlarge ]
Parcels (in purple) within 2,500 feet of the ocean but not in a storm surge-risk zone
As a provider of services to the lending industry, First American sees a flood risk management product as a valuable asset. “Every time you buy a piece of property, if the bank that’s making the loan is a federally backed bank, which most of them are, they are required by law to determine whether that property sits within the flood rate maps that the national flood insurance program puts out,” said senior product manager, Jason Sears. “And so we have a large marketshare within that service industry, so we provide most of the about 65% of the better marketshare today of flood determination.” Consequently the company has a rich background in GIS because determining whether a property sits within the floodplain boundaries is a GIS function.
First American’s flood plain business utilizes several different applications, some are off the shelf, some are internal and proprietary. Right now, they run roughly around a million flood determinations a month depending upon the marketing conditions.
“For us to be efficient and cost effective we need to make sure those things hit automatically so it comes into our system and the GIS software automatically is able to tell us whether that property is in a flood zone or not. We have a lot of geocoding technologies and we have employed a lot of different databases to help us do that automatically. Around 90% of the time or better we’re able to get an automatic determination. The property comes in, hits our database, pops back out to the lender, says yes or no, this is in a flood zone. If it is, it tells them the flood zone they’re in. We have to be extremely accurate on that because we insure every flood determination that we do, so basically if it’s a residential policy and we say you’re not in a flood zone, and it turns out you are, and you’re flooded, we have to cover just as though you had the flood insurance.”
Click to Enlarge ]
Illustrates how the boundary of the flood waters (white line) extended beyond the 100-year floodplain (blue line) in Findlay, Ohio
In addition to having highly accurate geocoding capabilities, First American is able to utilize and gather a lot of data resources. Last April the company announced its intent to build the first national parcel database. By the end of 2008, First American will have 110 million parcels for the U.S. identified and captured in their database. You will be able to attach an address and a centroid with all the parcel boundaries. “There is a lot of interest from a whole lot of different industries: oil and gas, insurance, utilities,” said Sears.
First American recently acquired Proxix Solutions, provider of a geocoder called PixPoint, into which they have integrated all their parcels. Additionally Proxix’ insurance products around natural risk, such as wildfire, coastal surge, earthquakes and tax products fit naturally into First American’s focus.
With the acquisition of Proxix, First American could also incorporate its property data and tag it to parcels. Putting this information together with GIS, risks and scoring capabilities, the company has formed a new organization called First American Spatial Solutions.
According to Sears, flood zone determinations were developed by FEMA in support of the National Flood Insurance Program which informs an owner if a piece of property is in or out of a flood zone for the purposes of qualifying that person for National Flood (subsidized ) insurance. “That’s fine, if that’s the business you’re in,” Sears pointed out. “However, for a number of insurers, they’re looking for something that is more flexible than that program, and more robust, meaning they’re dealing with multi-million dollar structures. For commercial insurers, the National Flood Insurance Program doesn’t even begin to address the cost of working with the claim, so they need something better than that. Something more extensive, accurate, more granular in determining the level of risk.”
Important factors for these insurers:
1) Accuracy – FEMA’s flood zone maps and flood zone determination product reflect the most extensive engineering effort nationwide to date. However, these are not good tools for determining accuracy and insurers say they need more reliable data. “Around 25-30% of flood claims filed every year are on properties outside that flood plain,” Sears said. In the case of specific storms and catastrophes, the percentage has ballooned up as much as 60% of properties that flood based on flood claims filed that are outside the flood plain.
2) Better understanding of whether high or low risk in a flood plain.
3) Flood risk as opposed to flood exposure. Flood risk is probability that an event is going to occur. Flood exposure is anticipated or forecasted losses due to flood within any given year. What’s the worst case scenario? “The way the insurance industry looks at it, they look at return periods, so they’re asking the question, was it a 500-year return period or every 500 years? What would the worst event in terms of financial consequences be?”
Sears stated that in order to determine this, you not only need to know the probability that an event is going to occur, and the frequency, you also need to know the severity, i.e., what’s the depth of flooding in a given spot in a given building, say 2 ft. or 4 ft. “You need to know what does 4 ft. mean, in terms of damage. Is that 20% of damage to the building? Half the value of the building? Will business shut down for a few days? What kind of contents are in that building?”