Welcome to GISWeekly!
GISWeekly examines select top news, picks out worthwhile reading from around the web, and special interest items you might not find elsewhere. This issue will feature Industry News, Top of the News, 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 giscafe-editor@IBSystems.com
Susan Smith, Managing Editor
CoreLogic’s Annual Storm Surge Report Reflects the Times
By Susan Smith
New Orleans Parcel Maps
In a recent interview with Dr. Howard Botts, executive vice president and director of database development for CoreLogic Spatial Solutions, information and geospatial technology solutions, he spoke of the annual update of the company’s storm surge report and what those enhancements look like. In keeping with the times, the storm surge report includes 1) the huge dollar exposure – the $300 billion in residential areas that would be affected by storm surge and hurricanes in the coming hurricane season, which is predicted to be a more active season, 2) the storm surge vs. FEMA flood zone information, 3) is the change the U.S. Army Corps of Engineers has implemented in New Orleans for category 1 and 2 hurricanes is going to provide a lot more protection in the greater New Orleans area.
GISWeekly: Can you explain the importance of these developments?
Howard Botts: What we’ve done is annually we update our surge report, this year the biggest single change was the Army Corps of Engineers finishing the levee and other projects for the storm surge barrier in New Orleans. The biggest single change was the completion of their storm reduction project – that had a significant impact in reducing the number of houses exposed to extreme risk in the area. The other thing is that we’ve gotten a lot more parcel information between this year and last year and enhanced property values and we’ve expanded the geographic coverage of the particular cities we looked at, to give a more comprehensive look at Mobile, Ala for example, the entire urban area, or the Miami to Palm Beach area. We extended that out to the counties, the dollar value exposure went up in virtually every market.
New Orleans change 2010-2011
The other significant change we made this year as reflected in the report, was looking at a comparison of properties in each market that were in a storm surge zone or in a FEMA 100 year flood plain. What we’ve discovered working with lots of clients is there is a false perception on the part of homeowners that if they’re not in a FEMA 100 year flood plain that they would have to get national flood insurance in order to get a home mortgage and...they are safe from coastal flooding issues. In fact we found they couldn’t be more wrong. If you take Virginia Beach as a most extreme case, where 87% of the residential properties in the greater Virginia Beach area are in a surge inundation area that are not in a FEMA flood plain zone. Only a very small fraction of the people in that market are actually protected by national flood insurance. A lot of insurance companies that use our data are mentioning to customers that they are in a surge risk zone and that you may want to consider buying national flood program insurance which is very inexpensive if you’re out of the 100 year flood plain. Of all the takeaways in this report this is most striking.
Virginia Beach 3D view
We’re talking about huge numbers of properties in this case, Virginia Beach would be 253,000 properties that are exposed to surge but not in a flood zone. You find that’s the most dramatic – Tampa is another 177,000 properties are exposed, Long Island almost 200,000. I would’ve thought there would have been more correlation, but they are significantly different.
The other big change we’ve seen, that started with Katrina but was reinforced by the recent Japanese tsunamis, was the greater awareness on the part of oil and gas utilities as to their vulnerability. A lot of them started to use our datasets to evaluate enterprise risk management; in particular what parts of infrastructure of an area will be damaged by storm surge, what areas are going to be inaccessible? Do we have power production or oil storage adjacent to water which would be impacted? Do we have enough mediation around those seawalls and other things that would prevent critical infrastructure or things that would be potentially high pollution sources from being damaged or destroyed?
GISWeekly: Will you build something more specific to those who are outside the floodplain?
HB: We have probably four groups of end users. We have both datasets -- CoreLogic is the largest provider of flood certification. We have all the FEMA flood zones and we work with the majority of the mortgage companies to do the flood certification when you buy a property. So we have the most up to date FEMA flood zone data and then you marry that with our storm surge inundation data. Together that really provides a property by property level because we’re modeling this very granular 10 x 10 meter or 30 x 30 meter so for any individual property we can tell you are you in or out of either of those. So our end users will utilize this to then either build proprietary models to evaluate their exposure or they will work with individual customers to make them aware of the big incentive for insurance companies.
When we look at property and casualty and homeowners, they do not cover flood waters as part of your homeowner policy and they thought that would give them a level of protection from non-wind damage in hurricane. Even though they were telling the homeowner that storm surge was what knocked your house off and destroyed your home and now you have only a slab, and that they won’t cover it. This has caused a lot of ill will where people believed it was wind not water that destroyed their homes. So insurance companies are now saying we know your house is at risk and we won’t insure it, but the federal government will insure it through their program insurance. That’s the major place where I see end users utilizing our data. More and more we’re finding out insurance companies we work with want to provide mitigation recommendations to customers. There’s often the sense that companies are looking at ways to increase rates, but given the softness of the insurance market these days most companies would rather work with the insureds and try to minimize the likelihood of damage. It is really only possible by the fact that we can identify risk at the individual property level.
GISWeekly: What role then is mapping playing in this?
HB: It’s huge. In the insurance space probably about 99.9% of our end users never see a map, it’s all automated spatial processes so you use all the processing power of a GIS, you just don’t get a map. Then for one/tenth of 1 percent which may be exceptions, they do want to see a map. The real power is in the spatial metaphor -- being able to drill down through all these different layers and extracting out what the risk is, and they feed those directly into real time pricing and quoting that deliver rates.
None of this is possible without GIS and geospatial processing.
GISWeekly: You’re using FEMA maps and other kinds of sources for the maps for the base database?
HB: For our storm surge datasets we’re talking about here, that’s a completely modeled dataset that we build and we look at offshore characteristics. We model hurricane speed, atmospheric pressure, a wide variety of things and then look at ocean bottom depth and we figure out the maximum amount of water that’s to be pushed up on shore. For the onshore piece we use digital terrain models. We use information on levees, railroad barriers, and other things. Most of the data we derive from this will generate internally now for flood certification. That’s really a government defined process where FEMA works with the local entities and does engineering studies to find the 100 year flood plain. In that case the government defines what the boundaries are, so in that case we would use FEMA data and we convert it to our particular models.