Why the 100-Year Floodplain Needs to be Rethought

Still from drone footage shot in Northwest Houston after flooding in 2016.

As of April 2017, there were just under 5 million active flood insurance policies under the National Flood Insurance Program. Since the program’s creation in 1968, the 100-year floodplain has been the standard tool used to assess the flood risk for a given property. Areas inside the 100-year floodplain have a 1 percent chance of flooding each year and are generally required to buy federally-regulated insurance by lenders. But that measure has been increasingly called into question, along with the role of the flood insurance program.

A recent study from a team of Texas researchers estimates nearly a quarter of insured flood losses nationwide come from properties outside the 100-year floodplain. That number goes up in coastal areas. In Harris County, for example, almost half of all repeat insured flood losses between 1978 and 2008 were located outside the 100-year floodplain.

The study, published online earlier this year and in print in the August issue of Natural Hazards Review, points out several major limitations to the current models used to map flood risk, including the fact that those models don’t take into account changing land use, like loss of wetlands or additional impervious surfaces — both important factors in the fast-growing Houston metropolitan area.

Focusing on losses incurred during five different storm events between 1999 and 2009 in the Armand Bayou Watershed, southeast of Houston, Russell Blessing, an urban and regional science graduate student at Texas A&M University, Antonia Sebastian, a civil and environmental engineering graduate student at Rice University and Sam Brody, a marine science professor with Texas A&M University at Galveston, found some $29 million in insured flood loss in two sub-basins in the watershed. Well over half were from areas outside the 100-year floodplain, so 80 percent of the claims and 75 of the insured losses were outside the floodplain. For the most part, the age of the home was not a factor in loss. Using a different mapping tool that compares the inundation depths over the duration of a storm to those upstream, the researchers were able to expand the 100-year floodplain to cover almost 2.5 times more claims.

“In general, flood claims that the floodplain failed to capture were located in areas characterized by poorly drained soils, high imperviousness, and relatively low roughness when compared with claims within floodplain boundaries,” the study explained.

Continuing to use the less accurate 100-year floodplain currently employed will likely only make the situation worse. In essence, the report noted, “it creates an environment in which riskier development decisions are more likely to occur in areas just outside of the floodplain which would actually be considered floodplain were they modeled more accurately.”

In the meantime, Brody, a Kinder Institute research fellow and the director of the Center for Texas Beaches and Shores, has launched a searchable website for Houston area homebuyers and homeowners to assess flood and other natural hazards risks via an interactive map. Currently covering Harris County, the Buyer Be-Where map will soon include Galveston County as well.


An earlier version of this story included the wrong number of active policies in the National Flood Insurance Program and misstated requirements for homeowners inside the floodplain to buy insurance. The story has been updated to reflect the changes.

Print Friendly, PDF & Email

Leah Binkovitz

Leah Binkovitz is Senior Editor with the Kinder Institute for Urban Research.


  1. The NFIP is a creation of Congress meant to insure against a peril which the private insurance industry refuses to insure against. The reason private insurance refuses is because there is no actuarially sound method to do so. Effectively, federal flood insurance is not pure risk insurance but actually a type of social insurance. To move the conversation forward on flood insurance reform, that fact has to be acknowledged first. By properly characterizing it as social insurance, we can start talking about how much more people in flood prone areas should pay.

  2. I think there is a typo – if 5 billion people had flood insurance the program might not be in debt 🙂

  3. “Areas inside the 100-year floodplain have a 1 percent chance of flooding each year and are required to buy federally-regulated insurance.”
    Technically, floodplain insurance is not required for these owners, but lenders will not finance them without it.

  4. The “Buyer Be-Where” map references incorrect data. Checking 415 White Wing Lane, the rating is low to very low. Flood is a risk factor 3. The lot abuts the bayou and has flooded several times in the last few years alone.

    I see the data is from Zillow. Zillow is known to be completely useless as a real-estate database.

  5. Flooding & Ground Water :

    Robindell neighborhood off of Braes Bayou and Hillcroft probably has been subsiding for years – the ground is lowering itself and “sinking” … The reason could be that residents are paying their water bill … YES their water bill represents their payment to the City for Water sucked up from under the neighborhood since they have a large well right next to Reamer street and 7 Acres retirement home.

    The result is that the neighborhood is sinking and flooding happens more often.

  6. Imagine paying for Flood Insurance in order to cover for the fact that when you are paying your water bill you are making your neighborhood more prone to flooding because ground water is removed under the area and the ground subsides. This is what is happening.

  7. The current models DO account for land use, etc., but not for ‘changing’ land use, etc. You would have to have some sort of dynamic model that was constantly updating to achieve anything close to that. Probably not technically possible at this time and definitely not economically possible. Not to mention that a regulatory boundary needs to be a somewhat static ‘line in the sand’ and cannot be moving all the time. Part of the problem is the regulatory floodplain and flood risk should be viewed as independent things for different purposes.

Leave a Reply

Your email address will not be published. Required fields are marked *