Homeowners whose properties sit outside federally designated flood zones often assume flooding is someone else's problem. No rule requires them to buy flood insurance, and the map suggests they are in the clear. The reality, however, may be very different.

Mounting evidence shows that significantly more homes are at risk of flooding than FEMA's 100-year flood maps indicate — roughly 6 million more, and 14.6 million properties in all, by one estimate. The official maps are commonly outdated, and some are two decades old. Even when updated, they typically rely on historical flood data rather than projections of future risk. The '100-year' terminology is easy to misread: it describes an event with a 1% chance of occurring in any given year, not one that arrives only once a century — a distinction that matters for anyone weighing whether a home outside a mapped zone is truly safe.

But there is another problem that redrawing the lines will not solve. FEMA's flood maps answer only one question: is the water likely to reach the neighborhood? They do not assess an individual home's risks, or what the impacts are likely to be when the water rises. Understanding that impact requires more comprehensive data, including the probable flood depth at the home, the building's age and construction quality, and insurance coverage.

Same street, different risks

Once those factors are considered, a home's true flood risk is far more complex than the map might suggest. The data shows how two houses on the same street can face radically different risks from flooding, even when located in the same flood zone.

One home may be newer, well maintained and covered by flood insurance. Another may be decades older, sitting at the low end of the same street where water runs deeper during a flood, in poor repair and uninsured. The same storm would leave the two households in very different predicaments, with very different paths to recovery.

Neighborhood averages can make it hard to tell which homes are likely to be hit hardest. Yet understanding an individual property's risks matters both to residents figuring out how to protect a home and to local officials deciding where limited flood-protection money should go.

Those are the conclusions of researchers who study how flooding affects homes and the people who live in them. To see how much house-by-house differences matter, the article's authors — Yilei Yu and James R. Elliott of Rice University and Dylan Connor of Arizona State University — together with colleagues Aaron Flores of Arizona State University and Sara Meerow of the University of Colorado Boulder, built a detailed flood risk score for nearly 1 million homes in Houston and across Harris County, where the city sits. The results are described in a new study published in the peer-reviewed journal Applied Geography.

The team's past research in Houston found that more than half of the 100,000 documented flood-prone buildings in the city currently lie outside official flood boundaries, and that these overlooked buildings are more likely to be in poor condition and uninsured, leaving them especially exposed when the floodwaters do come. The new research helps communities locate those homes.

Four factors that make a home vulnerable

To better understand each household's vulnerability, the researchers combined four types of information about each home:

• Water: How deep would floodwaters likely get at the home in a 100-year flood event, defined as having a 1% chance of occurrence in any given year? The team took a national flood model's map of estimated parcel-level flood depths across Harris County and recorded the estimated depth at each residence.
• Building: How old is the home, and how well has it been constructed and maintained? Older, lower-quality homes in poor repair tend to suffer more damage when floodwater enters the structure. County appraisal records collected by expert assessors can provide critical information for making such judgments.
• Mobility: How busy are the roads nearby? For each home, the researchers measured typical traffic within about a half-mile and within 3 miles. The busier those surrounding roads, the more a flood disrupts trips to work, school and hospitals, and the harder it is for first responders to get in and trapped residents to get out.
• Finances: How much would it cost to repair the home, and what share of nearby homes have flood insurance? Neither keeps water out of a home, but both reflect what it costs to recover.

The four criteria were chosen because flood research consistently highlights their importance for how hard a flood hits a community: how much water arrives, how well the building holds up, whether people can still get around, and how big the repair bill is for the owner. They are not the only possible measures; a city could use alternatives within each category, depending on available data.

What the study found

Homes in the same neighborhood can have very different flood vulnerability scores. The study found the scores differed more from house to house within a neighborhood than from one neighborhood to the next.

In practice, this means that knowing a neighborhood's flood risk does not necessarily tell how hard a flood could hit any individual home. A neighborhood could look safe on average while containing some of the county's riskiest homes, and there may be many well-protected homes in an otherwise risky neighborhood. Much of the problem, the researchers show, lies at the scale of buildings, not neighborhoods or ZIP codes.

What residents and communities can do

Homeowners can check several of these risks themselves. They can ask longtime neighbors or the city's flood plain office whether the home or the street has flooded before; look up the home's age in local property records and have an inspector assess its condition; find out whether the routes to work and school flood easily; and check whether they hold a flood insurance policy and what it covers. Most homeowner's policies do not cover flood damage, which is why flood coverage is a separate purchase.

Communities can apply the same approach with their own data to locate particularly vulnerable housing where neighborhood averages may disguise the need for help. They can start with existing address-level flood scores, such as First Street's, and add local records on buildings, roads and insurance. The extra details matter because officials have limited money for flood-proofing homes, improving drainage and helping families recover after floods. Better data on buildings and flood risk can help them invest wisely.

Data similar to that used for Houston exists across much of the United States, so other flood-prone cities could follow the same approach, putting house-by-house flood scores within reach of far more communities.

Floods do not stop at the lines on a map, and they do not hit every house on a street the same. The researchers believe it is time for residents and local flood authorities to prepare accordingly.

Yilei Yu is a Senior Researcher and Postdoctoral Fellow in Geography at Rice University. Dylan Connor is an Associate Professor of Geographical Sciences and Urban Planning at Arizona State University. James R. Elliott is a Professor of Sociology at Rice University.

This article is republished from The Conversation under a Creative Commons license. It was originally featured on Fortune.com.