As climate change intensifies severe rainstorms, the infrastructure protecting millions of Americans from flooding faces growing risk of failures, according to new calculations of expected precipitation in every county and locality across the contiguous United States.
The calculations suggest that one in nine residents of the lower 48 states, largely in populous regions including the Mid-Atlantic and the Texas Gulf Coast, is at significant risk of downpours that deliver at least 50 percent more rain per hour than local pipes, channels and culverts might be designed to drain.
“The data is startling, and it should be a wake-up call,” said Chad Berginnis, the executive director of the Association of State Floodplain Managers, a nonprofit organization focused on flood risk.
The new rain estimates, issued on Monday by the First Street Foundation, a nonprofit research group in New York, carry worrying implications for homeowners, too: They indicate that 12.6 million properties nationwide face significant flood risks despite not being required by the federal government to buy flood insurance.
The nation is set to pour hundreds of billions of dollars into new and improved roads, bridges and ports in the coming years under the bipartisan infrastructure plan that President Biden signed into law in 2021. First Street’s calculations suggest that many of these projects are being built to standards that are already out of date.
Matthew Eby, First Street’s executive director, said he hoped the new data could be used to make these investments more future-proof, “so that we don’t spend $1.2 trillion knowing that it’s wrong.”
The National Oceanic and Atmospheric Administration, the agency under the Commerce Department that produces the precipitation estimates used by planners and engineers across the country, declined to comment.
NOAA’s estimates are “the floor, not a ceiling,” said Abdullah Hasan, a White House spokesman. “States and localities often consider additional factors best suited to their local geographies when making project decisions.”
Every additional increment of global warming increases the likelihood of intense rain in many places for a simple reason: Hotter air can hold more moisture. But NOAA’s estimates of expected rainfall are only intermittently updated. And, as NOAA scientists described in a recent report prepared in collaboration with university researchers, the agency’s estimates assume that the intensity and frequency of extreme rain hasn’t increased in recent decades, despite ample evidence to the contrary.
The result, according to First Street, is that NOAA is substantially underestimating the risk of severe rain in some of the nation’s largest cities: Baltimore, Chicago, Dallas, Detroit, Houston, New York, Philadelphia and Washington among them. Other places where there are large differences between First Street’s rainfall estimates and NOAA’s include the Ohio River Basin, northwestern California and parts of the Mountain West.
In other areas, including those east of the Sierra Nevada and Cascade Range, First Street finds that NOAA is overestimating the likelihood of intense rain, implying that resources there might not be best spent on upgrading flood infrastructure.
NOAA and its predecessor agencies have been publishing data on expected rain and snow for decades. Its latest estimates, covering nearly every part of the country, are contained in a multivolume publication called Atlas 14. (Another set of estimates, called Atlas 2, covers the Northwestern states.)
Pick any point on the map, and the NOAA atlases tell you the probabilities there of various precipitation events — that is, a certain number of inches falling over a given span of time, from five minutes to 24 hours to 60 days.
But the atlas estimates are based on rain measurements collected over the past several decades, or, in some places, since the 19th century, “in a climate that just doesn’t exist anymore,” said Jeremy R. Porter, First Street’s head of climate implications research.
By contrast, First Street’s peer-reviewed methods for estimating precipitation use only rainfall records from this century, and only ones collected by the government’s most modern weather stations. (First Street plans to publish additional documentation on how it computed its new estimates on July 31.)
NOAA is working on updating its atlas estimates to better account for the warming climate. But the agency says its first data for Atlas 15 might be ready only in 2026.
First Street’s rain estimates also raise questions about the federal government’s guidance on flood risks to homes.
The Federal Emergency Management Agency maps areas of the country that it calculates to be at significant risk in a 100-year flood, or one with a 1 percent chance of occurring in any given year. FEMA’s maps guide decisions by builders, insurers and banks, and determine whether homeowners need to buy flood insurance.
But First Street’s data suggests that 17.7 million properties nationwide are at risk in a 100-year event. Of those, only about 5 million properties also fall into a FEMA flood-hazard zone. That means millions of other homeowners might be making decisions with an incomplete understanding of the true physical and financial risks they face.
In Houston, 145,000 properties lie in First Street’s 100-year flood zone but not in FEMA’s. New York has 124,000 such properties; Philadelphia, 108,000; and Chicago, 78,000.
In an emailed statement, FEMA said it welcomed outside efforts to improve the nation’s understanding of flood risk but cautioned that First Street’s assessments relied on data and methods that were different from its own.
“FEMA’s process is careful to neither understate nor overstate the current flood risk,” the statement said. “The accuracy of the flood data necessary to service the nation’s largest flood insurance program and the nation’s largest regulatory land use program is fundamentally different than the level of accuracy necessary to support First Street Foundation.”
NOAA began publishing Atlas 14 in 2004, which means that any drains, culverts and storm-water basins built since then might potentially have been sized according to standards that no longer reflect Earth’s present climate. But plenty of America’s infrastructure was laid down even earlier, meaning it was designed to specifications that are probably even more obsolete, said Daniel B. Wright, an associate professor of civil and environmental engineering at the University of Wisconsin-Madison.
“Certainly, updating Atlas 14 is something that needs to be done,” Dr. Wright said. “But the problem is huge, in the sense that there are trillions upon trillions of dollars of things that are based on horribly out-of-date information at this point.”
