NOAA retired the billion-dollar disasters database. It was the best water-damage record we had.

In May 2025 NOAA announced that its National Centers for Environmental Information would no longer update the Billion-Dollar Weather and Climate Disasters product, citing evolving priorities, statutory mandates and staffing changes. The record runs from 1980 to calendar year 2024 and stops there.

The archive remains available and NOAA describes the past reports and underlying data as authoritative. The methodology has since been picked up outside government, with Climate Central relaunching the database using the same approach.

The practical consequence for anyone writing about water damage is small but real: this is now a historical series. Citing it as though it were current is a quiet error, and one that will get more misleading every year.

Events causing at least one billion dollars in losses, adjusted for inflation to a consistent basis, across tropical cyclones, inland flooding, severe storms, winter storms, wildfire, drought and heat.

The frequency trend is not subtle. The 1980s averaged 3.3 qualifying events a year. The 2000s produced 59 events and the 2010s 119 — roughly 5.9 and 11.9 a year. The five years from 2020 to 2024 averaged 23.0. Across the full 1980 to 2024 record the annual average is 9.0, and losses over the final decade averaged around 140 billion dollars a year.

Part of that increase reflects more people and more valuable property standing in the path of the same weather. That is a well-documented component and it does not disappear because the headline number is dramatic.

A disaster catalogue measures money, not water in buildings. One hurricane that floods two hundred thousand homes and one drought costing the same amount each count once, and only one of them wets a house.

It also only ever saw the top of the distribution. The ordinary causes of indoor moisture damage — a supply line that fails while nobody is home, flashing that has leaked for two winters, a basement wall that takes water every spring — never appear in it, and between them they account for far more damp buildings than catastrophes do.

What it supported, and all it supported, was a narrow and useful statement: the number of American homes passing through a major wetting event in a given year is several times what it was in the 1980s. For assessing a specific property, a water event in its history matters — not as a verdict, but as one factor among several.

For the historical record, nothing needs to: the 1980 to 2024 archive is intact. For continuation, Climate Central's relaunch is the series to watch, with the caveat that a dataset maintained outside the agency that created it will take time to establish the same standing.

For the part that was always missing — non-catastrophic household water damage, which is most of the problem — there is still no public national series. Insurers hold that data as claims. Until someone publishes it in a form comparable across decades, the honest answer to how much water damage American homes experience is that nobody outside the industry knows.