Homeowners, farmers, and municipal budgets across the United States are absorbing the costs of more than a dozen weather events in 2026 that each caused at least a billion dollars in damage. But the federal government’s long-running official tally of those disasters no longer exists. NOAA’s National Environmental Satellite, Data, and Information Service ended updates to its Billion Dollar Weather and Climate Disasters product after calendar year 2024, leaving anyone tracking 2026 losses to piece together numbers from scattered federal datasets that were never designed to work alone.
Why the retirement of NOAA’s disaster tracker changes the math
For more than four decades, NOAA’s National Centers for Environmental Information maintained the only continuous, peer-reviewed federal record of billion-dollar U.S. weather and climate disasters, covering events from 1980 onward. That dataset, cataloged as NCEI Accession 0209268 with DOI 10.25921/stkw-7w73, drew on a specific pipeline: ISO property-loss estimates, FEMA public assistance records, USDA crop indemnities, and Commerce Department inputs, all cross-checked against one another before a final cost figure was published. The result was a single, comparable time series that Congress, insurers, and local planners relied on to measure whether disaster costs were accelerating.
That pipeline is now frozen. NOAA’s NESDIS notice states the disaster product is being retired with no updates beyond calendar year 2024. The dataset’s status is listed as archived and complete. Any claim that the U.S. has already logged more than a dozen billion-dollar disasters in 2026 therefore cannot come from the official NCEI series. It must come from alternative aggregators, whether insurers, academics, or state agencies, that lack the multi-source cross-check the federal product required.
The practical consequence is significant. The retired NOAA model did not simply add up insurance payouts. It blended private-sector loss data with federal disaster declarations and agricultural indemnities, then applied a consistent methodology documented through peer-reviewed research hosted on Drought.gov. Without that blending, any replacement tracker that relies on a single data stream, such as USDA crop insurance files or wildfire acreage alone, is likely to produce cost estimates with substantially higher variance from event to event. A Congressional Research Service report, designated IF12944, previously explained for policymakers how the public-private data fusion worked and why the methodology mattered for budget planning.
In practice, the old NCEI approach created a common language for very different types of disasters. A Midwestern drought, a Gulf Coast hurricane, and a Western megafire could all be compared on the same inflation-adjusted scale. Lawmakers used that record to argue over trends in climate risk, and federal agencies cited it when justifying investments in levees, forest management, and drought resilience. With the series now capped at 2024, those trend lines abruptly stop, even as the underlying physical hazards continue to evolve.
Federal data that still exists and what it cannot do alone
Two major federal data sources continue to publish raw inputs that fed the old NOAA model, but neither produces an integrated dollar-loss total on its own. The USDA Risk Management Agency maintains cause-of-loss files that record crop insurance indemnities by peril and geography. These files can show, for example, how much drought, hail, or freeze damage triggered payouts in a given state and year. They do not, however, capture property destruction, infrastructure repair, or uninsured losses, categories that often dwarf agricultural costs in hurricanes and flooding events.
The National Interagency Fire Center publishes wildfire activity statistics, including acres burned and incident counts. Those figures help confirm the timing and scale of fire seasons but stop well short of a dollar estimate. Suppression costs, structure losses, evacuations, and long-term health effects all require separate accounting that NIFC does not perform. In the old NOAA system, wildfire acreage was one input among many; FEMA public assistance data and state damage assessments filled in the rest.
FEMA’s OpenFEMA dataset on public assistance funded projects offers another partial view. It records federal disaster spending after a presidential declaration, but final cost figures often lag by months or years, and the dataset covers only the federal share of eligible expenses. Private losses, which typically represent the majority of total disaster costs, sit outside its scope entirely. Even within FEMA’s own ledgers, mitigation grants, temporary housing, and individual assistance programs are tracked separately, making it difficult to roll them into a single, timely estimate for any one storm or fire.
Other federal sources add still more fragments. The U.S. Army Corps of Engineers tracks spending on levee repairs and navigation channels after floods. The Small Business Administration reports on disaster loans to homeowners and firms. Health agencies monitor spikes in emergency-room visits during heat waves and smoke events. Each dataset captures a sliver of the economic impact, but none is designed to serve as the definitive ledger of total losses.
Gaps in 2026 disaster accounting and what to watch
The core problem facing anyone who wants an accurate 2026 disaster count is that no single replacement has inherited the full methodology NOAA used. The retired NCEI series cross-referenced at least four independent data streams before assigning a cost to any event. Alternative trackers that skip even one of those streams, particularly the ISO property-loss estimates or the FEMA cross-check, are working with an incomplete picture. The result is that two credible-looking tallies of the same storm or drought can diverge sharply, not because the weather differed but because the accounting did.
This matters directly for communities applying for federal mitigation grants, insurers setting regional premiums, and state legislatures debating building codes. A coastal county trying to justify stricter elevation standards may now have to commission its own loss study, rather than pointing to a nationally recognized NOAA figure. Rural electric cooperatives facing repeated ice storms or derechos could struggle to demonstrate that their cumulative damages meet thresholds for certain funding programs, because there is no longer a single, authoritative federal series to validate their claims.
For researchers, the retirement of the dataset introduces a break in what had been a continuous record. Studies that once extended trend lines forward each year must now either stop at 2024 or splice in numbers from private catastrophe-modeling firms or state reports, each with its own assumptions. That makes it harder to say with confidence whether billion-dollar disasters are becoming more frequent, or whether apparent increases reflect only shifting exposure, insurance coverage, or accounting practices.
In the absence of a direct successor, observers watching 2026 losses will need to pay closer attention to methodology. Does a given tally adjust all events to current-year dollars, as NOAA did? Does it include uninsured damages and indirect costs, or only insured property claims? Are agricultural losses, wildfire suppression expenses, and public infrastructure repairs counted on the same footing? Transparent answers to those questions will matter as much as the headline number itself.
The policy stakes are likely to grow. As Congress debates disaster aid reform and long-term climate adaptation, lawmakers will be working from a more fragmented evidentiary base than they had just a few years ago. Unless a new, coordinated federal effort emerges to replace the retired NOAA product, the “cost” of 2026’s billion-dollar disasters will increasingly depend on who is doing the counting-and which pieces of the damage they choose to leave out.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
