was destroyed. Open data saved it

Climate.gov was gutted when the Trump administration slashed NOAA's funding, pulling more than fifteen years of irreplaceable climate data, maps, educational materials, and government-commissioned research offline. What could have been a permanent erasure of the public record became something else — because a legal principle embedded in how the U.S. government publishes its data meant the information technically belonged to everyone, and a small team of former NOAA scientists refused to let it vanish.

The story of how open licensing stopped climate.gov's destruction from becoming an irreversible loss is a case study in why public-domain mandates for government data are not abstract policy debates. They are infrastructure. And right now, that infrastructure is holding on by a thread.

What Climate.gov Was — and Why Its Loss Matters

For well over a decade, Climate.gov served as the authoritative, publicly funded portal for climate science in the United States. Operated under the National Oceanic and Atmospheric Administration (NOAA), the site aggregated and presented peer-reviewed data, interactive maps, educational resources, and high-level summary reports in a format accessible to scientists, educators, journalists, policymakers, and curious members of the public alike. It was not simply a government brochure — it was a living, curated dataset updated continuously by agency staff.

Among its most significant holdings was the Fifth National Climate Assessment, the federal government's most comprehensive analysis of climate change impacts on the United States, assembled over years by hundreds of contributing scientists from across academia, government, and the private sector. When Climate.gov was pulled offline, that document was deleted from its primary public-access location. Also at risk were Arctic sea ice records — including the closely watched September minimum extent, measured as the total area of the Arctic Ocean at least 15% ice-covered — NOAA's archive of oral histories from people whose lives had been shaped by climate events, monthly climate indicator reports, and a vast library of teaching materials used in schools and universities across the country.

The site carried over fifteen years of accumulated knowledge. Its sudden removal was not a server migration or a scheduled maintenance window. It was, as technologist and publisher Ben Werdmuller put it plainly, an "act of vandalism."

The Mechanism of Destruction: NOAA Gutted, Climate.gov Taken Offline

The immediate cause of Climate.gov's shutdown was the Trump administration's sweeping cuts to NOAA's budget and workforce — described by Werdmuller as the agency's funding being "radically cut." Hundreds of NOAA staffers were laid off in late February 2025, and by early March of that year reporting indicated roughly 1,300 employees — on the order of 10 percent of the agency's total workforce — had been removed. Later in 2025, the Department of Defense signaled it would stop providing certain satellite weather data that NOAA scientists and forecasters had relied on, compounding the damage to the agency's operational capacity and its ability to maintain long-term dataset continuity.

The defunding of NOAA was part of a broader dismantling of federal scientific and environmental infrastructure that also saw the records and activities of DOGE become subject to their own transparency questions. Climate.gov, stripped of the staff who maintained it and the budget that kept it running, simply stopped functioning as a live resource.

What alarmed the scientific community most was not just the loss of future data collection — it was the disappearance of the historical archive. Data that researchers, teachers, and journalists had relied on for years vanished from its canonical location, with no guarantee it would ever return. The Fifth National Climate Assessment represented an enormous collective investment of scientific labor spanning multiple federal administrations. Its deletion sent an unmistakable signal about the current administration's posture toward federally produced climate science. Smaller but equally irreplaceable holdings — localized climate indicator reports, sea-level trend visualizations, multi-decadal temperature anomaly records — faced the same risk of quiet, permanent disappearance.

Why Open Data Was the Saving Grace

Here is where the story turns from catastrophe to something more instructive. Under U.S. law, works produced by federal government employees in the course of their official duties carry no copyright protection. They enter the public domain automatically. This is codified in 17 U.S.C. § 105, and it covers the entire corpus of material NOAA published on Climate.gov — every data visualization, every report, every educational module, every map, every chart. As Werdmuller summarized the principle, "US government data is public domain by law."

Why it matters: "Had it not been available under a permissive license, the administration's act of vandalism would have meant the data was gone for good. But because it was, the datasets can find a new home." — Ben Werdmuller, werd.io

That single legal fact changed everything. Because the data carried no restrictive license, former NOAA employees could legally copy it, host it, and republish it without seeking permission from the agency that had tried — whether intentionally or through institutional negligence — to let it disappear. The public-domain status of U.S. government data transformed what might have been a grieving process into a rescue operation.

This stands in stark contrast to what would have happened if the same information had been published by a private contractor, a foreign government agency, or even many nonprofits operating under standard copyright. In those cases, the original publisher controls redistribution rights, and a government decision to take content offline would be legally enforceable against anyone attempting to mirror or rehost it. The open-data framework was not just philosophically appealing here — it was the only mechanism that made recovery legally and practically possible. Without 17 U.S.C. § 105, the loss of Climate.gov would have been irreversible for anyone outside the government, because no one else would have had the legal right to act.

The Team Behind Climate.us: Scientists Become Digital Preservationists

The rescue was not carried out by a large institution or a well-funded coalition. It was led by Rebecca Lindsey, a former NOAA employee who understood both the scientific value of the data and the technical reality that it could be saved. She joined forces with two collaborators: Anna Eshelman, another former NOAA staffer, and Mary Lindsey, Rebecca's older sister and also a former NOAA employee. On the successor site, Rebecca Lindsey is listed as director and managing editor, Anna Eshelman as lead designer, and Mary Lindsey as lead data visualizer. This three-person core team — each of whom had spent significant portions of their careers building and maintaining the original resource — became the driving force behind the successor site.

The result of their work is Climate.us, which launched as a direct continuation of what Climate.gov had been — not a memorial or a dusty archive, but an active, functional, regularly updated resource. Structurally, Climate.us operates as a project of Multiplier, which its footer describes as "a tax-exempt nonprofit 501(c)3 umbrella organization" based in San Francisco. That organizational home gives Climate.us a legal and financial scaffolding that allows it to receive tax-deductible donations and enter into institutional partnerships. Even so, the site depends heavily on those donations to stay operational — a precarious position for infrastructure that carries genuine public importance.

What Climate.us Actually Contains

The successor site is not a stripped-down placeholder. It has rebuilt substantial portions of the original Climate.gov offering, preserving both the data holdings and the accessibility-first design philosophy that made the original so widely used:

• Global Climate Dashboard — tracks key indicators including Arctic sea ice extent, atmospheric CO₂ levels, mountain glacier mass, ocean heat content, sea level, spring snow cover, incoming solar radiation, and surface temperature anomalies, each displayed with sourcing and methodology notes.
• Data Snapshots / Maps & Data — regularly updated U.S. and global climate maps formatted for direct use in publications, presentations, and news reporting, with explicit public-domain licensing statements.
• Climate Explorer — an interactive map and graph tool providing historical and projected county-level climate data for the United States, designed for use by planners, journalists, and researchers without specialist GIS skills.
• Dataset Gallery — a curated library of climate datasets with full documentation, provenance information, and links to primary sources.
• Event Tracker — contextualizes extreme weather events within the broader long-term climate record, allowing users to assess whether a given event falls within or outside historical norms.
• Teaching Climate / Climate Literacy Guide — learning activities, standards-aligned curriculum materials, and professional development resources for K–12 and university educators, including a partnership with the CLEAN collection.
• ENSO Information Hub — dedicated resources on El Niño and La Niña cycles, their global teleconnections, and their downstream effects on regional weather patterns.
• Fifth National Climate Assessment — the full text and supporting materials of the assessment, restored to public accessibility after its removal from federal servers.

The site also links to the U.S. Climate Resilience Toolkit as a sister resource. In terms of scope and depth, Climate.us is not merely archival. Its News & Features section, in the site's own words, is "modeled after an online science magazine" — presenting complex data in accessible explainers and Q&As while retaining the scientific rigor of the original NOAA resource.

Open Data vs. Closed Data: A Structural Comparison

The climate.gov episode sharpens a debate that has long simmered in technology, science, and policy circles. The table below shows the concrete operational difference between open and closed government data when a hostile administration decides to act against its own published record.

Dimension	Open / Public-Domain Data (U.S. Federal, 17 U.S.C. § 105)	Closed / Copyrighted or Contractually Restricted Data
Legal right to mirror	Yes — anyone may copy and redistribute without permission	No — redistribution requires explicit rights-holder permission
Recovery after government removal	Possible by civil society, former employees, or volunteers acting immediately	Legally blocked without an explicit license grant
Cost of preservation	Storage and hosting only; no licensing overhead	Licensing fees on top of hosting costs, if redistribution is permitted at all
Dependence on government goodwill	Low — once published, the public holds rights in perpetuity	High — the government or its contractors retain control indefinitely
Risk from political transitions	Reduced — data can legally outlive the administration that created it	Severe — a new administration can effectively erase the public record without legal remedy
Ability to build derivative products	Unrestricted — researchers, journalists, and educators may adapt freely	Restricted — derivative works may require separate licensing agreements

This comparison raises an uncomfortable question: how much other federally produced data — from health agencies, environmental regulators, labor statistics bureaus, geological surveys — is technically public domain but practically inaccessible because no one has organized a preservation effort comparable to what Lindsey's team did for climate science? The Internet Archive and various academic data-rescue coalitions have begun systematic crawling efforts, but those efforts are reactive, underfunded, and dependent on the same political stability that proved illusory at NOAA.

The Fragility of the Fix and What It Exposes About Scientific Infrastructure

It would be easy to read the Climate.us story as a triumphant ending. It is not. The successor site is, by its own description, precarious. It relies on donations to operate. A three-person founding team, however skilled and dedicated, is not the same as a federal agency with a mandated budget, institutional continuity, legal authority to compel data reporting from states and localities, and operational relationships with satellite operators, ocean buoy networks, and international monitoring partners. Climate.us can preserve and present what existed. It cannot easily replace the data-collection infrastructure that NOAA was actively running.

This distinction matters enormously for anyone who works with longitudinal climate data. The climate record is not a static archive — it is a time series. Every month that NOAA's data-collection capacity is degraded or redirected, the continuity of long-term datasets is interrupted. Arctic sea ice measurements, ocean temperature readings, atmospheric CO₂ concentrations — these are scientifically meaningful precisely because they extend backward across consistent methodologies. Gaps in the series, or methodology changes forced by budget cuts, degrade the statistical value of everything that came before. A ten-year gap in a forty-year record does not merely lose ten years; it can compromise trend analyses across the entire dataset.

Werdmuller's framing of the rescue as a preservation effort is apt in a deeper sense than it might first appear — and it is worth extending that framing here. Journalism, at its best, is itself an act of preservation: of creating a record that can outlast the moment that produced it. The former NOAA scientists who built Climate.us are doing something that institutions normally do — maintaining continuity of knowledge across political transitions. The fact that three individuals with a 501(c)(3) umbrella and a donate button are now responsible for that continuity is a precise measure of how seriously the original infrastructure was damaged.

It also illustrates a risk pattern that technically literate readers will recognize from other domains: single points of failure hidden by apparent redundancy. Climate.gov looked robust — it was a U.S. government site backed by an agency with a multi-billion-dollar annual budget. But its resilience turned out to depend entirely on political will. The open-data framework provided a backup path, but that backup path required specific people with specific expertise to act before cached versions expired and institutional memory dispersed. This time, they did. Next time, that may not be guaranteed.

Key Takeaways

• Climate.gov was taken offline following the Trump administration's radical defunding of NOAA, which shed on the order of 1,300 employees — roughly 10% of the agency — by early 2025, with further weather-data-sharing restrictions signaled by the Department of Defense later that year.
• More than 15 years of climate data, including the Fifth National Climate Assessment, Arctic sea ice records, monthly climate indicators, and educational curricula, were at risk of permanent loss when the site went dark.
• U.S. federal data is legally public domain under 17 U.S.C. § 105, which meant former NOAA employees could legally copy, host, and republish the entire corpus without seeking permission from the agency.
• Rebecca Lindsey, Anna Eshelman, and Mary Lindsey — all former NOAA staff — formed the core team that built Climate.us as the direct successor, preserving the data and rebuilding most of the original site's functionality.
• Climate.us operates as a project of Multiplier, a San Francisco 501(c)(3) nonprofit umbrella organization, and depends on donations — making its long-term sustainability uncertain despite its critical public function.
• Open data licensing was the decisive variable. Under a closed or proprietary model, the administration's removal of the site would have been legally irreversible by any outside party. The recovery was only possible because of the public-domain principle codified in federal law.
• Data-collection continuity remains broken. Preserving the archive does not restore NOAA's operational capacity to generate new climate data, leaving long-term time series at growing risk of methodological gaps that will degrade their scientific value.

What Comes Next

The immediate crisis of the climate.gov data rescue has been met — the archive is online, the dashboard is running, and the educational resources are accessible to anyone with an internet connection. But the story of Climate.us is still being written, and several critical uncertainties remain unresolved.

The most pressing is financial. A three-person team running a resource of this scale on donations cannot be a permanent solution for data that serves millions of users ranging from elementary school teachers to Congressional Budget Office analysts. Climate.us will need to secure reliable institutional funding — from foundations, universities, or international scientific bodies — or risk the same kind of sudden shutdown that ended Climate.gov, this time for economic rather than political reasons.

The second uncertainty is operational. Former NOAA scientists can maintain and present historical data. They cannot, without substantial resources and legal authority, replicate the agency's satellite data agreements, its ocean buoy maintenance contracts, its relationships with the World Meteorological Organization, or its role in international climate monitoring consortia. If those functions are not restored within NOAA itself — either by a future administration or through emergency congressional action — the gaps will compound year by year, making the historical record progressively less useful for trend analysis even as it sits fully preserved on Climate.us's servers.

The third, and perhaps most consequential, uncertainty is political. The episode has handed advocates for open government data a concrete, documented, legally unambiguous case to bring to budget negotiations and legislative hearings: the difference between public-domain publication and closed-access hosting is the difference between a recoverable disaster and an irreversible one. Whether that argument gains legislative traction — in the form of stronger data-preservation mandates, dedicated archiving budgets, or requirements that agencies maintain redundant public mirrors of all published datasets — will determine whether the climate.gov episode becomes a cautionary tale that prompted systemic change, or merely an anecdote about three scientists who happened to act fast enough.

For anyone tracking how governments manage scientific infrastructure — or how they abandon it — Climate.us is now both an essential resource and a proof of concept: open data, properly understood, is not a feature of good governance but a precondition of it.