‘We’re losing our environmental history’: The future of government information under Trump

As the director of the National Security Archive’s Climate Change Transparency Project, Rachel Santarsiero is in the business of monitoring and facilitating the flow of information from the government to the public. What she’s seeing now, in the first weeks of President Trump’s second administration, is throwing the continuity of that process into doubt.

“We’re really losing our history here; we’re losing our environmental history,” Santarsiero told the Bulletin last week.

To some extent, government watchdogs, scientists, and climate and environmental activists were expecting this. During the first Trump administration, the use of terms like “climate change,” “clean energy,” and “adaptation” across federal environmental websites fell by 26 percent. In some cases, those terms were replaced by more ambiguous phrases like “energy independence” and “resilience”; other pages referencing climate change simply vanished.

But what Santarsiero and others are witnessing now goes far beyond that. Thousands of datasets have been removed from federal websites. Information on climate and the environment—from agencies like the EPA, the Council on Environmental Quality, NOAA, NASA, the State Department, and the Defense Department—has been deleted or become virtually impossible to find.

The administration’s wrecking-ball approach has raised profound questions about the integrity and future of vast amounts of information, public or not.

Under the first Trump administration, Santarsiero said, there were at least placeholder websites for federal groups like the Council on Environmental Quality. Now, the homepage (whitehouse.gov/ceq) redirects to whitehouse.gov, emblazoned with a photo of Donald Trump captioned “AMERICA IS BACK.”

The website for USAID, the agency responsible for administering US foreign civilian aid, now consists entirely of a letter to employees instructing them on how to retrieve personal belongings from their former office spaces. The Trump administration has fired or placed on leave all but 294 USAID staffers, out of more than 10,000.

“What we’re really concerned about is what happens to those records now that this agency has just been gutted,” said Santarsiero. “Typically, what happens when an agency is shuttered, or closed, [is that] whatever agency comes to replace it takes ownership of those records.” It seems like the State Department should now be responsible for USAID records, but their disposition is not clear.

In February, the Trump administration ended roughly 10,000 foreign aid projects, or 90 percent of the USAID’s humanitarian operations and half of the State Department’s, including at least 130 climate- and clean energy-related contracts. Senior USAID officials and other experts say cuts to foreign aid will increase global instability and result in hundreds of thousands of unnecessary deaths from malnutrition and disease. Finding information about those programs is now extraordinary difficult—the first six Google results for “USAID gov climate” all return an error message for “page not found.”

“We don’t really know what’s happening. There’s been absolutely no updates,” Santarsiero said. “I’ve sent a couple FOIA [Freedom of Information Act] requests to the State Department for USAID, [about] specific climate programs that were happening during the Biden administration. I’ve received no acknowledgement letter. It’s not surprising, but that just sets a precedent that is very worrisome for us about USAID records from past administrations. Where are those living? How do we get access to them? I think that is a very ominous bellwether.”

Librarians and archivists have sprung into action to download, save, and make available as much of this information as possible, but things may still fall through the cracks.

“The piece that worries me is that it is such a massive undertaking,” Santarsiero said. “It’s almost like, we don’t know what it is that we don’t know, we don’t know what it is that we’re missing.”

Climate in the crosshairs. Most of the climate and environmental data that’s disappeared since Trump took office has been equity-related, said Gretchen Gehrke, an environmental scientist and cofounder of the Environmental Data and Governance Initiative, where she leads the Website Monitoring Program.

The day after the inauguration, the administration took down the Climate and Economic Justice Screening Tool—a map that had layers of information from multiple agencies about climate and economic vulnerabilities. The tool was made to facilitate funding decisions for Justice40, a Biden administration initiative in which 40 percent of certain federal investments in climate and clean energy were supposed to go to economically disadvantaged communities. But, Gehrke said, it was widely used by environmental justice advocacy groups outside of government as well.

Fortunately, Public Environmental Data Partners, the coalition of groups working to preserve and make available federal environmental data, had downloaded the data before the inauguration, and because the map code was open source, they were able to recreate the tool and get it back online by January 24.

Gehrke said the Scope 3 Federal Emissions Inventory, which tracked emissions by the federal government, was also taken down. Bizarrely, Gehrke said, the Greenhouse Gas Reporting Program has been down for maintenance for more than three weeks, an unusually long time.

“It’s very possible that it’s down for maintenance,” Gehrke said. “I have never heard of a tool being down for maintenance for several weeks before.”

Gehrke also said that members of the coalition based in Canada have noticed that they can no longer access some pages, like the FEMA National Risk Index, a mapping tool that illustrates the risks from 18 natural hazards across the United States. The pages are still available in the United States, but it seems like the administration is cutting off access in foreign countries, which could undermine cross-border scientific collaboration and research.

Some pages had their server certificate removed, so visiting the URL doesn’t even return a 404 error (page not found). Without an error message, the crawler Gehrke’s group uses to monitor federal websites doesn’t flag a status change, making it harder to know when pages have been removed.

A frantic pace. Lynda Kellam, a digital archivist and librarian, said that her work as a volunteer for the Data Rescue Project has focused more on the social sciences, because groups like the Environmental Data and Governance Initiative have climate and environmental data covered.

Kellam said data and information are disappearing under the second Trump administration much faster than they did during his first term. “In 2017, we were doing data rescues until June, whereas if we waited that long to put together a data rescue event this year—you know, all the data is going away,” she said.

“They were very speedy right out the gate,” Gehrke agreed. “They have also done much more widespread and intensive information suppression than they did the first time around. They didn’t actually take down data last time. We feared that they would, and we did all these data rescue events in case they did, but they really didn’t touch data. This time is obviously very different.”

Kellam and Gehrke said the second Trump administration is also going so far as to change or edit data, including removing categories for people who identify as trans or nonbinary.

Archiving datasets is much more complicated than saving a single web page. “If you have the Excel file, that’s great, but you also need the documentation, you need the metadata, you need provenance information,” Kellam said. Automated crawls can miss pieces of that bigger picture, so it requires more human, hands-on involvement.

The Data Rescue Project maintains a spreadsheet of at-risk data to track what’s been saved, where it lives, and what still needs to be archived, and trains volunteers to do the actual archival work.

Gehrke is also focused on identifying and prioritizing the datasets to save and tools to replicate.

“Tons of people have said, ‘archive all of NOAA.’ That’s not happening,” Gehrke said. “Just as a brain teaser, we mapped out how much it would cost us, if we even could download ‘all of NOAA.’ It would cost us $500,000 in storage per month—just to host, cold storage, not even have access to it.”

Instead, the groups are relying on nominations by researchers and advocacy organizations to tell them what datasets are most important and then sharing the burden of archiving and publishing the information. “We don’t need 16 copies of EJ [environmental justice] screen,” she said. “We can do one copy of EJ screen and apply our efforts more effectively together elsewhere.”

Quite a lot of federal data collection is mandated by Congress. What is not mandated are the tools that the public can use to access that data. “NOAA’s climate data is congressionally mandated,” Gehrke said. “Whether or not we are able to access it is a different question.”

For tools that were not built on open-source platforms, Gehrke said she plans to submit FOIA requests to relevant agencies to ask for the code and all the documentation necessary to recreate them.

A snapshot in time. Much of the ongoing archival work would have happened regardless of who won the 2024 presidential election. Since 2008, the End of Term Web Archive has crawled the federal web domains every four years at the end of each presidential term, capturing and saving government websites on the Wayback Machine. Mark Graham, the director of the Wayback Machine, said they crawl before the election, after the election but before the inauguration, and then after the inauguration.

“This isn’t a political project,” said James Jacobs, a US government information librarian at Stanford who also works on the End of Term Web Archive (a collaborative effort among multiple institutions). “It’s a library and archives project.”

Jacobs’ work isn’t just about saving information but about giving the public the tools to find and access information—like extracting text from PDFs to make them searchable. His hope is to use the archive to create special collections organized by topic, like climate, the environment, and health.

Researchers can take data from the archive and analyze it, to see what’s been changed—or use the archives to find information that has since disappeared from federal websites.

“For example, this weekend, I said to myself, ‘I wonder how many web pages existed on US government websites that had the word monkeypox in them,’” Graham said. “Monkeypox being one word, pretty unique. I found 5,600. Then I said, ‘How many of those 5,600 are no longer available from those websites?’ And it’s 403. So, for some reason, it appears on the face of it, without going in and checking each one, that more than 400 pages related to monkeypox were taken offline by our government.”

Graham plans to regularly do that kind of rough analysis for the top 20, “to riff on George Carlin, naughty words.”

The Trump administration has made that project somewhat easier by providing lists of words and phrases that will, for example, get research flagged for review.

Suing for access. As with so much that the Trump administration is attempting, some of the fights over data access could be decided by the courts. Already, Earthjustice has sued the Department of Agriculture on behalf of the Northeast Organic Farming Association of New York, the Natural Resources Defense Council, and the Environmental Working Group, to try to force the agency to restore some of the “climate-related policies, guides, datasets, and resources” that have been removed from the web.

Last month, a federal judge ordered the Department of Health and Human Services, the Food and Drug Administration (FDA), and the Centers for Disease Control and Prevention (CDC) to restore websites and datasets essential for public health, after Doctors for America and Public Citizen sued the agencies. The Justice Department had tried to argue that the defendants hadn’t suffered “actual” injury because the information was available on the Wayback Machine, but the judge rejected that assertion on the grounds that the information was much harder to find. “These are injuries in fact,” US District Judge John Bates wrote. “These doctors’ time and effort are valuable, scarce resources, and being forced to spend them elsewhere makes their jobs harder and their treatment less effective.”

Other lawsuits targeting different agencies and datasets are likely to follow. But lawsuits are costly in both human resources and actual funds.

“We were asked to be a plaintiff in a lawsuit, but they wanted a super quick turnaround, and we just couldn’t make a decision that fast,” Gehrke said. “We’re a tiny organization, and so we definitely don’t have the funds to litigate.” But, she said, they could use their extensive expertise and evidence to support others pursuing litigation.

Gutted and privatized. Archiving can save a good amount of the information created before the Trump administration, but it’s possible that some long-running research and monitoring programs will grind to a halt. It’s also possible that more data collection and distribution will move outside of government.

This is disturbing to experts on both personal and professional levels. Kellam is concerned that the privatization of data and information will exacerbate inequalities in data access. Many institutions and individual researchers can’t afford the subscriptions costs to private data repositories.

Jacobs said that, as an avid sea kayaker, he relies on NOAA data to access the great outdoors safely. Right now, he still has access to information about tides, currents, and weather forecasts, but he’s not so sure that will still be the case going forward. And of course, Jacobs points out, it’s not just pleasure-seekers who depend on this information: So does the entire shipping industry.

Of course, most all Americans rely on access to NOAA data, just to know whether they need to take an umbrella when leaving the house or when to prepare for extreme weather events. Information from the CDC, FDA, and FEMA is vitally important to protecting citizens from a wide array of dangers.

Government data, on climate and the environment and so much else, is a vast and expansive resource that everyone relies on, whether they know it or not. And experts are worried about the future of this resource.

“Civil society does not have the capacity, the infrastructure necessary, the rights necessary, to collect a lot of the data that the government collects and disseminates,” said Gehrke. “Little nonprofits are not going to be sending up a satellite and collecting climate data. We are at the mercy of our government to collect this data for the public good, and if they choose not to, I don’t think there’s a lot that we can do about it.”

For now, Gehrke and others are focused on what they can do. “I am grateful for a place to throw my energy and to be doing a form of resistance that I can do,” she said. “I think that all of us need to be resisting in whatever way we can.”