LISTEN TO THIS ARTICLE

MONTREAL— Earth’s core may be as much as 1000°C cooler than thought, according to a new technique for measuring the temperature of molten iron under extreme pressures. The findings, presented here at the Goldschmidt Conference on 17 July, could help resolve a long-standing puzzle over how Earth’s magnetic field survived for billions of years before the solid inner core that helps power it today had formed.

“It looks like a rather elegant new solution to the problem,” says Bernie Wood, an experimental petrologist at the University of Oxford who was not involved with the work. However, he expects the results to spark debate among researchers who, after decades of efforts to recreate conditions at Earth’s center, have generally landed on higher temperatures.

“These temperatures are way cooler than anybody else’s results,” says Michael Walter, a high-pressure researcher at the Carnegie Institution for Science who led the research. “I suspect it’s going to take some convincing.”

Earth’s Mars-size core is composed mostly of iron, with a molten outer core surrounding a solid inner core the size of the Moon. The planet’s magnetic field is generated by the churning of molten iron in the outer core, and has existed for at least 4 billion years . Today, that dynamo is sustained mainly by the slow freezing of the inner core , which releases heat and buoyant light elements into the liquid metal above. But some models of Earth’s cooling history suggest the inner core began to form just over half a billion years ago, raising the so-called “new core paradox”: What kept the dynamo running for the 3.5 billion years before the arrival of the inner core?

The answer depends on the conditions deep inside the planet. Because the inner core is solid, it must not be hot enough to melt at the immense pressures it experiences. As a result, determining iron’s melting point under corelike pressures provides one of the best estimates of the core’s temperature. Many estimates put that temperature at more than 6000 kelvins.

Walter and his colleagues spent more than 5 years devising a new way to make that measurement. Like previous experiments, they squeezed tiny iron samples between the tips of diamond anvils to pressures of nearly 2 million atmospheres, equivalent to those in Earth’s outer core. But instead of heating the iron with a laser, they applied electrical pulses at microsecond intervals, offering better control. And instead of searching for signs of melting in the iron itself, they looked for the key pause in temperature that occurs when added energy goes into melting the iron rather than raising its temperature.

At the pressure of the inner core boundary, the team found, pure iron should melt at somewhere between 4420 and 5220 kelvins, more than 1000°C below previous estimates. Most of that uncertainty comes from extrapolating measurements made at outer core pressures to those at the inner core boundary, Walter says. After correcting for the lighter elements mixed into Earth’s core, the estimated temperature falls even lower, to a central estimate of 4284 kelvins.

The researchers then plugged these lower melting temperatures—as well as their estimate of how iron conducts heat under those pressures—into models of Earth’s thermal evolution. They found that cooling of the core alone could have powered the magnetic dynamo for most of Earth’s history, until the inner core began to freeze solid just over half a billion years ago, Walter says. “Our numbers make it more permissive to have a thermally driven geodynamo.”

The cooler temperatures have another implication: that the boundary between Earth’s core and mantle is also cooler than previously thought. That would make it harder for rocks there to melt, potentially undermining theories that clusters of giant blobs that sit on top of the core—identified by the unusually slow passage of earthquake waves through deep Earth—are made of melted rock, Walter says.

Dario Alfè, a mineral physicist at University College London whose work helped establish higher core temperatures, says the new measurements would represent a “jump back into the past,” reviving a debate that many researchers thought had largely been settled. But he remains unconvinced. “Whether or not this is correct is difficult to say, certainly not without proper inspection of the methods,” he says. “Frankly, I have spent enough of my career on this to want to reopen this discussion.”

LISTEN TO THIS ARTICLE

A federal judge ruled Friday that the White House Office of Management and Budget (OMB) cannot cancel active research grants solely because of a shift in the administration’s priorities, as it has done to billions of dollars in projects since President Donald Trump took office. But his administration may have already come up with another way to accomplish the same goal.

Last year, 23 states sued OMB and 11 federal agencies over their use of what is called the termination clause, part of the rules that govern how trillions of federal dollars are managed. The clause, which dates to the first Trump administration, allows the government to pull the plug on any award that “does not effectuate program goals, agency priorities, or the national interest as they exist at the time of the termination.” Since Trump took office again in January 2025, his administration’s attacks on universities and his antipathy to diversity, equity, and inclusion; climate science; and other topics have led to thousands of grant cancellations.

The 17 July ruling by U.S. District Court Judge Indira Talwani focused on the clause’s last phrase, which appears to let the government change the terms under which the grant was awarded. And although she upheld the government’s ability to terminate grants, she said applying after-the-fact conditions is illegal. “The court agrees with Plaintiffs and finds that the Termination Clause does not permit agencies to terminate grants based on program goals and agency priorities identified after grants were awarded,” Talwani wrote in her 32-page decision. At press time the administration had not said whether it would appeal.

The states did not ask for any grants to be reinstated. But Claudia Polsky, a University of California (UC), Berkeley law professor, says the ruling makes it much easier for universities that have lost grants to ask another federal court to do exactly that.

“It is essentially a declaration of liability that institutions can take to the Court of Federal Claims and ask to have their grants reinstated,” says Polsky, who won a suit last year on behalf of UC faculty whose grants had been terminated. “These states made a really smart strategic decision to seek clarity on the termination clause because it has been invoked so often by the Trump administration.”

The ruling also seeks to reassure institutions that they won’t lose future awards because of subsequent policy shifts, a practice that state governments see as an unwarranted power grab by the Trump administration. “The president and his allies cannot hold critical programs hostage to their personal whims and political ideologies, destabilizing the country by yanking essential federal funding,” said Attorney General Jennifer Davenport of New Jersey, one of the states that sued.

Even if the ruling is upheld on any appeal, however, the administration may attempt other ways to yank funding. Two days before Talwani’s ruling, for example, the Department of Health and Human Services ended funding for more than 70 research projects from its Agency for Healthcare Research and Quality (AHRQ). A spokesperson for the department said agency officials “determined that certain noncompeting continuation grants and jointly funded grants would not receive continuations awards.” That means the grants “were not terminated,” the spokesperson said, but rather that the scientists “were not awarded continued funding.”

Scott Delaney of Grant Witness, which tracks federal grants, calls AHRQ’s actions “termination in different clothes.” And Polsky notes a decision by the same federal judge who ruled the government had illegally terminated her clients’ grants. Ruling in a different case involving UC grants , the judge rejected the government’s argument that it could legally cut off funding because the grants had only been suspended, not terminated. “It sounds to me now like they’re just trying the same thing with some third verbal formulation,” she says.

The clause underlying the government’s claim that it can terminate a grant because the president’s priorities have changed was inserted into a revision of a massive document, called Regulation for Federal Financial Assistance, in the waning months of Trump’s first term. It was preserved despite tweaks to the document made in 2024 by then-President Joe Biden and remains in the most recent version proposed by OMB in late May that has attracted so much criticism from scientists and professional societies.

In ruling for the plaintiffs, Talwani said canceling a grant because of shifting priorities amounts to unfairly moving the goal posts because grant applicants are told in advance about an agency’s priorities and the goals of the program for which they are seeking funding. “It would be contrary to [any reasonable interpretation of] the Termination Clause to allow the termination of grants based on new priorities, including priorities not disclosed to the grantees,” she wrote.

It also defies logic. “Such an interpretation,” Talwani wrote, “would obviate any need to notify parties of program goals and priorities, given that agencies could change goals and priorities at any time.”

LISTEN TO THIS ARTICLE

Among the many unfounded claims made by U.S. President Donald Trump is his assertion that the 2020 census was biased toward Democrats. Although the criticism has barely registered with the public, scientists who work with federal statistical data are sounding the alarm on Trump’s proposed solution, which they see as the latest example of his administration’s efforts to politicize science.

“I realize that many of us have ‘outrage fatigue,’ but this particular outrage hits extremely close to home for the computer science theory community,” wrote Scott Aaronson, a computer scientist at the University of Texas at Austin, in a recent blog post .

Aaronson is referring to a 4 June directive from the Department of Commerce that caught agency scientists—and the external research community—by surprise. It bans the Census Bureau and the Bureau of Economic Analysis (BEA), Commerce’s two statistical agencies, from using time-tested methods designed to allow the maximum amount of data to be released from national surveys while protecting the privacy of respondents.

The new policy, Aaronson wrote, “bans essentially all modern techniques for preserving privacy in large datasets, including the 2030 census.” State-of-the-art approaches known as noise infusion—modifying a data set by adding random values—are now off-limits. Instead, the rule says surveys must rely on two older and less effective techniques to avoid disclosing a respondent’s identity. One, known as coarsening, rounds or aggregates data, such as by using 5-year intervals rather than annual increments. When that is insufficient to avoid revealing the identity of individuals or businesses, the directive suggests applying suppression: withholding certain data entirely.

The change marks the first time political appointees have dictated the agency’s statistical methods, census watchers say. It will affect the 2030 census, which is used to apportion seats in the House of Representatives by state every decade and determine the distribution of trillions of dollars annually in federal funds. It also affects a slew of economic and demographic surveys used by government agencies, businesses, researchers, and the public. And researchers say it ignores decades of improvements in survey research.

Economist Ron Jarmin, who retired in May after 33 years at Census, including two stints as acting director, says demographers will have to surrender crucial tools. “Noise infusion,” he says, “allows us to produce much more granular data on surveys” including on economic activity and jobs.

“It’s a throwback to the 1970s,” says statistician Sallie Keller, who last month returned to the University of Virginia after 4 years as chief scientist at Census. She warns that banning noise infusion will result in the release of incomplete and inaccurate data. “If your only choice is to do coarsening, aggregation, and suppression, the subpopulation sizes for which statistics will be able to be released are going to be very large, perhaps in the hundreds of thousands.”

The 4 June directive came from Commerce’s office of economic affairs, and knowledgeable sources tell Science it was crafted not by technical experts, but by departmental lawyers. It says only that “noise infusion is inconsistent with the Department’s … commitment to striking a balance of accuracy, confidentiality, objectivity, and relevance for each statistical product.” Commerce officials declined further comment on the directive.

However, many social scientists think the goal is to help achieve two of the president’s major policy goals—namely, helping Republicans win more seats in Congress and reducing immigration by restricting the rights of noncitizens. “The directive is not driven by science. It is driven by political interests,” says Cynthia Dwork, a theoretical computer scientist at Harvard University who helped develop one type of noise infusion, called differential privacy, that is now banned.

“Differential privacy is an especially strong form of noise infusion [that] gives agencies a formal way to measure and limit how much any one person’s information can affect a published result,” according to a recent blog post by researchers at the Data Foundation. But by definition it comes with trade-offs. Increasing the amount of noise boosts privacy but distorts individual data points, especially for data sets covering small geographic areas. Some demographers have asserted that using differential privacy might distort some raw data to the point of being unusable.

That academic debate is now being wielded for political ends. A 2025 article on the website of the Center for Renewing America , a conservative think tank founded by Russell Vought, the powerful head of the White House Office of Management and Budget, echoes the criticism of differential privacy. “The more noise introduced, the less accurate the data becomes,” it says.

The article, by anonymous center staffers, also asserts that using differential privacy in the 2020 census hurt Republicans by “tilting the political scales favorably toward Democrats for apportionment and redistricting purposes, and other ramifications.”

That claim is unfounded, counters Allison Plyer, chief demographer at the Data Center and former chair of Census’s scientific advisory committee, because the 2020 census methodology did not alter the population totals for any state. And in some cases, she adds, red states such as Texas and Florida only have themselves to blame if they think the final population counts were inaccurate. “Both those states refused to invest in ‘get out the count’ activities” to improve census response rates, Plyer says. “Studies show that states that invested in [such public outreach] did not have undercounts, and Texas and Florida did.”

The Center for Renewing America article also fingers differential privacy as an obstacle to the administration’s attempt to ask 2020 census respondents whether they are citizens and then exclude them from the final tally—a step that the Supreme Court ruled would violate the U.S. Constitution. “Even if the citizenship question is added to the [2030] Census … the algorithm [used in differential privacy] will be able to mask characteristic data, including citizenship status, preventing the ability to determine total voter eligibility or how to draw maps to exclude those populations,” the authors wrote.

Economist John Abowd, who as Census chief scientist during the first Trump administration pushed the bureau to adopt differential privacy for the 2020 census, believes the article describes what the administration hopes the ban will accomplish. “They actually want to track down people” so noncitizen individuals can be removed from any census products, Abowd says—a move that would flout laws meant to ensure the privacy of those who fill out federal surveys.

Keller thinks the directive will damage Census’s reputation as a top-flight national statistical agency. “What’s such a huge shame is that the Census Bureau had finally established itself to be on the leading edge of privacy protecting technologies,” she says.

Abowd, who testified as an expert witness when several states and immigrant rights groups sued over the proposed citizenship question, hopes the administration will reverse course. “I do not believe that they have considered all the consequences of the methods that they are using to enforce change on the statistical agencies,” he says. And a former senior Census official who requested anonymity expects the agency to push back. “Census Bureau will probably tell the political leadership, ‘OK, here’s what this will mean,’ and then I would not be surprised to see some walking back.”

But Census currently has a dearth of experienced professionals in leadership positions. Trump has yet to nominate anyone to replace statistician Robert Santos, who resigned as director in February 2025. George Cook, an institutional investor and Trump political appointee at Commerce, is overseeing Census in addition to serving as chief of staff for the department’s undersecretary of economic affairs. And last month the bureau named demographer Michael Lachanski, then a postdoctoral fellow at Georgetown University, to lead data, science, and policy—the No. 2 spot at the bureau. Lachanski is also chair of the agency’s committee that will implement the new directive.

In the meantime, social scientists are trying to spread the word. “We must collectively reject this anti-scientific approach to governing federal statistics,” Dwork, Abowd, and others wrote in a 2 July blog that encourages individual scientists to deliver this message to Congress: “This order will destroy the public data our nation relies on for important decisions … I want it to be rescinded.”

LISTEN TO THIS ARTICLE

Goats spend a lot of time bashing their heads into things. Winning a headbutting contest shows other members of the herd who’s boss—but bumping heads is also a way to play or explore new objects, so it’s often assumed goats must be protected against head injury by the shape of their skull and strong neck muscles.

A new study challenges that assumption. Researchers have found early signs of neurodegeneration in headbutting goats at just 1 year old , they report in a preprint posted to bioRxiv this month. The findings, they say, suggest goats could be a useful animal model for studying the link between head injuries and neurodegeneration seen in humans.

“I think the overall conclusions are pretty convincing,” says Ramon Diaz-Arrastia, a neurologist at the University of Pennsylvania who was not involved with the work. Traditionally, researchers have used rats and mice as animal models for studying brain injuries, and the field has become a “rodent monoculture,” he says. But rodent and human brains are very different, which could be partly to blame for the poor translation of preclinical findings to human trials, so it’s “really interesting to see this work in goats.”

Nicole Ackermans, a neuroscientist at the University of Alabama, first identified signs of neurodegeneration in wild headbutting animals in 2022, finding that the brains of dead muskoxen contained an abnormal version of the protein tau . In humans, these tangled tau proteins are associated with neurodegenerative diseases such as Alzheimer’s, and can form after head injuries. “I thought, ‘OK, this is something really interesting to look at for modeling brain damage,’” she says. But tracking how damage accumulates over time in wild animals is not feasible. That’s where the goats came in.

The team acquired three 6-month-old male goats—Alvin, Simon, and Theo—from farms and moved them to a nearby agricultural college where cameras recorded them for 6 months to track their headbutts. Each month, the researchers collected samples of blood, saliva, and brain fluid to look for biomarkers of brain injury and, at the start and end of the study, scanned the animals’ brains using positron emission tomography–MRI to spot any structural damage. “We just put them in the human scanner,” Ackermans says. “It caused a bit of a ruckus when we went down the hallways and there’s, like, two horns poking out of the gurney.”

Over the 6 months, the three animals made a whopping total of 227 headbutts a day. At the end of the study, the goats, only about 1 year old, showed no sign of large-scale brain damage in the scans, and no memory impairment when navigating a simple maze. However, it appeared they had accumulated mild brain injury: Some biomarkers—particularly GFAP, a protein released by the brain’s immune cells in response to injury or disease—exceeded normal concentrations for healthy humans. When the team euthanized the animals and examined their brains, they also found active immune cells—a typical damage response—and abnormal tau proteins that weren’t present in two younger goats. It’s unclear whether these markers can be directly compared across species, but it looks like there is “some trauma happening,” Ackermans says.

The mild effects make sense in animals that young, Ackermans says. She expects they gradually accumulate injuries over time, just as athletes exposed to regular head impacts suffer increasing damage as their career progresses, with cognitive effects emerging later in life.

Understanding how brain damage affects goat behavior over the long term will be crucial, Diaz-Arrastia says. Holly Vickery, an animal behavior and welfare scientist at Harper Adams University, says she isn’t aware of any dementialike symptoms in dairy and breeding goats, which can be kept into their early teens. But they may not headbutt as frequently as the animals in the study.

Ackermans says working out the “tipping point” for when brain injuries in goats develop into full-blown neurodegenerative disease is next on her to-do list. “Is it 100,000 headbutts?” she asks. Or perhaps goats “have to headbutt really hard, and the ones that don’t headbutt really hard don’t really do much [damage]?”

Answering those questions could help reveal how brain injury progresses to disease in humans. Unlike rodents, goats naturally express proteins such as tau and beta amyloid—another key player in human neurodegeneration—and their folded brain structure better resembles that of humans. This is important, Diaz-Arrastia says, because larger folded brains are affected differently by impacts than rodents’ smaller, smoother brains. And if goats self-inflict damage, there’s no need for researchers to induce injuries.

Alan McElligott, an animal behavior and welfare scientist at City University of Hong Kong, is wary of jumping the gun when it comes to using goats in brain injury research. The history of biomedical research is replete with findings from animal models that did not translate to humans, he says. “I think we need a high threshold before embarking on this.”

LISTEN TO THIS ARTICLE

Richard Grant, a primary care physician with Kaiser Permanente, had been waiting since the fall of 2025 for the Agency for Healthcare Research and Quality (AHRQ) to release the latest funds for the multiyear grant it awarded him in 2024. A few days ago, when an email from the agency showed up, he thought the money was finally coming. Instead, AHRQ was notifying him of the grant’s cancellation. The email, he says, was “the final nail in the coffin.”

The agency’s justification was essentially that his research now didn’t align with its priorities, which the letter noted includes “solution-oriented approaches in health disparities research.” Yet Grant’s project focused on how to better coordinate care for people with multiple chronic illnesses, such as diabetes and high-blood pressure, who face social barriers to care—such as not being able to afford transportation or food.

Similar terminations of ongoing grants went out this week to dozens of other scientists funded by AHRQ, a small federal agency focused on studying the quality of U.S. health care. The projects’ topics ranged from better coordinating kidney transplantation to reducing delays in diagnosing blood clots.

As of Friday afternoon, more than 70 projects that had won an estimated $185 million in funding commitments had received these “nonaward” notices, according to AcademyHealth, a research advocacy group that has been tracking them. That’s roughly half of all the active grants AHRQ had listed this week .

“This does not make sense,” says AcademyHealth CEO Aaron Carroll. The funding for these grants has already been appropriated for this fiscal year, in a “budget that was passed by both houses of this Congress, signed by the current president of the United States,” he says. “If I were Congress, I’d be livid.” Although AHRQ has recently been a target of Republicans who see its work as wasteful, the Republican-controlled Congress had approved $345 million in funding for the agency for this fiscal year.

The grant terminations are the latest disruption in a year of turmoil at the agency. Since 2025, it has lost much of its staff and canceled all open grant competitions . In August 2025, physician groups filed a lawsuit to force the agency to disburse tens of millions of dollars in unspent 2025 funds.

Asked about the notices sent out this week, a spokesperson for the Department of Health and Human Services, which oversees AHRQ, said the grants “were not terminated” but “they were not awarded continued funding.”

Data scientist Scott Delaney of Grant Witness, a group formed to track the termination of grants by President Donald Trump’s administration, says these “non award” notices are “terminations in different clothes.” He suspects the notices are worded this way because the administration is looking for other legal strategies to effectively cancel grants. “They were really bad at terminating grants in a lawful way … last year,” he says.

The Trump administration has consistently acted to eliminate staff and programs, “effectively dismantling AHRQ,” says Elisabeth Uphoff Kato, who was deputy director of the agency’s Center for Evidence and Practice Improvement until she resigned earlier this year.

Most of the projects given notice this week had probably already been halted because grant funds hadn’t been disbursed, Delaney says. But with these official agency letters clarifying the grants’ status, he adds, researchers might now be able to file lawsuits seeking the money’s restoration.

LISTEN TO THIS ARTICLE

On 5 November 2024, Chinese-born U.S. seismologist Youlin Chen was preparing to fly home to Boston after a monthlong visit to China, where he celebrated his mother’s 80th birthday and gave talks at two universities. For years, Chen, who became a U.S. citizen in 2011, had taken annual trips to China to see family, lecture, and meet with collaborators on projects for his employer, EMR Solutions & Technology, a Florida-based consulting firm.

But that day state security officers arrested Chen at Beijing’s international airport. Officials interrogated him more than 100 times about his research, says Chen’s wife, Yufang Rong, also a seismologist. He was often forced to sit on a hard stool with his hands and legs bound, she adds. Chinese authorities charged Chen with espionage on 1 May 2025 and he remains in detention without a trial, as Reuters first reported earlier this week.

One strand of Chen’s work used Chinese data to analyze North Korea’s six underground nuclear tests for the U.S. Department of State. But none of his work is or was classified, says EMR Senior Vice President Hafidh Ghalib. “He never had a security clearance,” so the company had “no concerns for his safety.”

“This is a really tragic case and one of great concern,” says nuclear nonproliferation expert Siegfried Hecker, a former director of Los Alamos National Laboratory now at Stanford University.

Chen’s detention has heightened alarm over a separate clampdown on the kind of geophysical data he analyzed. Under a policy posted in 2016, seismic networks and other government-funded projects in China were expected to deposit raw observations and processed data in a national data platform operated by the China Earthquake Networks Center. Although CENC restricted access to some sensitive data, other records could be downloaded freely or were available to domestic or foreign users after registration. But on 26 March, CENC’s data platform announced it would no longer provide raw seismic observations or basic information about monitoring stations, except in specially approved cases. Citing data security and national security, it said users would instead receive processed data or customized analyses.

That move, and Chen’s detention, have sent a chill through international seismology. Three U.S. seismologists who have worked with Chinese data declined to be interviewed, citing concerns that their collaborators in China could get in trouble. “This must put tremendous fear into Chinese seismologists who have been collaborating with foreign colleagues,” says Frank von Hippel, a senior research physicist with Princeton University’s Program on Science and Global Security. As one U.S. seismologist put it, “We’re petrified for them.”

Placing China’s raw data off-limits could impede studies of earthquake hazards, fault systems, and Earth’s interior, and make Chinese findings harder for outsiders to reproduce, say the seismologist and others. But Chen’s own research demonstrates how difficult it is to separate civilian and security applications of such work.

Chen had openly collaborated with Chinese government and university seismologists for well over a decade. A collaborative study he led , published in Geophysical Journal International in July 2024, used more than 31,000 waveforms recorded by 346 stations across eastern Asia, including China’s national seismic network, to produce maps of how seismic energy is absorbed by the crust. Such analyses can help estimate the size of seismic sources and improve predictions of ground motion and earthquake hazards. The paper disclosed funding from the U.S. Air Force Research Laboratory and the State Department and thanked CENC staff for providing data and technical support.

The same methods, however, can aid nuclear test monitoring. A December 2020 technical report Chen authored under a contract from the State Department’s arms control bureau sought to improve estimates of the magnitude and yield of North Korea’s nuclear tests and distinguish their signals from natural earthquakes. Ghalib says that, under Chinese rules then in force, Chen analyzed the restricted waveforms in China and did not take the raw records out of the country.

No public evidence links Chen’s detention to CENC’s subsequent data restriction. But the strategic sensitivity of China’s seismic records is illustrated by two small signals detected near China’s Lop Nur nuclear test site in June 2020. In February, U.S. officials alleged that they resulted from a concealed low-yield nuclear explosion; China denies the accusation. International monitoring data were insufficient to determine whether the signals were natural or explosive, and seismologists have said finer grained records from China’s much denser network might resolve the question.

“There are no doubts that seismologists worldwide, including us at EMR, have examined seismic waveforms from events near Lop Nur recorded by stations outside China,” says Ghalib, also a seismologist. Science found no evidence that Chen analyzed the signals using waveforms from Chinese seismometers.

In March, the State Department determined Chen is wrongfully detained. At a media briefing this week, a spokesperson for the Ministry of Foreign Affairs of the People’s Republic of China rejected that designation, stating, “There is no instance of wrongful detention.” A 16 July commentary on the Chinese portal 163.com meanwhile portrayed Chen’s November 2024 visit in a more sinister light. It claimed Chen took “professional seismic detection equipment … into strategic mountainous regions and areas surrounding key civil air defense facilities. There, he surreptitiously deployed temporary monitoring instruments to collect extensive, classified geological data firsthand.”

Rong says such claims are false. “These are crazy stories. He never went to those places and never touched any instruments,” she told Science .

Rong, a seismologist for the insurance firm FM, says she has never collaborated with her husband. But after news broke of his detention, her own collaborators in China sent her frantic messages asking her to remove her name from a paper to appear in the October issue of Engineering Geology . Rong says she tried, but it was too late—the article had already posted online on 26 June.

Chen’s detention has prompted the State Department to consider discouraging U.S. citizens from traveling to China, says Eric Lebson, chief strategy officer at Global Reach, a nonprofit that works to free U.S. citizens held abroad. The department, he says, is also mulling labeling China a State Sponsor of Wrongful Detention. Created by executive order last year, the designation has so far been applied to Iran and Afghanistan. This “could have very serious implications for China,” Lebson says. The designation can lead to sanctions, travel restrictions, export controls, and other penalties.

In the meantime, U.S. and Chinese cooperation in seismology hangs in the balance. “Many of our members, including those in China, depend on data collected and shared by regional and global seismic networks for their research,” the Seismological Society of America, of which Chen is a member, said in a statement.

U.S. embassy staff who have visited Chen and spoken with his lawyer—who was not allowed to see Chen until early this year and is not permitted to discuss the case—have told Rong her husband has lost about 18 kilograms. “We tried to give him shoes for exercise,” she says, “but that was not allowed.”

LISTEN TO THIS ARTICLE

Early last year, the U.K. government announced a drastic cut in overseas development funding to compensate for a planned increase in spending on national defense. Aid that included support for health and education in lower income countries would plummet by 40% in 2027, from 0.5% to 0.3% of national income , it said. On the heels of a sudden halt to most foreign aid by the U.S. government, the announcement prompted fears of further harm to vulnerable populations. But U.K. leaders didn’t say how cuts would be distributed, leading to more than a year of uncertainty for countries dependent on the money.

Yesterday, the U.K. Foreign, Commonwealth & Development Office (FCDO) finally published its country-by-country funding plans as part of its annual report . African nations shoulder the biggest cuts, with reductions amounting to more than 90% by 2029 for Malawi, Mozambique, Kenya, and Tanzania.

The new figures “confirm what many of us studying aid have warned and feared—a retreat that is heaviest for the countries least able to absorb it,” says Dan Banik, a political scientist at the University of Oslo. He notes that the worst hit African countries are home to millions already living in extreme poverty. “These cuts, made to free up money for defense … are not a technical rebalancing. They are in fact a political choice with immediate human consequences.”

The cross-party parliamentary body known as the U.K. International Development Committee, which previously criticized the U.K. government for failing to release the numbers sooner, didn’t like what it saw. “It is chilling to consider the consequences of these decisions,” wrote the committee’s chair, Sarah Champion, in a statement yesterday .

Cuts will be smaller, though still substantial, for some countries hit by conflict or other instability: Lebanon and South Sudan face U.K. aid reductions of about 40% between 2025 and ’29, for example, while the Democratic Republic of the Congo, which is grappling with an outbreak of Ebola , faces a 35% cut.

The U.K. government says that although it’s distributing less, it will spend the money better—investing in international organizations such as Gavi, the Vaccine Alliance , for example, as well as boosting R&D and private industry overseas. There had been “difficult decisions,” U.K. Foreign Secretary Yvette Cooper acknowledged in the report’s foreword, but the funding reduction “does not mean walking away from our values or responsibilities.”

Some polling of the U.K. public last year suggested the cuts have been popular: One survey found majority support for a reduction in overseas aid to boost national defense, with greatest backing among older and more conservative voters .

The government has also said the changes respond to nations’ wish to move away from aid dependency and donor paternalism toward partnerships that support long-term, sustainable development. In a statement to Science earlier this month, FCDO described the shift as “modernising our approach to development.” It added: “This means that we’re focusing on using UK expertise and multilateral investments to strengthen health systems and support locally-led reforms,” including through some projects funded before the cuts came into force.

But it’s unclear how effective this approach will be in some poorer countries such as Malawi, which faces a drop from more than £50 million per year to £5 million by the end of the decade (see Science ’s recent feature on the country’s effort to cope with aid cuts). Some officials there are particularly skeptical of the United Kingdom’s planned move away from direct support for health care services to so-called technical assistance, such as consulting on health policy or training. Owen Chikhwaza, director of reproductive health services in Malawi’s Ministry of Health, tells Science Malawi’s biggest problem is a lack of money, not poor efficiency or other problems that could be fixed with technical assistance. “As much as we need to optimize what we have, we need the actual support—the actual dollars to support the health care,” he says.

Global health experts agree. In countries with low income and high rates of disease, “the default should be funding the government to expand services,” says Katherine Klemperer, a policy analyst at the Center for Global Development. “Technical assistance takes money away from this,” she says, and could end up benefiting U.K. providers of technical assistance at the expense of those countries’ citizens.

In her statement, Champion said it was unclear whether other elements of the U.K.’s modernized approach, such as funding for international agencies or private sector investment, “fully makes up for the impact and flexibility being lost by shrinking bilateral budgets.”

Saskia Osendarp, executive director of the Micronutrient Forum and co-author on a commentary in Nature last year warning of aid cuts’ impacts on child nutrition worldwide , says some countries and charities are trying to find new ways to plug the growing funding gaps with innovative sorts of funding and new types of partnerships. “But these changes take time, and won’t be enough to cover up for [foreign aid] reductions in the past 2 years,” she adds, noting the cuts come at a moment of massive geopolitical tension and other crises worldwide. “It is just too sudden, too soon.”

LISTEN TO THIS ARTICLE

Satchidananda Panda, a prominent biologist known for his work on circadian rhythms, light sensing, and intermittent fasting, will resign next month after 22 years at the Salk Institute for Biological Studies. His planned departure, which has not been officially announced outside of Salk, comes after the institute contracted an employment law firm to investigate allegations of sexual misconduct by Panda. Panda denies the allegations, which were made earlier this year.

Current and former lab members, and other Salk employees, told Science the investigation stemmed from a complaint by a researcher in Panda’s lab. Science has learned it is at least the second time in recent years that Salk has investigated Panda following harassment allegations. Salk told lab members on 1 July that the investigation had concluded but would not share its findings. Salk employees say Panda has not been on campus for several weeks.

In an email sent to faculty and seen by Science , Salk President Gerald Joyce said, “Salk leadership has been informed that Dr. Panda intends to resign from the Institute, effective August 13, 2026.” Panda, who is in his mid-50s, told lab members and faculty in a separate email that he plans to write a book about how recent advances in circadian rhythms research will shape human health. “This is the right moment to pause, take fresh stock of these shifts, and redefine the goals for the next phase of my career,” Panda wrote. He has authored two books on circadian biology and is one of the most cited researchers in the world, according to the journal analysis firm Clarivate. In 2023, he was elected as a fellow at AAAS, which publishes Science .

Multiple Salk employees, including lab members, told Science they were worried Salk’s decision to let Panda resign without making the investigation public could allow him to join another institution where colleagues have no knowledge of the past concerns about his behavior.

In an emailed statement, Salk confirmed that an independent investigation had taken place after a research assistant raised “workplace concerns.” “Because Salk is committed to protecting the privacy of its personnel and encouraging an environment where all employees are empowered to speak up and raise concerns, we will not comment on the nature of the allegations, the workplace investigation, or the investigative findings,” the statement said.

In 2018, Salk suspended famed cancer biologist Inder Verma after Science reported allegations from eight women—six of them at Salk—of forcible sexual contact and other inappropriate behavior. Verma resigned later that year. The same year, Salk settled three lawsuits brought by senior female faculty who alleged widespread gender discrimination at the institute .

Science spoke with six researchers with knowledge of the situation, both in Panda’s group and other labs at Salk and the University of California San Diego, which is affiliated with Salk. Most spoke on condition of anonymity out of fear of career repercussions. They say the latest investigation concerned allegations of sexual misconduct brought against Panda by a research assistant in his lab.

Tiffany Chin, a graduate student at Salk who left Panda’s lab in April, said the research assistant approached her in late February. She told Chin that Panda had been harassing her, including giving her massages in his office. She said on one occasion she turned down a foot massage, and he later screamed at her that she would never make it in science, which “scared her,” Chin told Science in a text. “It made her realize that unless she complies with him that she’d lose her letter of recommendation, might get fired, etc.” The research assistant also showed Chin texts Panda sent her that she found inappropriate. She told Chin she planned to make a formal complaint against Panda to Salk’s human resources (HR) department.

Other people who spoke with the research assistant said she reported that Panda had drunk alcohol with her in his office and made sexually explicit references in conversation. One person Science spoke with witnessed Panda regularly meeting with the research assistant for hours in his office—far more time than he usually spent with his postdocs and graduate students. “She was the favorite among the lab personnel—[the favorites] were mostly women,” the person says.

A source with knowledge of the situation says in recent years Salk has conducted at least one other formal investigation into allegations of sexual misconduct by Panda. The source says an employee then in Panda’s lab alleged that while the two were traveling to a conference, he suggested she share a hotel room with him. The source says Salk took disciplinary action against Panda at the time.

In an email, Panda denies the “allegations and accusations. … I am deeply committed to maintaining a safe, respectful, and professional environment for everyone in my research team and across the Institute. These values are central to my leadership.”

Laura van Rosmalen, a postdoc in Panda’s lab, reached out to Science and said she finds Panda to be a good mentor to all lab members. “I see that he’s very supportive to the people in the lab and caring for everyone,” she says. The lab “really feels like a family,” she says.

After the February incidents, the research assistant went on extended leave and then resigned from the lab, according to multiple people.

According to emails seen by Science , Salk contracted attorney Renée Schor to conduct an independent investigation into the allegations. Beginning in March, Schor interviewed lab members in an inquiry that stretched over several months.

On 1 July , Salk’s HR department met with Panda’s lab to tell them the investigation had concluded and Panda was resigning, but that they would not share the outcome of the investigation, according to multiple lab members present at the meeting.

All the sources Science spoke with who were at the meeting say they came away with the impression that they were not to discuss the investigation. Two trainees at Salk independently reached out to Science afterward, concerned about the lack of transparency. “I believe that a senior scientist that abuses his position of power should not be able to move on to the next thing with his reputation fully protected—all while his victim(s) and lab members suffer the consequences,” Adam Rochussen, a postdoc in Panda’s lab who attended the meeting, wrote in an email. According to Rochussen, Panda told him it would be career “suicide” if lab members didn’t go along with his official story.

A second trainee at Salk expressed concern that vulnerable employees, especially international ones who rely on Salk to sponsor their visas, felt silenced by the decision to allow Panda to resign without revealing the investigation’s nature or outcome. “This dynamic can create an environment where people feel unable to ask questions or share information,” they said.

Jennifer Freyd, a psychologist and president of the Center for Institutional Courage who studies sexual harassment, echoes this sentiment. Allowing a person who commits harassment to resign quietly, she says, “is really harmful to the people who’ve been victims of this person but also bad for the institution and the next institution where he shows up. It’s a very destructive pattern to do that.”

Freyd notes it is sometimes necessary to keep information private to protect alleged victims. But she says institutions can work with lawyers to be transparent about an investigation’s results, at least internally or by reporting the outcome to funders. In a statement, Salk said, “the Institute complies with its reporting obligations.”

Rochussen says at least nine of the lab’s 26 employees have left since the beginning of the year, and that many cited their discomfort after learning about the allegations as their reason. The departures included three of Panda’s graduate students, including Chin, who asked for reassignments within Salk and affiliated institutions, he says. Multiple sources told Science some lab members—especially staff scientists and graduate students—have struggled to find new placements on short notice, in part because research and training funds are tight across the institute amid federal budget cuts.

Salk did not answer questions about what will happen to Panda’s grants, including a 10-year, $2.5-million-per-year grant from the Wu Tsai Human Performance Alliance for a project he led, and several grants from the National Institutes of Health.

In its statement, the institution said: “In view of Dr. Panda’s resignation from the Institute, we are arranging a transition plan to ensure those in the lab, and their scientific research, are disrupted as little as possible. We are working with the relevant funding agencies, both public and private, to support this effort.”

Malawi— It’s a quiet Monday morning as community health worker Fanizo Simenti arrives at the makeshift clinic in the village of Katenje, in central Malawi. The small, unfinished brick building stands in a clearing, its corrugated iron roof contrasting with the thatch on the surrounding houses. The four small patient rooms are unfurnished, except for a wooden bedframe in one reserved for expectant mothers. “We are looking at cement so we can have a good floor, and plastering,” Simenti says, gesturing at the uncovered ground and walls. “If we could have a better bed, and a mattress, that would be good.”

Simenti hopes this health post can address the village’s long-standing need for accessible medical care, particularly for pregnant women. He had been providing basic services such as prenatal counseling and childhood immunizations under the shade of a tree, but for births and other needs, many mothers travel more than 20 kilometers to Khola Health Centre, the nearest government-run facility—an unbearable walk in intense heat or heavy rains, and expensive to reach via motorbike- or bicycle-taxi.

A couple of years ago, Simenti proposed the community build its own facility. It would provide patients shelter and privacy, and with government support, it could even be upgraded to an official center, like Khola, with equipment and approval to provide birth assistance and other services. Gift Banda, a local farmer and businessperson, tells Science he was moved to contribute by seeing how people suffered trying to reach Khola. The community decided, “Let’s do our own clinic here,” he says.

Simenti estimates they’ve since raised about 20 million Malawi kwacha, about $12,000, including from local leaders, tradespeople, and a politician. And although the building is far from finished, he already takes care of some people within its walls. One local mother, Magret Banda (who is not related to Gift Banda), comes here for family planning and other services with her 3-year-old daughter, and says she now makes fewer trips to Khola. She also helps clean the floor in the building so other women and children don’t cough on the dust.

Such grassroots efforts have taken on new significance since early 2025, when the U.S. government abruptly withdrew most foreign aid and dismantled the U.S. Agency for International Development (USAID), a major backer of maternal outreach clinics and other health services across Malawi. Economists estimated the country could lose more than $170 million that year , with U.S. support for sexual and reproductive health all but wiped out. Separate cuts by other foreign donors, as well as worsening fuel shortages and rampant inflation, are now further squeezing the country’s health system.

Projects such as Simenti’s may be one way for Malawi to lessen the impact of these cuts and bring public health services closer to the people that need them. One official in Kasungu district, where Katenje is located, says staff have started to encourage other villages to emulate the approach, and Simenti says he’s been asked for advice by health workers who “want to have a similar achievement.” But the initiative also reveals the limits of this kind of local solution. The rising cost of living, worsened by the fall in foreign funding, has made it harder for people to chip in. “We depend on [donations] from people” in the community, Simenti says. “We don’t force them.”

Malawi’s broader health system now faces similar uncertainty. The aid reductions highlight a need for more reliable funding, government officials say. But new policies to generate that money—for example through new taxes—and to improve how the health system uses it could take years to bear fruit. Like other sub-Saharan countries reckoning with the seismic shift in foreign assistance, Malawi stands at a crossroads, says Dan Banik, a political scientist at the University of Oslo who has studied the role of aid in the region. It may be a chance for the country to “reset” its health sector into something more resilient, he says, but it’s unclear how far these efforts can go—and what happens to health workers and patients during the transition.

THE CUTS OF 2025 highlighted a deep vulnerability in Malawi’s health system. External donors have for years accounted for well over half of the country’s health spending —among the highest share of any low-income country in the world—and up to 90% in sectors such as sexual and reproductive health. Although Malawi’s government pays the salaries of community health workers like Simenti, donors have contributed to their training as well as drugs and equipment, outreach clinics, and support with health financing and management.

That assistance, much of it channeled through nongovernmental organizations (NGOs) and international agencies, has been instrumental in improving the health of Malawi’s mothers and children, says Jeevan Sharma, a development anthropologist at the University of Edinburgh. The maternal mortality ratio, for example, fell from somewhere between 750 and well over 1000 deaths per 100,000 live births in the 1990s to an estimated 224 in 2024—well below the average for sub-Saharan Africa (see graphic, below ). But the underlying challenges persist. The rate of teenage pregnancy , which increases risks for both mother and baby, is among the highest worldwide. Poor roads, bad weather, fuel shortages, and cash-strapped facilities can hinder medical access for the 80% of Malawians who live in rural areas. Treatable infections, blood pressure disorders, and postpartum bleeding remain major killers on the country’s maternity wards.

The strain is visible at secondary health facilities such as Kasungu District Hospital (KDH), a complex of low buildings in the district capital. On this Monday morning, lines of people stretch into the parking lot; mothers breastfeed babies on the floor as nurses steer gurneys through the crowded corridors. Pregnant women often arrive here late, not knowing when to seek help and then struggling to reach the hospital. Once here, they face further delays, says Emmanuel Golombe, Kasungu’s director of health and sanitation services: He’s seen emergency cesarean sections held up because the hospital’s only operating theater was occupied by nonmaternity patients. “The mother has to wait,” he says. “Sometimes it brings complications—maybe a ruptured uterus, maybe fetal death.”

Last year’s cuts hit donor-funded programs that had reduced pressure on the hospital and the rural communities it serves. Among them was MOMENTUM, an $80 million USAID program that procured equipment, supported health worker training, and funded visits for hospital staff to remote communities to provide family planning, prenatal counseling, and many other services. Amref Health Africa, an international NGO that received $28 million of the funds, estimates it supported more than 1500 clinics across the five districts it worked in, reaching more than half a million people before MOMENTUM was axed. Cuts from other countries stand to make matters worse. The United Kingdom’s plans to slash aid funding by 40%, for example, threaten major initiatives in reproductive health. Golombe and others say they’re waiting to learn exactly what the cuts will mean for their facilities; one U.K. government assessment has predicted it will deprive 250,000 Malawians of access to family planning .

The disruption has prompted fears Malawi could lose hard-earned gains, and groups such as the National Organization of Nurses and Midwives of Malawi have warned of a possible rise in maternal mortality. But it’s also sparked calls for change. “We have been relying so much on partners … that when the partners fold their hands, immediately our health care system is on its knees,” says Donald Makwakwa, executive director of the Family Planning Association of Malawi. His organization had already lost USAID funding in 2017, when President Donald Trump’s first administration blocked support for organizations that provided or advocated for abortion care, and was hit again in 2025. “What is currently happening should not be looked at from a negative point of view only,” Makwakwa says. “It’s a launchpad to think differently in terms of how, as a country, to finance our health care.”

IN LILONGWE, MALAWI’S sprawling capital, officials scrambled to respond to USAID’s withdrawal. “No one expected it,” says Andrew Likaka, who became chief of health services at the Ministry of Health after a new Malawian government came to power late last year. “We quickly sat down to reprogram how to work without the [lost] donors, without the NGOs.”

One approach has been to try to make health care more efficient—combining services such as HIV prevention and treatment with reproductive care, for example. But these efforts can only go so far. “We can optimize all we want, but it’s just not enough,” says Owen Chikhwaza, director of reproductive health services at the ministry.“We need the actual dollars to support the health care."

In the short term, much of that money still needs to come from foreign countries and philanthropists, officials say. But they’d like to change how it enters Malawi’s health system. Donors have tended to channel funds through NGOs and international agencies, bypassing Malawi’s government, because of concerns about corruption and mismanagement. Likaka and others argue this approach has led to duplication of health efforts, poor coordination with government priorities, and unnecessary spending on high salaries and other overheads. Direct payments would avoid these problems and help make Malawi more independent in the long run, they say.

Many global health experts back the idea. “Provision of financial support to the Malawian government directly … would substantially reduce costs, and … enable the government to develop its own capacities, making it more resilient to future aid shocks,” says Katherine Klemperer, a policy analyst at the Center for Global Development.

The most substantial direct agreement so far was signed with the United States earlier this year. Under the memorandum of understanding , the U.S. government will provide $792 million to the health sector over the next 5 years, while Malawi has pledged to increase its own health spending by $143 million. (More than 30 other countries have signed similar memoranda, for an estimated total of about $14 billion in U.S. funding .) Details of the agreement remain tightly guarded, despite calls for transparency from local organizations; U.S. deals with neighboring countries have been sharply criticized for clauses permitting U.S. access to minerals and health data. But Likaka insists Malawi’s own agreement protects citizens’ interests and confidentiality. “We did the best for the country,” he says. “We agreed in the interest of the nation.”

Malawi is also accelerating plans drawn up by previous governments to raise and better manage money within the country’s borders. The government has already increased certain taxes on income, goods, and services, and introduced electronic tax reporting systems to improve collection. The Parliamentary Committee on Health is pushing for additional “sin taxes” on products such as alcohol, and toll booths on roads. Proceeds should go into a special “national health fund” that would ensure they don’t get siphoned off to other sectors, says the committee’s chair, Anthony Masamba.

Plans to give health facilities more autonomy to generate and manage their own funds are underway, too. For example, district hospitals like KDH have been told to open new paying wards that offer patients more space and privacy while subsidizing care for others. Primary health centers are starting to receive direct funding from the government, bypassing delays and bureaucracy at the district level and helping centers buy what they need when they need it. Likaka wants to see a bigger role for private investors and health care providers, too.

The government’s work is already having an impact and will ultimately put Malawi’s health sector in a better position, Chikhwaza argues. “We have recollected ourselves” after last year’s disruption, he says.

Health advocates are more cautious. The health sector still receives less than 10% of the national budget , despite a decades-old commitment, shared by most African nations, to reach 15%, says George Jobe, executive director of the Malawi Health Equity Network, which represents some 200 organizations in the country and supports the government with budget analysis. Although Jobe supports many of the government’s reforms, he warns they must be implemented fairly—without taxes further squeezing struggling communities, for example, or paying wards exacerbating inequity. “The vulnerable, the poor must not be victimized,” he says. Makwakwa agrees. “It’s high time” to increase domestic funding, he says, “but those resources must be generated in a manner that does not punish the citizens.”

OUTSIDE LILONGWE , local officials and health workers say they support the government’s efforts but add that gains from such measures have so far been uneven. KDH opened a paying ward late last year, reportedly generating nearly $100,000 in its first 6 months. It hasn’t hurt care for nonpaying patients, staff insist. Instead, the additional revenue has helped pay off some of the hospital’s debts and provides a buffer during delays in government funding, they say. Golombe says the hospital recently used the revenue to buy critical drugs, including dexamethasone, which can be used to protect the lungs of premature babies, and pethidine, a painkiller used in cesarean section patients.

But other services have withered. Outreach clinics to remote communities have become infrequent, and mothers are paying the price, says Jaison Banda, a community health nurse and family planning coordinator at KDH. Emily Mvula, who lives in a village west of KDH, had her third child in 2025, a few months after MOMENTUM suddenly ended. She paid $9 to travel the 15 kilometers to KDH for her last prenatal appointments, a sum not every woman can afford. Now, she says she faces difficulties accessing family planning. She sometimes takes her baby daughter to a Catholic-run center, about a 2-hour walk, but it charges for some services and doesn’t offer contraception. She wishes the outreach clinics would come back.

Plans to give primary health centers more autonomy have faced challenges, too. Makiyoni Health Centre, a rural facility east of Kasungu in Salima district that was also hit by MOMENTUM’s termination, now receives direct government funding, but it’s falling short of the clinic’s needs. Staff here say they requested $20,000 last year for critical improvements such as patching holes in the maternity ward’s ceiling, renovating dilapidated latrines, and replacing solar-powered lights that don’t last through the night, leaving some babies to be delivered by flashlight. They learned they’d been allocated less than $4000, and even that they couldn’t access until late June, says Issah Omar Nicks, the nursing officer overseeing the maternity ward. Work has only just commenced on the latrines, he adds.

Nicks and his team have been exploring other ways to raise money in the meantime. They started a garden last year and have been growing and selling potatoes, he says. They also solicit a fee of about $0.60 for health passports—paper booklets normally issued by the Ministry of Health to log patient data—from people who can afford it. (Makiyoni isn’t the only health center to do so, officials tell Science , but the charges are controversial because the passports are important for tracking a person’s health.) Nicks estimates that in total, the ward’s initiatives have raised about $450 in the past couple years.

The funds have helped pay for cleaning supplies as well as medicines and stationery for record keeping. But they can’t solve some of the hardest challenges facing this and other rural facilities. Transport of patients in particular has become a daily struggle, as many district ambulances sit wheelless in parking lots for want of repairs or fuel. Nicks recounts the story of a mother they recently tried to transfer to Salima District Hospital during a difficult labor. They waited 2 hours for an ambulance, which only arrived after a local politician’s intervention. The mother was eventually transferred but neither she nor the baby survived, he says.

Staff hope help is on the way in the form of a system called m-mama. Created by the Vodafone Foundation and partners in 2013, it connects emergency calls with private vehicle owners, like a medical Uber, to get women and newborns to health facilities quickly. M-mama is already up and running in some other sub-Saharan countries; Vodafone and other donors are spending about $10 million to launch it in Malawi.

M-mama suggests one way outside support could continue: with donors helping establish services the government adopts and the private sector helps deliver. The system is designed to ultimately be “owned and operated and funded by national governments,” at a fraction of the cost of traditional ambulances, says Lee Wells, global director of m-mama for Vodafone. The launch date has been the subject of some confusion—health staff in Salima and Kasungu say they expected the service from January—but Wells says five districts should get m-mama from September, with those two districts joining the following month.

Other donors are exploring different ways to continue supporting Malawi’s health. Amref Health Africa recently received money from UNITAID, an agency funded primarily through a special tax on airline tickets, for a 4-year initiative designed to make maternal medicines and diagnostics more accessible . Hester Mkwinda Nyasulu, Amref’s country director for Malawi, believes NGOs will continue to play an especially important role in communities where government services are limited. He and Jobe want to see the government strike its own deals with NGOs to help fill gaps. “If we complement each other well, we’ll be able to reach more people,” Nyasulu says.

The health system that emerges from these changing relationships could be very different from what came before, Banik says. “We’re in a phase between the old order and a new order. We don’t really know what that new phase is going to look like, but it’s certainly not going to be a return to the past.”

WHILE POLICYMAKERS and advocates debate this transition, work continues as before in Katenje. As Tuesday morning rolls around, the ground inside the makeshift health clinic has been cleaned, a weighing scale hangs from the ceiling, and an open register lies on a small blue table set up to one side. Simenti offers family planning sessions most Tuesdays, but today he is running malnutrition screening for young children from a local school first. Just after 8 a.m., dozens arrive with their carers. Simenti tries to reassure a few crying kids. “They think I’m going to immunize them,” he says.

KDH’s Banda has caught a lift with Science to Katenje to see how construction is progressing. The building is on the small side, he says, and there’s not much space around it should the government want to upgrade it to an official primary health center. Nevertheless, he sees promise. It’s in a good location, with a big surrounding population and no other center nearby. The community could have a strong case for receiving government support, he says.

For now the project lacks money, like many plans for Malawi’s health services. Simenti hopes funds will come from somewhere—local contributors, NGOs, or philanthropists—to help finish what they started. But he’s proud to see the health post being used, even in the interim. For people here, building it has been a big commitment, he says. “Now it’s the time to give them the services.”

Alzheimer’s disease has long been seen as a tale of two proteins: beta amyloid, which forms sticky plaques in the brain, and tau, which in its diseased state creates tangles inside neurons. Antibodies that clear beta amyloid have been approved to treat Alzheimer’s, but it was tau that made headlines this week, as researchers presented key new details about a drug that reduced production of the protein and slowed cognitive decline in a recent clinical trial.

At the Alzheimer’s Association International Conference, neurologist Catherine Mummery of University College London presented results from a phase 2 trial testing diranersen, a drug developed by Biogen to lower the body’s production of tau, in more than 400 patients with early-stage Alzheimer’s. On the study’s main measure of cognition, participants getting diranersen saw as much as a 26% slowing of decline—about on par with the effect seen in earlier trials of approved antiamyloid drugs. (Biogen had announced in May the drug slowed cognitive decline but did not say by how much.)

Although the presentation sparked enthusiasm from Alzheimer’s researchers, it also drew attention to puzzling aspects of the trial results. A potentially worrisome side effect emerged at high doses, and contrary to expectations, patients taking the lowest of three possible doses saw the greatest benefit. For these reasons, the findings represent “a double, not a home run,” says neurologist Adam Boxer of the University of California San Francisco, who this month launched a clinical trials platform to try different antitau therapies in Alzheimer’s.

Diranersen belongs to a class of drugs known as antisense oligonucleotides, strands of RNA that dampen activity of specific genes. It interrupts production of tau by binding to the messenger RNA that encodes instructions for producing the protein, and must be injected directly into the cerebrospinal fluid to reach the brain. After 18 months, participants in all three dose groups had between one-third and one-half as much tau in their cerebrospinal fluid as when they started the trial, whereas levels increased slightly among people receiving placebo. Imaging done on a subgroup of participants showed the drug also reduced tau tangles in the brain.

Glenn Harris, who directs research partnerships for the Rainwater Charitable Foundation, a group that supports research on brain diseases involving tau, says many in the field had higher expectations for the drug’s cognitive benefit. But the findings create momentum for tau therapies at a time when several other candidates—mostly antibodies targeting abnormal tau—have failed in Alzheimer’s trials. “I think the enthusiasm is highest with this one because [it works by] shutting off the faucet of [tau] production or at least slowing it,” he says, apparently allowing the brain to clean out tangles.

At the meeting Mummery addressed the finding of decreasing cognitive benefits at higher doses, which caused the trial to fall short of the dose-dependent effect the investigators had chosen as its primary endpoint. The low dose appears to be “showing the most consistent positive clinical effect” across neuropsychological tests, Mummery said. Biogen is moving forward with a larger clinical trial, but has not revealed which doses it will investigate.

About one-quarter of people in the higher dose groups experienced a “confusional state” after diranersen was administered, a side effect that lasted up to 1 week. This was unexpected, Boxer says, and might be an off-target effect of the antisense molecules. “But a troubling possibility,” he says, is that it is related to tau reduction. “Maybe we can’t knock down tau without some effects.”

The cognitive benefit seen with diranersen is not enough to argue it should replace the approved antiamyloid antibodies, says Washington University in St. Louis neurologist Eric McDade, who co-directs a research group that tests Alzheimer’s drugs in families with inherited forms of the disease. Combination treatment with both drug types appears to be “the strongest way forward,” he suggests. “Maybe there is a synergistic effect—but that needs to be tested now.”

A few trials are testing antitau and antiamyloid drugs in combination, though none involve diranersen. One, by McDade’s group, uses an antitau antibody, and Boxer’s group is working with a vaccine that prompts the immune system to attack abnormal tau.

The modest cognitive effects in the trial may speak to the complexity of the disease process, Boxer says. “There’s been a long-standing hypothesis among Alzheimer’s researchers that amyloid lights the fire of the disease, but that it’s tau [fueling neurodegeneration],” Boxer says. These data “provide pretty clear support for that.” But Boxer thinks other abnormal proteins seen frequently in Alzheimer’s brains—among them alpha synuclein and TDP-43—might be stoking the flames, too, possibly lessening the beneficial effects of tau lowering.

And the disease involves a lot more than proteins, researchers have increasingly come to acknowledge. “We know that there’s a huge vascular component [in Alzheimer’s],” Harris says. “There’s a huge neuroinflammation component and neuroimmune component and synaptic health component.” Fortunately, he says, “There are pipelines for those therapies as well.”

LISTEN TO THIS ARTICLE

For 3 decades in the mid–19th century, the remote Atlantic Ocean island of St. Helena served as a place of liberation. In 1807, Great Britain had passed a law abolishing the Atlantic slave trade within the British empire; in the decades that followed, British naval ships intercepted hundreds of vessels carrying enslaved Africans across the ocean. From 1840 to 1872, some of the people freed from these slave ships were taken to St. Helena—a small island about halfway between the coast of Brazil and Africa—to recuperate before they either returned to Africa or traveled on to British colonies such as Jamaica.

But thousands of these liberated Africans, weakened by malnutrition and disease, died shortly after arriving on the island. In 2007, construction work on St. Helena’s first airport uncovered dozens of their graves. A new analysis of their remains, published today in Science , reveals where they came from, providing historians and descendant communities with a fine-grained look at regional impacts of the slave trade.

The data show many were trafficked from Africa’s western coast—but also reveal slavers swept up adults and children from deep within the continent’s interior. “The findings offer an important window into the history of the slave trade,” says Daniel Domingues da Silva, a historian at Rice University who was not involved with the new research.

The graves in the study were found in a part of the island called Rupert’s Valley, where recently freed Africans were housed after disembarking. Historical documents refer to a cemetery at the site. “It became clear pretty quickly these were the remains of liberated Africans,” says Hannes Schroeder, a geneticist at the University of Copenhagen who led the new research. “The question came up, where did they come from? And if these people were liberated captives brought to St. Helena, should they be repatriated to wherever they came from in Africa?”

Some of the liberated Africans survived and made homes on St. Helena, where their descendants make up part of the island’s 4000-strong population. But because those buried in Rupert’s Valley died soon after arriving, “most of them don’t have descendants,” says Helena Bennett, head of the St Helena National Trust and co-author of the new paper. “The island decided it would be good to understand where they came from.”

To that end, the island’s government permitted researchers to look for genetic material and analyze isotopes—which can provide clues to where someone grew up—in teeth from the cemetery. In 2023, DNA analysis linked some of the individuals in the cemetery to people living in modern-day Gabon and Angola, an area historians recognized as a former hub for the slave trade.

However, centuries of colonial rule and the slave trade itself reshaped the human geography of Africa, meaning genetic connections to modern populations can’t tell the whole story. “Communities that existed back then aren’t the same today,” Domingues da Silva says.

That’s where strontium isotopes came in. Incorporated into tooth enamel as it’s formed, strontium isotopes found in water and incorporated into plants “reflect where they were at that specific time in their lives,” says study co-author University of California Santa Cruz geochemist Vicky Oelze. “Even if they were displaced, the signature of where they grew up is locked in their teeth for eternity.”

First, Oelze and her team had to map out where different isotope signatures are found throughout sub-Saharan Africa, requiring her to contact dozens of botanists, geologists, and archaeologists to accumulate more than 1000 soil and plant samples. They published their map in 2024 . When they plotted the isotopes found in the teeth of 152 people buried at Rupert’s Valley, the data showed most individuals came from the West African coast. But some grew up much farther east, suggesting they were forced to journey more than 1000 kilometers after being enslaved.

Using molecular evidence to pinpoint people’s origins is “a first for the study of the transatlantic slave trade,” says Purdue University isotope geochemist Clement Bataille, who was not part of the research team. “I’m really glad this is happening.”

The teeth weren’t done yielding their secrets. Using lasers to burn off thin layers of enamel, the researchers traced how isotope values changed as the individual teeth gradually formed during their bearers’ childhoods, reflecting movement from place to place. In some cases, the values revealed that “individuals were moving when they were between 7 and 9 years old,” says Raquel Fleskes, an anthropological geneticist at Dartmouth College who was not part of the research team. “It’s not just adults; it’s also children who were ripped from their homes.”

The data helped St. Helena decide what to do with the excavated remains. A repatriation effort “wouldn’t have been reaching out to just one country; it would have been multiple countries which have changed since the 19th century,” says Bennett, who traces part of her ancestry to the island’s liberated African community.

Instead, the community chose to rebury the remains on St. Helena. Each was interred in a casket built by local schoolchildren. In a 2022 ceremony, members of the community sang Amazing Grace as the caskets were lowered into graves. Today, the graves are marked with hundreds of stones from the nearby beach, each painted with the handprint of a St. Helena child.

Each year since, community members have visited the gravesite or held an event focused on the reburied individuals on 23 August, the International Day for the Remembrance of the Slave Trade and its Abolition. “We are a migration island, formed by settlers, sailors, soldiers, and slaves,” Bennett says. “Liberated Africans are accepted as part of our identity and history—we are connected to their story, because they shaped our story.”

LISTEN TO THIS ARTICLE

Astronomers have discovered more than 6000 planets orbiting other stars—many of them small and rocky like Earth. But despite years of searching, firm evidence that such worlds have stable atmospheres has remained elusive. Now, researchers report in Science that the relatively nearby super-Earth LHS 1140b appears to be leaking helium gas into space , implying it still has a substantial atmosphere billions of years after its birth.

“This is an extraordinary result if confirmed,” says astronomer René Doyon of the University of Montréal, who was not involved in the work. “The signal is definitely there.”

The search for alien atmospheres, especially around temperate planets where liquid water can exist, is key to finding planets that could support life. Astronomers have found plenty of gas giant exoplanets, similar to Jupiter or Saturn, but in addition to lacking a rocky surface, many orbit so close to their stars that they are scorchingly hot, and thus uninhabitable.

Detecting atmospheres around small, rocky planets is especially difficult. The bright light from the planet’s parent star makes spotting the world hard enough. Detecting the far fainter signal of its atmosphere is harder still. To make things easier, astronomers have focused on small dim stars known as M dwarfs which, luckily, seem to have plenty of rocky planets around them. In many studies, scientists look for distinctive frequencies of light absorbed by atmospheric gases when planets pass in front of their home star.

But astronomers have not turned up anything convincing so far. For example, the seven Earth-size planets orbiting an M dwarf star known as TRAPPIST-1, just 41 light-years from Earth, have so far displayed no signs of atmospheres. That’s led many to worry that M dwarfs, which are often turbulent when young, blast away the atmospheres of their planets.

To settle the question, JWST’s managers have devoted 500 hours of telescope time to determining whether any small rocky planets around M dwarfs have atmospheres . They’re targeting nine exoplanets around various stars using an indirect method to detect the presence of atmospheres. JWST measures the temperature of the star-facing side of the planet. If this dayside is as hot as expected, the planet is probably a bare rock. But a cooler than expected dayside suggests winds are whisking heat to the nightside of the planet, implying an atmosphere.

The campaign recently released data from its first target , the Earth-size planet GJ 3929b, 52 light-years away. Initial analysis of those data, posted on the preprint server arXiv last month, reveals a world that shows every sign of being a bare rock . Still, the authors can’t exclude the possibility of a thin atmosphere.

The team publishing today took a different tack. Leader Collin Cherubim, a planetary scientist at Harvard University, had built a computer model of how exoplanetary atmospheres evolve over time. In particular, the model looked at how lighter elements, such as hydrogen and helium, would tend to drift up to the planet’s outer atmosphere and be the first to be boiled off by the heat and outbursts of the host star. As a result, heavier gases such as nitrogen and oxygen—held more tightly by the planet’s gravity—become enriched lower down.

Cherubim applied the model to LHS 1140b, a well-studied planet 49 light-years away whose density is slightly lower than Earth’s. That means it likely has a substantial atmosphere or is wrapped in water or ice, or a mixture of all three. Earlier studies of the world firmly ruled out a dense hydrogen atmosphere, although inconclusive hints of heavier gases such as nitrogen were found. Cherubim’s model predicted that LHS 1140b would have an upper atmosphere dominated by helium. “I was just following the numbers,” he says. “My model predicted escaping helium and a lot of it.”

To find that helium, the team used the 6.5-meter Magellan Telescopes in Chile, which recently had a new high-resolution spectrometer installed. “It was sensitive to this helium line in the near infrared, so it was kind of perfect,” Cherubim says. In 2024, the team observed the LHS 1140 system on a night when LHS 1140b and another smaller and closer in planet, LHS 1140c, both passed in front of their star. When the larger planet transited, the researchers saw starlight dropping in brightness at a wavelength known to be absorbed by helium. When the smaller planet crossed, they saw no such signal.

The team repeated the observation the following year and did not see the same signal, suggesting the helium loss varies with time. Doyon points out that M dwarf stars sometimes show the same helium absorption signal because the gas exists in their own atmospheres. But he says it would be “quite a coincidence” for variable helium in the stellar atmosphere to mimic a planetary signal precisely during a transit.

The variability is still a puzzle, Doyon says. “This begs for more observations by other instruments and independent analysis to be absolutely sure.”

Many teams are zooming in on LHS 1140b, because of its many charms. Besides being nearby, it could be hiding other gases in its lower atmosphere, such as nitrogen and carbon dioxide—much like Earth. And if those gases provide an element of greenhouse warming, it could possess a liquid water ocean warm enough to swim in. “I expect many, many telescopes to take a look at it,” Doyon says.

LISTEN TO THIS ARTICLE

MONTREAL— A meteorite discovered 4 years ago in the Sahara Desert by Sahrawi prospectors may be the oldest ever identified from Mars, preserving a fragment of the planet’s crust from more than 4.1 billion years ago. Although analyses are still unfolding, the nearly 800-gram rock, which split into eight pieces, is already offering an unprecedented glimpse of the planet’s watery youth—and challenging traditional views about how its crust formed.

The meteorite, called Teghaza 001, “will revolutionize the way that we think about early Mars,” said Lee Saper, a geochemist at NASA’s Jet Propulsion Laboratory (JPL), in a talk this week here at the Goldschmidt geochemistry conference. One study, examining hydrogen trapped in the rock’s minerals , appeared last month in Science Advances ; another, establishing its age, is under review at Science and is available as a preprint, and more are underway.

Chunks of Mars blasted into space by impacts occasionally make it to Earth as meteorites. Because the martian atmosphere has a distinctive fingerprint of noble gases, researchers can distinguish martian meteorites from other space rocks by tiny amounts of gas trapped in them. Although most of the planet’s surface is likely more than 3.5 billion years old, nearly all of the several hundred known martian meteorites are relatively young volcanic basalts—hard, durable rocks that better withstood their violent ejection from Mars and passage through Earth’s atmosphere.

Until now, only one known martian meteorite, 4.1-billion-year-old Allan Hills 84001 , discovered in Antarctica in 1984, preserved crust from that first chapter of martian history. (Another meteorite, nicknamed Black Beauty , contains individual mineral grains as old as 4.4 billion years, but assembled into a rock much more recently.) Allan Hills provided critical insight into early Mars, but relying on one sample has always been uncomfortable, says Lydia Hallis, a planetary scientist at the University of Glasgow. Teghaza changes that. “We’ve doubled our old meteorites,” she says. “There will be a lot of people, including me, who want to get a piece of this.” (The main mass of the meteorite is owned by the Maine Mineral & Gem Museum, which provided the small sample used in the studies.)

Teghaza is rich in zircons, mineral crystals prized by geologists because tiny amounts of uranium trapped within them decay into lead at a known rate, providing a precise clock. Work led by Yang Liu, a planetary scientist at JPL, dates the meteorite to more than 4.1 billion years ago. The rock may actually be older, however. When the mineral grains melt or are exposed to fluids, their clocks are reset to zero. “There’s more of a story to this rock than most other martian meteorites,” says Christopher Herd, a geologist at the University of Alberta who also works on the rock sampling efforts of NASA’s Perseverance rover.

Perhaps Teghaza’s biggest surprise is its composition: an unusual richness in silicon. “It’s very strange, we don’t expect that,” says Eva Scheller, a planetary scientist at Stanford University. Most of Mars’s surface is made up of basalts, the dark volcanic rocks produced by direct melting of the mantle, much like Earth’s ocean crust. By contrast, lighter, silica-rich rocks such as granite generally require repeated episodes of melting and chemical separation to evolve from basalt. On Earth, those processes are closely associated with plate tectonics, leading many researchers to assume that Mars, which lacks active tectonics, has little granitelike crust.

That picture has begun to change . Mars orbiters have spotted hints of granitelike rocks exposed by impacts. The density of the crust, inferred from marsquakes captured by NASA’s Insight lander, is lower than that of pure basalt. Other martian meteorites contain bits of quartz and related silica-rich minerals. And Perseverance, exploring terrain similar in age to Teghaza, last year discovered a quartz-rich outcrop, dubbed Plankeholmane, that looks almost exactly like granite, Liu reported earlier this year at the Lunar and Planetary Science Conference.

Taken together, the findings suggest that even without plate tectonics, Mars—and perhaps distant planets—can nevertheless generate complex magmatic systems capable of producing compositionally diverse crust. The repeated recycling of crust into magma and back again could have helped create the rich chemical assemblages that are necessary for life, says Tobermory Mackay-Champion, a geophysicist at the University of Bristol who led a study published last month in Nature Astronomy proposing such a system. “Compositionally diverse crust may be far more common on rocky planets than previously thought.”

Teghaza also strengthens evidence that Mars began to dry up very early in its history. Hydrogen, the key ingredient in water, provides one of the best tracers of the planet’s water. The meteorite’s ratio of hydrogen to its heavier isotope, deuterium, closely matches measurements from Allan Hills. Both indicate early Mars contained a higher ratio of hydrogen to deuterium than it does today. Because lighter hydrogen escapes to space more readily than deuterium, the changing ratio implies a large loss of hydrogen, and hence of water.

Mars’s ancient magnetic field would have shielded the atmosphere from charged particles in the solar wind, reducing the escape of hydrogen. But after the field disappeared about 3.9 billion years ago, researchers believe water loss accelerated. The new measurements, which also include records of the primordial ratio the planet formed with and preserved in its mantle, show this loss was already well underway 4.1 billion years ago. “This is the product of rapid hydrogen loss from the juvenile martian atmosphere,” Saper said during his talk.

Researchers expect Teghaza to yield more surprises. The rock clearly records alteration by water or other fluids in the crust, but the list of possible sources is long. More scrutiny of this piece of early Mars could reveal whether the planet resembled the volcanically active, water-altered landscape that many geologists envision occurred on Earth during its first several hundred million years, says James Day, a geochemist at the Scripps Institution of Oceanography. So far, he adds, “that’s what it appears to be.”

LISTEN TO THIS ARTICLE

Hiding out-of-sight on our produce, pipes, and even teeth are tiny bacterial cells enmeshed within a sticky web. These microbial clusters—known as biofilms—are more resistant to antibiotics and other weapons. Now, researchers have discovered another surprising way they survive: ejecting some of their cells as if they were escape pods jettisoned from a doomed sci-fi spaceship, researchers report this month in Nature Microbiology .

“It’s a major interesting finding,” says Jing Yan, a microbiologist at Yale University who wasn’t involved with the study. The work, he says, sheds light on a little-understood biofilm mechanism and proposes new ways those clusters might be dispersing cells.

As is the case with many scientific discoveries, this one came as a surprise. Researchers at the University of California San Diego (UCSD) starved a large biofilm of hay bacillus, a type of bacterium commonly used in probiotics and agriculture, for 16 hours, expecting to see it fall apart completely. Instead, they saw cells escaping the biofilm through a single path, as if flowing away on a river.

These escapees weren’t just passive projectiles. Unlike other cells in the biofilm, motile cells are shaped with a taillike motor called a flagellum, which allows them to swim away toward healthier environments. “I’m a Star Wars nerd. If you ever watched the original 1977 movie, C3PO and R2D2 eject themselves from the ship. It’s exactly like that,” says Gürol Süel, p rofessor of molecular biology at UCSD and leader of the study .

After releasing these cells, the original biofilm didn’t collapse, the team found. If healthier conditions were to return, the community could start to grow again.

“This way you’re hedging your bets because the other cells are not leaving,” Süel says. “They are still holding down the fort.”

The researchers also identified how the ejection mechanism works. Concentrated within the center of a biofilm, motile cells can secrete enzymes that produce a gummy “hydrogel” known as polyglutamic acid, which can absorb 1000 times its weight in water. As the hydrogel absorbs water, it swells, forcing those inner motile cells outward through the biofilm’s exterior, like volcanic lava erupting through Earth’s crust. This biological mechanism has only been observed once before. Jellyfish rely on the same force to release their stinging needles.

The discovery could have implications across medicine. By better understanding what triggers the secretion of hydrogel-producing enzymes, scientists could then force their release. In an experiment, Süel and his team observed a biofilm with motile cells genetically modified to overproduce those enzymes. The biofilm blew apart.

That could make biofilms far easier to kill without relying on antibiotics, which, beyond coming with a list of adverse side effects to humans, also decrease in efficacy with time as biofilms build resistance.

“Here’s an instance where we have no need for antibiotics,” Süel says. “We’re using the physics of hydrogels to disrupt biofilms without having to resort to any type of antibiotics or harsh chemicals.”

LISTEN TO THIS ARTICLE

June’s European heat wave, in which temperatures above 40°C led to the deaths of more than 10,000 people, made headlines not just for its severity, but for what it said about climate change. Last week, a group called World Weather Attribution (WWA) linked it firmly to global warming, saying such a heat wave would have been “virtually impossible” 50 years ago. It was a high-profile claim from a field called extreme event attribution, which seeks to gauge how global warming impacted a particular heat wave, flood, or storm. The field is still evolving, according to a report released today by the National Academies of Sciences, Engineering, and Medicine (NASEM).

“We need to make sure that we are clearly communicating the uncertainties of any given study,” says James Hurrell, an atmospheric scientist at Colorado State University and chair of the NASEM committee that wrote the report. “But I think the report is also trying to make the point that we can reduce some of the uncertainties.” It recommends researchers apply multiple attribution methods, adopt common standards, and communicate inherent uncertainties.

In the past decade, attribution science has advanced rapidly , with the annual number of studies more than doubling since 2012, according to the report. Better climate models, longer observational records, and more sophisticated statistical techniques have all improved researchers’ abilities to detect climate change’s influence. “The first attribution studies that come to my mind appeared in the early 2000s,” Hurrell says. “So, it’s a pretty new endeavor.”

It has moved well beyond academic journals. Groups such as WWA now routinely release assessments within days of disasters, without peer review. Attribution studies are also increasingly cited in lawsuits seeking to hold fossil fuel companies financially responsible for climate-related damages. Some researchers are attempting to trace portions of those damages to emissions from specific sources and companies.

The field also faces political scrutiny, which dogged the NASEM committee. Republican lawmakers and groups allied with the fossil fuel industry questioned whether some members had conflicts of interest because of ties to climate advocacy organizations and sought their internal communications through public records requests. Hurrell brushed off the criticisms, saying the report reflects the consensus opinions of the committee as a whole. “We really didn’t let the outside external chatter influence us in any way,” he says.

Because every weather event arises from a combination of many influences, including natural variability, assessing the role of global warming is challenging. Most attribution studies compare climate model simulations of today’s greenhouse-warmed world with simulations of a counterfactual world without human warming. Scientists then compare how often a given type of event occurs in both models, and how similar it looks.

A more computationally demanding method feeds data from a specific event into a weather model while removing the effects of greenhouse gases. It generates the atmospheric conditions surrounding the event in this hypothetical world, which can then be compared with actual observations.

Other methods forgo modeling entirely. Davide Faranda, a climate scientist at Paris-Saclay University, compares data on weather patterns leading up to a recent extreme weather event with historical analogs from the mid–20th century, predating most warming. The challenge is the quality of the available historical data, which is especially limited in the Global South, the report finds.

Confidence in the outcome of these methods depends on the type of event in question, the report concludes. Heat waves and heavy rainfall are closely linked to rising temperatures and yield the most confident attribution results. But hurricanes and severe storms are shaped not only by warming, but also by complex atmospheric dynamics that climate models still struggle to capture . And disasters such as wildfires can defy modeling, because they depend on human behavior, such as land management and accidental or deliberate ignitions.

Different attribution studies can also produce substantially different results for the same event. A 2025 study comparing analyses of a record-breaking 2021 heat wave in the Pacific Northwest found that some methods characterized it as a one-in-1000-year event, whereas others suggested an event of similar magnitude might occur every 50 years or so. “The sort of simple declarations are always suspicious because they’re likely just one of many valid approaches,” says Philip Mote, a climate scientist at Oregon State University who served on a 2016 NASEM review of attribution science.

The committee argues that confidence increases when studies combine different approaches to validate their results. It also recommends investment in higher resolution climate models and “a common overarching framework with recommended best practices.” And it urges groups that produce rapid attribution studies to routinely submit their analyses for peer review.

No one should expect certainty, says Friederike Otto, a climatologist and WWA’s founder. The purpose of attribution science is not to claim that climate change alone caused an extreme event, but to calculate how it increased its odds or intensity, she says. “If you say smoking causes cancer, you also mean that smoking dramatically increases the likelihood of you developing cancer.”

LISTEN TO THIS ARTICLE

One of the largest and most complete Tyrannosaurus rex fossils yet discovered was purchased by an anonymous bidder in a Sotheby’s auction in New York City for $50 million on Tuesday, making it the most expensive fossil of any kind ever sold. The 67-million-year-old T. rex measuring 11.5 meters in length and nicknamed “Gus” (pictured above) was unearthed on private land in Harding County in South Dakota, between 2021 and ’23. The fossil commanded such a high price because it’s unusually well preserved: It has a nearly intact skull, almost all its rib bones, and a relatively rare set of wishbones. Its price tag bested that of “Apex” the Stegosaurus, which sold for $44.6 million in 2024. Shortly after the auction, scientists and curators voiced their disappointment on social media. Some bemoaned the fact that pricy sales like this make it more expensive for museums and research facilities to acquire fossils, while others noted they incentivize fossil poaching on public or Indigenous land.

LISTEN TO THIS ARTICLE

Svalbard in Norway— The images captured in 2017 looked wrong. Against the pale winter sky above the Norwegian archipelago of Svalbard, tiny green smudges hovered beside an auroral curtain. The fragmented auroralike emissions (FAEs) were only a few kilometers across and lasted less than 1 minute. Most puzzling of all, they seemed to ignore invisible magnetic field lines that normally define the contours of the northern lights. “They really are a weird little thing,” says Katie Herlingshaw, a space physicist at the University Centre in Svalbard (UNIS).

A major new instrument may finally provide answers. This fall, EISCAT AB, an international organization headquartered in Sweden, plans to unveil EISCAT_3D , a $65 million next-generation radar system that will allow researchers to probe the upper atmosphere as never before. Together, the arrays being assembled across northern Scandinavia promise to be “the most advanced incoherent scatter radar in the world,” says Thomas Ulich, head of science for EISCAT AB, offering a panoramic view of the ionosphere—a high-altitude layer of electrically charged plasma—as it is roiled by solar storms.

Because the contours of Earth’s magnetic field funnel energetic particles into the atmosphere near the poles, northern Scandinavia is a good place to observe those disturbances, which aren’t just a scientific mystery. They can disrupt high-frequency radio links, increase atmospheric drag on low-flying satellites, and make signals from GPS and other space-based navigation systems flicker as they pass through disturbed plasma. EISCAT_3D should help scientists understand those threats. It will also track growing swarms of space debris, another hazard for spacecraft.

Those practical payoffs give EISCAT_3D urgency. And for researchers, the instrument’s debut comes at a scientifically tantalizing moment. “We’re in a golden age of auroral discovery,” says Stephan Buchert of the Swedish Institute of Space Physics.

Many of the surprises have come from a global network of aurora chasers. Amateurs have had eyes on the aurora since the 19th century, when ship captains and monks, for example, mailed observations to scientists in Scandinavia, says Maxime Grandin, a space physicist at the Finnish Meteorological Institute. Nowadays, he says, thanks to social media and smartphone images with accurate timestamps, “citizen scientists have transformed auroral science.”

The “dawn of this new era,” Grandin says, was the discovery of STEVE , a narrow ribbon of mauve light that can stretch hundreds of kilometers across the sky and appears farther south than ordinary aurorae. The spectacle attracted widespread scientific attention only after aurora buffs shared images online in the early 2010s. By combining photos with satellite measurements, researchers eventually showed that STEVE is associated with a narrow channel of heated plasma and fast-moving charged particles in the upper atmosphere.

The auroral menagerie continued to grow as amateurs helped reveal wavelike “dunes” and rapidly moving green streaks that are seen together with STEVE, dubbed the “Picket Fence.” Researchers on Svalbard recently added another curiosity: a faint pinkish “ghost aurora” that may involve highly excited nitrogen and oxygen molecules. FAEs add still more intrigue.

Aurora enthusiasts began to report fragments about a decade ago, and they remain unexplained. Hundreds have been spotted near Svalbard, whose location may be key to solving the puzzle. The archipelago sits beneath the magnetic field’s dayside cusp, the funnellike region where charged particles from the Sun stream into Earth’s upper atmosphere, triggering aurorae. Yet the FAEs seen there appear to arise differently, from processes within the ionosphere itself. “We don’t understand some of the fundamental physics,” Herlingshaw says.

EISCAT_3D is expected to peel back the curtain and expose the inner workings of FAEs and other newfound auroral features. With backing from its member nations—Finland, Norway, Sweden, China, Japan, and the United Kingdom—the organization has installed a transmitter-receiver array in Norway that aims powerful radar beams at the upper atmosphere, as well as thousands of receiver antennas in Finland and Sweden to capture the weak echoes scattered from free electrons in the ionosphere.

Current incoherent scatter radars probe the atmosphere along narrow lines of sight—like “drilling a hole in a glacier,” Grandin says. “You might be missing something superinteresting a couple kilometers away, or even a few meters away.” In contrast, EISCAT_3D’s trio of antenna arrays will “paint the sky,” as project documentation puts it, building 3D images of the upper atmosphere and allowing scientists to view phenomena such as FAEs in their full spatial context. “You can start looking at the energetics of the entire upper atmosphere,” Grandin says.

Barring “unforeseen complications,” Ulich says, a partial configuration of EISCAT_3D will be up and running late this year—just in time to catch the tail end of the Sun’s most active phase in its 11-year cycle. “We’re at solar max now,” says Lisa Baddeley, a space physicist at UNIS, “but it’s going to start to peter out soon.”

EISCAT_3D managers are currently working out how to allocate use of the radar to scientists from nonmember countries. They expect high demand, because EISCAT_3D is bound to reveal phenomena no one has yet imagined, Herlingshaw says. “It will be an absolute revolution.”

LISTEN TO THIS ARTICLE

MONTREAL— A glassy swath of lava discovered on the floor of the Atlantic Ocean doesn’t fit any existing explanation for how volcanoes form. Named the Balerion Lava Field after the gargantuan black dragon from the TV series Game of Thrones , it may represent a previously unrecognized form of volcanism lurking across Earth’s still mostly unexplored sea floor.

“This may be a common but invisible form of volcanism,” Froukje van der Zwan, a geochemist at the King Abdullah University of Science and Technology, said while presenting the results this week here at the Goldschmidt Conference.

Most volcanic activity is linked to the movements of tectonic plates. In subduction zones, such as those ringing the Pacific Ocean, slabs of oceanic crust sink into the mantle, releasing water that melts the overlying rock into reservoirs of magma that drive eruptions. Elsewhere, magma oozes out at midocean ridges, where plates pull apart to create new crust.

Volcanoes can also erupt far from plate boundaries, such as in Hawaii. In many places, these “intraplate volcanoes” have been linked to mantle hot spots powered by plumes of heat rising from deeper in Earth—in some cases all the way from the edge of the planet’s core. But the Balerion Lava Field—located in the northwestern Atlantic far from any subduction zone, midocean ridge, or known hot spot—fits none of these categories.

Van der Zwan and her colleagues first became interested in the region after acoustic data from a previous research cruise revealed a layer of hard rock buried just beneath the ocean sediment. That was odd: The crust there had formed about 20 million years ago and should have been blanketed by thick sediments.

Then, during a 2017 research cruise to the area, the researchers mapped the topography of the sea floor, dredged up rocks, and extracted several cores. They found the lava covered about 360 square kilometers of sea floor in a remarkably flat sheet and was glassy, suggesting it erupted as hot, runny magma. Dating shells preserved above and below the lava indicate it erupted less than 50,000 years ago—far younger than the 20-million-year-old crust beneath it. Its chemistry also defies existing volcanic classifications.

“The compositions are wacky. They are so far away from what we expect,” says Isra Ezad, an experimental petrologist at the University of Western Australia who wasn’t involved with the research. “There seems to be no surface deformation associated with it. It’s like someone secretly just put in some lava without there ever being any reason why.”

The researchers were binge-watching Game of Thrones during the 6-week cruise when they named the field. Van der Zwan says they settled on Balerion after finding a crystal of olivine in the lava that looked like a dragon. But a less fantastical explanation for the field has proved elusive.

Their leading explanation is that a block of rock called pyroxenite sank from the uppermost mantle deep into Earth before being carried up again by the mantle’s slow circulation. Because pyroxenite melts more easily than the surrounding mantle, it could have created a buoyant pocket of magma that eventually erupted onto the sea floor. How such a block could have cycled through the deep mantle so far from a subduction zone is unclear.

It’s also unknown how widespread this type of volcanism might be. Searches for submarine volcanoes have focused on topographic highs, potentially overlooking flat lava fields like Balerion. If common, such eruptions could provide hard anchors for deep sea creatures. “A lot of life needs a rock,” van der Zwan says. They could also alter the balance of how carbon and sulfur cycle between Earth’s interior and surface. The team has already identified another site about 400 kilometers away with a similar acoustic pattern to Balerion.

The find adds to a growing list of volcanic discoveries that challenge traditional ideas. Improved maps of the sea floor have revealed numerous tiny “petit-spot” volcanoes that appear to form through small fractures in oceanic crust as it bends ahead of a subduction zone. And earlier this year, Sébastien Pilet, a petrologist at the University of Lausanne, and his colleagues proposed that Italy’s famous Mount Etna might represent yet another form of intraplate volcanism. “We still have a lot of discoveries to make,” Pilet says. “These are things that open a lot of new questions about lava in Earth.”

LISTEN TO THIS ARTICLE

Deep in the lush forests of Lomami National Park in the Democratic Republic of the Congo lives a rare monkey that has long evaded scientists and even most local communities. Only few have witnessed the black, orange-mouthed primate quietly perching high in the canopy.

Now, following years of determined fieldwork, researchers have concluded the elusive animal–christened “Likweli,” the name given to it by the local Balanga people— is a new species . It’s only the fifth species of African monkey discovered in the past 75 years, and the second from Lomami, researchers report today in PLOS ONE . The discovery, they say, is a reminder of the importance of conservation in a region under threat from hunting pressures and habitat loss.

“It’s very exciting, but also very sobering to be finding a new species of primate when tropical forests are under so much stress right now,” says Thomas Gillespie, an ecologist at Emory University who was not involved with the paper.

The Likweli monkey was first caught on camera in 2008, when a research expedition snapped a blurry photo of a black-pelted primate they didn’t recognize. But it wasn’t spotted again by scientists until a decade later, when the Lukuru Wildlife Research Foundation’s TL2 Project, a conservation initiative focusing on the Tshupa, Lomami, and Lualaba river regions, established a field base in the area. When researchers spotted and photographed the monkeys again, they quickly realized they were dealing with an animal they didn’t know about.

Likweli belongs to the genus Colobus , and like other primates in the group it has four fingers. But “the orange mask around the lips … and the way [the monkey’s] hair is spiky” is different from other Colobus primates, says Junior Amboko, a conservation biologist and Ph.D. student at Florida Atlantic University (FAU) who co-authored the paper. Still, it wasn’t clear whether Likweli was a subspecies of an existing monkey, or a completely new species.

Over the next 4 years, Amboko and his team scoured the national park for the animals, observing them in their natural habitat and recording their vocalizations. They also consulted communities in the surrounding area for information on the species’ behavior and to aid in their search. They found that Likweli was little known among the local population—out of 52 villages the researchers visited, only eight communities recognized it.

During their fieldwork, the scientists also acquired dead specimens destined for illegal bushmeat trade that had been confiscated from hunters by local patrols. This allowed the team to study the animals’ bodies and extract genetic samples to compare with those from a nearby population of related monkeys.

The researchers found the Likweli monkeys are much smaller Colobus species, and have a unique orange-cream patch surrounding their mouth area, as well as a white patch on their backside. An analysis of the animals’ vocalizations also revealed the Likweli’s calls are distinct from those of their relatives. Of all Colobus species, they were most similar in appearance to the black colobus ( C. satanas ), but the genetic analysis showed the two species diverged some 5 million years ago.

Overall, the study points to “multiple lines of evidence that [we] were dealing with a new species,” says Kate Detwiler, a primatologist from FAU and co-author of the paper. The team gave the monkey the scientific name C. congoensis , after the Congo Basin itself, because they wanted to recognize the biodiversity in the area, Amboko says.

The researchers recommend that C. congoensis be listed as endangered, because it is an endemic species that exists in a small geographical range, which they estimate at about 1700 square kilometers. Habitat loss from hunting or land cultivation in the buffer zone around the national park could further threaten the monkey, the researchers write. “Just being endemic creates opportunities for extinction,” Gillespie says, “so it means that the national park … now deserve[s] even more attention.”

LISTEN TO THIS ARTICLE

It was named Oak City for a reason. Even a decade ago, the streets of Stellenbosch, South Africa, were lined with towering English oaks several hundred years old. European settlers had brought the species with them centuries ago, and it seemed to thrive.

But now the city’s oaks are dying, destroyed by the polyphagous shot-hole borer, an invasive beetle that infects oaks and at least 600 other plant species. Females drill a toothpick-size hole, lay eggs, and inoculate the wood with a fungus that produces fruiting bodies—food for the beetles after they hatch. The fungus and the beetle tunnels clog the tree’s water conducting channels, sometimes killing it. “We are losing a lot of the urban forest canopy,” says David Richardson, an invasion biologist at Stellenbosch University (SU). But the damage extends far beyond Stellenbosch.

The world’s trees are already under siege from a plethora of invasive insect pests—the emerald ash borer, the hemlock woolly adelgid, the European spruce bark beetle, to mention a few. But as their names suggest, many specialize on a single group of trees. The polyphagous shot-hole borer is an all-purpose threat, infecting many woody plants. What’s more, all it takes is one female to start an infestation, because unfertilized eggs can develop into males, which the female can then mate with.

Native to China, Taiwan, and Vietnam, the chia seed–size insect readily stows away in wood pallets, shipping crates, and plants, and it has reached not just South Africa, but a half-dozen other countries around the world. Now, a study published this week in the Journal of Pest Science uses genetic tools to map its invasions and forecast future spread.

Piecing together the pest’s global trek has been challenging for researchers because several almost-identical shot-hole borers are loose on the planet. The polyphagous shot-hole borer ( Euwallacea fornicatus ), for instance, was in California for a decade before geneticists at the University of California (UC), Riverside sequenced its DNA and realized it was a new arrival. Previously, the pest had been misidentified as another less voracious invasive species, the tea shot-hole borer ( E. perbrevis ). “They are so damn small and similar,” says Jiri Hulcr, a forest entomologist at the University of Florida.

The new study has brought some clarity. An international team of researchers scoured the scientific literature, looking for reports of beetle sightings that used DNA analyses to distinguish the two species. The paper’s authors used the genetic data to build family trees and reconstruct how the insects’ likely spread, tracing at least six independent introductions of each species from their native ranges in Asia to other locales. “This study adds evidence showing how difficult it is to prevent the movement of nonnative insects,” says Angela Mech, a forest entomologist at the University of Maine.

The team also looked at the habitats and climate where the beetles are found now to project where they might show up in the future. The polyphagous shot-hole borer thrives in a broader range of temperatures and in drier climates than the tea shot-hole borer, and it seems to be a more aggressive colonizer. The team forecasts it could ultimately spread to parts of the Mediterranean region, the southeastern United States, Madagascar, nearly all of eastern Australia, and other places. The borer is “a perfect storm in progress,” says Esteban Ceriani Nakamurakare, a forest entomologist at Argentina’s National Scientific and Technical Research Council and co-author of the new paper. If left unchecked, it “will fundamentally alter ecosystem structure and landscape dynamics wherever it gains a foothold.”

Ornamental and fruit trees seem to be particularly vulnerable. In South America, for instance, where shot-hole borers have spread along 3500 kilometers of coast since the pest was first detected in Brazil in 2020, Ceriani Nakamurakare and his colleagues have found that landscaping favorites, imported species such as box elders and plane trees, are common hosts. It’s not clear why. Ceriani Nakamurakare notes that because such trees are planted in parks and gardens, they get plenty of water, which appears to make them more attractive to the beetles. Others think stressful urban environments may weaken trees and make them more vulnerable. Hosts “are almost never healthy trees,” Hulcr points out.

The damage is not limited to urban forests, however. After the species arrived in South Africa in 2012, it felled trees in Stellenbosch, Cape Town, and other urban areas. But “it has now established strongly in many of South Africa’s native forests and has started killing some species” there as well, says SU ecologist Francois Roets, who thinks the long-term consequences of these forest invasions will be “substantial.”

In the U.S., the beetle has moved from Southern California, where it was initially detected, to the San Jose area. That’s fueled fears it may one day reach the almond, pistachio, and other crops in California’s fertile Central Valley. “It is a serious threat,” says Shannon Lynch, a forest pathologist at UC Davis who has been testing which tree species in California might be vulnerable.

Various groups are studying how to combat the pest’s spread. Paul Rugman-Jones, a UC Riverside entomologist, is exploring whether parasitoid wasps that are natural enemies of this borer can be used to control it in California. In Australia, Murdoch University entomologist Wei Xu recently succeeded in rearing the fungus and beetle pair in the lab—a development he hopes will help him zero in on control methods such as chemical attractants or repellents.

Australia spent $40 million pruning or removing 5000 trees in Perth after shot-hole borers were detected in the city in 2021. The effort failed to get rid of them, but according to a study published in January the beetle has yet to reach the best habitats, where its population could skyrocket. “Prevention and early detection are critical,” Xu says.