The reason is that these fires do not only burn trees. In many boreal regions, flames can spread downward into thick layers of carbon rich soil beneath the forest floor. These soils, known as peat, contain partially decomposed plant material that has accumulated over hundreds or even thousands of years. Because the cold, wet conditions of the far north slow the breakdown of organic matter, these landscapes store enormous amounts of carbon underground.

Satellite Data May Miss Underground Peat Fires

According to the study, many widely used models that estimate wildfire carbon emissions fail to fully capture this underground burning. Most of these models rely heavily on satellite observations of visible flames and are based largely on fires that occur at lower latitudes. As a result, they may overlook slower, less visible fires that smolder deep within peat and organic soils.

“Many of the fires that matter most for the climate don’t look dramatic from space,” said study lead author Johan Eckdahl, a postdoctoral scholar in Berkeley’s Energy and Resources Group. “Peatlands and organic soils can smolder for weeks to years, releasing enormous amounts of ancient carbon.”

Reconstructing Emissions From Swedish Wildfires

The research, published in the journal Science Advances, analyzed 324 wildfires that burned across Sweden in 2018. Eckdahl and his colleagues combined detailed national forest records with direct field measurements to reconstruct the amount of carbon released by each fire.

Using these data, the team created a detailed map of wildfire emissions. Their analysis showed that local conditions such as climate, vegetation, and soil characteristics strongly influence how much carbon is stored in forests and how much is released during a wildfire.

Major Differences From Global Fire Models

When the researchers compared their reconstructed emissions with six widely used global wildfire models, they discovered significant discrepancies. In some places, the models overestimated carbon emissions. In other areas, especially where fires burned deep into soil layers, emissions were dramatically underestimated.

For instance, the models predicted higher emissions in the county of Gävleborg, where intense fires burned through dry forests and were clearly visible from satellites.

However, the situation was very different in neighboring Dalarna County. There, lower intensity fires burned quietly into thick layers of organic soil and were less noticeable from space. In that region, the models underestimated carbon emissions by as much as 14 times.

“Sweden is a very large country, but it’s quite small compared to Siberia and Canada,” Eckdahl said. “We may be severely underestimating the impact of the recent extreme fire seasons in these regions.”

Field Measurements Reveal Soil Carbon Loss

To measure how much carbon wildfires release from soil, the research team collected data from 50 locations affected by fires in 2018. Nineteen sites experienced high intensity fires, while 31 had lower intensity burns.

At each site, the researchers measured the thickness of the organic rich soil layer — which can vary from a few inches to many feet — and collected soil samples. By comparing carbon levels in burned soil with samples from nearby unburned forests, the team calculated how much carbon had been emitted.

“Once you’re out there, it’s a simple task — just dig some holes — but the hard part is getting to the sites,” Eckdahl said. “Sweden has a good network of forest roads, but in Siberia, I hear it’s a real trek, which is one reason why we’re severely missing measurements from that region.”

Expanding Research to Fire Prone U.S. Forests

Eckdahl is now working with colleagues at UC Berkeley and other institutions as part of the Western Fire & Forest Collaborative to apply similar research methods in forests across the Western United States.

Although forests in the western U.S. generally do not contain the same thick peat soils found in northern boreal regions, several other factors still influence wildfire emissions. These include regional climate patterns, the types of trees and vegetation present, and soil conditions. Eckdahl plans to study the role of soil microbes such as bacteria and fungi and how they contribute to forest recovery after wildfire.

“Forests in the Lower 48 and those far up north may look very different, but they share the common currency of carbon,” said Eckdahl. “By improving our understanding of how this element flows between the land and the atmosphere, we can better anticipate the impact of future fire regimes in a warming world and design smarter strategies to reduce climate risks on society.”

Lars Nieradzik of Lund University and Louise Rütting of the Brandenburg University of Technology are co authors of the paper.

The disease the AI targets is acromegaly, an uncommon condition that usually appears in middle age. It is caused by excessive production of growth hormone, which leads to enlarged hands and feet, changes in facial appearance, and abnormal growth of bones and internal organs. Because the disorder develops gradually over many years, it can be difficult to recognize early.

If untreated, acromegaly can lead to serious health problems and shorten life expectancy by about 10 years. “Because the condition progresses so slowly, and because it is a rare disease, it is not uncommon to take up to a decade for it to be diagnosed,” says Kobe University endocrinologist Hidenori Fukuoka. He adds, “With the progress of AI tools, there have been attempts to use photographs for early detection, but they have not been adopted in clinical practice.”

A Privacy Focused AI Approach Using Hand Images

When the research team reviewed existing AI studies, they found that many systems depend on facial photos to identify disease. However, facial recognition can raise privacy concerns for patients. To address this issue, the scientists chose a different strategy.

Yuka Ohmachi, a graduate student at Kobe University, explains, “Trying to address this concern, we decided to focus on the hands, a body part we routinely examine alongside the face in clinical practice for diagnostic purposes, particularly because acromegaly often manifests changes in the hands.”

To strengthen privacy protections, the researchers limited their images to the back of the hand and a clenched fist. They intentionally avoided palm images because palm line patterns are highly individual and could reveal identity. This careful approach helped recruit a large number of participants. In total, 725 patients from 15 medical institutions across Japan contributed more than 11,000 images used to train and test the AI model.

AI Outperforms Experienced Specialists

The team reported their results in the Journal of Clinical Endocrinology & Metabolism. Their AI model demonstrated very high sensitivity and specificity when identifying acromegaly from the hand images. In direct comparisons, the system even performed better than experienced endocrinologists who evaluated the same photographs.

“Frankly, I was surprised that the diagnostic accuracy reached such a high level using only photographs of the back of the hand and the clenched fist. What struck me as particularly significant was achieving this level of performance without facial features, which makes this approach a great deal more practical for disease screening,” says Ohmachi.

Expanding Medical AI to Other Conditions

The researchers now hope to adapt their system to detect additional medical conditions that produce visible changes in the hands. Possible targets include rheumatoid arthritis, anemia and finger clubbing. Ohmachi says, “This result could be the entry point for expanding the potential of medical AI.”

Supporting Doctors and Improving Access to Care

In real clinical settings, doctors rely on far more than hand images when diagnosing patients. Medical history, lab tests and physical exams all play important roles. The Kobe University researchers see their AI tool as something that could assist physicians rather than replace them. In their study, they describe the technology as a way to “complement clinical expertise, reduce diagnostic oversight and enable earlier intervention.”

Study lead Fukuoka says: “We believe that, by further developing this technology, it could lead to creating a medical infrastructure during comprehensive health check-ups to connect suspected cases of hand-related disorders to specialists. Furthermore, it could support non-specialist physicians in regional healthcare settings, thus contributing to a reduction of healthcare disparities there.”

The research received funding from the Hyogo Foundation for Science Technology. The project also involved collaborators from Fukuoka University, Hyogo Medical University, Nagoya University, Hiroshima University, Toranomon Hospital, Nippon Medical School, Kagoshima University, Tottori University, Yamagata University, Okayama University, Hyogo Prefectural Kakogawa Medical Center, Hokkaido University, International University of Health and Welfare, Moriyama Memorial Hospital and Konan Women’s University.

The findings appear in the journal Global Ecology and Conservation. The research was carried out by primatologist Professor Anna Nekaris OBE of Anglia Ruskin University along with collaborators from the conservation group Plumploris e.V. and the University of Western Australia. Their work examined the fate of Bengal slow lorises (Nycticebus bengalensis) that were released in Bangladesh.

Slow Lorises and the Illegal Pet Trade

Slow lorises are known for their large eyes and gentle looking faces, features that have unfortunately made them popular in the illegal wildlife trade. Because of this demand, they rank among the most heavily trafficked primates in the world.

All slow loris species are listed by the International Union for Conservation of Nature as Critically Endangered, Endangered, or Vulnerable. Their threatened status means they are frequently rescued and later released as part of conservation efforts aimed at rebuilding wild populations.

Tracking Bengal Slow Lorises After Release

Despite these good intentions, the new research shows that release programs can sometimes end tragically. Scientists fitted nine Bengal slow lorises with radio collars and followed their movements after releasing them into a national park in northeastern Bangladesh. The park has been used for previous wildlife releases.

The results were stark. Only two of the nine animals survived after returning to the forest. Three died within just 10 days of release, and four more died within six months. Researchers recovered four of the seven bodies, and all showed evidence that they had been killed by other slow lorises.

Territorial Conflicts and Venomous Bites

Slow lorises are extremely territorial animals. They are also the only venomous primates in the world, using specialized teeth to deliver a toxic bite. The animals recovered during the study had obvious bite wounds on the head, face, and digits, indicating that deadly territorial encounters were responsible for their deaths.

The research also revealed that animals kept in captivity for longer periods tended to survive for fewer days after release. In addition, the released lorises moved around more and appeared more alert than wild Bengal slow lorises normally do.

The two animals that survived traveled across larger areas than those that died. This pattern suggests that survival depended on leaving established territories and avoiding confrontations with resident lorises.

Rethinking Wildlife Rescue and Release

Large and charismatic animals such as big cats often receive intensive monitoring after they are released. In contrast, many smaller species are not closely tracked, meaning the outcomes of their releases frequently remain unknown.

The researchers stress that successful wildlife releases require careful planning. Evaluating the suitability of the release site and the condition of each animal is essential. Detailed habitat assessments, long term monitoring, and rehabilitation guidelines tailored to each species can improve the chances of success.

Senior author Anna Nekaris OBE, Professor of Ecology, Conservation and Environment at Anglia Ruskin University in Cambridge, England, and head of the Little Fireface Project, said: “It’s assumed that returning confiscated or rescued animals to the wild is always a positive conservation story. Our research shows that for highly territorial species like slow lorises, releasing them into areas that are already densely populated can be a death trap.

“Many rescued endangered species are often released because the public expects it, but for animals such as the Bengal slow loris, this is not always the best course of action. Without fully understanding the animal’s behaviour, its time spent in captivity and the density of resident populations at the release site, reintroductions may do more harm than good.”

Lead author Hassan Al-Razi, the team leader of Plumploris e.V. Bangladesh, said: “Rescue and release have become an increasingly common practice in Bangladesh. Many wild animals, including slow lorises, are rescued and subsequently released back into the wild.

“However, in many cases, these releases are conducted inappropriately. For forest-dwelling species, release sites are often selected based on logistical convenience rather than ecological suitability. As a result, certain forests have effectively become dumping grounds for rescued animals and are no longer appropriate release sites.

“Although our research has focused on the Bengal slow loris and demonstrated the consequences of such practices, we believe similar patterns likely affect many other species.”

This approach has produced major breakthroughs. However, it has not fully explained a central feature of human thinking: how all these separate systems come together to form a single, unified mind.

Researchers at the University of Notre Dame set out to address that question. Using advanced neuroimaging, they examined how the brain is organized overall and how that organization gives rise to intelligence.

“Neuroscience has been very successful at explaining what particular networks do, but much less successful at explaining how a single, coherent mind emerges from their interaction,” said Aron Barbey, the Andrew J. McKenna Family Professor of Psychology in Notre Dame’s Department of Psychology.

General Intelligence and Connected Cognitive Abilities

Psychologists have long observed that skills like attention, memory, perception, and language tend to be linked. People who perform well in one area often perform well in others. This pattern is known as “general intelligence.” It influences how effectively individuals learn, solve problems, and adapt across academic, professional, social, and health settings.

For more than a century, this pattern has suggested that human cognition is unified at a deep level. What scientists have lacked is a clear explanation for why that unity exists.

“The problem of intelligence is not one of functional localization,” said Barbey, who also directs the Notre Dame Human Neuroimaging Center and the Decision Neuroscience Laboratory. “Contemporary research often asks where general intelligence originates in the brain — focusing primarily on a specific network of regions within the frontal and parietal cortex. But the more fundamental question is how intelligence emerges from the principles that govern global brain function — how distributed networks communicate and collectively process information.”

To explore this broader perspective, Barbey and his team, including lead author and Notre Dame graduate student Ramsey Wilcox, tested a framework known as the Network Neuroscience Theory. Their findings were published in Nature Communications.

The Network Neuroscience Theory Explained

According to the researchers, general intelligence is not a specific ability or mental strategy. Instead, it reflects a pattern in which many cognitive skills are positively related. They propose that this pattern stems from how efficiently the brain’s networks are structured and how well they work together.

To evaluate this idea, the team analyzed brain imaging and cognitive performance data from 831 adults in the Human Connectome Project. They also examined an independent group of 145 adults in the INSIGHT Study, funded by the Intelligence Advanced Research Projects Activity’s SHARP program. By combining measures of brain structure and brain function, the researchers created a detailed picture of large-scale brain organization.

Rather than tying intelligence to a single brain region or function, the Network Neuroscience Theory views it as a property of the brain as a whole. Intelligence, in this framework, depends on how effectively networks coordinate and reorganize themselves to handle different challenges.

Barbey and Wilcox describe this as a major shift in perspective.

“We found evidence for system-wide coordination in the brain that is both robust and adaptable,” Wilcox said. “This coordination does not carry out cognition itself, but determines the range of cognitive operations the system can support.”

“Within this framework, the brain is modeled as a network whose behavior is constrained by global properties such as efficiency, flexibility and integration,” Wilcox said. “These properties are not tied to individual tasks or brain networks, but are characteristics of the system as a whole, shaping every cognitive operation without being reducible to any one of them.”

“Once the question shifts from where intelligence is to how the system is organized,” Wilcox noted, “the empirical targets change.”

Intelligence as Whole Brain Coordination

The findings supported four main predictions of the Network Neuroscience Theory.

First, intelligence does not reside in a single network. It arises from processing distributed across many networks. The brain must divide tasks among specialized systems and combine their outputs when necessary.

Second, successful coordination requires strong integration and long-distance communication. Barbey described “a large and complex system of connections that serve as ‘shortcuts’ linking distant brain regions and integrating information across the networks.” These connections allow far apart areas of the brain to exchange information efficiently, supporting unified processing.

Third, integration depends on regulatory regions that guide how information flows. These hubs help orchestrate activity across networks, selecting the right systems for the job. Whether someone is interpreting subtle clues, learning a new skill, or deciding between careful analysis and quick intuition, these regulatory areas help manage the process.

Finally, general intelligence depends on balancing local specialization with global integration. The brain performs best when tightly connected local clusters operate efficiently while still maintaining short communication paths to distant regions. This balance supports flexible and effective problem solving.

Across both groups studied, differences in general intelligence consistently matched these large-scale organizational features. No single brain area or traditional “intelligence network” explained the results.

“General intelligence becomes visible when cognition is coordinated,” Barbey noted, “when many processes must work together under system-level constraints.”

Implications for Artificial Intelligence and Brain Development

The implications extend beyond understanding human intelligence. By focusing on large-scale brain organization, the findings offer insight into why the mind functions as a unified system in the first place.

This perspective may also explain why intelligence tends to increase during childhood, decline with aging, and be especially vulnerable to widespread brain injury. In each situation, what changes most is large-scale coordination rather than isolated functions.

The results also contribute to debates about artificial intelligence. If human intelligence depends on system-level organization rather than a single general-purpose mechanism, then building artificial general intelligence may require more than simply scaling up specialized tools.

“This research can push us into thinking about how to use design characteristics of the human brain to motivate advances in human-centered, biologically inspired artificial intelligence,” Barbey said.

“Many AI systems can perform specific tasks very well, but they still struggle to apply what they know across different situations.” Barbey said. “Human intelligence is defined by this flexibility — and it reflects the unique organization of the human brain.”

The research was conducted with co-authors Babak Hemmatian and Lav Varshney of Stony Brook University.