Researchers at the 天美影视传媒 and partners at Carnegie Mellon University are collaborating with the U.S. Department of Energy’s on a new project to help scientists use artificial intelligence to better understand the universe.

Modern astronomy is producing more data than ever before. Powerful telescopes 鈥� like the 鈥� and other instruments around the world are collecting detailed information about billions of stars, galaxies and other cosmic objects. These observations help researchers investigate some of the biggest mysteries in science, including the nature of dark matter and dark energy, how the universe evolved over time and what it is made of.

But there is a challenge: Much of the data from those projects is stored in different formats, housed at different institutions and difficult to combine. As a result, scientists often spend significant time preparing data before they can begin analyzing it.

鈥淭he scientific opportunities and the data analysis challenges are incredible,鈥� said , head of physics at Carnegie Mellon University.

The new effort, led by Mandelbaum and funded by the U.S. Department of Energy鈥檚 , will create a shared data service to allow researchers to seamlessly access and combine information from multiple astronomy experiments. Rather than moving massive datasets from one location to another, the system will allow the data to remain where it is stored while making it available through a central platform.

鈥淎 new generation of telescopes and surveys will each change the way we understand our universe,鈥� said co-investigator , a UW professor of astronomy and director of the eScience Institute. 鈥淏ut it is when we bring these data together to look at the universe from a unified perspective that these discoveries will be truly transformative.鈥�

The data infrastructure developed as part of this project will be available to the astronomical community at the SLAC-hosted Rubin Observatory鈥檚 U.S. Data Facility and via the American Science Cloud, which integrates the nation鈥檚 most advanced high-performance computing systems, scientific facilities, data resources and production capabilities into a single, coordinated AI-driven system. The project extends the UW鈥檚 investments in Rubin Observatory and the UW , which are supported by Charles and Lisa Simonyi and led by and .听

The infrastructure also will support a growing area of AI known as foundation models. These AI systems are trained on large and diverse datasets, enabling them to recognize patterns and connections that might otherwise go unnoticed.

In astronomy, foundation models could help researchers analyze many different types of observations at once, including images, measurements of light from distant objects and records of how those objects change over time. By bringing these data sources together, scientists hope to uncover new insights about the universe more quickly and efficiently.

Ultimately, the team hopes to transform the vast collections of astronomical data being gathered today into a long-lasting scientific resource, helping researchers answer some of humanity’s most fundamental questions about the origin, evolution and makeup of the universe.

Other UW co-investigators include , a research scientist and engineer in astronomy. Other co-investigators include , director of engineering for the LINCC project and , senior staff scientist at SLAC and at Stanford鈥檚/SLAC鈥檚 Kavli Institute for Particle Astrophysics and Cosmology.

Additional modeling will be provided by Francois Lanusse of the French National Centre for Scientific Research. Their work is an extension of the Schmidt Sciences-supported LINCC Frameworks program, a partnership led jointly by CMU and the UW.

This story was adapted from a press release by .

Approximately 90% of the U.S. population . A significant portion of the water supplying those systems comes from forested lands, which means that policies impacting forests also impact our water access.

In 2001, the Clinton administration passed the , blocking 60 million acres of national forest land from development to limit industrial timber harvest and preserve forest ecosystems. Although popular with the public, the roadless rule drew immediate criticism from timber and related industries. Last summer, the federal government announced plans to rescind it.

A new study from the 天美影视传媒 and Conservation Science Partners, , highlights the potential consequences of losing those protections by mapping how the roadless rule protects rivers.

鈥淭he roadless rule supports the drinking water supply for 25 million Americans and offers critical protection of wildlife habitat and recreational assets. In short, rivers in roadless areas are essential for both people and nature,鈥� said lead author , a UW professor of aquatic and fishery science.

To show how forests impact freshwater, the researchers looked at nearly 110,000 square miles of national forest representing 2,488 officially designated 鈥渞oadless areas.鈥� They cross-referenced the roadless areas map with a recent study assessing river protections nationwide to see where rivers and roadless areas overlap, and therefore which rivers were vulnerable to losing protection.

The researchers found that more than 80,000 miles of rivers in the continental U.S. receive some protection from the roadless rule. Of those protected segments, nearly 62,000 miles of river are protected by only the roadless rule. That water reaches 25 million people across the country, often at downstream distances far from roadless areas.

Research shows that forested lands provide higher quality water because soil microbes and plant roots filter contaminants before water arrives at treatment facilities. Cleaner water requires less processing, reducing potential treatment costs for public utilities. Some water utilities are investing in watershed protection as a way to save money and limit chemical use as demand for water rises.

鈥淔orest cover is well recognized for generating economic benefits by avoiding the large capital costs of water treatment plants needed to ensure clean, safe drinking water for people,鈥� said Olden.

Roadless areas are also vital strongholds for sensitive aquatic species, Olden added. At-risk species such as the use protected habitat for spawning and raising young. Hunters and anglers also value roadless areas because they support such productive fish and wildlife habitat and offer unparalleled opportunities for outdoor recreation.

After the U.S. Agriculture Secretary Brooke Rollins announced the plan to rescind the roadless rule last year, more than half a million people submitted comments during the public comment period, which ended in September. According to the Center for Western Priorities, more than expressed opposition to the plan.

Earlier this year, Republican lawmakers attempted to by attaching it to the Wildfire Prevention Act, which supports prescribed burns and forest thinning to fight megafires. Their to justify clawing back protections claimed that rescinding the roadless rule would allow for better forest management, but . Research shows that roadbuilding in formerly roadless areas is more likely to increase fire risk.

鈥淭o be clear, the rule does not block any management action that supports forest health, wildfire mitigation or recreation,鈥� Olden said. 鈥淚n fact, energy projects, transmission lines and mining development remain permitted within roadless areas.鈥�

Roadbuilding and logging can cause sediment build up in lakes and rivers, which must be filtered out. Chemicals from construction can also end up in the water supply. Reductions in forest cover resulting from rescinding the roadless rule may compromise water quality in the U.S., among other negative consequences for animals and ecosystems.

The U.S. Forest Service says it is reviewing public comments and a proposed rule and draft statement of environmental impact this year.

鈥淎ny decision to rescind or downgrade the roadless rule that may put forested lands at risk requires careful consideration of the numerous benefits they offer to people and nature,鈥� said Olden. 鈥淥ur study offers data to inform such decisions.鈥�

This study was funded by the 天美影视传媒 and from a contract from American Rivers to Conservation Science Partners.

For more information, contact Olden at olden@uw.edu.

Arguably the most important crop on Earth, rice has been cultivated for roughly 10,000 years. It鈥檚 a staple food for more than half the global population, with about 90% cultivated and consumed in Asia.

But a new 天美影视传媒 study, recently published in , warns that this essential crop is in danger. Due to rising temperatures driven by climate change, projections show that Asia鈥檚 major rice-producing regions may soon pass the thermal limits that have remained consistent throughout the crop鈥檚 history.

Using satellite maps, agricultural records, archaeological data and climate projections, researchers found that domesticated Asian rice has never thrived where the mean annual temperature exceeds 28 degrees Celsius 鈥� 82 degrees Fahrenheit 鈥� or where the warm-season maximum temperature exceeds 33 C, or 91 F.听

Climate projections estimate that, by the end of this century, the land area exceeding these temperature limits could expand by 10 to 30 times in Asia鈥檚 major rice-producing nations. This would create unparalleled challenges in a region where more than a billion people rely on rice cultivation for their livelihoods. While rice breeding programs offer some hope, the researchers found that even the major rice subspecies won鈥檛 thrive in the projected temperatures.

UW News spoke with , an archaeologist, UW associate professor of anthropology and co-author of the study, about the research and what it means for the future.

How did you become interested in this topic?

JDG: A lot of my research focuses on how climatic events have shaped people鈥檚 ability to farm or grow crops in environments around the world. One of the places I previously worked was the Tibetan Plateau, and some of my early research documented that there was a cooling event around 4,000 years ago that halted Tibetans’ ability to grow two critical crops, which were . They shifted to wheat and barley after that, and they’ve been growing them ever since.

I became interested in applying the same research to understand how our current unprecedented moment of climate change may impact crop distribution. I work in Asia, particularly China, and I鈥檝e worked on rice for most of my career. Our team found that the same cooling event that affected Tibet had a major impact on rice genetics. In fact, it led to the development of cold-adapted temperate rice, which is the same type of rice that people rely on for subsistence today in Japan, northern China and Korea. It’s the short-grained sticky rice that’s cold tolerant, as opposed to the original form of rice that was a semi-subtropical cultivar.听

We then became interested in the types of challenges rice will face moving forward. We pulled records of everywhere that rice has ever been cultivated in Asia throughout human history 鈥� and all the climatic conditions under which it鈥檚 been cultivated 鈥� and compared that to the types of situations that we’ll face under global warming today.

What did you find as you started looking toward the future?

JDG: Basically, we found that large areas that are major rice producers are going to face rising temperatures that are unprecedented in the history of rice cultivation. Over the course of the past 10,000 years and its domestication, rice has been adapted to cooler conditions and not to warmer conditions. We used a wide variety of forward-looking climate projection models, and all of those models seem to converge on the same point: Large parts of primary rice growing regions around the world are expected to surpass the known temperature limit where these crops can be cultivated.

I use the word unprecedented, and I don鈥檛 use it lightly. Another term I could apply to this would be no analog. There is no known situation over the course of rice鈥檚 cultivation where rice grew in regions which had such high mean annual temperatures. This crop has simply never experienced this before, so there is no data for how it will react. But we do have thousands of years of data saying that to date, it hasn鈥檛 been cultivated in temperatures like these. In fact, there are only two parts of the world today that have mean annual temperatures that are similar to those that we expect will occur in major rice producing regions of the world: the Sahara desert and parts of the Arabian peninsula. There is a reason these areas are largely desert. Most people think about water but temperature is a critical reason, too.听

I can鈥檛 underscore enough how writing this study felt. Millions of people live in this region and depend on temperatures as we have known them to continue to live and farm there. This is beyond devastating, and we are beginning to see the impacts of processes like this already.听

What are some recent examples of rising temperatures causing problems with crop cultivation?

JDG: In 2023, of all non-basmati rice. Sona masoori rice, or the type of rice grown across most of low altitude South Asia, had such great losses due massive heat waves that the government stopped all exports. People were panic buying rice that year, even in the U.S.

With climate change, the temperature is not increasing in a completely linear fashion. But the average is increasing over time. As the average increases, there’s a higher probability of these extreme events occurring. It’s already happening in our lifetime, and this study is solid evidence for why addressing climate change should be an absolute top priority for all of us. Over a billion people on the planet are rice cultivators, and that is their primary means of livelihood. For a fourth of the world鈥檚 population, rice is a main staple in the diet.

We could have written this paper for so many crops, including wheat, corn and others. What we’re dealing with here is that plant photosynthesis just doesn’t function well above those temperature limits. We鈥檙e running into fundamental limits of photosynthetic biochemical process plants, which are temperature limited. There are not that many types of plants that can sustain life under conditions like that, and certainly they are not our major economic plants.听

We live in an era where we have all experienced climate change. In 2026, most of the globe has experienced an extreme heat event. We understand how difficult this is for us as mammals to live through, and yet some of us have the privilege of escaping indoors or even to air conditioning. Plants, on the other hand, cannot move to escape the heat. I鈥檓 sure many here in Seattle remember the . The plants in our yard are still recovering from those short few days. That event caused billions of dollars in losses in the agricultural sector in the Pacific Northwest. Some berry and soft fruit farmers experienced nearly 100% crop losses. By midcentury, conditions like this could occur every five to 10 years and could have a huge impact on all plants, including the ones we rely on for food.听

This also critically highlights why we need to expand rather than contract our dietary breadth. Sadly, the opposite of this is happening due to industrial farming practices. Humanity is relying on an increasingly narrow range of species. We are essentially putting all our eggs into one or just a few baskets when we need crop diversity.听

What do climate projections take into account, and what can be done to change the path we鈥檙e currently on?

JDG: For this study, we used multiple climate projections based on what different countries鈥� carbon commitments will be moving forward. What we found is that even for climate scenarios where there is a strong global commitment to sustainability-focused growth, international cooperation, and an eventual transition to net-zero emissions, major rice growing regions are still impacted (SSP 1-2.6 on our maps). These impacts expand dramatically with other climate scenarios which assume less concerted action, and sadly this is consistent with where we are headed today which is probably somewhere between SSP 3- 7.0.

Our actions can change the course of what scenario we might be looking at with these maps. We have the technology to move forward with more climate-friendly solutions and many countries around the world are trying to take the lead while we lag behind. For instance, China, where I work, has made massive investments in public transit and railways. Nearly everybody in China drives an electric vehicle. I did not see a single gas-powered car last time I was there. We could do that here at home, and we’re not. Our politicians are making active choices to halt this type of action and at the same time we have the highest per capita emissions in the world. We could and should do much more. It can feel hopeless, particularly for those of us who live in a country where action from our politicians has been in a decades-long gridlock for meaningful change. But we shouldn鈥檛 stop lobbying for change.听

We can also look at what steps we can make in our own lives to lower emissions. It鈥檚 worth noting that the vast part of emissions come from the top 1%, and that portion of the population can really drive meaningful personal action. Simple steps for those that have the means and access can be switching to solar if you own a home or taking public transit where you can.Asking honestly, how am I contributing to this and what can I do differently can always help. For me, the single largest part of my personal emissions was flying and I ceased a large part of all my noncritical travel for that reason. Emissions from short-distance flights and private jets contribute hugely to this issue.

No country will be immune to a crisis of this magnitude, and it is an issue that deserves global attention. While the impacts may initially appear distant to readers from the U.S., our interconnected economies mean that we will also be affected through global trade systems and shifting agricultural dynamics. The same environmental and socioeconomic processes influencing rice production in Asia will have comparable implications for crops such as corn in lower-latitude regions of the U.S. and even here in the Pacific Northwest, as we saw with the last heat dome. Addressing these challenges requires recognizing their global scope and engaging proactively with the evidence before us.听

Other co-authors of the study were of the University of Florida and and of New York University.听

The study was funded in part by grants from the Zegar Family Foundation and the NSF Plant Genome Research Program.

For more information, contact d’Alpoim Guedes jguedes@uw.edu.

From a mountaintop in Chile, under clear dark skies, the Simonyi Survey Telescope at the NSF鈥揇OE Vera C. Rubin Observatory has officially begun the Legacy Survey of Space and Time (). The 10-year survey will create the most comprehensive, cinematic record of the universe in history. Over the next decade, Rubin will observe the entire southern sky every few nights to create an ultra-wide, ultra-high-definition time-lapse record of our universe.听

鈥淭he decision to officially begin the LSST was made after a period of system optimization and a careful operational review of technical readiness, data system performance and scientific validation,鈥� said , a 天美影视传媒 professor of astronomy and head of LSST. 鈥�The 天美影视传媒 Rubin team played a central role in optimizing the observatory and helping prepare it for the start of full survey operations.鈥�

The Simonyi Survey Telescope鈥檚 unique design combines enormous light-collecting power, the ability to move rapidly across the sky and a wide field of view. The attached 3,200-megapixel camera 鈥� the largest digital camera in the world 鈥� is now capturing a new, detailed image approximately every 40 seconds. Using a telescope with this speed and sensitivity, Rubin is capable of catching faint objects and fleeting events with reliability and consistency every night.

Over the next decade, Rubin will illuminate a treasure trove of discoveries: pulsating stars, supernova explosions, the fossil record of galaxies, clues to the mysteries of dark energy and dark matter, and entirely new phenomena we鈥檝e never seen before. Some cosmic processes unfold slowly, unpredictably or incredibly rarely, which is why a 10-year survey is essential. By returning to each point in the sky about 800 times over a decade, Rubin data will provide the scientific community with deep, time-rich views needed to uncover subtle events, capture moving objects and study the accelerating expansion of the universe.

This milestone follows the Rubin First Look event that took place in June 2025, which was followed by final commissioning work, an operational readiness review and the beginning of the alert stream.

Each night, Rubin collects approximately 10 terabytes of data and produces as many as seven million alerts of changes in the night sky. These alerts stream to : automated systems that sort and classify these changes so scientists can act quickly. UW researchers led by , research associate professor of astronomy, developed the alert pipeline.

鈥淎stronomers have already used Rubin’s public alerts to discover and follow up hundreds of transient phenomena during the early optimization period,鈥� Bellm said. 鈥淲e can expect many more exciting discoveries with the start of the full survey.鈥�

Not only is Rubin helping to unlock the mysteries of the distant universe, it is also the most powerful solar system discovery machine ever built. By taking about a thousand images every night, Rubin is compiling a detailed census of our solar system, including millions of asteroids and comets. In just a month and a half, during early optimization surveys, Rubin discovered over 11,000 never-before-seen asteroids, including 33 near-Earth objects and 380 trans-Neptunian objects.

When the LSST is complete, the final dataset will contain billions of objects with trillions of measurements, all accessible through regular data releases. This is the first time so much astronomical data will be available to so many people, opening the door to new kinds of discovery by both scientists and the public. Rubin invites anyone in the world to engage with its data and explore the dynamic universe in ways never before possible.

鈥淚t is amazing and humbling to be here at this time and place as we start the Legacy Survey of Space and Time, after more than two decades of incredible work by our dedicated team,鈥� said Bob Blum, director of Rubin Observatory at NSF NOIRLab. 鈥淩ubin Observatory is for everyone; the LSST will change how we do astronomy and astrophysics, allowing researchers anywhere to participate in cutting-edge science.鈥�

Visit to follow the status of the LSST in real time.

Rubin Observatory is jointly operated by NSF NOIRLab and SLAC. Observatory operations听are funded by the U.S. National Science Foundation and the U.S. Department of Energy鈥檚 Office of Science.

For more information, contact Ivezi膰 at ivezic@astro.washington.edu.

This story was adapted from a press release by .

New images of cigar-shaped M82 galaxy capture millions of stars

The Messier 82 galaxy, known as M82 or the Cigar galaxy, has long fascinated researchers with its astronomical rate of star formation 鈥� approximately 10 times faster than the Milky Way. Researchers have pored over grainy, low-resolution, images taken by previous generations of telescopes, which weren鈥檛 powerful enough to see through the thick cloud of dust surrounding the galaxy. The , however, can pierce straight through with extremely sharp vision. That enabled a team of astronomers from multiple institutions, including NASA and the UW, to capture new high-resolution images. Posted June 23, the images include more than 16.5 million individual stars and provide the clearest look yet at M82鈥檚 , the flattened central hub that contains most of the galaxy鈥檚 stellar mass. That could help scientists understand how M82 formed and for how long it has been producing stars so prodigiously.

For more information, contact team member a UW research professor of astronomy, at benw1@uw.edu.

All images are included in NASA鈥檚

New study maps pollution disparities by state and sector across almost 20 years

Air quality in the United States has improved markedly since the landmark Clean Air Act passed in 1970. However, the gains have not been equally shared: Today, communities of color and low-income communities are exposed to disproportionately more air pollution than the overall population. In in Science Advances, UW researchers created the first comprehensive map cataloging how air quality inequity has changed per state and economic sector from 2002 to 2019. The study confirmed that, despite improvements in overall air quality, pollution tends to be concentrated in Black, Hispanic and low-income communities. The findings include specific state-level opportunities for improvement across 11 sectors 鈥� for example, disparities in construction-related emissions in Florida increased significantly during the study period. The findings and resulting database could help policymakers across the country prioritize environmental justice projects.

For more information, contact senior author , UW professor of civil and environmental engineering at jdmarsh@uw.edu.

The other UW co-authors are , , and . A full list of co-authors is .听

Researchers observe ultrafast chemistry happening in real time

Molecules are not static. Instead, they are having little dance parties 鈥� their atoms wiggle and twist around in space. Occasionally, upon receiving a burst of energy, the bonds holding atoms together in a molecule can break and reform with the atoms in a different configuration. While the number of atoms stays the same, the orientation of these atoms determines a molecule’s chemical properties 鈥� an important part of its identity. In , a UW-led team witnessed firsthand, and for the first time, a molecule turning into its “alter ego.” The researchers observed a hydrogen atom, also known as a proton, jump to a new position by bonding to a different atom in the same molecule. This process, which happens within a few millionths of billionths of a second, is important for various fundamental processes, including photosynthesis, and when DNA acquires mutations. To understand why, and how, this happens so fast, the researchers developed a new tool that probes molecular structure on an ultrafast timescale. They were able to use this technology to detect how the molecule’s wiggles allowed the proton transfer to happen. These findings will help researchers test existing theories about these ultrafast chemical dynamics and develop new molecules for clean energy processes.

For more information, contact senior author , UW professor of chemistry, at mkhalil@uw.edu.听听听

Co-authors , and completed this work while at the UW. Funding information is .

Random events leave lasting signature on the atmospheric methane record, new study shows

Methane is a powerful greenhouse gas with a complicated life cycle. It鈥檚 released into the atmosphere by both natural and industrial processes, and there are multiple pathways by which it鈥檚 broken down. Recently, atmospheric methane levels have reached record highs but the rate of accumulation has been somewhat inconsistent over time. To understand why, researchers are looking at climate records preceding the industrial era, via ice cores. These deep cylinders of glacial ice document slow swings in atmospheric methane levels spanning decades, or even centuries. This pattern is typically associated with gradual climate change, but in , UW researchers show that it doesn鈥檛 have to be. Instead, they reveal that short-term, random events, such as fires or changes in wetlands, can spark gradual shifts. Not only does this clarify the historical record, but it also adds nuance to modern trends.

For more information, contact senior author , UW doctoral student of atmospheric and climate science at emei@uw.edu.

The other UW co-authors are and . A full list of co-authors is .

Data spanning nearly half a century shows that endangered southern resident killer whales are spending less time in inland waters, whereas their larger cousins, Bigg鈥檚 killer whales, are increasingly present in Puget Sound.

The National Oceanic and Atmospheric Administration southern resident killer whales as endangered in 2005 after rapid population decline in the late 1990s. Now, , split into three pods: J, K and L. Bigg鈥檚 鈥� sometimes referred to as transients 鈥� are more common, but difficult to count because they travel in smaller groups over wider ranges.

Looking at data from 鈥檚 Sightings Archive between 1978 and 2022, 天美影视传媒 researchers modeled migratory trends based on observations from researchers, recreational boaters and whale watchers. They found that K and L pods are visiting Puget Sound less often, but the J pod remains well represented. The data on Bigg鈥檚 corroborates recent results showing a steady increase in inland waters.

The results in PLOS One.

鈥淲e do see increasing transient presence over time, but we don鈥檛 see a definitive decline or overall increase for the southern residents. Their presence here is much more variable,鈥� said lead author , a UW postdoctoral researcher of marine and environmental affairs.

Key behavioral and subtle physical differences . The southern residents eat salmon, while Bigg鈥檚 prey on seals, porpoises and other marine mammals. Seals and sea lions rebounded in Washington after the Mammal Protection Act, which may have drawn Bigg鈥檚 killer whales to inland waters, but that doesn鈥檛 explain the changing distribution of southern residents.

Because southern residents are organized into tight matriarchal societies led by female elders, researchers believe that social cues may play an important role.

鈥淒oes J pod know something that K and L don鈥檛? Or vice versa? We like to think about which pods have really old grandmas left and who’s teaching them where to go,鈥� said co-author , a marine mammal specialist at the National Oceanic and Atmospheric Administration (NOAA), West Coast Regional office.

Policies to protect southern residents typically apply to all pods. With K and L spending more time in coastal waters, NOAA for southern residents in 2021 to include 16,000 square miles of marine waters between the U.S. and Canada border and Point Sur, California.

Measures like the , which encourages commercial ships to slow down where whales are present, aim to mitigate the impact of noise. Boats are also of the southern residents.

Indications of changing habitat have prompted some to question the need for such regulations in Puget Sound, but these results underscore their continued importance.

鈥淓ven though we鈥檙e seeing less of K and L pods, we still have to think about how our actions impact J pod. They鈥檙e still hanging around,鈥� Koehn said.

The study also notes that southern residents and Bigg鈥檚 are sharing habitat more often, though it isn鈥檛 clear whether they mingle or avoid each other. This raises questions about their relationship and underscores the importance of accounting for both in management decisions.

鈥淗aving more transients around could be good for the southern residents, because they eat marine mammals that also eat salmon,鈥� Rand said.

But if the southern residents avoid the transients, their increased presence could be disruptive. Researchers are actively studying threats to the southern residents 鈥� including prey availability 鈥� to support the imperiled population.

This analysis wouldn鈥檛 have been possible without consistent contributions from citizen scientists. People who report whale sightings using apps like Whale Alert help researchers provide data to policymakers, which can be consequential for the whales.

鈥淭his study quantitatively shows things that people have been suspecting,鈥� Rand said. 鈥淭here are more transients here in Washington, but the southern resident鈥檚 story is a bit more complicated.鈥�

Additional co-authors include of the Whale Museum and of NOAA Fisheries Northwest Fisheries Science Center.

This study was funded by Washington Sea Grant, NOAA Fisheries West Coast and the Puget Sound Partnership.听

For more information, contact Rand at zrand@uw.edu.

At the end of every presidential term, the preserves that administration鈥檚 web presence as a vast trove of documents and webpages. The archive began in 2008, with George W. Bush鈥檚 second term, and runs up to 2024, collecting images, text, graphs, redacted pages and other media. So while it contains important public information, finding that information in the glut can prove difficult.

A 天美影视传媒-led research team created , an efficient search system for PDFs from the End of Term Web Archive. Users can look up exact keywords, like 鈥淔AFSA,鈥� or use a semantic search, which finds documents on a topic even if the exact search terms don鈥檛 appear on the page. A visual search option lets them query for qualities like 鈥渞edacted documents,鈥� “aerial photographs鈥� or 鈥減ie charts.鈥� The system can currently search the 10 million PDFs hosted online during Donald Trump鈥檚 first term; the team plans to expand it to the whole archive.听

Because researchers used highly efficient artificial intelligence models to read the documents, processing all the PDFs costs less than $1,500, or about $1 per 47,000 pages. By comparison, Google might charge consumers .听

The team will July 5 at the Annual Meeting of the Association for Computational Linguistics in San Diego.听

鈥淭he End of Term Web Archive is immensely important to historians, journalists and the American public,鈥� said senior author , a UW assistant professor in the Information School. 鈥淏ut many of these digital archives are getting so big 鈥� just announced its trillionth page archived 鈥� that finding information is the real challenge.鈥�

The team worked with PDFs because they are a ubiquitous file format and can contain text, charts and images 鈥� a mix that is challenging for existing search systems but makes the documents ideal candidates for GovScape鈥檚 multimodal search.听

鈥淛ust as library classification systems group books on similar topics on the same shelf, these embeddings group similar pages with one another based on their visual and textual content,鈥� Lee said.

Researchers then built different indexing systems for the three kinds of search. The keyword search uses a basic index 鈥� similar to a book index 鈥� for all the text. If a user types in 鈥淔AFSA,鈥� the system finds all the pages the word appears on.听

For semantic and image searches, the system takes the user鈥檚 search term and creates an embedding. It then compares this embedding with the indices created from the embeddings of PDF pages and identifies the closest matches, which are returned as search results.听

鈥淥ur next goal is to cover all of the 70 million PDFs in the entire End of Term Web Archive 鈥� everything from 2008 to 2024,鈥� Lee said. 鈥淥ne of the challenges moving forward is how to efficiently search at that scale.鈥澨�

Because government archives contain 鈥渆very file type under the sun,鈥� Lee said, future work might expand to documents such as spreadsheets, images and HTML pages.听

鈥淚’m really excited about the prospects for better access to government information with projects like GovScape,鈥� Lee said. 鈥淏eing able to actually find relevant information is vital to the health of democracy and to the functioning of society.鈥�

Co-authors include of Boston University, who completed this research as a doctoral student in the Paul G. Allen School of Computer Science & Engineering; and , who completed this research as UW master鈥檚 students in the Information School;,,, , and , all students in the Allen School; of Harvard University; of the Massachusetts Institute of Technology; of the University of North Texas; and of the American Institute of Physics.听

For more information, contact Lee at bcgl@uw.edu.

Why is it important to study mosquitoes?听

Willem Laursen: Mosquitoes have been an enduring scourge of humanity for millennia. Their bites are a nuisance to humans and animals alike, and ancient texts describe illnesses consistent with mosquito-borne diseases, such as malaria, long before the source of transmission was understood.

Globalization and climate change are expanding the geographic range of many mosquito species, and their increasing resistance to insecticides threatens the long-term effectiveness of current control strategies. As a result, we urgently need new approaches for controlling mosquito-borne disease.

If we can better understand the genetic and sensory basis of mosquito behavior, we might be able to find new opportunities to disrupt disease transmission. Critical behaviors such as host seeking and blood feeding are highly specialized and difficult to model in other organisms, making it essential to study these mechanisms directly in mosquitoes themselves.

Andrea Durant: These mosquito-related problems are not just for humans. Warmer winters and early-season snowmelt have led to massive swarms of mosquitoes coinciding with wildlife migration, which changes foraging patterns in the Arctic tundra and forces animals like caribou to use precious energy reserves on evading these mosquito-blackened skies. Mosquito swarms are also a big problem for agriculture, particularly cattle herds.

What do you study?

AD: My lab studies how mosquitoes maintain a stable internal environment when faced with changing external conditions. Mosquitoes start their life as an egg that is deposited in or near water, and the larval, or juvenile, stages are aquatic. Unlike the terrestrial flying adult mosquito that has agency in its choice of residence, a mosquito larva is tied to wherever it hatches 鈥� it must survive and develop there, or die.

Sometimes the aquatic reservoirs where an adult female has selected to lay her eggs can be quite extreme, such as very polluted freshwater and seawater. We study specialized adaptations that allow these larvae to survive 鈥� most mosquito species require clean freshwater for larval development. Our goal is to reveal how mosquitoes have been able to successfully expand their habitats to places like urban sewage systems and salty coastal habitats.

WL: In my lab, our research focuses on understanding how mosquitoes sense things at the cellular level. We are trying to determine what proteins mosquitoes use to detect human-associated cues, such as heat and humidity. By identifying the cellular and molecular machinery mosquitoes use to find hosts, food sources, mates and egg-laying sites, we hope to better understand how specialized behaviors, such as blood feeding, evolve, and to uncover new targets for controlling the transmission of mosquito-borne diseases.

Jeffrey Riffell: My lab studies the 鈥渉ow鈥� of mosquito biting behavior. We also study how they visit flowers and plants 鈥� yes, they can pollinate certain plants! 鈥� to understand their natural behaviors. By learning more about mosquito physiology and behavior, we would like to develop new tools for traps and ways to control mosquitoes around people’s homes.

Tell us what it’s like to be someone who studies mosquitoes.

JR: Mosquitoes, all day and all the time. Although we try to minimize the potential for mosquito biting in the lab and in our field sites, you have to grin and bear it when dealing with these little vampires.听

WL: Being around large swarms of mosquitoes all day does desensitize me a bit. Sometimes I will be out hiking or camping with family members and I won’t be paying much attention until I start hearing complaints about the mosquitoes. Working with mosquitoes also leads me to do funny things, such as collecting sweat or wearing a nylon stocking for days to collect human odors for behavioral assays.

Rearing transgenic mosquitoes in the lab is a bit like ranching: We have to keep track of large herds of animals. Because the life stages live in different environments, we have to constantly shuttle them around between water-filled trays, for the larvae/pupae, and cages, for the terrestrial adults. We also have to move the adults around on a specific schedule to make sure they have access to our artificial blood feeders. Some lab members jokingly put a sign on the door that says “Welcome to The Ranch.”

AD: Willem is to a rancher as I am to a protagonist in “Swamp People.” We often venture outside of the lab to hunt mosquitoes in their natural habitat in urban and peri-urban areas. Sometimes we find ourselves in picturesque places like the beautiful pillow basalt coastlines of the San Juan Islands. Most often, I can be found headfirst in a nutrient-rich septic system in someone’s backyard filled with mosquito larvae or marching into the fray of massive swarms of saline-tolerant mosquitoes that await in tidal marshlands and mangrove forests.

What is the coolest mosquito fact you know?

WL: There are over 3,500 species of mosquitoes, with vastly different appearances, life histories and host preferences. Many are generalists. A few strongly prefer humans and some feed from cold-blooded animals like frogs or earthworms. The large amber-encased Toxorhynchites elephant mosquito shown in the movie “Jurassic Park” feeds on other mosquito larvae and doesn鈥檛 actually drink blood at all.

JR: I like These mosquitoes are very pretty, and they shoot their eggs into tree holes.

What鈥檚 one thing you wish people understood about mosquitoes?

AD: The incalculable misery that mosquitoes exert on humans and other animals certainly overshadows any appreciation for the importance of mosquitoes in nature. Many species of mosquitoes are critical to biodiversity and are actually fundamental to the food chain. There are numerous examples of areas with reduced breeding success and animal survival because there have been effective vector control programs and non-targeted mosquito eradication efforts.

For more information, contact Laursen at wlaursen@uw.edu, Durant at durantan@uw.edu and Riffell at jriffell@uw.edu.

If you shop on Google Flights, you get a quick comparison for different itineraries: One flight鈥檚 carbon emissions may be average, while another鈥檚 are 14% higher. But if you go shopping for a new laptop, you likely won鈥檛 find quick, comprehensible information on different models鈥� sustainability bonafides, despite the of producing and discarding electronics. In part, that鈥檚 because understanding a device鈥檚 emissions is difficult and time-consuming, even for experts.听

天美影视传媒 researchers developed an artificial intelligence system that automatically estimates the environmental impacts of making different electronic devices. The system uses AI agents 鈥� programs that perform tasks autonomously 鈥� to comb through publicly available data and conduct life cycle assessments, or LCAs. The system achieves an average error rate of 5%-19%, similar to the accuracy of LCAs conducted by experts.

The team June 12 in Nature Electronics.听

AI agents have recently grown increasingly capable of performing complex tasks. Today’s agents can search the web and pull information about electronic parts from product descriptions, images and documents.听

鈥淪ome of our previous research made me curious about how LCA experts perform environmental assessments 鈥� and whether that process could be automated,鈥� said lead author , a UW doctoral student in the Allen School. 鈥淪o to understand the bottlenecks firsthand, and then built a system that emulates these interactions with two AI agents. Each of them mimics different roles in the LCA process.鈥�

One agent acts as a sort of analyst, defining what information needs to be gathered and how it will fit together. It also reviews results for accuracy. The second agent is more like an engineer. It scrapes publicly available data for information on an electronic device鈥檚 components. That might entail sifting through spreadsheets, or looking up images of the insides of devices and taking chip information from them 鈥� including from sources not typically used for LCAs, such as and posts on.听

The two agents work in a loop. The first sets the scope, the second gathers information. The first then looks that information over and might send the second agent searching again, and so on. The agents then reference to convert the complete list of parts to carbon estimates.

The team also developed a new method to bypass this detailed data collection and directly estimate carbon footprints. For common devices like laptops and smartphones with publicly available carbon footprint reports, they found that products with similar specs like screen size and processors clustered around similar carbon values, because only a handful of companies make specialized parts for all these devices. So an unknown device’s footprint can be represented as a weighted average of similar products.听

They also use this to estimate the carbon for materials not in LCA databases. For example, a new type of sustainable plastic could be estimated based on plastics with similar properties and chemistry.

鈥淲e tried this 鈥榥earest-neighbors鈥� approach and found that for materials, it鈥檚 actually better than the standard approach of a human picking the single closest entry,鈥� said Zhang. 鈥淲hen estimating missing emissions factors in a test, the average error for our method was 23%. Human experts had an average error of 143%.鈥澨�

The authors note that while the aim of the system is to help reduce carbon emissions overall, running AI models requires energy, so they鈥檝e taken several steps to mitigate its impact. They use small AI models that aren鈥檛 as energy-intensive as general-purpose models. They also start the process by running a search to see if the device鈥檚 estimated emissions have already been calculated. If so, it can stop there. If the system does need to call its AI models repeatedly, estimating a device鈥檚 carbon footprint is currently on par with the emissions generated by brewing a cup of tea.

The team plans to collaborate with companies in the future to help automate their workflows.听

鈥淎 lot of big companies have sustainability teams that perform these LCAs,鈥� Iyer said. 鈥淥ur hope is that automating this will actually free up their time, so they can spend their time reducing the carbon footprint of the products themselves, instead of hunting down elusive stats.鈥澨�

Co-authors include , a UW student in the Allen School;, , a UW postdoctoral researcher in the Allen School; , a UW doctoral student in the Allen School; , a UW professor in the Allen School; of Wesleyan University, who completed this research as a UW doctoral student in the Allen School; of the University of Notre Dame; of Northeastern University; and of Brown University, who completed this research as a UW assistant professor in the Allen School.听

This research was funded by Amazon Research Awards and the National Science Foundation. Zhang was supported by the .

For more information, contact Iyer at vsiyer@uw.edu and Zhang at zzhihan@cs.washington.edu.

Quantum materials are a class of exotic materials with special properties that are governed by rather than . Those properties 鈥� like , and unusual forms of magnetism 鈥� often originate in the tiny repeating patterns of atoms inside crystals, but through clever engineering they can be observed and controlled at a more human scale. Quantum materials are helping to power the quickly growing field of , and could find their way into future generations of energy-efficient electronics.听

Designing new materials from the atomic scale up, however, requires intense modeling and simulation. Some materials may appear ordinary when viewed as small clusters of atoms, yet reveal new and useful properties when their atomic building blocks repeat and interact over larger distances. Researchers must be able to accurately predict behaviors at large scales in order to find materials with practical applications 鈥� otherwise designing new materials is a slow and costly trial-and-error process.

In the past 50 years, supercomputers have helped materials scientists solve some of those thorny prediction problems, but two recent studies from the 天美影视传媒 demonstrate how newer computing techniques can help researchers sniff out promising quantum materials to pursue. , published June 2 in the Proceedings of the National Academy of Sciences, shows how researchers can use artificial intelligence to simulate dozens of sheets of atoms stacked in intricate patterns, a process that produces complex and potentially useful quantum behaviors. , published June 8 in Nature Communications, shows how quantum computers can create a self-improving design loop by discovering new materials that could themselves be components of future quantum computers.听

鈥淲hat is exciting is that AI and quantum computing are beginning to change not just what problems we can solve, but how we do research,鈥� said , a UW associate professor of materials science and engineering and the senior author of both studies.

These two new tools 鈥� AI and quantum computing 鈥� are complementary in that they each excel at a different kind of simulation problem. With the right training, an AI model can act as a fast and relatively inexpensive surrogate of a supercomputer, extrapolating the behavior of huge material systems from a relatively small dataset. Cao and collaborators used this approach to stack virtual sheets of atoms on top of one another over and over 鈥� a process that created completely new phenomena that were absent on a smaller scale, but would have been impractical to model by traditional supercomputing. From there, researchers can try to make the most promising materials in the lab to prove out the simulations.

Quantum computers, on the other hand, are essentially powered by the same quantum phenomena 鈥� like entanglement 鈥� that Cao and other materials researchers want to study. Such phenomena can be difficult to simulate using traditional computers or AI systems, but quantum computers are naturally suited to the task. In the study, Cao and his team used a quantum computer to study an exotic phase of matter known as a .听

Moving forward, Cao and his team plan to further build out their datasets and eventually develop models that can simulate a much wider range of materials. They also hope to combine their AI and quantum computing systems into a more powerful and flexible hybrid tool.

鈥淭he next step is to bring these tools together,鈥� Cao said. 鈥淲e can use AI to guide quantum simulations, and quantum computers to generate new data and insights that improve AI models.鈥�

鈥淲e are at the start of a new era,鈥� said , UW professor and chair of materials science and engineering and co-author of both studies. 鈥淥ur field is fundamentally changing. Things that were literally impossible a couple of years ago are now becoming routine. And we are only beginning to see what AI and quantum computing will make possible for quantum materials.鈥�

was led by , a UW doctoral student of materials science and engineering. was led by , a UW doctoral student of physics. A complete list of authors is included with the studies.

The authors acknowledge the support of Amazon and the Department of Energy.

For more information, contact Cao at tingcao@uw.edu.