Preview data from the point to a wave of discoveries that may complete the census of the Milky Way’s neighboring galaxies — including its faintest companions.

Using early data from Rubin, a pair of recent studies have identified two ultra-faint and previously unknown satellites of the Milky Way. One study, in Research Notes of the AAS, identifies a galaxy called Aquarius IV, located in the Milky Way’s distant outskirts. The second study, in The Astrophysical Journal Letters, identifies Rubin-GC1, a star cluster located on the far side of the Milky Way, inside a murky region shrouded by cosmic dust.

Large galaxies are often orbited by satellites — small galaxies or star clusters — but many are extremely faint and difficult to find. The Simonyi Survey Telescope’s 3.2 gigapixel sensor — the largest digital camera ever built — can bring many more of those faint satellites in view. More than 20 researchers at the ĚÇĐÄÔ­´´ have worked to get in-focus images from the telescope, validate data ahead of releases and maintain an accurate catalog of known stars, galaxies and other phenomena. 

“Finding something as faint as Aquarius IV comes down to how clean your stellar catalog is,” said , a UW postdoctoral scholar in astronomy and co-author of both studies. “A handful of misclassified galaxies or spurious detections can wash out the signal entirely, so much of our work at the UW was about making sure that when Rubin tells you a point of light is a star, you can believe it.”

The Rubin Observatory began its 10-year Legacy Survey of Space and Time () in June. One of its main missions is to map what is known as the Milky Way’s “halo” — the cloudlike sphere of gas, stars and dark matter that surrounds the main part of the galaxy — and find previously hidden satellite galaxies, star clusters and other substructures. There are about 65 known satellites of the Milky Way, yet there is speculation that hundreds more Milky Way satellites remain undetected.

“What I find especially exciting is that these discoveries are just the beginning for Rubin,” said , a graduate student of astronomy at Yale University and lead author of the Aquarius IV study. “These new satellites will give us new opportunities to test models of galaxy formation, trace the assembly history of the Milky Way and investigate the nature of dark matter.”

For the two new studies, astronomers analyzed Rubin Observatory preview data taken between April 2025 and January 2026. 

“Excitingly, these two objects came out of test data covering just 15% of the full LSST footprint,” Ferguson said. “Now that the LSST is underway, we’ll be able to push to fainter and more distant systems than any survey before. Aquarius IV and Rubin-GC1 are a preview of what the next decade will look like.”

Beyond the significance of the discoveries themselves, the researchers say their findings may provide valuable information about the nature of dark matter, the invisible scaffolding that makes up most of the matter in the universe. Ultra-faint galaxies such as Aquarius IV, which have a fraction of the stars that a larger galaxy contains, are almost entirely composed of dark matter.

Likewise, the researchers say, Rubin’s early data may yield new insight into the life cycle of some of the universe’s hardest-to-spot galaxies.

“These discoveries are incredibly promising,” said , a professor of astronomy and physics at Yale University who co-authored both studies. “Continued discoveries will allow us to complete the census of satellite galaxies around the Milky Way, which is a key to understanding how the faintest galaxies form.”

UW co-authors include , an assistant professor of astronomy and , a postdoctoral scholar in astronomy. A complete list of co-authors is with the .

Operations of the Vera C. Rubin Observatory are funded by the U.S. National Science Foundation and the U.S. Department of Energy’s Office of Science.

For more information, contact Ferguson at pferguso@uw.edu.

Adapted from a press release by .

Source

Antarctic blue whales appear to be creeping back from the brink of extinction after 50 years of protection. A new study estimates the total population at 4,018, up from 418 after the last reported whale catch in 1973.

Still, their population pales in comparison to pre-whaling levels, when hundreds of thousands of blue whales swam in the Southern Ocean.

Antarctic blue whales were the primary target of commercial whaling in the 20th century. Each whale can weigh more than 400,000 pounds, making them the largest animals known to Earth and a profitable source of whale oil. Killing them was banned by the International Whaling Commission in 1964, but illegal whaling by Soviet fleets continued for another 10 years. By the time it stopped, it was almost too late for the species.

“No whale was hunted quite to the extent that Antarctic blue whales were hunted,” said lead author , a postdoctoral scholar in the UW School of Marine and Environmental Affairs. “We brought together historical whaling data and contemporary abundance estimates to create a full picture of just how heavily people hunted them and how they are doing now.”

The results were .

Tracking whale populations helps researchers determine whether they are recovering from whaling, but counting whales is a challenge. Although large enough to spot from space, Antarctic blue whales spend limited time on the surface. Researchers rely on a combination of photo identification and line transect surveys to estimate the total population.

Line transect surveying involves travelling in a straight line and counting animals sighted en route. In this case, a ship sails along an axis looking for whales. The number of sightings and distance from the boat can be used to calculate density, and therefore abundance.

However, these methods are imperfect and expensive. The International Whaling Commission surveyed the entire Southern Ocean in expeditions from 1978 to 1984, 1985 to 1991, and 1991 to 2004. This data was used to generate the last circumpolar abundance estimate for this population, but few line transect surveys have been completed since.

Rand and colleagues created a model that accounts for all available data sources, including historical catch data, counts from these line-transect surveys and more recently, photo identifications. The researchers estimate that there are now 4,018 Antarctic blue whales, representing 1.9% of the pre-whaling population. Whaling reduced the population by 99.8%, and researchers estimate that it is now growing at a rate of 4.5% per year.

One could argue that humans have some degree of moral responsibility to remedy the effects of whaling, but that’s not the only case for rehabilitation.

Antarctic blue whales are “ecosystem engineers,” Rand said. When they die, their enormous bodies sink thousands of feet to the ocean floor, where entire communities feed on their corpses. This process, called whalefall, stimulates large scale nutrient cycling and is thought to sequester carbon.

Antarctic blue whales live long lives and give birth to just one calf at a time, and only every two to three years. The extent of decimation means a long, slow road to recovery. Although their population is far from what it once was, the picture that emerged from this study is cause for cautious optimism.

“I think this is relatively realistic for the kind of population growth we would expect from them,” Rand said. “I think it’s probably an open question whether they will make it back to pre-whaling levels.”

Antarctic blue whales are still listed as critically endangered, and it isn’t clear how they will respond to environmental change. Tracking their numbers is the only way to determine whether they are rebounding or struggling under new pressures.

Rand hopes that future technological advances, such as passive acoustic monitoring, will make surveillance easier. Researchers use recordings of whale song to estimate density but haven’t yet figured out how to infer the total number of animals from acoustics.

“Antarctica is a big place, and it’s not just one country’s ocean,” Rand said. “We really need more monitoring and continued monitoring to figure out what is going to happen to this species in the future.”

Other co-authors are , professor of aquatic and fishery sciences, and Douglas Kinzey of the National Oceanic and Atmospheric Administration; and of the Australian Marine Mammal Centre; Koji Matsuoka of the Institute of Cetacean Research; and Ken Findlay of the University of Pretoria in South Africa.

This study was funded by the National Oceanic and Atmospheric Association.

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

Source

Artificial intelligence models replicate, and sometimes amplify, all sorts of human biases. In the last few years, ĚÇĐÄÔ­´´ researchers have found AI has erased female characters from kids’ stories, favored job applicants of certain races and genders, and swayed human decisions.Ěý

Much bias research has focused on text and visuals. Yet people speak to AI models with increasing regularity, and AI tools are now used to screen job applicants and score oral presentations. Now, a UW-led team of researchers found that state-of-the-art AI models were biased against speakers with certain accents.Ěý

The team investigated how state-of-the-art multimodal language models, including two from Google and one from OpenAI, evaluated human speech in job interviews, conference presentations and spoken English proficiency tests. The models scored responses from voices speaking English with five different accents: American, British, Chinese, Indian and Nigerian. The AI models generally ranked Chinese and Nigerian accents the lowest, Indian in the middle, and American and British the highest. Even when the content was exactly the same and the models could accurately transcribe the speech, the systems ranked both the delivery and content higher for the American and British speakers.Ěý

The team Oct. 25 at the Conference on Empirical Methods in Natural Language Processing in Budapest, Hungary.Ěý

“People are now trusting audio AI models as sort of third-party evaluators, for hiring, for oral exams in education, for certain English tests in immigration — contexts where fairness really matters,” said co-lead author , a UW doctoral student in the Paul G. Allen School of Computer Science & Engineering. “So I was curious how models perceive speech from people who don’t speak with so-called ‘standard’ American accents.”

The team chose accents from five regions that have large populations of English speakers and where the researchers also had connections. Researchers then tested the accents in three different domains — hiring, presentations and English tests. To study differences at scale, they had AI voices from read the same pieces of text to the AI models being tested. They validated the results with 52 human speakers from the five countries.Ěý

After some testing, the team focused the study on three proprietary AI modelsĚý — Google’s Gemini 2.5 Pro and Gemini 2.5 Flash, and OpenAI’s GPT-Audio 1.5 — and one open-source model, Qwen 3 Omni.Ěý

“We wanted to check these state-of-the-art models,” said senior author , a UW associate professor in the Allen School, “because these are what companies would deploy in a real-world setting.”

The team had the AI voices read various scripts — interview responses, a class presentation, a conference presentation and English proficiency test responses — with the five different accents in both male and female voices. The four AI models then rated the responses on a scale of one to seven for both delivery and content, and gave a brief explanation for their scores.Ěý

Here are samples of AI voices reading the same hiring passage:Ěý

British

Nigerian

Indian

American

Chinese

Researchers repeated the interview response test with the 52 human participants, having them read four scripted responses each and respond to questions without a script.Ěý

The results favored American and British accents — with a greater disparity between the human accents than between the AI accents.

To better understand the results, researchers analyzed features of the accents. They then measured their “acoustic distance” and found that accents further from Americans’ were often ranked lower.Ěý

The team found this apparent hierarchy manifested in the models’ explanations, too.Ěý

“Even though they repeatedly claimed Chinese and Nigerian speakers ‘mumbled’ or were ‘difficult to understand,’ the models were actually very good at transcribing the accented speech,” Ahia said. “They understood the speakers, but they just had this implicit notion of hierarchy among them based on accent.”

This bias can seep into models during training in various ways — possibly through data that describes accented speakers as hard to understand, through data sets that underrepresent certain populations and accents, or through training methods that reinforce conventional expectations about how a professional speaker should sound. English, for instance, is the official language in Nigeria and is the teaching language in Indian schools, yet both accents were rated lower than British and American.

“One clear direction for future work is to create more representative training and evaluation data sets that could improve the accuracy and fairness of language models,” Tsvetkov said.Ěý

The team plans to expand the research by studying more human accents — including regional variations within countries — and by looking at how AI models are analyzing audio in medical contexts, such as automated emergency response.Ěý

“For all of these, we hope that developers build and publish evaluations of their models that don’t focus only on average performance, but also break it down into subpopulations, dialects and so on,” Tsvetkov said. “We don’t want the average performance to be 97% accurate for the general population, but for a smaller group only 20%.”

Additional co-authors include of Koel Labs, who completed this work as a UW undergraduate in the Allen School; of University of Wisconsin–Madison, who completed this research as a UW postdoctoral scholar in the Allen School; and of the University of Toronto; , a UW doctoral student in the Allen School; of the CISPA Helmholtz Center; of LMU Munich, who completed this research as a UW postdoctoral scholar in the Allen School; of Ohio State University; and , UW vice provost for AI and professor in the Allen School.Ěý

This research was funded in part by gifts from Google and OpenAI.Ěý

For more information, contact Ahia at oahia@cs.washington.edu and Tsvetkov at yuliats@cs.washington.edu.

Source

The ĚÇĐÄÔ­´´â€™s 2026 incoming class reflects the University’s commitment to expanding access to a UW education for students from across Washington and from a broad range of communities and backgrounds.

Washington residents make up three-quarters of this year’s incoming class, and nearly four out of 10 are first-generation college students, according to the 2026-27 official enrollment count, which occurs 10 days after the start of the fall quarter. UW’s ongoing outreach to Washington high schools and communities is reflected in a student body that includes residents of 38 of Washington’s 39 counties.

There are additional indicators in the incoming class that the UW’s efforts to expand access and decrease disparities are making headway: The number of first-generation first-year students is up 4% from last year and there was a 5% increase in Pell Grant recipients.

“Seeing so many Washington students choose the UW reinforces our commitment to making a world-class education accessible to people from every corner of our state. Students from across Washington bring diverse experiences and perspectives that enrich our campus and foster collaboration,” ĚýUW President Robert J. Jones said. “As a public university serving the entire state, we have a responsibility to reflect the people and values of the communities we serve. That’s what it means to be the university for °Â˛š˛őłóžą˛Ô˛ľłŮ´Ç˛Ô.”

This year’s incoming class has a total of 12,552 first-year and transfer students enrolled across all three campuses, a 3.5% increase from last year.

Washington residents make up 75% of the incoming class across all three campuses, a 2.1% increase from last year. Of these students, 6,245 of the incoming first-year and transfer students at the UW in Seattle are Washington residents. At UW Bothell, 1,692 incoming first-year and transfer students are Washington residents. There are 1,455 Washington residents among the incoming first-year and transfer students at UW Tacoma.

There are 7,475 new first-year students at the UW in Seattle and 1,744 new transfer students. At UW Bothell, 1,149 new first-year students enrolled, alongside 651 transfer students. At UW Tacoma, 748 new first-year students enrolled, joining 785 new transfer students — a combined record high for the Tacoma campus. Of these transfer students, 1,468 in Seattle came from Washington community colleges, while 538 transfers at Bothell and 623 at Tacoma came from Washington community colleges. The total, 2,629, marks an increase on all three campuses.

The total number of students across all three campuses this year is 63,866. Enrollment at the UW’s Seattle campus is 52,338, 6,458 students enrolled at UW Bothell — the highest for this campus — and 5,073 are pursuing degrees at UW Tacoma. (The numbers reflect that three students are enrolled at more than one campus.) Of the 63,866 enrolled students across all three campuses, 46,987 are undergraduates (36,636 in Seattle, 5,938 in Bothell and 4,413 in Tacoma) and 16,879 are pursuing graduate or professional degrees (15,702 in Seattle, 520 in Bothell and 660 in Tacoma).

The number of international first-year and transfer students across all three campuses is 1,077, 99 more than 2025. In total, there are 7,333 international students enrolled across the three campuses, which reflects about a 7% drop from last year and correlates with national international student enrollment declines. Of all these international students, 6,951 are in Seattle, 222 are at UW Bothell and 160 are at UW Tacoma.

The official enrollment count — a snapshot in time — will later be presented to the UW Board of Regents.

For more information, contact Victor Balta at balta@uw.edu.Ěý

Source

It’s a dizzyingly busy time for scientists searching for life on distant planets. To date, astronomers have and are searching the atmospheres of the most promising worlds for chemicals that might be byproducts of life. are particularly interested in atmospheres like Earth’s that trap enough heat to support liquid water on the planet’s surface, which scientists consider the key factor in determining whether a planet can support life as we know it. 

But scientists also want to distinguish between habitable and inhabited worlds. To do that, they need to know what sorts of atmospheric conditions on a planet can be explained without life, so that they can identify planets with atmospheres that can only be explained by life’s influence.

To provide that planetary baseline, ĚÇĐÄÔ­´´ researchers built a complex virtual Earth with one crucial tweak: no life.

in The Planetary Science Journal, the researchers simulated the Earth’s geochemical systems for 5 billion years — from its formation 4.5 billion years ago through roughly 500 million years in the future. The simulation modeled the exchange of chemicals between the core, mantle, crust, surface and atmosphere over time. It accurately predicted 19 planetary properties like heat flow through the mantle, but left out the influence of life on those systems. The simulation can be generalized to predict how an atmosphere might evolve on any rocky planet, and can help pinpoint evidence of life in atmospheres that don’t fit the model.

“This model can help us see how habitable but non-living planets might mimic living planets,” said lead author , who completed this research as a UW doctoral student of astronomy and is now a postdoctoral fellow at UC Riverside. “When we interpret observations of an exoplanet’s atmosphere, this tool will help us determine the probability that what we observe indicates life on the surface.”

Gilbert-Janizek’s work builds on a model developed in the lab of senior author , a UW associate professor of astronomy. Earlier versions of the model simulated the core, mantle, crust and stellar radiation, but did not include the surface or atmosphere. 

A planet’s atmosphere is shaped by many nonbiological factors, including heat and gases from the interior and from the planet’s star. But the atmosphere of a distant planet is also where astronomers hope to spot explicit evidence of life.

There are many ways that life might change an atmosphere: On Earth, plants and animals pump steady streams of oxygen and methane into the sky. Those two elements destroy each other when they mingle. Thus, an alien astronomer studying our planet might suspect that something alive is replenishing the supply of both gases.

That logic led human astronomer Gilbert-Janizek to add surface and atmosphere modeling to the existing Earth simulation, completing the picture of how chemicals move through all of the Earth’s systems. 

The result is a model that can be tuned to study any rocky exoplanet that astrobiologists find interesting. Researchers can plug in basic details about that planet and run the simulation to see whether the model can reproduce the observed atmosphere without the influence of life. If they can’t get a match, the planet in question gets even more interesting.

“Dr. Gilbert-Janizek’s work represents an exciting step toward understanding where life can exist in the universe,” Barnes said. “After more than a decade of work by many scientists, Dr. Gilbert-Janizek has finally completed a model of the Earth that simultaneously reproduces properties of Earth’s core, mantle, surface and atmosphere. That calibration gives us confidence that future simulations of exoplanets will be accurate.”  

Gilbert-Janizek says the new and improved model can also help inform the design of future telescopes built to study distant planets, like the proposed . 

“Realizing the ambitious goals of a telescope like the Habitable Worlds Observatory requires decades of work across many generations of scientists,” Gilbert-Janizek said. “I’m so grateful to join this lineage and one day pass the torch to the next generation so that we can continue chipping away at this question of whether or not we are alone in the universe.”

Other UW co-authors include former astronomy doctoral students , now a postdoctoral researcher at the University of Oklahoma and , now a high school teacher in the Seattle area.

A complete list of co-authors is .

This research was funded by NASA.

For more information, contact Gilbert-Janizek at sgilbert@ucr.edu or Barnes at rory@astro.washington.edu.

Source

The bivalve-rich beaches of Hood Canal were all summer after detection of a marine biotoxin that causes , a life-threatening foodborne illness that affects the nervous system.

Given that Washington’s commercial shellfish industry employs nearly 2,400 people and brings in more than $240 million each year, a monthslong closure during peak season had sweeping consequences.

Detecting harmful algal blooms early is critical for consumer safety and trust in the industry. The Washington State Department of Health oversees a that monitors toxin levels in shellfish tissue from various growing areas. The effort is supported by an unofficial partner: , a program launched by National Oceanic and Atmospheric Administration in 2006 that migrated to Washington Sea Grant several years later.

Unlike the DOH, SoundToxins screens the water for harmful algae, which can expedite detection. When SoundToxins monitors detected Ěý — a species of phytoplankton that produces the toxin that causes PSP — they quickly relayed the information to the DOH, which closed the impacted marine areas before anyone got sick.

Although this type of advance warning can save lives, the future funding of the program is uncertain. Washington Sea Grant relies on a patchwork of research grants and federal and state funds to operate SoundToxins. Recent cuts to research funding and the current state of Washington’s budget have illuminated the precariousness of funding for vital programs like this. Washington Sea Grant remains committed to working with state and federal partners to find the right funding pathway to keep SoundToxins running.

A quantified the value of SoundToxins with survey data. UW News spoke with program manager about the results and the program.

How is the monitoring program and the early detection of harmful algal blooms (HABs) valuable?

Michelle Lepori-Bui: Harmful algal blooms can be really detrimental to the shellfish industry as closures have a direct impact on revenue and because recalls and illnesses affect the reputation of shellfish as a safe and healthy food source. Robust monitoring programs and early warning of these events help protect public health and minimize these impacts. There are many benefits to early detection of HABs to the general public and recreational and commercial harvesters that can be hard to quantify. Farmers can benefit by modifying harvesting activities to minimize losses, avoiding recalls, and maintaining trust in the market, while shellfish consumers and coastal communities benefit from safe local food sources and working waterfronts.

How does the new study demonstrate the value of SoundToxins?

MLB: This survey sought to capture the many tangible and intangible benefits of the program and distill them into a single economic metric using an approach called ‘willingness to pay.’ The study found that shellfish growers value the program with a willingness to pay $533/year for phytoplankton monitoring. With changing environmental conditions, shellfish mortality events, and heat waves, growers are facing an extremely challenging production environment. The finding that they are willing to pay this amount, even during these challenging economic times, shows how valuable this program is to coastal communities.

How does SoundToxins support the state-led effort?

MLB: The DOH’s biotoxin monitoring program is a series of steps in compliance with the National Shellfish Sanitation Program that starts with monitoring for biotoxins in mussel tissue. Mussels typically pick up toxins faster than other species, so they are a good indicator, but other species are tested as well. If toxins are detected, they can initiate recreational and commercial closures.

We’re not part of the official regulatory process, but our work can guide the DOH’s efforts through early warning. We maintain a database populated with volunteer data and a map that uses red, yellow and green color coding to report HAB presence on the coast and in Puget Sound and surrounding waters. We notify the DOH via email when something changes and it uses that information to prioritize certain samples or request additional samples.

Who collects the data?

MLB: SoundToxins is a very cost-effective program because we have really amazing monitors, partners and volunteers. A lot of different folks do the actual monitoring. We work with tribes, growers, environmental education groups and individuals at 35 different sites throughout the region. We provide them with equipment and training and support the database, and they collect and analyze the samples, posting everything online within 24 hours.

What happened this summer? How was it different from years past?

MLB: We got an alert from our monitor in early June that they found Alexandrium cells in a sample from Hoodsport, which is pretty far south in Hood Canal. That alert went straight to the DOH and it reacted quickly, requesting more samples from various parts of Hood Canal.

Historically, there haven’t been many incidents of PSP in south Hood Canal. There were a few blooms in the early 2000s and a big bloom in 2015 that went further south than usual, close to Hoodsport but not to the end of Hood Canal. That was considered unprecedented and this year surpassed that.

It’s understood globally that harmful algal blooms are increasing in frequency, severity and impact. This is partially related to changing climate conditions. Also, phytoplankton like Alexandrium respond to nutrients in the water. The input of nutrients from human activities can influence blooms. We have seen things showing up in places they haven’t been before and increasing in intensity.

A lot of people are working on models to predict harmful algal blooms, but it’s complicated, especially in our area where there are a lot of complex water bodies. We need to continue doing the research to try to understand things better, but in the meantime, monitoring is critical.

Co-authors include Michelle Lepori-Bui, marine water quality specialist at Washington Sea Grant; , water quality specialist at Washington Sea Grant; Vera Trainer, a UW research scientist in environmental and forest science and director of the Olympic Natural Resource Center; of Woods Hole Oceanographic Institution; and of NOAA; and Jerry Borchert and Tracie Barry of WA DOH.Ěý

This research was funded by the NOAA National Centers for Coastal Ocean Science Prevention, Control, and Mitigation of HABs.

For more information, contact Lepori-Bui at mdtlb@uw.edu.Ěý

Source

The U.S. House of Representatives this past summer that would make Daylight Saving Time (DST) permanent in the United States — meaning there would be no turning back the clocks each fall. Nineteen states, including Washington, . If the bill is ultimately signed into law, the change would go into effect in those states.

The bill is currently stalled in the U.S. Senate. are a central issue because permanent DST would lead to darker mornings from November to March. This would mean darker commutes for everyone, but how much would it affect children going to school? ĚÇĐÄÔ­´´ researchers did the math for 14 out of the 19 states that preapproved the change. Using school start times, enrollment data, and latitude and longitude measurements, the team calculated just how many days each child would be commuting to school in the dark and then added the trips together.

In Washington, there would be almost 35 million more trips to school in the dark under permanent DST compared to the current system of switching between DST and Standard Time twice a year. The other 13 states followed a similar trend, with Georgia having the highest deviation from the current system — 71 million more trips. When the researchers extrapolated these results for the rest of the country, it added at least 2 billion more dark commutes to school.

The team also analyzed traffic collision data from the U.S. and found that it was roughly twice as likely for a fatal crash involving a school-aged pedestrian to occur during dark mornings than after sunrise.

The researchers Oct. 29 in Current Biology.

“It’s called the ‘Sunshine Protection Act’ because you get an extra hour of light at the end of the day. But you will take that hour from the morning. You could call it the Sunshine Destruction Act,” said senior author , UW professor of biology. “These dark mornings will inequitably affect children who have to go to school at a certain time. It’s a very tangible metric that anybody from any state or socioeconomic background can relate to. Nobody wants their kids to go to school in the dark.”

Dark mornings affect everyone. Exposure to bright light in the morning helps set our circadian clock, which helps regulate many facets of our lives, such as our mood, sleep and appetite. Without that morning light, our bodies struggle to regulate these key functions.

“When we send kids to school in the dark, we are depriving them from an opportunity to get that morning daylight,” said lead author , a UW postdoctoral scholar in biology. “We were asking what’s the cost for these children. We wanted actual numbers, a real number of kids based on the geographic location of each school.”

The study revealed the areas that would be the hardest hit by the switch: communities on the western edge of their time zones, such as Forks, Washington, and communities with early school start times. For example, Mississippi and Alabama schools have an average start time of earlier than 8 a.m. and both balloon from about 2 million dark trips for each state under the current system to more than 20 million dark trips for each state under permanent DST.

“Some districts probably never experience a dark commute under the current system, and, if this passes, suddenly they will wake up to this new reality that their kids are going to school in the dark,” RodrĂ­guez Ferrante said. “Keep in mind that we used the actual start time listed for each school, and a lot of schools offer breakfast or other things before school, so the kids could be getting there even earlier.”

The researchers argue that permanent DST would also undo the benefits of having a delayed start time for high school students, which is intended to account for teenagers’ shifted circadian clock compared to kids and adults.

“I’ve heard the argument, ‘If we’re in permanent DST, we can just delay the school start time to address these issues,'” de la Iglesia said. “Well, if we do that, you will end up going to work later. And everything is going to get delayed. You will get out of work an hour later, and you will miss the opportunity to enjoy your evening light. And essentially, we’re just going to be realigning everybody to solar time. That, by definition, is Standard Time.”

Daniel Zhang, a high school student, and UW biology undergraduate students , , Jessie Lu, , Elena Kully-Rivera, , are also co-authors on this paper.

For more information, contact de la Iglesia at horaciod@uw.edu and RodrĂ­guez Ferrante at guadarf@uw.edu.

Source

Pulling the strings: How stem cells in the developing fruit fly brain organize their DNA from outside the nucleus

Cells are very particular about the proper locations of their contents. For example, DNA is packaged in the nucleus in a highly specific way, but researchers are still trying to understand how this happens. For stem cells in the brain of a larval fruit fly — model organisms used to study processes such as early development — the DNA is organized so that the center of each strand, called the centromere, is at the top of the nucleus. In a paper in the Journal of Cell Biology, UW researchers describe how proteins outside the nucleus work alongside proteins inside the nucleus to control the position of centromeres. The researchers observed this mechanism by imaging stem cells within larval fruit fly brains and then using machine learning to measure how centromere locations change when specific proteins are removed. The research team also followed the locations of two specific genes that are far away from the centromeres. Both genes lost their usual positions in the nucleus when the centromeres strayed from their normal location, suggesting that proper centromere positioning may be important for orienting the rest of the DNA in the nucleus. Studying how and why DNA is organized in these stem cells can help researchers better understand how stem cells are regulated and how brains develop.

For more information, contact lead author Jennifer Taylor, UW postdoc in biology, at jatay@uw.edu.ĚýĚýĚýĚýĚýĚýĚý

The other UW co-author is . A full list of funding is .

Micro language models let AI run instantly on small devices like smartwatches

Though artificial intelligence can refer to many technologies, the current AI boom has been driven by large language models, or LLMs. They power chatbots and AI agents by running predictions based on ever-increasing amounts of variables called “parameters.” Current cutting-edge models are estimated to have several trillion parameters, making them far too large to run on slower phones or small devices like smartwatches. In a paper at EMNLP 2026, UW researchers introduced “micro language models,” which run with 8-30 million parameters. These models are small enough to run on devices with limited computing power and give users instant responses that fill the lag before the device can access an LLM through the cloud. For instance, a user could ask their smartwatch a question and the micro language model would start responding instantly; then after four to eight words, the LLM would complete the response. The micro language model made mistakes in 4% to 8% of responses that the other model corrected seamlessly before continuing. In a user study, nine out of 10 participants preferred the paired system to the lagging responses from only a cloud model.

For more information, contact ulm@cs.washington.edu or visit .

The UW co-authors are , , and . A full list of co-authors and funding is in .

Policy gaps impact migrant fishers in Washington state

The commercial fishing industry is key to Washington’s maritime economy, but there are concerns about exploitation and forced labor among fishers. While much of industry is regulated by the Washington Department of Fish and Wildlife, exemptions in U.S. law mean some vessels aren’t subject to the same citizenship requirements. A research team from the UW Harry Bridges Center for Labor Studies recently focused on understanding the experiences of migrant fishers pursuing albacore tuna, a highly migratory species, who dock or unload at ports in Washington state. Researchers found the exemptions allow migrant fishers to work without authorization or visas, leaving them vulnerable to forced labor and exploitation. Meanwhile, complicated policy and regulatory frameworks make it difficult to address workers’ rights violations. The researchers gathered information from 15 migrant fishers, 14 U.S. citizen fishers, and more than 35 subject matter experts. Every migrant fisher interviewed by the team reported experiences aligned with the International Labour Organization’s indicators of forced labor, and both migrant and U.S. citizen fishers reported difficult living and working conditions. Approaches to combating forced labor are varied and uneven, the researchers found, and additional work is needed to better understand and support the needs of Washington’s fishers. Along with the full report, the team also published .

For more information, contact , research director at the Bridges Center, at rerstad@uw.edu.

The other UW co-authors are , and .

Burrowing shrimp are on the move in a mudflat near you

Burrowing shrimp are a pest well known to shellfish growers. These pale-bodied creatures excavate mud as they dig into the ground, creating piles of sediment that suffocate oysters and clams growing on the surface. In estuaries on the west coast, shrimp beds reach densities of hundreds of shrimp per square meter, but the boundaries of beds change. in Estuarine, Coastal and Shelf Science shows that shrimp beds move up to 100 feet each year. The study took place at Willapa Bay over the course of three years and is the first to quantify how rapidly the boundaries of these beds shift over time. Researchers still don’t know why the shrimp move, but they observed a clear pattern in the ecosystem. Shrimp change the texture of sediment by eating organic matter and releasing finer particles, so sediment becomes muddier and more compact once they leave. After several years, these abandoned beds were colonized by eelgrass, a marker of ecosystem health. This “engineering” effect was less evident in areas where moving water keeps mud from settling. Studying these rapid shifts and the types of sediment left behind could ultimately help shellfish growers better protect their product.

For more information, contact co-author Jennifer Ruesink, a UW professor of biology, at ruesink@uw.edu.ĚýĚýĚýĚýĚýĚýĚý

The other UW co-authors are Wesley Hull, Haleh Mawson, and . A full list of co-authors and funding is .

Source

Scientists may add a new dwarf planet to the official roster of our solar system.

The distant, icy body, provisionally dubbed 2026 ED138, orbits our system beyond Pluto. It was spotted earlier this month by a scientist combing through data from the Simonyi Survey Telescope at the . Initial measurements place ED138 among the largest dwarf planet candidates ever discovered beyond Neptune; if confirmed, it would join the five dwarf planets recognized by the : Pluto, Ceres, Eris, Haumea and Makemake.

The results were presented at a press conference today by , a senior astrophysicist at the Harvard & Smithsonian Center for Astrophysics, and by , a ĚÇĐÄÔ­´´ professor of astronomy and leader of Rubin’s solar system team. Holman first spotted the candidate and Jurić contributed additional observations.

“If confirmed as a dwarf planet, ED138 would join a very small family of worlds,” Holman said. “Rubin will image the whole southern sky again and again for 10 years, so any others hiding in the places earlier surveys skipped are now within reach.”

“Finding an object this big, roughly 1,000 kilometers across, is something that happens maybe once in a generation,” Jurić said. “It’s the largest trans-Neptunian object found in about 20 years, and Rubin turned it up while we were still commissioning the telescope.”

On the evening of Oct. 13, Holman was running data collected by Rubin earlier in 2026 through a search code he developed with colleague Kevin Napier. The code is built to look at billions of detections of moving objects, compare them and find the detections that correspond to the same object. Using this method, Holman uncovered ED138, and he immediately noticed that its brightness and distance indicated that it is unusually large.

“My first reaction was one of excited disbelief,” Holman said. “Objects like this are very rare, so I immediately started checking whether I or the code had done something wrong.”

To remove any doubt, Jurić searched through Rubin’s catalogs and images and found 20 more observations, going back to March 2026. The team also spotted ED138 in archival images from other surveys going all the way back to 2016, allowing the team to trace more than ten years of ED138’s orbit. 

ED138 is currently at a distance of about 46 astronomical units (), where 1 au is the average distance between the Earth and the sun. For reference, dwarf planet Pluto orbits at an average distance of 40 au. This classifies ED138 as a trans-Neptunian object () — objects that orbit the sun at a distance greater than Neptune’s.

Scientists study the size, composition and orbital trajectories of TNOs to understand how the solar system formed, how the giant planets migrated over time and how smaller objects have been scattered throughout this evolution. 

Not all TNOs are considered dwarf planet candidates, with the distinction being in the size and shape of the object. To be a dwarf planet an object needs to be massive enough for its own gravity to pull it into a round shape. Based on ED138’s distance, brightness and estimated surface reflectivity, the team estimates that its size is between 600 and 1,500 kilometers; icy bodies this large are expected to be round. Further observations are needed to measure its exact size and confirm its nature as a dwarf planet. 

ED138 was found in a region of sky close to the plane of the Milky Way — a crowded area filled with millions of stars. To most telescopes, ED138 has been lost among the myriad background sources, which is likely why it was not discovered until now. Rubin Observatory’s regular and continuous monitoring of the entire Southern Hemisphere sky, combined with powerful algorithms that sift through millions to billions of points of light to identify those corresponding to small moving objects, enabled the team to spot ED138 within the crowded field.

ED138 joins the rapidly growing sample of solar system objects discovered by Rubin since it began collecting early data in preparation for the (LSST), which began in June 2026. In April, before LSST formally began, the Rubin solar system team announced observations of over , which included roughly 380 new TNOs. At the time, the team called this large submission “the tip of the iceberg” and expressed their excitement for the discoveries to come once LSST began. With the discovery of ED138, those hopes are coming to fruition.

“Rubin is giving us an unprecedented capacity to explore the sky through a combination of speed, depth, and precision,” Jurić said. “And the discovery is just the beginning: Once we identify a new object, we can search back months, and sometimes even decades, finding it in images where nobody knew to look. This ability to both discover and ‘precover’ objects is transforming how quickly and accurately we can determine their orbits and understand the distant solar system.”

The public can interact with this discovery on , an interactive web-based app that provides an intuitive way to explore the structure of our cosmic backyard in three dimensions and in real time.

Operations of the Vera C. Rubin Observatory are funded by the U.S. National Science Foundation and the U.S. Department of Energy’s Office of Science.

For more information, contact Mario Jurić at mjuric@uw.edu.

This story was adapted from a press release by NOIRlab. More images and video clips are available on the .

Source

There’s still time to return your ballot for the 2026 general election, but if you’re planning to vote by mail, you may want to hurry. That’s according to a research brief Ěýby a team led by the ĚÇĐÄÔ­´´, which found that ongoing changes to U.S. Postal Service mail processes could double the percentage of ballots that arrive too late to be counted.Ěý

Researchers estimate the mail delays could cause tens of thousands of otherwise valid ballots to be rejected. These mail delays disproportionately impact rural voters and those in outlying suburban communities.Ěý

“This could affect 20,000 or more voters who put their ballots in the mail ahead of election day,” said research lead , professor of public policy and governance at the UW. “That may sound like a small number of voters in the grand scheme of things, but it could be really consequential, particularly in rural and outer suburban communities.”

The research brief — part of a led by the UW Evans School of Public Policy and Governance — has not yet been peer-reviewed. Researchers plan to submit it to an academic journal in the coming months.Ěý

State law requires every ballot postmarked by Election Day to be counted. That system relied on the Postal Service’s practice of postmarking mail on the date it’s received. But new policies first rolled out in October 2023 shifted how the agency collects and processes mail.

Under the program — part of an — mail in ZIP codes more than 50 miles from a regional center is collected only once a day. That mail may not be postmarked until it reaches a processing facility, which could be the next day or even later.

RTO was implemented in about two dozen Washington ZIP codes in February 2024 and has since been expanded to . Rural voters are particularly likely to live in an RTO ZIP code, including over 90% of rural voters who voted in the 2024 general election.

To project the delays’ effects on ballot rejection rates, researchers analyzed voter-level ballot data and county-level rejection data from the Washington Secretary of State. They found that between 2020 and 2025, about 11% of primary ballots and nearly 5% of general ballots were postmarked before Election Day, but not received until after. While counted in past elections, these last-minute ballots are those most likely to be impacted this year.

Researchers also found that the late ballot rejection rate has risen significantly since RTO was implemented in 2024. Across Washington, the county average late ballot rejection rate was 1.1% in the 2025 general election — about twice as high as the typical rate in general elections. In that election, the ballot rejection rate for voters in RTO ZIP codes was 3.2%, more than double the rate for those in urban areas.Ěý

It’s difficult to project exactly how many voters are at risk of having their ballots rejected in 2026, but Allard said he expected the number to rise.

“I think we’re at risk of having 1% to maybe even 1.5% of all ballots rejected for being late due to postal service delays, which would be unprecedented statewide,” said Allard, who is the Daniel J. Evans Endowed Professor of Social Policy.

He encouraged voters to consider using or putting their ballots in the mail at least a week before Election Day, which this year is Tuesday, November 3. Voters taking ballots to the post office also can ask their postal worker to postmark their ballot envelope on the spot.

“It shouldn’t fall only on the voters to solve this problem,” Allard said, “but there are steps we can all take to ensure our ballots are received on time.”Ěý

Co-authors include , UW doctoral student of sociology; , UW doctoral student of public policy and governance; and of the University of California, Berkeley.Ěý

For more information or to reach the researchers, contact Alden Woods at acwoods@uw.edu.Ěý

Source