Experts available: LA protests, LLM hallucinations, hurricane season, wildfires and air quality, and more
The Virginia Tech media relations office has the following experts available for interviews this week surrounding issues in the news. To schedule an interview, please contact mediarelations@vt.edu.
Virginia Tech experts available to discuss headlines in the news
LA unrest prompts questions about state, federal power
As protests against ICE that began over the weekend continue in Los Angeles, the federal government continues to send more uniformed personnel, adding 700 Marines to join the 2,000 National Guardsmen already deployed without the consent of California Governor Gavin Newsom. Karen Hult, a scholar and expert in the presidency and U.S. federalism, can help explain the authority of the state and federal government when it comes to these actions, as well as the history of such actions in the past. [Available starting Tuesday afternoon.]
President Trump’s “big beautiful bill” faces Senate pushback
A Trump-endorsed bill advancing tax cuts and border security that passed the House of Representatives is facing pushback in the Senate. Political scientist Karen Hult can discuss the significance of the bill’s progress, what is likely to change, and how long it might be before the bill returns to the House. Read more here. [Available starting Tuesday afternoon.]
Are AI responses truly impossible to predict?
Recent news coverage asserts that even the companies that developed large language models (LLMs) don’t understand the choices AIs make, such as sometimes inventing fictional answers to questions. Computer engineer and machine learning expert Walid Saad said the truth is more nuanced. “It is generally challenging to ascertain how the output of LLMs will behave, but this concern is rapidly becoming alleviated. First, there are a number of rigorous mathematical approaches to provide what we call ‘explainable’ or ‘trustworthy’ AI. Second, we can design AI algorithms that understand the data and thus make choices based on reasoning that can be predicted and interpreted.” [Available Thursday and Friday.]
An above-normal hurricane season
The National Oceanic and Atmospheric Administration has predicted an above-normal hurricane season in 2025. Weather expert Stephanie Zick is available to discuss the upcoming hurricane season and what weather models have predicted. In addition, geophysicist Manoo Shirzaei can discuss his research into the growing vulnerability of high-income nations to flooding, and sociologist Liesel Ritchie can discuss her research examining the impacts of storm damage and hurricane preparedness issues.
Wildfires and air quality
With meteorological summer upon us, wildfires in Canada have already started burning, sweeping smoky air across the eastern United States, and the National Oceanic and Atmospheric Association’s summer weather forecast predicts a hotter, more humid summer in the mid-Atlantic. Air quality expert Gabriel Isaacman-VanWertz advises the public to keep an eye on the Environmental Protection Agency’s AirNow map to look at the Air Quality Index (AQI) and the Fire and Smoke map to stay aware of whether outdoor air quality is safe during these extreme weather events.
Tips to avoid ticks this summer
Everyone heading outdoors to enjoy the warmer weather needs to be aware of ticks and the serious illnesses they transmit. In Virginia, common ticks like the blacklegged or deer tick, the lone star tick and the American dog tick carry diseases such as Lyme disease and Rocky Mountain spotted fever and can pose other dangers. Emergency medicine physician Stephanie Lareau shares tips on prevention and treatment of tick bites. “I’m seeing the impacts of tick-borne diseases,” she said, “and I urge everyone to take preventative measures.”
Research Highlights
Researchers use deep learning to predict flooding this hurricane season
The 2025 hurricane season is forecast to be more active than ever, with potentially devastating storms whose heavy rainfall and powerful storm surges cause dangerous coastal flooding. Extreme water levels — like the 15 feet of flooding Floridians saw during Hurricane Helene in 2024 — threaten lives, wash away homes, and damage ecosystems. But they can be difficult to predict without complex, data-intensive computer models that areas with limited resources can't support.
A recent study published in Water Resources Research by civil and environmental engineering graduate student Samuel Daramola, along with faculty advisor David F. Muñoz and collaborators Siddharth Saksena, Jennifer Irish, and Paul Muñoz from Vrije Universiteit Brussel in Belgium, introduces a new deep learning framework to predict the rise and fall of water levels during storms — even in places where tide gauges fail or data is scarce — through a technique known as “transfer learning.”
The framework, called Long Short-Term Memory Station Approximated Models (LSTM-SAM), offers faster and more affordable predictions that enable smarter decisions about when to evacuate, where to place emergency resources, and how to protect infrastructure when hurricanes approach. For emergency planners, local governments, and disaster response teams, it could be a game-changer — and could save lives. More here.
Groundbreaking study maps the movements of marine megafauna
A sweeping new study published recently in the journal Science is helping pinpoint where whales, sharks, turtles, and other ocean giants need the most protection, and where current efforts fall short.
Led by Ana Sequeira of Australian National University and supported by the United Nations, the research synthesized data from 12,000 satellite-tracked animals across more than 100 species. It reveals how marine megafauna move globally and where their migratory, feeding, and breeding behaviors intersect with human threats such as fishing, shipping, and pollution. Virginia Tech contributed to the effort, called MegaMove, a massive collaboration of nearly 400 scientists across more than 50 countries. The project used biologging data collected using satellite tags to inform a new blueprint for ocean conservation.
“This is one of the largest marine tracking data sets ever assembled,” said Francesco Ferretti, a marine ecologist at Virginia Tech who contributed to the study. “It’s not just about drawing lines on a map. We need to understand animal behavior and overlap that with human activity to find the best solutions.” More here.
