BUFFALO, N.Y. — Jacob Tristan was enjoying fireworks at Niagara Falls when he realized the answer to a research mystery. 

The Texas A&M University senior had been noticing a recurring seismic signal around the Niagara River every night at about 10 p.m. as part of the summer research program at the University at Buffalo’s Center for Geological and Climate Hazards.

“It was while watching the nightly fireworks display that I was like, ‘Wait a minute. This could be what we're seeing in the data,’” Tristan said.

It was exactly the type of discovery that the Research Experience for Undergraduates (REU) program at UB is designed to foster. Supported by the National Science Foundation and the State University of New York Chancellor’s Summer Research Excellence Fund, the nine-week program allowed 15 undergraduate students from UB and across the country to use Western New York as a living laboratory to explore climate, atmospheric and geological challenges. 

Other research projects with a local focus included studying what drives lake-effect snow and improving long-distance cycling along the Erie Canalway Trail.

“We want to make as many of the research projects as tailored to Western New York as possible because we want students to think about the place that they live in,” says Sophie Nowicki, PhD, director of the Center for Geological and Climate Hazards, SUNY Empire Innovation Professor in the UB Department of Earth Sciences and a core faculty member at the UB RENEW Institute. “We’re trying to bring down research to something that's not abstract, but something that you can smell, touch and feel.”

Tristan and other students in the REU program presented their research at the inaugural summer poster session and reception hosted by the center and the College of Arts and Sciences. 

His project involved deploying seismic sensors to better understand how local geology can affect earthquake shaking, a phenomenon known as site amplification. Earthquakes in the eastern United States can cause shaking over much greater distances than comparably sized earthquakes in the West, in part because seismic waves can travel more efficiently through the East’s older, less fractured rock.

He and his advisers placed seismic sensors along the Niagara River, as well as on the UB North Campus. They expected sensors near Niagara Falls to detect weaker ground vibrations overnight, when some of the river’s water is diverted upstream of the falls to U.S. and Canadian power plants, but were surprised that the nightly fireworks were powerful enough to show up “beautifully” in their data. 

“You could see each individual firework being shot during the event, as well as a practice firework several minutes before the show,” Tristan said.

The chance to use Niagara Falls itself as part of his research made the experience all the more memorable. 

“Being so close to Niagara Falls and doing seismic research on the falls, it’s really cool,” he said. “It's not something that you get to do every day.”

UB geology major Alexandra Avellaneda wasn’t as close to her research subject but got a lot out of it all the same. She used satellite radar imagery and a deep-learning artificial intelligence model to analyze crevasses on Greenland's Sverdrup Glacier over an 11-year period. However, the coverage of large cracks in the ice varied greatly from year to year, with no clear long-term trend. 

“I felt I did all this work and then there were no results,” she said. “But when I talked to my adviser, she gave me so many other things to look at and I realized that no, this is valuable.”

Avellaneda also sharpened her knowledge and experience with modeling. She’d worked in the Glacier Modeling Lab the previous two semesters, but this was her first experience training her own model.

“I used to have no interest in models, coding or anything like that, but this summer I did a lot of it and felt super motivated,” she said. 

UB computer science major Mridul Wadhwa was looking to combine computer science with geology. He got that opportunity through modernizing a model used to predict river temperatures, eliminating cumbersome data-conversion steps and making it easier for researchers to use. 

Wadhwa used AI to assist with tasks such as testing and modifying code but found that human oversight remained essential. 

“You have to know how to use it correctly and not fully rely on it … but I think integrating AI with computer science can really make your workflow better,” he said.

Wadhwa said he thinks the project — as well as learning from professors, exchanging ideas with other students and attending workshops — will prepare him for graduate school. 

“We were getting out in the real world and learning new things that we didn’t know about,” he said. “It was really helpful.”