Geography undergraduate researches satellite estimates of Antarctic meltwater

UNIVERSITY PARK, Pa. — As he prepares to graduate in May, James Rowley is completing an honors thesis focused on one of the most remote and rapidly changing places on Earth: Antarctica. A Penn State Schreyer Honors Scholar majoring in geography and economics with a certificate in geographic information science, Rowley is studying uncertainty in satellite estimates of Antarctic surface hydrology as an undergraduate research assistant in Penn State’s Cryosphere and Climate Lab, led by Luke Trusel, associate professor of geography.

The lab studies how polar ice sheets in Antarctica and Greenland are responding to climate change using satellite observations, climate models and artificial intelligence (AI)-driven approaches. Rowley works with Trusel and geography doctoral student Mahsa Bahrami on research related to supraglacial lakes, pools of meltwater that form on the surface of Antarctic ice shelves during the austral summer. In his thesis, Rowley compares satellite-based algorithms used to detect those lakes across Antarctica and examines the sources and magnitude of uncertainty in those estimates.

“Understanding how many supraglacial lakes there are in Antarctica is important because they may help explain how quickly ice shelves collapse into the ocean,” Rowley said.

Rowley began his research through Penn State’s Undergraduate Research Opportunities Connection, or UROC, program during his second year. He started with data annotation work connected to Bahrami’s research and later used those annotations to help train a deep learning algorithm. That experience grew into his Schreyer thesis, which compares automatic supraglacial lake classification algorithms and examines sources of uncertainty in the results.

He said the project has underscored both the importance and difficulty of studying Antarctica.

“What I’ve learned over the past couple years is how difficult it is to say anything with certainty about Antarctica,” Rowley said. “It is huge, it is far away and there is still so much we do not know.”

That uncertainty is part of what makes the work meaningful. Scientists are still working to better understand how meltwater on ice shelves may contribute to instability and collapse, and improving the tools used to detect and compare supraglacial lakes can help strengthen that understanding, according to Rowley.

For Rowley, the research has also changed the way he thinks about academic work.

“You realize pretty quickly that research is what drives everything,” he said. “Being in that space and seeing the work people put in day and night gave me a whole different level of respect for it.”

He said working in the lab added a valuable layer of learning beyond the classroom, giving him direct experience with the research process.

“Being in the lab showed me the difference between writing a paper for class and doing real-world research,” Rowley said. “In research, you cannot write a sentence without serious thought and serious literature behind it.”

He said the experience also shaped how he approaches questions, problems and his own role in the research process.

“While conducting research, you learn what critical thinking actually is,” Rowley said. “You learn to ask questions, accept being wrong sometimes and build confidence in your ability to contribute.”

This spring, Rowley presented his research at the American Association of Geographers (AAG) annual meeting, where he shared preliminary findings from his thesis and connected with others working in related areas of geography and remote sensing. One exchange stood out.

“One of the wildest moments was having someone come up while I was presenting and tell me he created the NDWI (Normalized Difference Water Index), which is foundational to the work I do,” Rowley said. “Meeting the person whose work underpins so much of my research was kind of insane.”

The experience also gave him a broader view of the research community.

“One thing that really surprised me was how interconnected the labs are,” Rowley said. “From the outside, you imagine everyone competing, but in reality, a lot of people know each other, stay in touch and work together. ... It feels almost like a big family, at least in the cryosphere. That was one of the most surprising parts for me.”

As he finishes his thesis, Rowley said he would encourage other students to explore research opportunities early, even if they are unsure whether research will shape their long-term plans.

“If you are going to get involved in research, start early,” he said. “When you are younger, it is a great time to spend even five hours a week helping with annotation or data processing and learning how research works.”

He said the experience can be valuable well beyond academia.

“Even if you do not stay in research, doing it for a semester or two makes you smarter, helps you build connections and gives you a better sense of how to read studies and think critically about them,” Rowley said.