MANKATO — It’s like running a marathon, but for the brain.
Fueled by cans of Red Bull and coffee from little to no sleep over a 24-hour period at Minnesota State University, 250 students from universities across the Midwest proposed solutions to water quality problems facing Minnesota.
MSU hosted the 8th annual Midwest Undergraduate Data Analytics Competition Saturday and Sunday – drawing top students from leading universities in Minnesota, Iowa, Wisconsin, Illinois and other Midwestern states to crunch numbers and derive a story from that data.
Event chair and MSU Computer Science Assistant Professor Rajeev Bukralia said Mankato’s conference had the largest turnout yet, with 60 teams competing.
“The universities send their best of the best students,” Bukralia said. “The competition has a very high reputation, particularly programs that teach data analytics. Employers recognize it.”
Each year a different sponsor provides a data set pertaining to a real world issue. This year, the Minnesota Pollution Control Agency took the lead. When students arrived on campus Saturday morning, they had no prior knowledge of the data they would be analyzing. As they filled the Ostrander Auditorium at the Centennial Student Union, they were presented with hard data on Minnesota’s watersheds that make up the streams, creeks and rivers that flow into even larger rivers and eventually the ocean.
“This year’s competition is especially important because we were able to get a data problem that has a value for the whole state and the citizens of the state,” Bukralia said. “The competition is about analyzing data in such a way that you can tell an interesting story about the problem.”
When MSU was selected to host the event a few months ago, Bukralia approached Kimberly Musser, associate director of the Water Resources Center at MSU. She said the MPCA had troves of data on watersheds.
“It’s a nationally renowned example of a state that’s doing an excellent job doing long-term monitoring of their water,” Musser said. “I’ve talked to researchers in other parts of the country who say, ‘we love working in Minnesota because there’s such an excellent data set, tools and techniques.’”
Lee Ganske oversees the statewide river monitoring network for MPCA and provided the students with data related to more intense precipitation affected by climate change, and the state’s water quality; it was the student’s job to find patterns and potential solutions to that data.
“Our water quality issues in Minnesota are in many cases a function of how the land is used and also the natural characteristics of the landscape, the soils, the geology,” Ganske said. “In order to appropriately manage our rivers to protect them, we need to continue to build our understanding of what is it about the nature of watersheds that really influence water quality.”
Even with barely any sleep, students emerged from spending the night in an assigned classroom full of enthusiasm as they presented their findings to 90 judges, 50 of whom were industry professionals from big Twin Cities-based companies like General Mills and United Health – companies that have an urgent and unfilled need for data analysts who can make sense of increasingly complex statistics.
“These companies are looking for top talent, and they’re struggling,” Bukralia said. “They want to harness that data to find insights into it to create value and innovations to solve difficult problems. Their appetite for students who have those skills is huge.”
The students were able to present those skills in front of potential employers. Skills in computer science, math, science, and information technology are important, but equally important – and a skill the judges were focusing on – is the ability to think creatively and explain complex statistics in a way that’s understandable. That’s one of the most significant takeaways for MSU Senior Alycia Holwerda, who was competing in MUDAC for the third year with four other MSU student teammates.
“I learned how important it is to tell a story in data science,” Holwerda said. “When you’re talking to someone from a client perspective, they’re not going to know all the coding or statistics behind it. You want to be able to relate it in terms they’re going to understand, make it engaging and have visualizations that are intriguing.”
That approach paid off on Sunday, with the team winning the data visualization award. They used visuals to explain the correlation between nitrate levels, poor water quality and chemical fertilizers used for agriculture in watersheds in southern Minnesota not found in the forests of the north.
“People liked what we showed them and they understood it right away,” teammate and MSU student Nayeem Imtiaz said. “They found our model to be creative.”
Ganske said giving students a real world problem like diminishing water quality helps them identify the causes of those problems, and ultimately leading to potential solutions.
“That problem is largely a function of our increasing concern over the impacts of climate change on rivers and water quality in general,” Ganske said. “We’re asking students to delve into that a little more and see what they can find in terms of relationships there. That will ultimately shape the way management activities on the landscape play out. It will shape policy.”