Skip to main content
The team stands outside the Millennium Sciences Building
science-journal

Penn State team wins Global Frog Discovery Challenge

9 September 2022

A group of Penn Staters placed first in the Global Frog Discovery Challenge, a competition that attracted over 9,000 participants across more than 100 countries. Referred to officially as "the Challenge," the program is part of the Better Working World Data Challenge, sponsored by the Ernst and Young Global Limited organization (EY).

Frogs were the focus of the competition because frogs are an indicator species — a go-to for scientists researching the environmental health of ecosystems. Species distribution models are among the most widely used ecological environmental regulation and conservation tools worldwide. Participants in the Challenge were tasked with developing the best possible methodology, given available public data for determining where specific frog species would be found. Improved methodologies could yield real improvements in global efforts to combat ongoing losses in biodiversity.

Penn State’s winning team consisted of Fuhan Yang, Thu Tran, Chiara Vanalli, and Emily Howerton, all doctoral candidates in the department of biology, along with Weiming Hu, machine learning postdoctoral researcher at the Center for Western Weather and Water Extremes, Scripps Institution of Oceanography, University of California, San Diego. Hu graduated from Penn State in 2021 with a doctorate in geography.

Image
The team stands outside the Millennium Sciences Building
The Sweet Frogs Team of Emily Howerton, Fuhan Yang, Chiara Vanalli, Thu Tran, and Weiming Hu.  Credit: Penn State. 

The Penn State team entered the competition under the whimsical moniker "Sweet Frogs," a name appropriate not only for the Challenge's goal of accurately predicting frog species' location, but also for this particular group of friends, who enjoyed frequenting a local Sweet Frog frozen yogurt establishment together.

"We thought [the Challenge] sounded interesting — like a riddle — and we thought it might be fun to do over the summer," explained Yang.

Team Sweet Frogs didn't expect to win; in fact, they entered the Challenge just a month before the competition closed. Additionally, the group shared that it took weeks to download the massive data sets consisting of field data, satellite imagery, and complementary geospatial data from NASA and other resources needed to start to try out their models.

In the first round of the competition, wherein all participating teams’ models were reviewed and scored against existing models, Sweet Frogs scored well enough to move to the semi-finals.

For the next round, teams had to write a report and submit a brief video explaining their model and results. To the team’s surprise, after a panel of judges from the EY, NASA, Microsoft, and other scientific entities weighed in, Sweet Frogs came out on top.

This video was submitted to EY's Global Frog Discovery Challenge by the winning "Sweet Frogs" team of graduate students from Penn State University. It explains the team's unique approach for using publicly available data to accurately predict the locations of various frog species that serve as key indicators of biodiversity within specific ecosystems. Credit: The Sweet Frogs Team.

"This experience emphasized the value of good-quality data, what we can learn from such data sets using quantitative methods, and the importance of storytelling and communicating our work," Tran commented. "I would love to continue applying my skill set and contributing to meaningful interdisciplinary projects like this one and those in my Ph.D. program at Penn State."

"This experience reinforced my dedication to using machine learning and data science for environmental problems and conservation," Hu shared. "They provide a set of powerful tools to analyze the large volume of publicly available data. I'm super excited to see its successful application to frog detection and conservation."

"It does not matter how complex the problem is; if we bring people together, a solution is possible,” added Vanali. “This is the case for many of the challenges our world faces, like habitat destruction, climate change, and food insecurity. In my future career, I would like to apply my computational and analytical skills to contribute to 'a better working world.'"

Howerton shared, "How exciting it can be to work on meaningful, real-world problems in a collaborative environment. This insight is important as I explore what comes next for my career."

Yang agreed. "It is rewarding to know how our work helps to answer real-world questions."

The team will receive a cash prize $6,000, and their winning data model will be available free of charge to support efforts by researchers, ecologists and governments.

Maciej Boni, associate professor of biology at Penn State and graduate adviser to Sweet Frogs' team members Yang and Tran, was thrilled by their team’s success.

"This entirely young scientist-led effort shows us the promise of giving grad students independence at an early age,” Boni said. “Coming in first in an international contest like this is a huge accomplishment. I am incredibly proud of my students and all the students involved."

The Challenge is part of EY's commitment to innovate and use technology to address some of the world's largest environmental and climate change problems. The annual event is open to students and early-career data scientists worldwide.