Jia Yu, assistant professor in the School of Electrical Engineering and Computer Science, in collaboration with researchers at the University of Maryland has received a $1.25 million National Science Foundation grant to develop a data platform that will help researchers who are studying the vast and poorly understood Arctic to better manage their data.
The Arctic is critically important in understanding climate change, but it’s also a great example of a big data challenge, says Yu.
The region is undergoing rapid change with rising air and sea surface temperatures and decreasing snow and ice cover. The Arctic also affects the rest of the world — its snow-covered, white surfaces reflect light and act as the Earth’s air conditioner, and Greenland’s melting ice sheets are important to the understanding of sea level rise around the world.
Because of its remoteness, though, researchers often have to rely on satellites to monitor climate variables and to conduct their science, which means they can end up with overwhelming amounts of data. They also collect varying data from a variety of sources. To analyze the terabytes or petabytes of data has meant that scientists have had to access expensive super computers to do their work, said Yu.
“A lot of people are working on polar science right now,” he said. “The Arctic is so far from us, but it affects the entire planet. It’s a really important problem and I hope our work will help to introduce some promising outcomes.”
The WSU and University of Maryland team are developing a low-cost data platform that will allow researchers to link their own computers into a cluster rather than using super computers to more easily analyze the geospatial data in the Arctic.
Their algorithm will be able to load data from many heterogenous sources, creating a uniform system that is more useable for researchers. At the same time, their system will run analytics, cross validating data from different data sources and finding coincident data from different sources to help the scientists find correlations. Finally, the algorithm will use data mining to automatically detect geospatial patterns in the data.
Yu’s team includes a diverse group with a variety of scientific backgrounds, including a computer scientist, oceanographer, climate scientist, glaciologist, and an ecologist. He hopes to build a community of users for the software and make it easy for the scientific community to use. He will be presenting the project at the upcoming American Geophysical Union conference in early December.
