CU Boulder joins interdisciplinary team of researchers aiming to understand the future of these imperiled regions of the world

The world’s polar ice sheets continue to shrink—and sea levels continue to rise—because of climate change, which can lead to coastal flooding and changes to the ocean’s circulation patterns. And though scientists have a lot of data about how ice sheets have changed in the past, they can’t easily use that data to predict how they’ll evolve in the future.

A new interdisciplinary project, funded by a five-year cooperative agreement from the National Science Foundation, aims to address that challenge by bringing together Arctic and Antarctic researchers, data scientists, computer scientists and other experts by creating a new institute, which will be housed at the University of Maryland, Baltimore County. 

By sifting through and analyzing extensive polar ice sheet data, collaborators working within the newly formed Institute for Harnessing Data and Model Revolution in the Polar Regions (iHARP) hope to be able to model the speed and intensity of polar ice melt over the next few decades to help inform policy decisions.

“Ice sheets are already changing very quickly in the current climate, and they’re expected to change more in the future climate, but how much and how fast, those are really the big questions,” said Jan Lenaerts, an assistant professor in the Department of Atmospheric and Oceanic Sciences (ATOC) at the University of Colorado Boulder and co-principal investigator for the project.

“That ultimately directly relates to a rise in sea level on a global scale, and that has huge implications for millions of people on earth who live along the coastlines.”

They also hope to better understand where that melting is likely to be the most severe.

“How does the East Coast of the U.S. compare to the West Coast compare to southeast Asia to Africa—who are the winners and losers of ice sheet mass loss and sea-level rise?” Lenaerts asked.

As part of the new institute, CU Boulder is joining forces with eight other institutions in seven states: University of Maryland Baltimore County, University of Alaska Fairbanks, University of California Irvine, Amherst College, Bowie State University, Universities Space Research Association, University of Minnesota Twin Cities and University of Texas Austin. Government agencies and private companies are also participating. 

The NSF awarded $13 million for the project, which is part of the federal agency’s Harnessing the Data Revolution Big Idea initiative. CU Boulder’s research team, which includes Lenaerts and Rajashree Tri Datta of ATOC, as well as Alison Banwell of the Cooperative Institute for Research in Environmental Sciences (CIRES), will receive around $1 million. The funding will also support graduate and undergraduate student research.

Right now, polar researchers struggle to make sense of the vast amounts of data collected by remote sensing instruments on satellites and aircrafts over the last two decades. These ice sheet observations are scattered across space and time, which makes it difficult to create a cohesive, continuous observational record, Lenaerts said.

“It’s chaos basically,” he said. “We need to stitch together all these observations and that’s very hard to do by humans. If we can feed this into some sort of huge algorithm, we can make more sense of the data hopefully. That’s where this project comes in.”

As part of the iHARP team, polar researchers from CU Boulder and other institutions will serve as subject matter experts, while data and computer scientists will offer their expertise in new techniques like deep generative adversarial networks, physics-informed machine learning, causal artificial intelligence, data assimilation and scalable algorithms. 

“This is a virtual environment that will hopefully facilitate lots of interdisciplinary science,” Lenaerts said. “This institute is really bringing together domain scientists—people like us who study ice sheets and polar climate change—with data scientists, or scientists who have no idea about ice sheets but know a lot about how to work with data in a meaningful way. We are giving them data, and they are giving us interpretations that we can then hopefully use to improve our science and the understanding of ice sheets in general.”