ASU AI project analyzes big data to help analysts find solutions to Arctic warming

Author
Published

May 7, 2023

photo by Andreas Weith, CC BY-SA 4.0, via Wikimedia Commons

The Arctic is facing a climate crisis that’s threatening the region, its people and the rest of the world. And while solutions to this crisis are available, like many parts of the polar region, they are just out of reach.

For years, satellites and drones have collected an avalanche of scientific data from even the most remote and unexplored areas of the Arctic. But the problem is that there is too much information, and it’s almost impossible to interpret.

One Arizona State University professor hopes to change that.

In August, Wenwen Li and her partners were awarded a $1 million research grant to help scientists learn to use artificial intelligence to address the pending disaster in the Arctic. Li is the principal investigator on the project.

“The problem in the Arctic is so urgent,” said Li, a data scientist trained in computer science and earth system science at ASU. “We need to resolve it as soon as we possibly can.”

Big data

Li, a professor in ASU’s School of Geographical Sciences and Urban Planning, describes the daunting scale of relevant scientific information as “the big data” challenge.”

That may be an understatement.

NASA’s climate change data repository is expected to have 350 petabytes of data by 2030 — the equivalent of about 10 billion single-spaced typewritten pages per day. And the satellite company MaxarTechnologies has more than 125 petabytes of global imagery. (That’s comparable to the total number of letters delivered by the U.S. Postal Service in 25 years.)

And that’s just part of the problem.

Another element is that the data gathered is often incomprehensible to the scientists that need it most.

The computers currently in use by the Arctic science community don’t have the capacity to secure and interpret data on that scale, and doing so is essential for them to solve the serious problems the Arctic is facing.

“Data alone will not help,” said Li, who directs the Cyberinfrastructure and Computational Intelligence Lab on ASU’s Tempe campus. “We need the ability to analyze huge amounts of data and receive useful scientific information from it. This process needs the support of artificial intelligence. But many Arctic scientists do not have the skills needed to work with artificial intelligence.”

Li and her partners will use the research grant to develop a cyber training program for Arctic scientists, and those from other disciplines, to access, study and ultimately interpret these complex volumes of data with the use of artificial intelligence.

The project, titled “Cyber 2A: CyberTraining on AI-driven Analytics for Next Generation Arctic Scientists,” runs from March 2023 through February 2026.

“The cutting-edge methods of using artificial-intelligence-driven analytics will give Arctic scientists the opportunity to make exciting new discoveries about what is really happening in the Arctic,” Li said.

The undertaking is a collaboration between Arizona State University; the University of California, Santa Barbara; University of Illinois, Urbana-Champaign; and the Woodwell Climate Research Center. The team is made up of experts in cyberinfrastructure, high-performance computing, artificial intelligence and Arctic science.

The polar problem

Since 1979, climate change and warming throughout the Arctic are taking place at a rapid rate: four times that of other parts of the world, according to a report from the Finnish Meteorological Institute.

And there is no part of the planet untouched by this warming.

As the ice melts throughout the 177.6 trillion square-feet of the region, there is a trickle down effect. Infrastructures once securely built on permafrost, a frozen layer of stable soil, are now sinking.

The unstable infrastructure impacts everything from the economy to the ability of animals — polar bears, walruses, Arctic foxes, caribou — to hunt and retain their habitat.

As the ice disappears, so does its reflective powers, allowing the sun’s energy to enter and be retained by the planet. The melting ice also releases methane, a greenhouse gas, that leads to more global warming. And on and on it goes until it seems that the frozen frontier may soon be facing its final years.

Predictions in the journal Nature Climate Change say that by 2040, there will be no more ice in the Arctic. And because the Arctic plays an important role in moderating the global climate, it will have dire consequences for the rest of the planet.

Scientists, geologists and others studying these problems cannot keep up with the thaw across the global Arctic and the vast amount of unanalyzed research data associated with it. The result is that informed decisions for policymakers and quick actions are impossible to make.

Team takes on the challenge

That’s where Li and her team come in.

The grant will help scientists trained in AI-related analytics to accurately predict real time changes and ultimately find solutions to global warming and climate change in the Arctic.

The free training will be in-person, online and through a monthly webinar series and is open to both scientists and educators.

An Arctic-AI research network will be established for sharing ideas and resources. And all training materials will be deposited in the Arctic Data Center’s Learning Hub to ensure access to scientists, professionals and developers in the Arctic science and geoscience domains and beyond.

Inclusivity is a major part of the effort. A recruitment plan is underway to create a strong and diverse STEM research workforce providing opportunities for minorities, economically disadvantaged groups, women, members of the Arctic Indiginous community and more.

“The cybertaining training grant will empower a new generation of Arctic researchers and leaders in utilizing all the data that is being collected across the Arctic,” said Anna Liljedahl, co-principal investigator for the project from the Woodwell Climate Research Center. “Not just a handful of data processed close to a decade after it was collected, which is how much of the science is currently operating.”

And ultimately, Li and her peers hope to do their part in restoring the environmental health of a piece of the planet.