Sea ice offshore of Nome breaks up, February 2025. Ben Townsend/ KNOM

Researchers turn to Indigenous knowledge to improve Bering Strait sea ice forecasts

For subsistence users in the Bering Strait region, knowing when sea ice might break apart can be a matter of life and death. Researchers at the University of Michigan are developing a forecasting tool designed to predict those conditions, particularly near the shore, where traditional computer models often struggle.

John McClure is a research engagement specialist with the Cooperative Institute for Great Lakes Research (CIGLR). Last winter the CIGLR team traveled to Gambell, Savoonga and Nome to learn how it could further improve its tool. 

He said the team made it back home with plenty of fresh ideas. 

“Thinking about observations of the sun and the moon and clouds and animal behavior and ice color, those sort of things that aren't involved in some of those more technical models, but are very much so involved in how a lot of other folks navigate sea ice,” McClure said. 

The CIGLR group released a report in September summarizing their findings. One key recommendation was to better incorporate Indigenous knowledge into the forecasting model.

While the model relies heavily on satellites looking down at the ice from space, generations of subsistence hunters have studied conditions from the opposite perspective — looking up from the ice itself. From below, the color and appearance of clouds can reveal what’s happening on the ice miles away, offering clues that satellites and computer models don’t always capture.

Nome resident Ukallaysaaq Okleasik helped facilitate the workshops. He said local expertise can infuse the models with more realism, which also makes the end result more usable. 

“Seeing how that collaboration, really an underutilized strategy right now,  that could be a real advancement in the beta model and other research in the Arctic,” Okleasik said. 

The researchers say their model already does a good job predicting overall sea ice coverage. But conditions near shore are much harder to forecast.

Ayumi Fujisaki-Manome, the project’s lead researcher, said that’s partly because there isn’t enough reliable data to verify what’s happening in those areas.

“That might be where we could rely more on indigenous knowledge holders, like subsistence hunters, people on the ground,” Fujisake-Manome said. “They may be able to help with real-time verification of the model, where satellite data have limitations.”

Workshop participants also recommended incorporating more information about ocean currents, tides and water temperatures, along with adding Indigenous languages and terminology.

Work on the tool is ongoing, and the researchers plan to return to the region to ensure the feedback has been incorporated as expected. The project’s federal funding is set to end next July, but the researchers hope to host the tool on another website. They’re also exploring partnerships with agencies that could help maintain and improve the tool over the long term.

Did you enjoy this Climate & Environment/Science story?

Consider supporting our work by becoming a one-time or recurring donor.

Scroll to Top