Artificial intelligence is being deployed to help protect endangered Dolphin and Union caribou as they migrate across melting Arctic sea ice. In a new study published in Ecological Solutions and Evidence, scientists at the British Antarctic Survey (BAS), in partnership with The Alan Turing Institute, WWF, and the Government of Nunavut, unveiled a system that could provide early warnings to prevent disruptions by ice-breaking vessels.
The AI model forecasts when sea ice conditions are suitable for caribou to begin their autumn migration south from Victoria Island to mainland Canada—an annual journey that depends on the formation of stable sea ice.
“Many Arctic animals' life histories are intrinsically linked to sea ice,” said Dr. Ellen Bowler, lead author and BAS machine learning scientist. “As ice becomes less stable, AI tools can help local teams act fast to protect critical migration routes”.
The system builds on IceNet, a machine learning model developed by BAS and The Alan Turing Institute that predicts sea ice extents up to three months in advance. By integrating GPS data from caribou collars with satellite ice observations, the tool provides migration forecasts that can guide conservation efforts and short-term limits on icebreaker activity.
The project could extend to other Arctic species—including polar bears and walruses—offering a new AI-driven approach to mitigate climate impacts.
“This is about helping people and nature adapt together,” said WWF-UK’s Chief Polar Advisor, Dr. Rod Downie. “We need to urgently tackle the climate crisis and we need to plan for change so that people and nature can continue to thrive. AI-based conservation tools will have a role in helping us to achieve that”.