Monitoring whale migration from space

Whales often swim deep below the water’s surface, but a lot can be learned about their migration routes and behaviours by watching them from hundreds—or even thousands—of kilometres above the Earth’s surface. Making use of high-resolution satellite technology, Whale Seeker is working with WWF and other collaborators to pilot the use of high-resolution satellite imagery for monitoring Arctic whale migrations. As EMILY CHARRY TISSIER explains, collaborations like these are essential for enhancing marine conservation efforts in the face of growing threats from climate change and ship traffic.

For millions of years, whales have embarked on epic seasonal migrations to and from the Arctic in search of areas that are rich in food and that offer ideal temperatures and conditions for mating and calving. The backdrop to these voyages was once pristine, icy water and a cacophony of natural sounds—from the eerie creaking of ice to the clicks, whistles and songs that marine mammals use to communicate.

But as the climate crisis has heated up, the backdrop has shifted, and is now also characterized by melting ice and rising industrial activity. The accompanying hazards—from ship strikes to underwater noise pollution—are making whales’ migrations more difficult.

Yet their migrations are critical, not just for the whales’ own survival, but for the marine ecosystem and the cultural heritage of Indigenous Peoples, who face threats to their cultures, livelihoods and food security as the environment transforms.

Studying whales from space

Whale Seeker is using advanced satellite imagery and machine learning to test the utility of high-resolution satellite imagery for monitoring whales and safeguarding their future.

To detect the whales, Whale Seeker uses a combination of high-resolution satellite images and a proprietary artificial intelligence (AI) platform trained to recognize the unique shapes and shadows of whales near the surface of the water. These images, captured from hundreds of kilometres above Earth, are filtered to identify areas where whales are likely to be found. Then, machine learning algorithms scan the imagery for visual patterns that match known whale characteristics, such as body length and fin shape.

This process dramatically reduces the time and effort required to detect whales in expanses of ocean that are too large to survey from a plane, drone or boat. These tasks once took days or months of manual work to complete, but can now be done in a matter of hours. The result is a rapid stream of data that we can share with marine biologists, environmental regulators and shipping companies.

In Arctic regions where remoteness or ice cover limits access for human observers monitoring from boats or small aircraft, this satellite-based approach opens new possibilities for understanding and protecting marine mammal populations with minimal intrusion.