AI Accurately Predicts Bird Migrations

Scientists at Cornell and UMass Amherst team up to track feathered flights.

Cornell’s eBird online project lets bird watchers around the world report data on bird sightings. Researchers now have a massive amount of data on where to find specific birds at specific times of the year, but much less is known about the actual movements in between.

In a recent study, scientists at  Umass Amherst and Cornell’s eBird team have developed an AI-based tool called BirdFlow that predicts migration patterns based on eBird data.

The team trained an AI with the eBird data, then used data from another team of scientists to verify that the predictions were accurate. The project used the Bridges-2 supercomputer at the Pittsburg Supercomputing Center to help manage the massive amounts of data.

According to project co-leader Adriaan Dokter, “We’ll be able to unravel the routes that birds take, from their breeding grounds to stopover points to wintering grounds and back without having to capture birds and attach tracking devices. Understanding these connections is critical to learning why some populations are doing poorly and some are doing well.”