Who’s flying through your neighborhood?

Using citizen science and machine learning to predict bird migration

By
Heather Kamins
A flock of birds in flight, silhouetted against a pink and blue sky

Even the most casual observer can see the changing population of birds outside their window from season to season. And as it turns out, casual observations can add up to a much bigger data pool for researchers to use in predicting when and where birds will migrate—knowledge that helps with everything from conservation efforts to (human) aviation safety.

Accurately predicting where migratory birds will go next has proven to be one of the toughest jobs in biology, but UMass computer scientists have taken on the task in collaboration with biologists at the Cornell Lab of Ornithology. Thanks to their eBird project, researchers have been able to gather data from hundreds of thousands of people around the world who collectively contribute over 200 million bird sightings per year.

eBird trend map of Massachusetts and surrounding areas for the ruby-throated hummingbird

A sample trend map from the dynamic eBird tool

“eBird data is amazing because it shows where birds of a given species are every week across their entire range,” says Dan Sheldon, professor of information and computer sciences, “but it doesn’t track individuals, so we need to infer what routes individual birds follow to best explain the species-level patterns.”

That’s where computer science comes in. Sheldon and his team used the eBird data to create a new predictive model capable of accurately forecasting where a migratory bird will go next. The model, BirdFlow, provides ecologists with data they have not been able to access previously, and will eventually create visualizations and other educational material for the general public.

“Birds today are experiencing rapid environmental change, and many species are declining,” says Benjamin Van Doren, a postdoctoral fellow at the Cornell Lab of Ornithology and a co-author of the study announcing the new tool. “Using BirdFlow, we can unite different data sources and paint a more complete picture of bird movements, with exciting applications for guiding conservation action.”

Get involved!

We’re on the lookout

Share your most intriguing nooks, niches, coordinates, or curiosities on campus or anywhere in the region. Email magazine@umass.edu and we’ll investigate!