Songbirds socialise on the wing during migration
by University of Illinois at Urbana-Champaign

The nocturnal sky is alive with the presence of migrating songbirds, soaring in the millions along routes shaped by evolutionary history. However, recent research from the University of Illinois Urbana-Champaign suggests that these migratory paths may not be entirely instinctual. Analysis of over 18,300 hours of recorded flight calls indicates that songbirds may communicate with other species during migration, establishing social bonds and potentially sharing information about their journey.
The study titled "Social associations across species during nocturnal bird migration" was published in Current Biology on January 15.
"We cannot definitively ascertain the content of their calls, but it is possible that birds emit sounds while flying to indicate their species, age, and sex. It is reasonable to hypothesize that these calls may also pertain to navigation or the identification of suitable resting locations," stated Benjamin Van Doren, the lead author of the study and an assistant professor in the Department of Natural Resources and Environmental Sciences at the College of Agricultural, Consumer and Environmental Sciences at Illinois. Van Doren initiated this research at the Cornell Lab of Ornithology.
Previous work by the study's co-authors at the University of Maryland Center for Environmental Science, Appalachian Laboratory, indicated that birds tend to associate with other species at stopover locations during migration. However, until now, there was no evidence to suggest that different songbird species interact or communicate vocally while in flight.
While Van Doren acknowledges that innate patterns and memory continue to play significant roles in migratory behaviour, he advocates for a re-evaluation of songbird migration from a social perspective.
"In recent years, there has been a growing acknowledgment of the significance of social information in bird migration, yet most research has focused on species that migrate during the day or in family groups," he remarked.
The social context appears to play a significant role in species such as hawks and storks, which gather in large numbers during their daytime migrations. Young birds acquire behaviours by observing their peers and their navigation techniques, rather than solely from their familial connections.
However, these visual signals become unavailable at night, which is when most songbirds migrate. This observation prompted Van Doren to explore alternative social cues. He was fortunate to have access to acoustic recordings of autumn nocturnal bird migrations collected from 26 locations over three years in eastern North America.
"These nocturnal acoustic recordings provide a unique insight into this largely unseen yet immensely significant movement of birds - hundreds of millions flying over the U.S. on any given night during migration," Van Doren remarked. "This phenomenon often goes unnoticed as it occurs while we are asleep."
The extensive 18,300-hour acoustic dataset would have posed significant challenges for processing and analysis prior to the advent of AI. However, a machine learning tool enabled Van Doren's team to swiftly identify the distinctive flight calls of 27 species, including 25 well-represented songbirds.
After identifying the species, the team analyzed the frequency of specific calls occurring together over intervals of 15, 30, and 60 seconds. They discovered that the associations between species were stronger than would be expected by random chance, regardless of the time interval examined.
In seeking to understand these associations, the researchers found that the wing lengths of the species and the similarity of their calls were the most critical factors. Conversely, birds that formed bonds during stopovers did not maintain those connections while in flight, nor did they necessarily travel with closely related species or those that preferred similar habitats.
"Species with comparable wing sizes were more inclined to associate, as wing length is directly related to flight speed. If two species have similar wing structures, they are likely to fly at similar speeds, making it easier for them to remain together," Van Doren explained.
It is conceivable that the vocalizations of certain species have evolved to resemble one another over time due to social connections, or that species producing similar calls are more inclined to associate with one another.
Van Doren observes that the 25 species studied represent only a small fraction of the songbird species that migrate at night, with some not vocalizing during their flights at all. He and his research team intend to conduct further investigations, which will include attaching miniature microphones to individual birds to monitor their interactions with flight companions during migration.
Nevertheless, these initial findings prompt several thought-provoking, albeit speculative, ideas. For instance, short-lived songbird species that cannot depend on parental guidance may instead rely on social bonds with others to navigate their journeys. Furthermore, the alarming decline in avian biodiversity due to climate change and habitat destruction could threaten the partner species that migrate together.
"This research challenges the long-standing belief that songbirds migrate independently, solely guided by instinct," Van Doren remarked. "Understanding the implications of these social relationships - not only for migration but also for other biological aspects - will be crucial in addressing the risks they encounter in an evolving environment."
16 January 2025
Share this story