Flying cameras are giving biologists an all-encompassing view of migration that reveals how social interactions motivate the animals’ every move.
Ecologists Andrew Berdahl, a Santa Fe Institute fellow, Colin Torney of the University of Glasgow, and colleagues flew drones to capture footage of Dolphin and Union caribou, a Canadian herd, as the animals crossed from Victoria Island to the Canadian mainland in the last stage of their fall migration.
Scientists have long pondered the dynamics of animal migrations, but they’ve had limited ways to study them. “Tracking collars revolutionized ecology by allowing researchers to quantify animal movement in a rigorous way,” says Berdahl of the widely used GPS and radio-collar technology. But GPS doesn’t capture the entire herd nor the dynamics between animals—providing only a snapshot, rather than a full-picture view.
A Better View
In order to truly understand migration in detail, he says, “we also need to know the social context driving movement decisions. And for that we need to see the behavior of [more] of the herd at the same time.”
So in 2015, the team put cameras in the sky. They captured 12 hours and 40 minutes of footage of the running caribou herd (in short clips, limited by drone battery life). They then fed the information into a computer vision program that identified the unique caribou in each frame, then linked the animals’ positions across frames to get their individual trajectories. This allowed the scientists to see each animal’s behavior in relation its neighbors—visualizing what were “essentially traffic rules for caribou.”
The footage revealed great variation in the interactions within the herd—who paid attention to whom, who stuck close to whom, and whose moves were most influential on others—illustrating the importance of age, sex, social status, and reproductive status in the animals’ movement choices. For example, bull males were the least likely to take cues from others, while young calves tended to be copy-cats.
It also showed that caribou were more influenced by herd members in front of them than by those beside or behind them. This unequal influence supports traditional Inuit knowledge that as herds travel, a subset of lead animals effectively determines the path of the group. Presumably, Berdahl says, more experienced animals take up front positions that let them teach younger individuals the migration route.
“It’s very exciting to be able to use new technology to obtain quantitative information about large animals [on the move],” says Iain Couzin of the Max Planck Institute for Ornithology in Germany, an expert on collective behavior not involved in the study. “These are very difficult data to get. Drones are opening up a new opportunity for us biologists, giving us access to systems that are incredibly hard to study.”
The team’s methods and findings are reported today in Philosophical Transactions of the Royal Society, B.
The study authors see their technique as a means to examine how social information is used at different life stages, how this behavior varies throughout the year, and whether some individuals are persistent leaders or followers.
“Every individual step an animal takes is a decision,” says Berdahl, that depends on social cues in concert with environmental ones such as obstacles and incentives along the path. He hopes that zooming in on those cues can offer clues to the motivations and pressures that drive caribou movement, plus insight into the leadership dynamics of herds—especially as some are forced to forge new paths in a rapidly changing environment.
Meanwhile, there’s great potential for drones to gather additional layers of information—such as high-resolution, 3-D images of the migration landscape—to put the animals’ behaviors into even more detailed context. “These interactions don’t happen in a barren environment,” says Couzin, who, in his own research, looks forward to taking “a hybrid approach, using different technologies at different scales to get a more comprehensive picture” of animal movements in natural settings.
“This is a developing methodology that could be applied to many animal systems,” Berdahl adds. “I’d be tempted to say it will revolutionize studies of this kind.”