The scenes start out innocently enough, often with a springtime stroll through Alaska’s Nelchina River Basin. But without warning, things turn grim: tableaus of blood and gore, usually with an unlucky caribou calf at the center.
Such is the video footage collected by scientists over 3 years from cameras strapped around bears’ necks, offering the first “bear’s eye view” of life in this bucolic but harsh reserve. One of the team’s main findings: These bears kill a lot more than we think they do. A whole lot more.
“It was really exciting because it’s the kind of thing you know occurs,” says Christopher Brockman, a wildlife biologist with the Alaska Department of Fish and Game (ADFG) in Palmer and lead author of the study.
Figuring out the “kill rate” of large carnivores is notoriously difficult, says study author Bruce Dale, a wildlife science coordinator at ADFG in Palmer. It’s hard tracking them for long periods over large distances, and it’s also easy to miss a kill. A brown bear might spend a couple days eating an adult moose, compared with just 40 minutes for a caribou calf. So preferred tracking methods—like twice-daily aerial observations—can’t capture the full scope of bears’ hunting and feeding habits. They also can’t capture individual variation in kill rates, resulting in wildly different estimates for entire populations.
To come up with a more accurate estimate, Brockman and his colleagues decided to try something new: outfitting the bears with camera collars and GPS trackers. Similar to security cameras, the collars filmed 10-second clips every 5 to 15 minutes for more than a month, from mid-May to late June. “We were focusing on the period of time when the [moose and caribou] calves are most vulnerable to predation,” Brockman says. He and his team collared 17 bears in total from 2011–13.
Just seven collars provided the researchers with decent data—the others either fell off, didn’t work properly, or were chewed off their mothers by meddlesome bear cubs. But the ones that did work gave the team more than 100 hours of footage—36,376 clips in total. From those, the scientists reconstructed the bears’ days. Most of their time was spent resting or traveling (60.5% and 21.3%, respectively), and just 6.3% was spent feeding. There were even a few instances of bear mating caught on camera, Brockman says.
By carefully analyzing the most gruesome footage, the researchers were also able to identify the bears’ prey. More than half of their meals came from moose or caribou calves, whereas vegetation made up nearly 20%, and adult moose made up just over 12%, they report this month in the Wildlife Society Bulletin. But there were also some unusual items on the menu: snowshoe hares, swans, and even other brown bears. In one case, a 10-year-old male killed—and ate—a 6-year-old female bear.
Overall, the bears killed an average of 34.4 moose and caribou calves over 45 days. That’s far higher than average kill rates from previous studies using other methods, including aerial observation. Compared with one 1988 study in which scientists counted an average of 5.4 moose calf kills from the air in a different part of Alaska, the new study found an average of 13.3 moose calf kills. The new study also found wide variation in the number of calves killed by any one bear, with one killing 44 calves in 25 days and another killing just seven in 27 days.
That matters because kill rates are often used to manage wildlife living in protected areas. For example, if too many calves are being killed by bears, then removing a few of the predators could have a big impact on allowing the moose and caribou population to increase. Alternatively, Brockman says if the management goal is to increase bear numbers in an area, it may be important to pay attention to whether there are enough calves around in the spring to support more of the predators.
Still, the study has its weaknesses. The low numbers make it hard to draw any firm conclusions about predation rates, says Martin Leclerc, a Ph.D. student at the Université de Sherbrooke in Canada who has studied how female brown bears use human settlements to shield their cubs. He also points out that the bears chosen in the first year of the study were known calf killers and could thus represent a bias in the overall numbers.
But the use of camera collars could greatly improve knowledge about population dynamics in the future, Leclerc says. “The technological development will really help biologists and ecologists to have a more precise understanding of predator-prey relationships.”