
William Dunker knew where the deer was despite the thick bush and rough terrain on Afognak Island in Alaska. Even so, it was a difficult operation. The Sitka black-tailed deer on the Kodiak Archipelago just southwest of Anchorage were skittish around humans. If Dunker darted this one, it might run into a nearby waterway and drown as the tranquilizer took effect. Or it could run up a steep cliff and fall off, getting injured or even meeting its end.
Making matters worse, it was October, and the island was full of large Kodiak bears (Ursus arctos middendorffi) eager to put on fat for the coming winter hibernation. A tranquilized deer would make an easy meal for a hungry bear. So might Dunker himself.
Dunker, a biologist with the Alaska Department of Fish and Game, and his team were tranquilizing the animals so they could place GPS collars on them and learn more about their seasonal movement and foraging behavior. Once he was close enough to a deer, he would often make fawn distress calls in an effort to elicit a response that would reveal the deer’s location. But he was worried that predators may also home in on the sound. “He felt he was just bait, going into the field,” said Shannon Finnegan, who was working with Koniag Native Corporation at the time.
But despite being in the truck, Finnegan had an extra set of eyes in the sky—in the form of a thermal drone—watching for bears and guiding Dunker to the deer to dart.
“It just made the whole procedure more efficient,” Finnegan said.

Spooking Sitka
The Sitka black-tailed deer (Odocoileus hemionus sitkensis) population in the Kodiak Archipelago is important for both Indigenous and non-Indigenous Alaskans, who harvest the animals for meat and for sport. Scientists already know its population fluctuates each year based on climate conditions and other factors. Wildlife managers want to learn more about the best ways to respond to these population swings. For example, could they modify habitat to lower the impact of unsuitable weather years on the species?
But first, they needed to learn more about the population. Finnegan, Dunker and their colleagues published their findings in a study recently in the Wildlife Society Bulletin.
They set out by road in a truck on Afognak Island with the goal of fitting 25 individuals with GPS collars. They didn’t have much luck, initially—they couldn’t find many deer to target, and those they did see got spooked easily, taking off before Dunker could dart them.
“After three days, we were getting very disheartened,” Finnegan said.
But then, the team realized that they had a drone. Finnegan had an operator’s license, but she wasn’t super enthusiastic about its prospects. “We kind of fell into the drone world reluctantly on my part,” she said. “I thought they were a bit overrated.”
Nonetheless, the team gave it a try. “As soon as we started doing that, our whole operation changed,” Finnegan said.
Suddenly, they could see that deer were all around them—the thick vegetation found on much of the island just hid them. “There were a lot more animals near us than we realized,” she said.
Out of the 10 deer they initially collared for their study, the team ended up capturing eight with the help of the thermal drone.
“We’ve had a lot more success capturing the deer faster than they had been able to do in the southeast [of Alaska],” she said.
Drone advantages
The team fitted a quadcopter—a drone with four helicopter-like propellers—with a thermal camera, so not only could they spot deer, but they could also be selective on the type they wanted to target for their study. For example, they could focus in on adult females, pregnant females or adult males. They could take a GPS signal of the deer and guide Dunker on the ground via radio as he closed in. Finnegan could also tell him to get the dart ready once he was close enough, even if he couldn’t yet see the target.
Once Dunker was close enough, he conducted fawn distress calls, which would occasionally elicit a response. And after he took the shot, Finnegan could follow which way the deer ran. Usually, wildlife managers wait 10 minutes before approaching the animal after the dart hits, but sometimes the animal goes down more quickly. By following it via drone, Finnegan could notify Dunker exactly when it was safe to approach.
Drone tracking could also help in other situations. Deer will sometimes run into a river or up a steep cliff, which could present serious danger for the animal if they pass out in a deep waterway or fall to their death. While that didn’t happen in this study, Finnegan said that drone operators could warn people on the ground if something like this may be happening. The tracker in the field could then try to scare the animal off in another, safer direction or get to the animal before it drowns.

Having eyes in the sky could also help keep the team safe from bears. Finnegan recalls that Dunker would sometimes ask five or six times if there were any bears as he homed in on a target deer. From the air, she could sweep the area while the team conducted its fieldwork on the sedated deer, watching for predators until the deer came back to consciousness.
Finnegan said that thermal drones might have other applications in wildlife management, too. It could help track problem bears that enter towns in Alaska, for example, making it safer for the professionals who set out to subdue them. She also points to efforts in Oregon where wildlife managers have been using drones to haze wolves (Canis lupus) away from livestock.
She was hesitant to use the tool in the beginning, but she hopes that, like her, others will get more comfortable with the gear.
“That mindset is starting to change,” she said.
This article features research that was published in a TWS peer-reviewed journal. Individual online access to all TWS journal articles is a benefit of membership. Join TWS now to read the latest in wildlife research.