Grizzlies on the Go
As Grizzly Populations Grow and Their Territories Expand, One UM Researcher is Hunting the Precise Pathways They Take to Move Between Ecosystems
By Jacob Baynham
In the fall of 2020, a 4-year-old male grizzly killed some chickens at a property in the foothills of the Flint Creek Range southeast of Drummond. To prevent it from becoming habituated to humans, Montana Fish, Wildlife and Parks captured the bear, put a GPS collar on it, and released it in the Blackfoot Valley, at the southern edge of the vast Northern Continental Divide Ecosystem, home to more than 1,000 grizzlies.
The bear, nicknamed Lingenpolter for the community near where it was found, didn’t stick around for long. It soon made a beeline for the location of its capture. There was one major obstacle: Interstate 90. According to its GPS data, Lingenpolter made multiple attempts to cross the busy thoroughfare that fall, each time retreating, seemingly to scout for a better spot. In a map of its movements, the bear looks to be bouncing off the interstate. Eventually it stayed north of I-90 and wintered in a den in the Mission Mountains.
Next spring, Lingenpolter woke up and tried again. Finally, after more than 40 failed attempts, the bear crossed I-90, likely beneath a bridge. Once across the interstate, Lingenpolter returned to the Flint Creek valley and the Anaconda Pintler Wilderness, not far from the Bitterroot Ecosystem where no grizzlies are known to live full time. Grizzly bears are highly individualized in their behavior, but Lingenpolter’s movements illustrate the lengths these animals can travel between ecosystems and the hurdles they can cross along the way.
“There’s lots of evidence that bears are exploring new areas to settle, and evidence that their range is continuing to expand,” says Sarah Sells, assistant unit leader for the Montana Cooperative Wildlife Research Unit at the University of Montana.
A year before Lingenpolter’s capture, Sells was finishing her Ph.D. in wildlife biology at UM and thinking about a project for her post-doc. Sells had studied bighorn sheep and wolves and was interested in research with a conservation impact. So she went to Hilary Cooley, the UM-based coordinator of the Grizzly Bear Recovery Program for the U.S. Fish and Wildlife Service, and asked her about the biggest unanswered question in grizzly conservation.
Cooley says although much was known about the four populations of grizzlies within six designated ecosystems across Montana, Wyoming, Idaho and Washington, little was known about the routes bears might choose to travel between those ecosystems. Dispersal between ecosystems is an important part of grizzly recovery to avoid isolated “island populations” that could suffer from in-breeding and genetic defects.
In 2021 Sells designed a study and set to work. Building on previous research that modeled connective pathways for male grizzlies from the Greater Yellowstone Ecosystem, Sells selected 65 collared grizzlies — 46 female and 19 male — in the Northern Continental Divide Ecosystem with a collective dataset of GPS locations that spanned 20 years.
For the first phase of her research, Sells and her team used actual GPS data from bear collars to observe how each bear moved across the landscape, selecting for variables such as greenness of vegetation, terrain ruggedness, proximity to the forest edge, density of riparian areas, density of buildings and distance to secure habitat. The collars gave a fix for a bear’s location about every three hours. Between each fix, Sells could see how far the bears traveled and the angles of their turns.
Grizzly bear habits differ from one individual to the next. Using all the data, Sells built movement models for each bear and then ran repeated simulations that predicted their habitat use across the ecosystem.
“It’s kind of like a video game world where you have all these habitat layers built up,” she says. “You can think of dropping a bear into that world and asking them to use their movement model to select steps. They keep walking along as their model helps them identify what habitat responses they would have in real life.”
Using those simulations, Sells determined which variables the bears were selecting for when they moved. She applied this data to a color-coded map where dark blue parts indicated high-use areas and red parts indicated low use. Sells then compared this map to a larger dataset of grizzly GPS fixes and found that the simulations effectively predicted habitat use.
In the second phase of her study, Sells tested if her models of grizzly movements in the Northern Continental Divide Ecosystem could predict movements of bears in other ecosystems, too. She compared her habitat-use models with hundreds of thousands of GPS fixes on more than 200 collared bears in the Cabinet-Yaak Ecosystem, the Selkirk Ecosystem and the Greater Yellowstone Ecosystem.
The results were clear. “These models performed really well even though they were for bears in completely different areas,” Sells says.
Now that she knew how to predict habitat use in one ecosystem and that those models transferred well, Sells turned to the third phase of her study: mapping the pathways grizzlies might take when traveling between ecosystems. To do this, she applied her models of habitat selection to the territory that separates western Montana grizzly populations. She then simulated both directed movements, where a grizzly was put on the landscape and asked to walk to a selected endpoint in a neighboring ecosystem, and undirected movements, where a bear had no defined endpoint. Sells conducted hundreds of simulations for each bear and plotted them on maps.
The simulations showed that grizzlies would favor mountainous routes and river valleys to travel between ecosystems. While there’s no way to immediately test the accuracy of these pathways, Sells has mapped 127 verified outlier grizzly sightings since 2010, which match up nicely with the pathways she predicted. Over time, as more data and outlier sightings become available, Sells and her team will be able to measure the accuracy of their findings.
Sells is gratified that her research is already informing decisions in grizzly bear conservation. Montana Fish, Wildlife and Parks is one agency putting the findings to use.
“It helps us in terms of identifying areas we might want to protect as bear habitat or movement habitat,” says Cecily Costello, one of the study’s co-authors and a research wildlife biologist for Montana FWP. “We can predict where we really need to have people working to help pave the way for bear presence. It also helps us with outreach and education with the public.”
Costello has studied black bears and grizzlies for more than 30 years. “The maps really formalized a lot of what we suspected,” Costello says. “But they also have that independent credibility that allows us to say there’s science behind this. This is where we think it’s most likely that bears are going to move.”
In preparing for the expansion of grizzly bear territories, and possible dispersal between populations, Costello says humans will need to learn to coexist with grizzlies by securing their chicken coops, garbage, beehives and other attractants.
“We always want the bears to love the wilderness,” she says. “That’s where they’re most secure and separated from us. But we can’t guarantee that that’s exactly where the bears want to be.”
Indeed, private land is key to grizzly conservation, says Jim Williams, partnerships manager and wildlife biologist for the Heart of the Rockies Initiative, a group that works with land trust agencies to protect private land from future development. Private land makes up 50% of the area between grizzly recovery zones, and many animals — not just grizzlies — use it for migration corridors.
“As different species face different precipitation and temperature patterns, they sometimes need to move to find food,” Williams says. “In order to move, they need space. Most of these spaces are private, either ranches or farms.”
Williams already has used Sells’ research. The Gallatin Valley Land Trust was working with a property owner near the Tobacco Root Mountains and the Highlands Range, a prime connective area between the Northern Continental Divide Ecosystem and the Greater Yellowstone Ecosystem. The property owner wanted to protect the land from future subdivision. Using Sells’ predicted pathway maps overlaid on the property, Williams was able to find conservation-minded donors willing to finance an easement on that land.
“These projects not only are good for bears,” Williams says, “Sarah’s paper is also helping keep working agricultural families on the land.”
Another way Sells sees this research helping grizzlies is identifying places to install wildlife crossings on the interstates and highways between ecosystems. Sells already is looking into possible wildlife crossings in the Bitterroot Valley and on I-90. Such measures would protect drivers and help bears like Lingenpolter safely cross busy motorways.
“Bears are hit on roads quite frequently,” Sells says. “It’s a behavioral deterrent. It’s something that’s scary, probably, to them. When they do try, they don’t always make it.”
Sells says she’s found her dream job at UM and hopes to continue her research into the unanswered questions of wildlife conservation, both with grizzlies and other species. She says her three-phase study to predict grizzly movements could be applied to other threatened species around the globe.
In her office in the Natural Sciences building, Sells keeps skulls of Pleistocene carnivores, including the extinct giant short-faced bear, on a bookshelf above her desk. Her windows look out into the Missoula Valley and Hellgate Canyon, once prime grizzly habitat, and a place where the bruins may, slowly, be starting to return.
“What’s interesting about bears,” she says, “is they can be around people and you’d never be the wiser. That’s a great thing. When bears can be in the vicinity and not be seen and causing problems, that helps with co-existence.” •
