Drool glinted in the sun as the cattle in a small enclosure on the McFarland White Ranch tossed their heads and, one by one, funneled into a steel chute designed to hold them still. Metal screeched as the gates opened and closed behind them. Some mooed as Harrison White approached, carrying a thick black polymer collar. 

White’s multigenerational family ranch lies a dozen miles away from tiny Two Dot, Montana, in the forested foothills where the Crazy Mountains slump into the Musselshell River Valley’s rolling plains. On this July morning, its cows were visiting an unusual kind of jewelry studio to be fitted with new, high-tech necklaces — GPS tracking collars that communicate with nearby base station towers. Via satellite, these stations connect to an app where ranchers like White and his sister, Lanie, can see the location of each collared animal. With a few clicks of a mouse, they can draw invisible fences. If a cow gets too close to one, her collar delivers a warning noise, then a shock.

After White buckled on her collar, the cow jumped out of the chute, bucking and swinging her head as she adjusted to its weight. In a few moments, she was nibbling nearby grass. 

About 620,000 miles of barbed wire and other fencing crisscrosses the West like lattices on a pie crust of an imprecise baker, dividing land and life. These fences enable people to dictate how public and private lands are grazed and how, or even if, migrating wildlife reaches its final destination. They are sometimes more trouble than they are worth, though. Just ask the ranchers who build and repair them.  

“Real fence is bullshit,” Lanie White said as she programmed the collars stacked in the bed of a nearby ATV. “It’s expensive, it’s damaging, and it fragments ecosystems.”  

The use of virtual fencing has expanded dramatically on Western land since 2021, when the technology became commercially available. More and more ranchers are using it to replace physical fences destroyed by wildfires, keep cows away from spawning fishor nesting birdsand direct cows to graze fire breaks.

The McFarland White Ranch was one of the first in Montana to try virtual fencing, starting in 2022 with one brand, Vence, before switching to another, named Halter. Vence was co-founded by a New Zealand livestock manager and partially bankrolled by a division of Shell, which wanted to use the technology to reduce overgrazing and increase carbon stored in the soil as part of the carbon offset programs it participates in. The pharmaceutical giant Merck’s Animal Health division acquired it in 2022. Halter also has roots in New Zealand, where the dairy industry’s practice of frequently rotating cows through small paddocks requires heavy investments in fencing.

Virtual fencing won’t replace barbed wire anytime soon; it doesn’t always work, and it’s hard to bet one’s livelihood on new technology. Yet the White family believes it can transform their ranch for the better, and they’re committed to realizing its potential. Virtual fencing, said Lanie White, is “the biggest ranching revolution that will occur in my lifetime.”

PHYSICAL FENCING HAS A LONG, tangled history in the U.S. In the early 1800s, colonists were drawn West by the promise of “free” land stolen from Native Americans and a belief in the new nation’s “manifest destiny” to control the continent. But the Great Plains lacked the wood and stone used for fencing back East, and when settlers tried stringing smooth wire between fenceposts, livestock pushed right through it.

By the late 1860s and early 1870s, the race to develop “spiked” wire fencing was underway. After Illinois farmer Joseph Glidden invented a new design in 1873, he patented his technique for attaching barbs to double-stranded wire as well as the machinery required to mass-produce the fencing. Glidden’s metallic imitation of a thorny hedge was cheaper and easier to string than anything else available, and sales exploded. Soon, settlers were slicing up the landscape. “So much of our fencing system in the United States is following section lines — the township and range system,” said David Augustine, a landscape ecologist with the Department of Agriculture’s Agricultural Research Service. “Almost every section has a fence on it, which rarely follows an ecological boundary.”

“Real fence is bullshit.”

Fencing has always been used to control both animals and humans; during the 20th century, armies defended trenches with barbed wire, while states use it to imprison people. “Control through violence and force, I think that’s the underlying basic meaning,” said Scott Abbott, a Utah Valley University professor and co-author, with Lyn Ellen Bennett, of The Perfect Fence: Untangling the Meanings of Barbed Wire. 

 Historians and scholars argue that barbed wire — along with railroads, plows and the telegraph — enabled the colonization of the West. “They were mechanisms of containment in a paradigm of domination,” said Dina Gilio-Whitaker (Confederated Tribes of the Colville Reservation), assistant director of the California Indian Culture and Sovereignty Center and a special adviser to the HCN board of directors. Fencing and the private-property ethos it upheld encouraged individualism. Together with the Dawes Act of 1887, which divided communal reservation land into individual plots, fencing facilitated tribes’ assimilation into a capitalist economy. 

Wildlife was also impacted, though managers weren’t always aware of it. Officials installed fencing along the northern boundary of Yellowstone National Park in the early 1900s to keep livestock out of the park and pronghorn inside it to protect them from poachers. (The population declined, and the fencing was removed.) But by the 1960s, concerns about pronghorn getting entangled and dying in fences began to appear in the scientific literature. In 1965, photos of pronghorn dying after allegedly being caught in Wyoming sheep fencing prompted Rep. Henry Reuss of Wisconsin to introduce legislation prohibiting the construction of fencing that impeded wildlife movement on public lands. (A Bureau of Land Management report at the time found that 4,804 miles of barbed wire and 1,340 miles of woven wire fencing had been constructed on public lands in the state.) Fiction reflected the shifting attitudes: In Edward Abbey’s 1975 novel The Monkey Wrench Gang, a river guide named “Seldom Seen” Smith says, “Always cut fence. That’s the law west of the 100th meridian. East of that don’t matter none. Back there it’s all lost anyhow. But west, we cut fence.” 

Despite environmentalists’ growing anger over the mistreatment of Western landscapes, Seldom Seen’s instructions were merely a wish list. There was no practical alternative to barbed wire fencing in the West — not until virtual fencing’s arrival in the West just four years ago. After more than a century and a half, could the dream of a fenceless West finally become a reality?

The idea of virtually controlling animals isn’t new. In 1973, “invisible fences” of buried wire, reinforced by shock collars, were patented for dogs. Researchers began experimenting with radio-wave activation in 1987. Scientists within the Department of Agriculture started designing virtual fencing for cows in the early 2000s, but it took more than two decades — and the efforts of overseas companies working with dairy producers — for virtual fencing to become commercially viable for the West’s beef cattle, which roam over large areas. 

The possibility that this new technology might serve both ranchers and wildlife has conservationists excited. The Nature Conservancy (TNC) is among the organizations offering financial support to Montana ranchers who implement virtual fencing. In partnership with the Southwest Montana Sagebrush Partnership, TNC is working with eight Montana ranchers who are using three different companies — Vence, Halter and Gallagher eShepherd. Jim Berkey, TNC’s High Divide Headwaters program director, sees the potential for a paradigm shift. If virtual fencing can ease the burden of physical fencing on wildlife, he envisions “vast potential changes across the landscape.”

FOR RANCHERS, VIRTUAL FENCING’S appeal increases with every wildfire. Fast-moving rangeland fires obliterate wooden fenceposts, destroying thousands of miles of astronomically expensive fencing that costs $15,000 per mile or more to replace. Virtual fencing costs vary, but collars (which are rented from the companies) and their batteries usually run about $50 per year, while brands that rely on base stations typically charge $10,000 per tower. Most base stations are easily relocated when pastures are changed; transmitting signals between a collar and a virtual fence requires about the same amount of cell coverage needed to send a text.

The economics of virtual fencing tend to favor large ranches, which rely on a lot of people to manage their herds and require many miles of fencing. Smaller operations that have less to gain may find it harder to invest in the necessary equipment. Ranchers may be able to share the upfront costs with nonprofits like TNC, the National Wildlife Federation or Rocky Mountain Elk Foundation; state departments of wildlife; private donors or family foundations; and, sometimes, federal agencies. Often, though, ranchers aren’t inclined to try it out until they’re faced with the even higher costs of wildfire recovery. 

The 2020 East Troublesome Fire near Granby, Colorado, spurred some of the first virtual fencing in the West and its first known use on a Forest Service allotment. A Natural Resources Conservation Service grant funded a Vence trial with six ranchers and close to 3,000 cattle starting in 2022. “At that time, nobody knew what virtual fencing was,” said Allegra Waterman-Snow, a district manager at the Eagle County Conservation District. “It was completely new.” 

More wildfires in 2022 prompted more ranchers to try virtual fencing. Dan Probert, an eastern Oregon rancher, saw 20 miles of fencing burn on the 20,000-acre Saddle Creek Forest Service allotment during the 2022 Double Creek Fire. Some of the allotment, which he’s permitted to graze, overlaps with the remote Hells Canyon National Recreation Area, and Probert estimated that the cost of flying or packing in the materials would have cost $20,000 to $25,000 per mile. “No one, including us and the Forest Service and a bunch of different wildlife groups, wanted to put the hard fences back in,” he said. 

Mike and Joy Wilson, who raise cattle in north-central Washington, have also lost physical fencing to wildfire, but they switched to virtual fencing when they leased 9,000 acres from the Confederated Tribes of the Colville Reservation, which had recently reacquired the parcel and hoped to help create a wildlife corridor between the Rocky Mountains and the Cascade Range. It had minimal fencing, and any new physical fencing risked disrupting the corridor. “With the ability to put virtual fences in, the wildlife never knows the fence is there, so it just made good sense,” Mike Wilson said. It also made sense financially: Neither landlord nor lessee wanted to pay for new fencing, and the nonprofit group Conservation Northwest was willing to fund two base stations.

Near Salmon, Idaho, the nearly 70,000-acre Diamond Moose allotment on the Salmon-Challis National Forest had minimal fencing before it burned in the 2022 Moose Fire. The Forest Service would have likely closed the entire allotment to grazing, perhaps for several years, while the vegetation recovered.

But when permittees Chyenne and Jay Smith rounded up their cattle after the fire, they noticed that the fire had scorched some forage but left large areas unscathed. They wondered if virtual fencing could be temporarily positioned around the unburned areas so they could continue to graze their cattle while the burned acreage recovered.

Melinda Ellison, an associate professor of range livestock research at the University of Idaho’s Nancy M. Cummings Research, Extension and Education Center, was already familiar with virtual fencing. She was actively seeking an opportunity to test it in the steep, rugged grazing allotments near Salmon when the Smiths approached her as the Moose Fire was being contained. Ellison teamed up with the couple on a last-minute hunt for funding and agency partnerships, eventually enlisting the Bureau of Land Management, Forest Service and Idaho Department of Agriculture. Ellison’s team has been working with area ranchers ever since, helped by grants from the Foundation for America’s Public Lands and the USDA.

It took time to teach the Smiths’ cows new routes — cows that stubbornly ignored the shocks were removed from the range — and the ranchers struggled to apply the technology to the landscape. In the beginning, Ellison and her students repositioned virtual fences on the Smiths’ behalf. “The software side of it is still moderately clunky, so you kind of have to be a little bit techie to be very proficient with it,” Ellison said. 

Hardware snafus included cows stepping on and damaging loose collars as they lowered their heads to graze; once, two bulls got stuck together when the carabiners on their collars interlocked. Other early adopters of virtual fencing faced similar problems, said USDA researcher Augustine. “The product was just not quite right,” he said. “It hadn’t been thoroughly validated on working ranches.”

The satellite research station’s cow-calf operation is now focused on virtual fencing’s practical applications. Three graduate students and one or two undergraduates from the University of Idaho work on the project, which spans nine ranches, 2,000 or more collars and 20 base stations.

Grant money pays for collar rental and batteries, and the Forest Service has purchased three base stations so far. Dwayne Rice, the agency’s Region 2 rangeland program manager, said that it prefers not to invest in collars. “We’re not in the livestock business, and we don’t want to be in the livestock management business,” he said. “That is the business of the rancher.” 

“Cows are kind of like people in that way; some are trouble-makers no matter how much you do.”

As of spring 2026, cattle on 270,000 acres of allotments within the Salmon-Challis were being managed with virtual fencing, said Kyle Nelson, rangeland management specialist for the Salmon-Challis National Forest. The Forest Service permits it as a temporary range improvement similar to a fence gate. “I think it’s been a great technology to implement,” Nelson said. “I see us continuing to use it.” 

Chyenne Smith’s computer is now integral to her morning routine. She starts the day at her kitchen table with coffee or tea and a map on her laptop showing her herd’s location. The arrangement of the cow-shaped icons on the screen can tell her whether they’re gathered around a water trough — a sign that the water is still flowing — or avoiding a salt lick, indicating that a replacement lick is needed. Rapid or unnatural movements may indicate an injured cow or the presence of a predator. Jay said that his and Chyenne’s time on the allotments is more targeted and purposeful, since they can see exactly where problems are occurring. “We wouldn’t like this if it wasn’t hands-on,” he told me.

Virtual fencing has made their jobs easier in other ways, the Smiths said. In the past, some cows would head to higher ground before the spring forage matured, while others lollygagged at lower elevations and missed out on peak forage, potentially violating the terms of their grazing permit. “We were able to fix both of those problems,” Chyenne said. Now, they use the technology to graze evenly or pick strategic areas that need resting. “It makes us better range managers,” Jay said. 

On a sunny spring day, I met a group of recently collared cows in one of the Smiths’ fields. Reddish clay painted the hills below the Beaverhead Mountains, where snow still clung to the peaks. The Smiths had placed a virtual fence near a physical fence so that the cows would learn to respond to their collars’ beeps and shocks as they would to a physical barrier. “Cows are kind of like people in that way; some are troublemakers no matter how much you do,” Chyenne said. The animals were hunkered down in bushes, but as we approached, they began to scatter. When they got too close to the physical fencing behind us, their collars began to beep. Slowly, they ambled off, returning to their pasture. They were quickly getting the hang of it.   

NOT EVERY TRIAL HAS GONE SO smoothly, though. The McFarland White Ranch near Two Dot, Montana, became one of the first and largest proving grounds for virtual fencing in 2022, when the Whites collared 1,800 cows with Vence technology. The Whites hoped that virtual fencing would help them contain their animals more efficiently, allowing them to more quickly reforest their property after the 2021 American Fork Fire. But shortly after they started, hundreds of collars started breaking and falling off — and with them, quarter-pound lithium batteries. Once the batteries died, the collars couldn’t be found. 

“I didn’t know it until it was a disaster,” said Lanie White, who worried that the batteries would rust, break open and leach heavy metals into the headwaters of the American Fork River and Big Elk Creek, both of which run through ranch property. The remaining collars needed new batteries every six months — more frequently than the Whites had been led to believe, they said. (Chyenne Smith, who is still using Vence collars, said they also lost collars initially but were able to find most of them. Now they lose only a few, she said.) According to Merck Animal Health spokesperson Jennifer Luddy, “System performance may vary by terrain, weather, infrastructure, grazing practices and frequency of use.” 

In February 2023, the Whites removed the remaining collars. Lanie White has since told neighbors and other ranchers about her negative experience; Vence declined to discuss details. “We take product performance concerns seriously and carefully review each case,” Luddy wrote in an email to HCN. Today, a heap of collars with dead batteries lies beside one of the Whites’ outbuildings. Another dozen are hung up on racks outside a barn — relics of a failed experiment.

In the fall of 2024, the Whites re-collared 350 cows. They used collars from Halter, and Lanie White said she noticed immediate differences. The cows calm down faster after being fitted with the new collars, she said, perhaps because they fit more snugly and don’t swing as much when the animals move.

The batteries in the Halter collars are charged by small solar panels, and the company claims each collar will last five to seven years — something the Whites and other ranchers appreciate. “Changing those batteries out is a pain in the ass. Solar is a huge improvement,” said Oregon rancher Probert, a Vence customer. Previous batteries, he said, drained quickly if cows interacted with fencing frequently, complicating rotational grazing in small paddocks. The newest Halter collars can transmit data directly to the Starlink satellite network, eliminating the need for expensive local base stations.

Behind the wire with a fence ecologist

The Whites have less institutional support than the Smiths, but they’re not on their own: Most of the current costs of their system are covered by the National Fish and Wildlife Foundation, and they’ve also received help from the Property and Environment Research Center (PERC), the Forest Service and the Rocky Mountain Elk Foundation. 

Representatives from Halter have helped the Whites troubleshoot their equipment by phone and in person, and in July, Morrison visited the ranch to help the Whites fit their herd with some 300 new collars. By the end of the month, he was scheduled to help ranchers across Montana and the Dakotas put on more than 8,000 collars.

After all the problems, I asked Lanie White why she continued to test virtual fencing. “We’re nothing if not tenacious,” she replied. If there’s a ranching revolution to be had — and she thinks there is — she wants to be part of it.

“I WOULD SAY VIRTUAL FENCING is here to stay, for sure,” said Paul Evangelista, a researcher who studies it at Colorado State University. Roughly 200 of the more than 900 Forest Service permittees in Region 2, which includes Wyoming and Colorado as well as the Great Plains, now use virtual fencing. 

The technology is spreading especially quickly in eastern Oregon, where the 2024 Lone Rock Fire destroyed 1,100 miles of fencing that would cost an estimated $25 million to replace. “It burned so much that ranchers truly did not have a place to put their cows,” said Dallas Hall Defrees, regenerative ranching program director for Sustainable Northwest, a Portland-based nonprofit that runs a regional regenerative ranching program. 

Private land made available for short-term grazing lacked fencing, rendering it basically useless to ranchers who needed to safely contain their cattle. But a project to install virtual fencing on 177,000 acres of private land across three counties in the Upper Rock Creek watershed is currently underway, led by Sustainable Northwest and its partners. Once complete, it will be the largest virtual fencing system in the state and likely the Pacific Northwest. “What we’re trying to do with virtual fencing is to have working lands and to do them better, to have rural economies, rural communities and ecosystems all function alongside each other,” Defrees said.  

“We wouldn’t like this if it wasn’t hands-on.”

More companies are entering the market. The initial players, Vence and Halter, have been joined by Gallagher eShepherd and a Norwegian company called Nofence, and the competition seems to be driving improvements. “The pace at which these companies are innovating and changing the technology to meet producer needs is happening pretty quick,” Waterman-Snow said. “I think virtual fencing is going to look pretty different in another five years.” 

Business for established brands is booming. At the Whites’ ranch in July, Halter territory manager Joshua Morrison rattled off numbers over a lunch of ground beef, potatoes and peas. He said that ranchers on 75 properties in his coverage territory, which stretches from eastern Montana to western North Dakota, had purchased 65,000 collars from Halter as of July 2026, and 40 more ranches in western Montana and Idaho were trying the technology. “It’s just blowing up,” Morrison said. 

While virtual fencing’s hardware and software improve every year, the technology still has its limitations, technical and otherwise. “It’s not a scalpel yet,” Defrees said. “It’s a steak knife.” Virtual fencing can’t prevent drought, invasive species or wildfires. And it’s not a shortcut to good land management. “We did not see that the simple fact of virtual fencing meant that your ground improved,” Waterman-Snow said. 

In the arid West, where thirsty cows will ignore beeps or shocks to get to water, ranchers who use virtual fencing must make sure their enclosed areas contain enough water and are a decent distance from streams and other appealing areas. “You can draw all the fences in the world with virtual fencing, but if you don’t have water, it ain’t no good,” said Koy Holland, a third-generation rancher in the Grasshopper Valley west of Dillon, Montana. 

Then again, not all ranchers, at an average age of 58, are comfortable with smartphones, much less an emerging technology with a still-clunky interface. “The learning curve is steeper for the humans than it is for the livestock,” said Tony Vorster, a researcher at Colorado State University.  

EVERYONE WANTS SOMETHING different from virtual fencing. Ranchers want to save money, graze more efficiently. Wildlife groups imagine a fence-free future where deer, pronghorn and even birds like grouse and owls can move unfettered. Virtual fencing companies see a largely untapped market. 

Ranchers are also intrigued by a new source of data on their cows. Some hope the GPS data from virtual fencing systems will help convince permitting agencies that they’re grazing their allotments responsibly and enable them to make more informed decisions about livestock numbers and grazing duration. Several ranchers said they hoped the data could protect them from the effects of lawsuits brought by conservation groups critical of public-lands grazing. “Each cow collects an astronomical amount of unbiased data about how they graze, how long they graze,” said Chyenne Smith. “It’s a really good tool if you’re not breaking laws and messing stuff up.”

Despite the dream of a wire-free West, virtual fencing has yet to bring about a meaningful reduction of physical fencing; the technology is just too young for ranchers or land managers to feel comfortable tearing down barriers. After less than two years of experience with virtual fencing, Holland chuckled at the suggestion that he take down his physical fencing. In his area, it costs between $17,000 and $20,000 per mile in labor and materials, and “it’s pretty hard to rip a fence down when it costs that much,” he said. “What we’re doing is throwing gates open.”

The Salmon-Challis National Forest, while further along in its virtual fencing experiment, also hasn’t removed physical fencing from its allotments. “It is definitely a discussion that we’re all kind of curious about, but I don’t think we’re quite there yet,” Nelson said. “I don’t necessarily think we’ll get to a point where we have no physical fences out there.”

What’s it like to be a cow?

Even ranchers in Eagle County, home to one of the nation’s first trials, aren’t confident enough in the technology, which only works 70% to 80% of the time in the area’s mountainous terrain. “That’s honestly just not quite enough assurance for them to be able to consider taking down fences,” Waterman-Snow said. Perimeter fencing, which keeps cows off roads and out of neighboring properties, will almost certainly remain in place for the foreseeable future. 

Virtual fencing may well eliminate the need for some new fence construction. But the West’s fencing legacy is proving harder to shake. As I left the McFarland White ranch, driving slowly on the bumpy dirt road, rain splattered my windshield. A few minutes later, I stopped abruptly at the sight of a woven-wire fence topped with barbed wire, a design that dates back more than a century and presents an impassable barrier to many species. 

Posted on the fence was a “Private Property, No Trespassing” sign. Directly beyond it stood a pronghorn with two fawns. While some Westerners were trying to do things differently, change was slow to catch on. I was looking at the past, the present, and the future, all within a few short miles.

This story is part of High Country News’ Conservation Beyond Boundaries project, which is supported by the BAND Foundation and the Mighty Arrow Family Foundation.

We welcome reader letters. Email High Country News at editor@hcn.org or submit a letter to the editor. See our letters to the editor policy.

This article appeared in the November 2026 print edition of the magazine with the headline “After the barb.”

Republish

Spread the word. News organizations can pick-up quality news, essays and feature stories for free.

Republish This Story

Republish our articles for free, online or in print, under a Creative Commons license.