There are more than 50,000 deer-vehicle collisions in Virginia each year. One local scientist has a low-cost solution.
May 3, 2013. It’s 7:50 in the morning. A 51-year-old man driving an SUV west on I-64 collides with a deer. The man is unhurt; police notify VDOT that they’ll need to remove the carcass.
October 13, 2014. Evening rush hour: 5:35pm. A 28-year-old woman is driving east on the same highway. From the police report: VEH#1 WAS IN THE RIGHT LANE WHEN A DEER CAME FROM THE LEFT AND STRUCK THE VEHICLE. THE DEER THEN FLED THE SCENE.
November 1, 2017. At 1:31am, two women have just driven over the Mechums River. A deer flashes into their headlights right before they feel the thump of the collision. The driver, who’s in her 60s, is fine, but her 21-year-old passenger sustains a “visible injury.”
July 26, 2017. Westbound lanes, just after midnight. VEHICLE 1 SWERVED TO MISS AN ANIMAL IN THE ROADWAY, LOST CONTROL, CRASHED INTO THE BANK AND ROLLED OVER. Visible injury to the 29-year-old driver.
Spend a little time looking through VDOT’s crash data, and it soon becomes obvious: A substantial portion of the mishaps—both minor and serious—occurring on Virginia’s roads involve a certain hoofed mammal. Odocoileus virginianus, the whitetail deer, causes enough crashes to earn its own code in the police reports. Alongside “1. Rear End”; “3. Head On”; and “8. Non-Collision” there’s “10. Deer.”
“Deer are not wired to learn to cross a road,” says Matt Knox, who leads the deer program for the Department of Game and Inland Fisheries. “They’re programmed to get away from wolves and panthers, but cars are a predator they haven’t evolved to deal with.”
It’s a problem for all parties involved. There are more than 50,000 deer-vehicle collisions—or DVCs, as they’re known to VDOT—each year in Virginia. Many of those result in death for the deer, and VDOT spends time and treasure dealing with the carcasses. For drivers, deer are a hazard to property as well as to life and limb. While Virginia isn’t one of the very worst states for DVCs (those honors go to West Virginia, Montana, and Pennsylvania), it’s still in the top dozen. A driver in Virginia has a one in 99 chance of hitting a deer, according to State Farm, which tracks deer-related insurance claims nationwide. Fatal crashes caused by deer are rare, but not unheard of.
The busiest time for DVCs is right around the corner: fall, which is rutting (mating) season for deer. Bridget Donaldson, a scientist with the Virginia Transportation Research Council—that’s VDOT’s research arm—has spent almost four years tracking deer activity and collisions along a stretch of I-64 west of Charlottesville. Of all the DVCs recorded during her study period, more than half occurred during October and November. During the rut, says Knox, “Their activity goes up dramatically, and that means they’re crossing roads.”
Donaldson (who’s based in Charlottesville) found that in the stretch of I-64 she studied, there were about nine deer-related crashes per mile per year. “That’s high,” she says. “That number can be decreased.” Fewer crashes would, obviously, benefit wildlife: not only deer, but bears and innumerable smaller animals. But how to prevent collisions, when our roadways cut right through forests and other prime wildlife habitat? Could there be a way to mitigate what seems like an unfortunate fact of modern life?
Deer in the road
In the fall, after bucks have shed the velvet on their antlers and dispersed from their chummy summertime “bachelor groups,” they find themselves restless and driven, propelled by the highest testosterone levels they’ll have all year. Does, meanwhile, have weaned the fawns they birthed in the spring and, as the days grow shorter, prepare to come into estrus, meaning they’re ready to breed. Searching for mates, both sexes crisscross the landscape more frequently than at any other time of year.
While bucks travel alone in autumn, does continue to move in social groups. Led by an alpha female—usually their mother or grandmother—younger does and fawns stick together, following the alpha whether her decisions are good ones or not.
“She controls when and where they move,” says Knox. “If she crosses the road, those other animals are going to cross regardless of traffic, because she’s the leader. If they’re crossing a bridge or overpass and the oldest doe jumps to her death, they all jump to their death.”
For drivers, this means that where there’s one deer in the road, there are often several—sometimes as many as a dozen. That increases the chaos factor during a road crossing, making it more likely that an encounter will end in a collision.
It doesn’t help that deer are neither nocturnal nor diurnal, but crepuscular: most active at dawn and dusk. Unfortunately, in the fall, those are often the same hours that people commute to and from work. While DVCs can happen at any time of day, the overlap of deer and human “rush hours” means there are that many more opportunities for fatal contact.
George Bragg, the third-generation owner of Bragg’s Body Shop in Charlottesville, is used to seeing cars come into his shop with deer damage, and recently, he’s noticed a trend. “Traditionally, we used to see a lot of deer hits in fall and winter,” he says. “Nowadays, in the last two to three years, it’s a consistent problem year-round.” He chalks up the increase in spring and summer collisions to Charlottesville’s steady pace of growth. “I attribute it to construction and development,” he says.
Donaldson says that the more fragmented forested habitat are for deer, the more often they’ll get hit—and maybe not for the reasons you’d assume. It’s not that deer need large swaths of unbroken forest, which human activity interrupts. It’s that they actually prefer edge habitats: places where forest meets field, or a creek breaks up the landscape—or a road cuts through the woods. Of that last type, there are more and more, as human developments extends its fingers into the landscape. “Deer love that edge, and use rights-of-way as part of their habitat,” Donaldson says. “They like early successional vegetation,” the plants that tend to grow back first when a forest is cut down. “They thrive where humans are, because we break up the habitat.”
Fencing them out
As more and more places see development, deer become used to the sights and sounds of roads and highways, walking and feeding along roadsides with no true understanding of how dangerous they are. “Deer raised near highways become habituated,” Donaldson says. “They haven’t evolved to understand the speed” of motorized vehicles.
This explains what sometimes looks to drivers like very stupid behavior—the classic “deer-in-headlights” situation, or the fact that deer sometimes run directly into the sides of vehicles, or leap off embankments onto a vehicle’s roof.
What deer can understand, as gardeners well know, are eight-foot-high fences.
When Donaldson joined the VTRC in 2003, she brought with her an interest in wildlife ecology, and she knew that, especially in the West, some roads were constructed with underpasses specifically for use by animals. Those are effective but expensive, and very tough to retrofit. “I thought maybe we could make better use of existing underpasses,” she says. Her idea was to study the effectiveness of roadside fences in funneling animals toward spots where they can safely cross under the road. If those underpasses were already built, VDOT would save the cost and hassle of new construction.
Donaldson identified two spots on I-64, between Charlottesville and Crozet, that could serve as wildlife crossings. One is ideal for deer: a tall, wide space under the Mechums River bridge with lots of visibility.
The other, located just east of the Ivy exit, is less inviting. It’s a “box culvert”—a concrete tunnel—that was built along with the interstate, back in the 1960s, to accommodate cattle on a farm the new highway had bisected. The culvert is 189 feet long and dark. From one end, the other side looks like a small square of sunlight framed by blackness. Donaldson knew from reading other studies that a culvert like this would not be a preferred crossing for deer, who hesitate to pass through confined spaces. But would they use it if they were forced to, by fencing?
Donaldson and her colleagues started out by studying deer activity near the two underpasses for two years before fencing was installed. In 2013 she placed game cameras every tenth of a mile, for a one-mile stretch centered on each underpass, and collected images of animals at the roadside: deer feeding, deer attempting to cross, even deer mating and sparring. Her cameras recorded more than 4,700 deer visits to the two sites.
As she’d expected, deer were more willing to walk under the bridge than to brave the box culvert. Also as she’d expected, they frequently crossed the road at both sites, regardless of having a safer passage available.
A mile of fencing went in at the box culvert in 2016, and another mile at the Mechums bridge in 2017. The fence is pretty simple: eight feet high, made of woven wire attached to metal stakes, with occasional “jumpouts” that let deer escape if they get trapped on the wrong side.
Simple it may be, but it seems to be working. In the mile surrounding the box culvert, the number of DVCs dropped from 16 (in two years of study) to just one in the two and a half years since the fencing went in. And the annual number of deer crossings through the culvert went from 148 to 745—a 505 percent increase.
“Now they realize they have to use it,” says Donaldson. She’s excited that the fence has done exactly what it was supposed to—channel deer to the safe crossing, even though it’s not what their instincts tell them to prefer. And it’s safe to assume that, at the two sites, it’s prevented more than 30 DVCs. Images of does bringing their fawns through the culvert suggest that a new generation of deer will learn to use it as part of their normal movements.
The post-fencing study at the Mechums bridge site won’t be fully complete until 2020, but so far it looks just as promising—no DVCs at all, and a marked increase in deer using the underpass.
Too many deer?
Donaldson’s study, she says, “is the most comprehensive evaluation of adding fencing to existing infrastructure. It’s the first time we’re heavily studying the effects of this.”
She plans to share her findings with other road ecologists around the country, and one thing she’ll emphasize is how cost-effective such a strategy can be. The two sections of fencing cost around $300,000 to install, and maintenance has so far been paid for by the VTRC’s research budget.
Wildlife conservation isn’t VDOT’s main priority here, says Donaldson. (Earlier this year, Virginia’s General Assembly failed to pass a bill that would have required studies of habitat corridors, to better understand how deer and other animals move through the landscape, and provide protection for those corridors. Several other states have passed similar legislation.) While deer do benefit from the new fencing, human safety and protection of property are more central to VDOT’s mission.
It’s hard to know the true cost of the roadkill problem to the state and to individuals, since many DVCs never get reported to the police. But for those unlucky drivers who do make contact with deer, Bragg says, it’s not unusual for the cost of vehicle damage to get into the four-figure range. “It doesn’t take much to do a lot of damage,” he says. “Vehicles today are made with crumple zones and a lot of plastic, and the sheet metal these days is very thin to make cars lighter. A good hit from a 200-pound animal can render $2,500 worth of damage to an average car.” State Farm reports an even higher figure: $4,341 per claim on average.
Bragg has seen vehicles totaled when deer have damaged roofs and windshields and set off airbags. “That can total a brand-new car,” he says. “We’ve seen that happen at less than 500 miles.”
If habitat is becoming ever more fragmented, and the number of roads and drivers in our area is increasing, deer and drivers will inevitably continue to encounter each other with great frequency. Seeing—and being nervous about—deer while driving is part of what drives the public’s perception that there are too many deer in the environment.
Matt Knox says that overpopulation of deer is a myth, or at least half a myth. “There’s two ways to look at this,” he says. “The first way is from a biological perspective. If you have so much food out there on the landscape, it can carry so many deer. From that perspective, the deer population is not overpopulated. The deer are healthy.”
But the biological carrying capacity is a different yardstick than human feelings. “The cultural carrying capacity is people’s tolerance for deer,” he explains. From that perspective, many people do believe there are just too many white-tails. Recreational hunting is the major means of limiting their population. In Virginia, hunters kill nearly 200,000 deer each year. Charlottesville has hired wildlife pros to cull 125 deer annually for the past two years on city property.
Dealing with the mess
Along with the fact that deer mow down people’s gardens and landscaping, deer-vehicle collisions are a major reason that many humans see them as a nuisance. There’s the danger and cost of hitting a deer, of course, but there’s also the unpleasant experience of seeing mangled carcasses on the roadside.
It may be unfair that people move into deer habitat, endanger them with our vehicles, and then profess disgust at seeing the bloody results. But from VDOT’s perspective, deer carcasses are another possible hazard to drivers, and removing them is just part of its charge. That task costs the agency more than $4 million annually.
VDOT and its contractors sometimes pick up carcasses as part of regular maintenance runs, and sometimes in response to calls from motorists. More than 55,000 carcasses must be dealt with each year. Most of these, including those picked up in Albemarle, go to landfills. But that solution is becoming more expensive, says Jimmy White, a colleague of Donaldson at the VTRC.
“As the technology in landfills has progressed over the years,” he says, “[adding carcasses] messes up what’s going on in the landfill.” This makes landfills less willing to accept dead deer; some charge high disposal fees when VDOT brings in carcasses.
In 2013, Donaldson and White studied several different systems that VDOT could use to compost deer at its own facilities. They tried composting in rotating drums, in windrows made of wood chips, and in forced-air compost bins. The forced-air bins worked best—they destroyed pathogens, broke carcasses down efficiently, and didn’t take up much space.
The half-dozen forced-air compost facilities installed during that pilot study are still in operation, says White, and are scattered around the state. Eventually he’d like to see more of them built. At a local VDOT headquarters outside Lynchburg—essentially a large yard where snowplows, mowers, and other maintenance equipment is stored—White shows off a set of compost bins that can break down hundreds of carcasses at a time.
It’s all very plain-looking: four concrete bins under one large shed roof. Each is about 10 feet wide and 18 feet deep, and has a swinging metal door across the front that opens to allow front-loaders to add animals and turn the compost. (One of the four bins is marked: PLACE DEAD ANIMALS HERE.) Along the floor, thin PVC pipes hiss with the forced air that helps the system work quickly.
To use the bins, workers line the bottom with sawdust and layer carcasses on top, back to front, all facing the same direction. They continue layering animals and sawdust as the pile builds up. “It might take a month to fill it,” says White, “and then it sits for another month until it’s finished. The neat thing is that in the bin that’s working, there’s no odor—it’s all balanced. The guys that operate it can tell by the odor if it’s out of whack.”
On this visit, on a hot summer day, there is no smell and just three animals—two deer and a dog—in the bin. An adjacent shed holds finished compost, looking like dark sawdust studded with bones and antlers, which will be added into new piles as a starter. In the busy fall season, White says, as many as 100 animals might be added to each bin before the whole system is at capacity.
“We create an environment so the microbes prosper and eat everything up in about 28 days,” says White. “It’s pretty fast.” Long probe thermometers measure temperatures inside the piles of up to 140 degrees during the hottest part of the cycle, and drainage channels collect the leachate, rich with beneficial microbes, that managers recycle back into the piles.
At a cost of around $150,000 per compost facility, the bins can save VDOT money over the long term, concluded Donaldson and White in their study.
That’s also a big part of the case Donaldson wants to make for fencing: that it’s smart management of public funds.
“While underpasses designed for wildlife are a great solution when new roads are constructed, for existing roads it can cost millions to dig up the road and install a new underpass,” she says. “It’s low-hanging fruit when we can make better use of existing structures—it’s inexpensive, and the mitigation has a profound effect. We hope that our projects in Charlottesville are going to be the model for the rest of the state.”
How not to hit a deer
Consider your timing. In fall, and at dawn and dusk, your chances of a DVC increase. Drive extra cautiously during those times.
Consider your position. Preferred crossings include creek bottoms and places with lots of vegetation on one or both sides of the road.
Slow down. “If you’re going 55,” says Matt Knox with the DGIF, “the chances of a DVC are much higher than if you’re going 25.” And if a collision seems imminent, don’t swerve. Jerking the wheel can make you lose control of your car, resulting in a much worse impact than hitting the deer. As Knox puts it, “Would you rather hit a deer, or an oak tree?”
Bears, too
When Bridget Donaldson took C-VILLE on a tour of the box culvert where VDOT’s new fencing is funneling wildlife, we saw lots of tracks in the tunnel. Deer were obviously using the tunnel, as were smaller animals like raccoons. Just before we left, we discovered bear tracks too.
Sure enough, later that day Donaldson discovered footage of a black bear emerging from the culvert.
An average of five bears per year died on I-64 between Afton Mountain and Charlottesville during Donaldson’s three-year study of the road. Like deer, bears are most likely to cross roads during the fall, when they’re feeding intensively to get ready for hibernation.
Though deer are the biggest problem for motorists and the main focus of Donaldson’s fencing study, bears benefit from the project too—as do the handful of drivers each year who would otherwise have collided with an animal that can weigh up to 500 pounds.