Bayfield resident Glenn Johnston drives through the Colorado Department of Transportation's wildlife-detection system about six times a week on his way to Durango and back. But he has yet to see the detection system work properly.
"To me, it seems like a complete waste of taxpayer money," Johnston said. "It gives so many false signals. The lights come on, but I've yet to see an animal within the area that it covers."
State highway officials acknowledge the system is flawed. But they haven't given up on the experiment.
In fact, they are extending the detection zone this month and tinkering with the technology in hopes of eliminating so many false positives, said Mike McVaugh, CDOT safety and traffic engineer.
"The system can't tell a car from a deer, so it would trigger," McVaugh said. "So we've tried to implement some of our traffic-signal technology to tell the system to disregard a vehicle."
When installing the system, CDOT tried to calibrate it to detect large animals but not small animals or large objects like cars.
The $1.2 million wildlife-detection system was installed during the summer of 2008 along a one-mile stretch of U.S. Highway 160, between mile markers 95.6 and 96.6, just east of the Florida River. It is an important migration corridor for deer and elk during the spring and fall.
The system is the first of its kind to detect large-animal movements near the edge of the road. When an animal is detected, it activates 40-inch-by-60-inch lighted signs that read "Wildlife Detected."
The system uses the same technology as perimeter security systems used by the military, prisons, airports and some private landowners. Cables were buried 1 foot deep, 30 feet from the highway, along both sides of the one-mile stretch of road. The cables emit an electromagnetic field, and when large animals pass through the field, a disturbance is detected and the warning signs are illuminated.
The system seems to work with deer and elk; the problem is the warning signs light up when no wildlife is present, said Nancy Shanks, spokeswoman for CDOT.
"We were getting a lot of false positives," she said. "The signs would go on when there wouldn't be anything there."
And now drivers like Johnston have learned to ignore the signs.
"I tend to disregard it, and I believe that is the feeling of many drivers," he said.
The system was installed by Senstar Corp. as part of a pilot program. The company since has revamped its technology in hopes of providing more reliable detection. Highway officials will be able to deactivate certain sections of the cable - for example, where it goes under a private driveway. That way, it won't activate when residents roll over it, McVaugh said.
"You can actually train the system to go: If something goes across this segment, disregard it," McVaugh said. "And we can do it by time of day. We can say at 6 a.m., disregard, but at 5 p.m., go active in case a deer crosses at night. That's what we couldn't do before."
Senstar is upgrading the existing one-mile system at its own expense, he said.
The new segment, which will cost taxpayers about $500,000, will add a half-mile of detection zone to the west end of the existing system. CDOT officials are confident the new technology and recalibration will work, Shanks said. Otherwise, they wouldn't be extending the system, she said.
"We all hope that it will be very successful and serve as another tool to reduce the animal-vehicle collisions and not completely limit the wildlife migration," she said. "This has never been done before, and we're very excited to be the first to apply it."
The detection system adjoins more than a mile of deer fencing, which is being installed this year. The fencing will force deer and elk to cross the highway where the detection system is located.
The fencing lines both sides of the highway. If animals become trapped between the fences, they can escape via one of four ramps built inside the perimeter.
The test zone covers one of the most dangerous sections of highway in Colorado when it comes to deer-car impacts, McVaugh said. Recent statistics were not immediately available, he said, but in 2003, there was an average of 14 deer-versus-car hits per half-mile per year.
Highway officials said the number of impacts is down since the research project was installed, but they couldn't quantify by how much.
"It's going to take two to three years of data collection before we can really say what it's doing," McVaugh said.
The 8-foot tall fencing works well to keep deer off the highway, but too much fencing interferes with migratory patterns.
"If we fenced off the whole highway corridor from Durango to Bayfield, we've bisected their entire livable habitat for the summer and winter ranges," McVaugh said.
The detection system has been deactivated while improvements are being made, but it is expected to be reactivated this fall before migration season.
"We're working really hard to make this thing as accurate as we can," McVaugh said. "The first project that people have questioned is a research project. It's not something that comes straight out of a box and works the best it can."