Some 700 to 800 moose are bludgeoned to death by motor vehicles each year in Alaska. The actual number of moose-related collisions is certainly higher --some moose aren't badly injured in a collision, even though the vehicle is, while other injured moose stagger into the woods and are never found. The total number of collisions is unknown because many people don't report motor vehicle accidents.
To put Alaska's moose-vehicle collisions in perspective, in recent years motor vehicle collisions from all causes have ranged from about 10,000 to 13,000 incidents annually. Motor vehicles hit moose two to three times as often as they run over pedestrians and bicyclists combined. However, unlike bike or pedestrian collisions, moose collisions are seldom fatal to humans.
Peak season for Alaska moose-vehicle collisions is December and January, when it's two to five times more likely than other months.
Attempts to reduce collision rates
For decades, scientists have tried to develop a practical way to reduce the number of collisions with deer, elk and moose.
Highway safety engineers, with the help of wildlife experts, have tried to modify the road-crossing behavior of deer, elk and moose. Warning whistles attached to vehicles, roadside reflectors, and chemical repellents either don't work or have unclear results. Providing supplemental food some distance from roads -- in other words, moose feeding stations like those occasionally funded by the Alaska Legislature -- are expensive, largely unproven, and cause other problems.
Permanently reducing the number of moose in zones surrounding roads will probably reduce moose-vehicle collisions, but any reduction in moose density substantive enough to reduce collisions will also reduce future sport and subsistence hunting opportunities.
Attempts to modify driver behavior with warning signs, flashing lights, or public awareness campaigns have been ineffective or remain unproven.
Fences, underpasses, overpasses
You can't avoid a moose until you can see it. Moose are more likely to cross roads where forest cover encroaches on the right of way, and these animals are less likely to be visible until too late. However, as much as they make sense, roadside clearing and lighting remain unproven techniques, and both methods are relatively expensive.
The only method of reducing moose-vehicle collisions that is reasonably effective and broadly accepted is fencing combined with special underpasses or overpasses to allow wildlife movement across the road barrier. Fences are one of the most expensive approaches to build and maintain, and a single road-crossing structure suitable for moose passage can cost millions of dollars.
On the busiest part of the Glenn Highway, between Anchorage and Eagle River, moose mitigation in the late 1980s and early 1990s -- which included widening and lighting the highway in addition to building fences -- halved the number of moose-vehicle collisions while average daily traffic volumes increased from about 35,000 to 50,000 trips per day.
Another often-suggested way to reduce moose-vehicle collisions is by lowering the speed limit. The faster you drive, the greater the braking distance. According to the Alaska Driver Manual, average braking distances, including reaction time, under ideal conditions for a vehicle travelling 50, 60, or 70 mph are 67, 94, and 122 yards, respectively. In other words, a relatively small increase in speed, from 50 to 70 mph nearly doubles a vehicle's braking distance.
Ideal conditions means no wet or icy pavement, excellent brakes, and an alert, undistracted, relatively young driver. Few drivers encounter moose under ideal conditions. Road conditions aside, many of us aren't fully focused on the road. There may be stress at work or home, alcohol consumption, old age, or distractions like texting, kids and pets. Drivers just aren't paying much attention. In the real world, 90 percent of drivers need a little more than 200 yards to stop a vehicle travelling 70 mph on dry pavement.
Crash severity also increases with a vehicle's speed. For example, the probability that a collision with a pedestrian will result in a fatality increases from 5 percent at 20 mph to 85 percent at 40 mph. Not surprisingly, hitting a moose, which weighs far more than the average pedestrian, at 40 mph or greater is extremely hazardous to both the moose and the unfortunate occupants of the vehicle.
Motor vehicle operators are supposed to slow down when conditions are less than ideal. But we've all witnessed many -- known as tailgaters, lane changers, and ditch divers -- who don't. There's even an official term for it: "Driving at a speed unsuitable for the prevailing road or traffic conditions."
Common sense dictates that vehicles traveling at slower speeds hit fewer moose. Not only do slower approach speeds substantially reduce braking distance, they give moose more time to react. However, not enough research has been conducted to determine whether a reduction in the speed limit results in fewer moose-vehicle collisions. Two studies have had mixed results.
Establishing speed limits by majority rule
Federal and state transportation agencies, including the Alaska Department of Transportation, strongly oppose reducing posted speed limits to avoid wildlife collisions. Their reasoning seems sound. Research has repeatedly shown that many drivers will ignore a posted speed limit less than what they consider to be a reasonable operating speed. Thus, highway agencies determine speed limits based on the observed speeds of 85 percent of drivers, ignoring only the fastest 15 percent, those deemed "reckless."
I've racked my brain to think of other laws with similarly permissive standards. And yet statistics confirm that fewer accidents occur when speed limits are based on the 85th percentile rule. The problem boils down to human nature. Some drivers will obey a speed limit set at a speed less than they think they can safely drive, while other drivers will ignore it. The subsequent dispersion of speeds -- some fast, some slow -- causes more accidents than more uniform speeds at the higher limit.
Even as a scientist, I've always resisted the urge to put unbounded faith in statistics. So I'm a little skeptical of the 85th percentile rule. It's a little too pat. Where was the research conducted that showed fewer collisions when the 85th percentile was used to set speed limits? It probably wasn't Alaska, where we can have snow and ice on the roads for five or six months each year. Did they measure operating speeds in summer or winter? Was the database collected in an area with lots of deer or moose collisions? Or is it a national average?
Visibility vs. braking distance
Once brakes are applied at any given speed, the distance it takes to stop is largely determined by forces outside of the driver's control. Roads are designed and speed limits are established to allow ample sight distance to avoid collisions under normal conditions. But what happens when a large object, like a moose, suddenly materializes on the road? Engineers can't account for the unpredictable movements of wild animals.
Moose make it difficult for drivers in several ways. Moose are much more unpredictable than other motorists -- or even pedestrians. They burst out of the tree line and charge full speed across the highway. Or they stand calmly on the side of the road until a car is near, then bolt across the road. They run partway across, then double back. Or they stand on a road and don't realize a vehicle is approaching until it's too late. They don't have brake lights or turn signals. They seldom use crosswalks.
In fact, moose are a lot like kids in this respect. In locations where unsupervised kids are likely to be encountered along the road -- like where they live -- speed limits are reduced. Speed limits near schools are reduced to 20 mph. Why? Because kids are unpredictable.
Moose also tend to be encountered more often in specific areas, like where a road crosses a stream or where lots of good browse lines both sides of the highway. Warning signs often identify these areas, but most drivers ignore the signs and at least some, "Donna the Deer Lady" in particular, just don't get it at all.
Avoiding moose, which can be difficult in the daytime, is compounded in the dark. Because their hair tends to be dark, moose reflect less light than deer. Deer can often be spotted by their eye shine which, like a reflective road sign, is visible up to quarter-mile away, well beyond the illumination of most manmade or natural objects. Moose eyes also reflect light, but moose don't normally stare at oncoming headlights like deer, so drivers rarely benefit.
Overdriving your headlights
Plenty of drivers run into moose in the daylight, but most moose-vehicle collisions occur at night for an obvious reason: visibility. Unlike owls and other denizens of the night, humans don't have great night vision; instead, we rely on our headlights.
A couple of Canadian scientists have measured the average distance that a moose can be spotted in headlights by using a life-sized specimen, a foam core fitted with a real moose hide and antlers. The beast was positioned perpendicular to approaching traffic on the centerline and on the right and left shoulders of the highway so it wasn't completely predictable. Results were reported in metric; I've rounded them off to yards and miles per hour.
The average detection distance was about 80 yards for the low-beam setting and 150 yards for high beams. Not surprisingly, the decoy was significantly more visible in the center of the lane than on the shoulders. Nevertheless, most motorists were unable to see a moose on the road shoulder at night, even with high beams, more than about 98 yards away.
The researchers concluded that motorists driving at night faster than about 45 mph are likely to be "overdriving the illumination capabilities of their headlamps for moose encounters." For the low beam setting, the maximum safe speed in moose country drops to about 40 mph; for high beams the maximum safe speed is about 50-55 mph.
Alert drivers in ideal conditions
I can't emphasize enough that these tests were conducted during ideal conditions. Participating drivers knew they were about to encounter the moose decoy and, in the words of the researchers, were "sober, alert, and unusually attentive drivers." Add ice and snow, add the fact that most commuters aren't as young as they used to be or alert as they ought to be, and you've got a problem. For the average driver, the moose will materialize much closer than the distance required to stop the vehicle. Paying attention is crucial. I can't tell you how many times I've watched drivers whiz past a moose standing less than 20 feet from the asphalt without a flicker of recognition.
Obviously, overdriving your headlights is driving blind. Yet I'm going to go way out on a limb and suggest that most Anchorage commuters are unwilling to drive 45 mph on the Glenn Highway. And that unwillingness encapsulates the problem: almost every Anchorage commuter using the Glenn Highway, for most of the winter, is depending on the dubious road-savvy of our local moose and a large dose of luck to make it to work and back home safely. Unfortunately, according to road-safety statistics, if you drive too slowly compared to other motorists, you are creating a hazard. Other drivers are far more dangerous to the average motorist than the occasional moose.
The lights lining the Glenn and other highways help a lot; however, moose are seldom visible behind the light poles and visibility isn't ideal between the poles. Moose aren't visible at all in the woods, and most of the time the tree line or the dark areas that coalesce behind the lights are only 20 to 30 yards from the highway. That's about two seconds for a running moose.
Trading minutes for lives
Because their legs are so long, in a full-on collision a moose typically slides across a vehicle's hood and crashes through the windshield. Some moose come to rest in the drivers seat.
Slowing down to allow yourself a little more wiggle room is less onerous than it sounds. The 22 miles of Glenn Highway between the Eklutna and Muldoon interchanges – the only commuter route coming into the Anchorage bowl from the north – takes 20.3 minutes to drive at 65 mph versus 24 minutes at 55 mph. The difference in time is 3.7 minutes. That doesn't sound so bad does it? Running into a moose, on the other hand, will cost days if not the rest of your life.
Given the inability of most drivers to see, much less brake in time to avoid a moose collision at night, the Canadian researchers suggested reducing speed limits seasonally or when it's dark. That's unlikely to happen. Instead, because human behavior is so intractable, the Federal Highway Administration and some state departments of transportation are still hoping to change moose behavior. As Temple Grandin says, "Engineering is easy -- it's the people problems that are hard."
For those who refuse to slow down, the technology now exists to "illuminate" moose with infrared sensors at over 400 yards and to automatically flip conventional headlights from high to low beams, depending on traffic. Unfortunately, few models are equipped with these devices. Far infrared thermal imaging, in particular, is an expensive option, adding up to $2,200 or more to the cost of the vehicle.
Rather than reducing speed limits, the Alaska Department of Transportation is prepared to erect more fencing, without providing the equally necessary moose-crossing structures. These overpasses or underpasses are critical; otherwise, moose will find a way around the fences, usually through the gaps provided at intersections.
The Alaska Department of Public Safety, which knows human behavior at least as well as the DOT, provides some very good advice for avoiding moose collisions in its Alaska Driver Manual. To paraphrase:
Nobody's going to force you to slow down for moose. You have to decide, in the privacy of your own self-propelled battering ram, how many lost minutes your life is worth.
Rick Sinnott is a former Alaska Department of Fish and Game wildlife biologist. The views expressed here are the writer's own and are not necessarily endorsed by Alaska Dispatch. Contact him at email@example.com