Topic:
MOTOR VEHICLES; ANIMALS; TRAFFIC ACCIDENTS; STATISTICAL INFORMATION;
Location:
TRAFFIC ACCIDENTS;

OLR Research Report


November 16, 2004

 

2004-R-0805

MOTOR VEHICLE CRASHES INVOLVING COLLISIONS WITH ANIMALS AND PREVENTION MEASURES

By: James J. Fazzalaro, Principal Analyst

You asked for the number of accidents In Connecticut involving collisions between motor vehicles and animals and examples of the types of countermeasures used in other states to reduce the likelihood of such collisions.

SUMMARY

There were 1,049 reported crashes involving motor vehicles and animals in 2002 and 1,216 such crashes in 2003. These represented 1.3% and 1.5% of all the reported motor vehicle crashes in those years respectively. Approximately 90% of all animal-vehicle crashes in Connecticut involve only damage to the striking vehicles, but one fatality resulted in 2003 and two resulted in 2002. In 2002, 30% of all the animal-vehicle crashes reported in Connecticut occurred in 10 municipalities. The 10 municipalities with the greatest number of reported animal-vehicle crashes in 2003 accounted for more than one-third of all such crashes.

The literature on the issue of animal-vehicle conflicts generally identifies 16 general categories of preventive countermeasures that have been used or evaluated over the years. Of these, exclusionary fencing and wildlife crossings have been widely used and evaluations of their effectiveness have been generally positive, but their effective implementation can be both complicated and expensive. Certain other measures such as vehicle mounted deer whistles and roadside reflectors that reflect vehicle lights off the roadway to act as a deterrent, particularly for deer, have also been used for a number of years, but research on their effectiveness in preventing animal-vehicle crashes has produced conflicting conclusions.

Almost all states have used roadside warning signs at one time or another, but passive warning signs that are up all year round have not been shown to have a major effect on reducing animal-vehicle crashes since motorists tend to ignore them in the absence of some evidence (such as an animal carcass) that reinforces the nature of the problem. Some states have also used lighted or animated warning signs as well as seasonally posted signs with mixed results. Among the newest use of warning signs as a countermeasure involves coupling active signs with dynamic animal detection systems that use electronic means to detect large animals near the roadway and activate flashing lights on the warning signs when an animal is detected.

Some of the other countermeasures that have been used to one degree or another, but which do not have an extensive history of evaluative research supporting their effectiveness include: roadway lighting implemented expressly to make animals more visible, speed limit reductions in animal crash areas, roadside vegetation management, “deer flagging” displays that use warning signs duplicating the warning postures of white-tail deer to attempt to warn off deer, intercept feeding where food sources attractive to certain species are located away from the roadway to intercept them while foraging, the use of alternatives to deicing salts on or near the roadway, hunting and herd management policies and programs, and the use of repellents on food sources or the roadway environment that either make roadside vegetation unattractive or mimic the presence of predators, and public information and education programs. Researchers have also identified the planning, design, and maintenance cycle that highways undergo as an opportunity to build consideration of the animal-vehicle crash problem into the implementation process.

Finally, a relatively new area of activity is the development of in-vehicle animal detection technologies that involve equipping vehicles with infrared sensing systems and driver displays that are capable of identifying animals near the roadside and providing the driver with a visual presentation of that information. Two such systems are currently deployed—one that is marketed to the truck and bus industry and another that was developed by the General Motors Corporation and is currently available as an option on the Cadillac Deville model line.

ANIMAL RELATED ACCIDENTS IN CONNECTICUT

Department of Transportation accident statistics for Connecticut indicate that there were 1,049 accidents in 2002 involving vehicles colliding with animals and 1,216 such accidents in 2003. Most of these accidents (89% in 2002 and 91% in 2003) involved only property damage to the striking vehicles, but two fatalities resulted in 2002 and one in 2003. In 2002, more than 38% of the crashes occurred in October, November, and December. In 2003, more than 46% of the crashes occurred during those three months. Animal-vehicle crashes represented 1.3% of all the reported motor vehicle crashes in 2002 and 1.5% of all the reported motor vehicle crashes in 2003.

Table 1 shows details for the animal related accidents in Connecticut for 2002 and 2003 and includes information on the 10 municipalities that experienced the greatest number of animal-related collisions in each year. The towns of Ridgefield, Woodbridge, and Waterford accounted for approximately 12% of all the animal-vehicle crashes recorded in 2002 and 2003.

Table 1: Connecticut Motor Vehicle Accidents Involving Collisions With Animals — 2002 and 2003

 

2002

2003

10 Municipalities With Greatest Crash Totals—2002

10 Municipalities

With Greatest Crash Totals—2003

Accidents

1,049

1,216

Ridgefield 53

Woodbridge 37

Guilford 35

Waterford 34

Madison 30

Coventry 28

New Milford 27

Stamford 26

Groton 24

Orange 24

Waterford 52

Ridgefield 49

Woodbridge 46

Coventry 43

Madison 42

Groton 40

Wilton 37

New Milford 35

Guilford 33

Glastonbury 30

Fatal

2

1

Injury Producing

116

104

Property Damage only

931

1,111

Total Vehicles Involved

1,081

1,242

Total Reported Accidents

78,673

80,857 (est.)

Animal Crashes as Percent of Total Accidents

1.3%

1.5%

PREVENTION MEASURES

The literature on preventing animal-vehicle crashes identifies 16 types of countermeasures that have been used to varying degrees around the world. Most of them have not been analyzed in any great detail with respect to their effectiveness. The 16 categories of countermeasures are identified below.

● Exclusionary Fencing

Any fencing along roadways is obviously intended to restrict access to the road, but its role with respect to preventing animals from reaching the roadway is complex. A 1992 study reported that 11 states had erected at least some fencing aimed specifically at excluding or restricting wildlife from roadways. More recently, a Transportation Research Board report identified many more states using this countermeasure (Interaction Between Roadways and Wildlife Ecology, NCHRP Synthesis 305, Transportation Research Board, 2002). States' use of specifically designed exclusionary fencing ranges across an array of different animal species including bear (Florida and Georgia); deer, elk, and moose (may states); turtles (Iowa and California); and reptiles and amphibians (Illinois, Iowa, and numerous applications in Europe). Michigan has experimented with electrified fencing in high deer crossing areas.

Properly designed, constructed, and maintained fencing is considered to be one of the most effective means of preventing wildlife from entering the roadway. This has been documented in a number of studies conducted in Pennsylvania, Colorado, Minnesota, Utah, Wyoming and elsewhere. Properly designed and maintained exclusionary fencing can be effective by itself, but is considered most effective when used in conjunction with wildlife crossings so that animal populations are not cut off from normal range and migratory movement.

Deer generally appear to present the greatest problems with respect to fencing. They will regularly test a fence and will quickly exploit any breaks or gaps in the fencing. They will also jump over fencing that is insufficiently high. While there is not universal agreement on the fence height necessary to prevent them from jumping over it, several studies suggest that fencing 7.5 to 8 feet in height can be effective, although not necessarily in all instances. Deer will also circumvent a fencing installation to follow a well-established feeding or migration path by

following it to the end. One study concludes that fencing must extend at least one-half mile beyond areas with high deer populations to discourage circumvention.

Fence maintenance is also a significant issue. Animals, and deer in particular, will attempt to dig under most fencing so regular examination of exclusionary fencing installations is required to repair tunnels and breaks caused by erosion, animals, falling trees, and human encroachment. Research has shown that deer can crawl through openings under a fence of no more than 10 inches.

Another issue associated with the use of exclusionary fencing is dealing with animals that get past the fencing and become trapped in the roadway. Dealing with this requires some type of escape route for the animal. Thus some fences include one-way gates or earthen escape ramps to provide the animal with the means to avoid entrapment.

Effective exclusionary fencing can be expensive to construct and maintain. A 1998 Iowa study estimated the cost of constructing an 8-ft. chain link fence on one side of a roadway at $42,000 per mile. It did not estimate continuing maintenance costs.

● Wildlife Crossings

In recent years, wildlife crossings have been among the most widely implemented animal-vehicle crash prevention measures. They are usually constructed to serve two purposes—to decrease the fragmentation of habitat and to make the roadway more negotiable by animals. They take several forms including bridge extensions, overpasses, and underpasses (culverts or tunnels) and have been used for all types of wildlife from large mammals such as bear and moose to small mammals, amphibians, and fish species. A recent Transportation Research Board survey showed that more than half of the states have installed at least one type of wildlife crossing with positive results. However, Florida, which was the first state to make use of wildlife crossings, is generally considered to be the leader in this area. Animal overpasses, in particular, are widely used in France, Germany, the Netherlands, and Switzerland, among other European nations. Underpasses are more common in the United States. Wildlife crossings are generally most effective at reducing animal-vehicle crashes when they are combined with exclusionary fencing that helps steer the animals toward the crossing, although certain species like bear may try to climb over the fences rather than use an underpass and others like coyote and badger will attempt to dig under them. A limited amount of research into the effectiveness of specially designed at-grade crosswalks for deer has also been done in a 1997 study in Utah (Lehnert and Bissonette, Effectiveness of Highway Crosswalk Structures at Reducing Deer-vehicle Collisions, Wildlife Society Bulletin, 25:809-18, 1997).

Florida, Massachusetts, Montana, New Hampshire, Texas, and Wisconsin use culverts for reptiles and amphibians. Nebraska and South Dakota use them for turtles. A number of states, including California, Illinois, Kansas, Montana, New Hampshire, New Jersey, Oregon, Texas, Utah, and Virginia use underpasses (usually culverts) for small mammals. Arkansas, Colorado, Wyoming, Utah, Michigan, Maine, Washington, and Kansas use larger culverts for deer and large mammals. Florida has had considerable experience with the use of underpasses for black bear and the Florida panther. North Carolina has also used underpasses for black bear. More than 24 states have used extended bridges over rivers and streams. These provide an area over watercourses that is adequate to accommodate aquatic species and still provide dry habitat for terrestrial species that rely on travel along these corridors. Florida, Hawaii, New Jersey and Utah make use of wildlife overpasses in certain areas.

Animal crossings must be carefully designed and placed to be effective and frequently must be tailored to the behavior of a specific animal species. This is especially true for underpasses where height and length, placement, noise levels, substrate, vegetative cover, moisture, temperature, light, and interaction with human use and use by other animal species can all critically affect whether animals will use the crossing. Size is an important consideration, but size thresholds that will encourage usage vary among animal species and there is no absolute rule. For some species, it appears that the openness of the passage (its size relative to the width of the road) may be more important than its height. Tunnels that allow the animal to see the opposite end of the passage encourage use by some species. While larger passages tend to be more effective, some small mammals actually prefer smaller spaces to travel through.

Placement is important with respect to both the distances between passages and, especially with respect to smaller mammals, the travel distance necessary to reach the passage. Some species are reluctant to use passages that do not have enough ambient light, but others that are sensitive to human disturbance such as mountain lion avoid passages that are artificially lit. Passages for amphibian species must have a moist substrate to encourage use, but standing water deters other species from using the passage. Amphibians also may avoid underpasses when the difference in temperature between the inside and

the outside of the passage is too large. Some mammals have trouble with passages where traffic noise is too great, particularly those that are sensitive to human disturbance.

The flooring of the passage can also be either an attractant or a deterrent to use. Small mammals prefer adequate cover on the passage floor for security while crossing while amphibians may be encouraged if the floor is designed to mimic streambed conditions. Bears and other species that prefer the forest environment usually respond better if the approach to the passage has considerable vegetation. On the other hand, experience with mountain goats shows that they prefer more visibility when approaching a passage. Small and middle-sized mammals are attracted to approaches that feature vegetation, rocks, and logs, but if the species is wary of ambush by predators, it may be reluctant to use a passage where the entrance is too confining.

Interaction with other species, whether human or other animals, can also affect whether certain animal species will use a passage. Deer, moose, and elk, as well as many carnivores, are reluctant to make use of a passage when it is either too close to a populated area or they can detect human activity within the underpass. Also, some research shows that use of a passage by predators may make prey species reluctant to use it as well.

Many of the issues associated with the use of wildlife passages, as discussed above, are analyzed in more detail in an evaluation done by Jackson and Griffin entitled “A Strategy for Mitigating Highway Impacts on Wildlife” which is enclosed with this report and available on the Internet at http://www.umass.edu/nrec/pdf_files/tws_strategy_ms.pdf.

● Deer Whistles

These devices have been around since the late-1970s. They attach to the front of a motor vehicle and, as air moves through them as the vehicle moves, they produce an ultrasonic noise that manufacturers claim will alert deer of the vehicle's presence at a range of up to one-quarter mile. Air-activated whistles work when the vehicle's speed is at least 30 miles per hour. Some recent versions of these whistles are electronically powered instead of air activated and thus can produce the ultrasonic sound regardless of the vehicle's speed. Deer whistles are produced by a number of different manufacturers. About 20 states appear to have some experience with their use.

The effectiveness of deer whistles is a subject of considerable debate. While there have been a number of studies of them, many have been anecdotal rather than scientifically designed and documented. A few before-and-after studies of whistle equipped government vehicles found reductions in the number of deer crashes after vehicles were equipped with the whistles. For example, after 55 patrol vehicles of the Onodaga County, New York Sheriff's Department were equipped with deer whistles, crashes with deer were reported to have decreased from approximately 10 per year to two in two years. Similarly, a test fleet analyzed by the Idaho National Engineering and Environmental Laboratory showed no crashes in the five years after equipping the fleet with whistles compared to an average of 17 crashes per year before they were installed.

On the other hand, a number of other studies, including some that have attempted to measure the hearing sensitivity of deer, have questioned whether deer can actually hear the sound frequencies the whistles emit and whether they can distinguish the whistle noise from the rest of the sound emanating from the moving vehicle. These researchers tend to take a fairly strong position that vehicle-mounted whistles are not effective.

● Roadside Reflectors or Mirrors

This countermeasure involves setting up reflectors or mirrors along the road edge to reflect the lights of approaching vehicles off the roadway to areas where animals may be foraging. The theory behind these reflectors is that the reflected light will immobilize the animals and prevent them from approaching more closely. The concept has been in use by at least 22 states going back for more than 30 years, but evaluations of their effectiveness have produced conflicting conclusions. Studies of their use in California, Colorado, Maine, Wyoming, and Ontario, Canada have questioned their effectiveness in reducing animal mortality. However, studies of their use in Iowa, Minnesota, Oregon, Wisconsin, and British Columbia, Canada found significant reductions in animal-vehicle crashes where reflectors were in use. One study conducted in Washington found them to be effective in eastern Washington, but another study conducted in western Washington found them not to be effective in reducing deer kill in that part of the state.

● Speed Limit Reduction

Reduced speed limits in known animal-vehicle conflict areas have been implemented or considered in several jurisdictions around the country. These speed limit changes are frequently accompanied by signs identifying the area as vehicle-animal conflict zone. Reduced operating speeds are theorized to provide a driver with additional time and distance to see and react to possible conflicts with animals. These benefits have been fairly clearly established for drivers approaching stationary objects or objects moving at a relatively uniform speed, but are less well known with respect to the less predictable movements of animals. As many as seven states appear to have used reduced speed limits as a crash countermeasure, but there have been few studies evaluating its effect.

Studies of reduced speed areas were conducted in Yellowstone National Park in the United States and Jasper National Park in Alberta, Canada. Researchers conducting these studies felt that the lower speed limits were related to fewer animal crashes. However, some of the evidence in these studies suggests that the effect varies with different animal species and the ways in which their behavior differs, in effect, it may benefit certain species more than others because of the different ways they relate and react to traffic. There has also been little research into the interrelationship of posted speed limits and vehicle operating speeds with other factors such as roadway design, topography, and known animal habitat.

Signs and Sign-related Technology (including Automated Animal Detection Systems)

Virtually all states have used animal crossing warning signs at one time or another. The most common application of warning signs is installation of “passive” signs displaying either a warning or an animal legend (deer, elk, moose, geese, cattle, salamanders, and others specific to a particular resident animal population) that conforms to the Manual on Uniform Traffic Control Devices. Generally, the most effective types of warning signs are ones associated with a constant hazard, i.e., a sharp curve, rather than a sporadic hazard, i.e., animal crossings. Thus passive animal warning signs left up year-round tend to be disregarded by motorists who do not reduce their travel speeds unless the warning sign is associated with evidence of the hazard, for example, an animal carcass in the general area of the warning sign. One Colorado study found that average vehicle speeds were reduced by almost eight miles per hour when deer carcasses were visible on the road shoulder near the warning sign (Pojar et al., Effectiveness of a Lighted, Animated Deer Crossing Sign, Journal of Wildlife Management, Vol. 39, No. 1, 1975, pp. 87-91). When not reinforced by actual driver experience of the potential danger, passive warning signs usually result in little adjustment in travel speed, particularly when motorists are familiar with the area.

Two variations some states have used to try and overcome the disadvantages of year round passive warning signs are the use of lighted and/or animated signs and temporary passive signs. In the first instance, the sign is enhanced with the use lights, flags, a lighted and animated figure of a deer or other animal, or some other device. The

Colorado study noted above found a small speed reduction of three miles per hour when enhanced signs were used but identified no decrease in animal crashes associated with it.

Other research has focused on the effect of seasonal signing which have been used in Utah, Michigan, and a few other states. These signs generally have a different design than typical passive signs and are erected or uncovered only during the times of year when animal-vehicle crashes are most frequent. In Utah, the use of the seasonal signing noting the area as a deer migration area (mule deer) was associated with both an average speed reduction of eight miles per hour and a 50% reduction in animal-vehicle crashes in the Spring and a 70% reduction in the Fall (Messmer et al., Modifying Human Behavior to Reduce Wildlife-Vehicle Collisions Using Temporary Signing, presented at “Wildlife and Highways: Seeking Solutions to an Ecological and Socio-Economic Dilemma”, Nashville, Tennessee, September, 2000, pp. 134-147). In Michigan, the signs were of a different design and message (noting the area as a high animal crash area) and involved a different deer species (white-tail deer). The Michigan study found no significant decrease in animal-vehicle crashes and a small but statistically significant decrease in vehicle speeds (Elizabeth Rogers, An Ecological Landscape Study of Deer-Vehicle Collisions in Kent County, Michigan, Prepared for the Kent County Road Commission, January 2004). At least one source believes that the differing results of these studies suggest that that temporary seasonal signing has its greatest effect when the resident animal species has more established and well known migratory habits (mule deer in Utah) and a local driving population that is aware of both these habits and their safety implications (Keith Knapp, Deer-Vehicle Crash Countermeasure Toolbox: a Decision and Choice Resource, Prepared by the Deer-Vehicle Crash Information Clearinghouse for the Wisconsin Department of Transportation, Report No. DVCIC-02, June 2004).

Perhaps the most rapidly growing application of warning sign countermeasure involves the use of an active warning in conjunction with a dynamic animal detection system. In this application, warning signs are equipped with lights that flash only when an electronic animal detection system senses a large animal near the roadway. An electronic beam along the road shoulder is broken when the animal enters and the lights on the warning sign begin to flash, notifying motorists of the presence of the animal. Animal detection systems are currently in use in Minnesota (infrared light), Wyoming (radar), Washington (laser), Indiana (radio waves), and British Columbia (heat detection cameras). Washington has also put into use a system where elk in an established herd near one highway have been equipped with radio collars that

activate the warning signs whenever any of the animals comes within one-quarter mile of the road. Dynamic detection systems are also use in Montana and in Yellowstone National Park.

● Hunting and Herd Reduction

Hunting and herd reduction measures are perhaps the most controversial that are applied to the animal-vehicle crash issue. They are based on the theory that increases in conflicts between vehicles and animals are at least partially due to the unchecked growth of many animal populations, particularly deer populations. Similar to the increase in human accident exposure that occurs as vehicle miles traveled increases, the greater the number of animals exposed to the roadway environment, the greater their exposure to conflicts with vehicles. However, many animal advocacy groups assert that the concentration of animal-vehicle crashes involving many large mammal species in the late-Fall months coincides with the onset of hunting season in many states. They argue that this makes the animals panicky and more likely to take risks to escape pursuit. They further argue that increases in the length of the hunting season or the number of animals to be harvested through these means may make the crash problem worse not better.

Conventional wisdom suggests that reducing the size of animal populations subject to conflicts with vehicles could reduce the number of such conflicts, but most of the research studies on herd reduction policies focus their analysis on the effects these have on the target animal population itself and, while the researchers may make references to decreases in such accidents, there appear to have been no studies that have rigorously examined hunting or herd reduction policies with the objective of establishing a quantitative relationship between the policies and a subsequent change in animal crashes in a large geographic area such as a state. A few studies or analyses have credited herd reduction programs for decreases in animal crashes in relatively small geographic areas such as parks or urban settings however.

Because of the lack of any comprehensive examination of the causal connection between herd reduction programs and subsequent reductions in animal crashes, there are a number of unresolved questions regarding their overall effectiveness. Beside their effectiveness over a large geographic areas, it is not clear how great a herd reduction is necessary to produce a substantial reduction in animal-vehicle crashes, how animal range and migration patterns influence the effect of herd reductions on crashes, and how to design a cost effective herd reduction program.

States which have been most active in the areas of hunting and herd reduction studies and programs and the herd size and density factors that may relate most closely to animal-vehicle crashes include Wisconsin, Michigan, Minnesota, Illinois, Iowa, Kansas, Pennsylvania, and Ohio. Not coincidentally, several of these states have both the largest deer populations and some of the largest problems with reported animal-vehicle crashes. For example, Michigan experiences about 60,000 deer-vehicle crashes a year and in Wisconsin, almost one in every six reported crashes involves a collision with a deer.

● Roadside Vegetation Management

Some believe that certain policies on vegetation management along roadsides may encourage or attract animals to forage close to the road and thus increase animal vehicle conflicts. While a large number of studies have apparently looked at what types of plants and vegetation certain types of animals prefer or resist, most of them are aimed at helping residential landowners develop animal resistant plantings rather than identifying what types of vegetation management policies may discourage roadside foraging. A few have examined how certain plantings attract animals to the roadside but did not draw any conclusions about how this affected crashes. No studies seem to have measured any changes in animal-vehicle crashes as a result of changing roadside vegetation policies. A few studies have looked at the crash impact of roadside vegetation clearance and identified some benefit to clearing low vegetation for certain distances from the road edge, but these have been limited in scope.

● In-Vehicle Technologies

These are systems installed in motor vehicles that combine sensing devices and visual displays aimed at helping drivers avoid animal crashes, particularly at night. They typically use an infrared “night vision” type of technology that senses nearby animals and shows them to the driver in the visual display. Application of this technology is very new but two systems are currently in use and others are being developed. One system, Bendix XVision, is designed specifically for trucks and buses to improve the driver's night visibility. The other, Cadillac Night Vision, is an option offered on new Cadillac De Ville models. Because of the relative newness of the applications, little evaluation has been done of their overall effectiveness. There is believed to be potential for this technology, but also some potential concerns regarding driver information overload or distraction.

● Deicing Salt Alternatives

Some have theorized that the use of road deicing salt (sodium chloride) can contribute to the animal crash potential because deer and most other animals are attracted to deposits of natural or artificial salt in their normal course of feeding. However, very little research has been done to help define the impact road deicing salts have in drawing animals into roadside conflict with motor vehicles. The most significant research on the subject occurred in Ontario, Canada in the early 1980s. It found that the existence of roadside saltwater pools was a significant attractant for moose in otherwise salt-deficient areas and that about half of all the moose-vehicle crashes that occurred on the road were at or near the pools (Fraser and Thomas, Moose-Vehicle Accidents in Ontario: Relation to Highway Salt. Wildlife Society Bulletin, Vol. 10, No. 3, 1982, pp.261-65). While, as this study indicates, road salt may factor into in animal-vehicle crashes at particular locations where the salt collects, not enough is known about this relationship at present. Some researchers suggest that use of salt alternatives for deicing might be worth considering in known high animal-vehicle crash areas if public safety and other issues such as cost permit.

● Public Information and Education

Public information and education programs on the animal-vehicle crash problem tend to focus on two areas—making the public aware of the nature and extent of the problem and of the areas in the state where the problem is most evident and informing the public of driving behaviors that they should follow to minimize the likelihood of such conflicts or the results of one should it occur. However, similar to public information programs in other areas of motor vehicle safety, it is very difficult to demonstrate how these programs produce a direct reduction in the number of crashes they target. It appears that there are no studies that connect a specific public information or education program directly to a reduction in animal-vehicle crashes. Rather, in most places, these campaigns are viewed as important components of crash reduction programs that involve other countermeasures as well.

Animal-vehicle crash information and education programs are usually targeted to the Fall and Spring (the highest and second highest concentrations of animal-vehicle crashes in most locations). Most of the states that have a significant number of animal-vehicle crashes produce some form of public awareness materials or information programs through internet websites, media spots, brochures, or by some other means.

● Roadway Maintenance, Design, and Planning Policies

The essence of this countermeasure is to build specific consideration of wildlife habitat and movement patterns into roadway maintenance, design, and planning activities. Maintenance decisions that could have an impact on animal-vehicle crashes include winter maintenance and roadside vegetation policy (discussed elsewhere in this report) and carcass removal policy (delayed removal of animal carcasses may draw scavengers to the location and increase the possibility of subsequent conflicts). Roadway design issues include posted speed limits, road curvature, road cross-section, and bridge height and length. Roadway planning issues that could take the potential of vehicle-animal conflicts into account include those relating to alignment and location and project programming.

A number of studies have looked at or modeled the roadway and adjacent land use characteristics that seem to be associated with known animal-vehicle conflicts. Some states, for example, Michigan, Washington, and Florida, appear to have combined this crash and animal carcass data with species and habitat information within a geographic information system to identify existing and future “wildlife critical” road locations that cut through animal habitats or habitat linking corridors.

Some researchers have proposed that planning for a roadway alignment should take into consideration its “road zone effect.” They define this as an area that could experience a negative ecological impact from the road's existence. They assert that the zone includes the area within approximately 1,000 feet of the roadway edge (Forman and Deblinger, The Ecological Road-effect Zone for Transportation Planning and Massachusetts Highway Example, Proceedings on the International Conference on Wildlife Ecology and Transportation, Report No. FL-ER-69-98, Florida Department of Transportation, 1998, pp. 78-96).

● Roadway Lighting

Statistics on deer crashes show that the majority of such animal-vehicle conflicts occur in the evening twilight and nighttime hours when deer are most actively foraging. Roadway lighting as a preventative measure is based on a theory that, since the most active period for deer foraging overlaps the time of minimum driver visibility, providing effective roadway lighting would make it easier for drivers to avoid animals. However, little research has been done on this approach. There appears to have been only one study that has attempted to evaluate the effectiveness of roadway lighting in reducing animal-vehicle crashes. While the researchers found that deer crashes were about 18% less when the lighting was on compared to when it was off, they felt that this difference was not significant. They also noted that when a taxidermy-mounted deer was placed in the emergency lane of a lit section, there was a reduction in average vehicle speeds of eight miles per hour (Beck et al., Highway Lighting to Prevent Deer-Auto Accidents. Final Report. Report CDOH-P&R-R-77-5, Colorado Division of Highways, 1977). Only two states are believed to have used roadway lighting specifically as a countermeasure for animal-vehicle crashes.

● Deer-Flagging Displays

White-tailed deer warn other deer of a potential danger by raising their tails to expose a white underside. Some scientists have theorized that crashes involving deer might be reduced if they were warned away from a road by use of signs or displays that mimic this warning behavior. There appears to have been two studies conducted in Pennsylvania testing this concept. The first study conducted in 1975 suggested that deer avoided areas where such displays were set up. A second study to verify and expand upon the results of the first study was conducted along I-80 in Pennsylvania in 1978. However, in the second study, the researchers found inconclusive results with reduced numbers of deer present in some cases and more present in others. No subsequent research designed to overcome the limitations of these earlier studies appears to have been conducted to date.

● Intercept Feeding

Intercept feeding involves the strategic placement of food supplies attractive to certain types of animals in locations that will keep them away from the roadside. Research on this countermeasure is minimal, however, it has been evaluated in tests involving deer populations conducted in Utah in the mid-1980s. The researchers concluded that intercept feeding might be an effective short-term measure that could reduce deer crashes by up to 50% during those portions of the year when the deer population was concentrated, but they did not believe this would be sufficient to justify the work and expense involved in implementing a full scale intercept feeding program. They felt that combining an intercept program with other countermeasures might increase its effectiveness. They also noted that there are continuing costs associated with any feeding program, it may make the wildlife dependent on the food, and it may attract more deer to the roadside.

● Repellants on Food Sources or Roadways

The use of various types of chemical or biological repellants as a means of deterring animal activity has been studied extensively on species such as white-tail and mule deer, caribou, and elk. However, these evaluations have been limited to studying their effects on groups of captive animals rather than animals in the wild. The studies generally show that repellants applied to food sources to make them taste unpleasant or, in the case of repellants such as predator urine applied to areas to create avoidance responses in the test animal populations, are effective in reducing feeding activity. The majority of these studies focused on treating food sources, but a few looked at the idea of using the repellents in sand and salt mixtures used for road maintenance or along the roadside.

A recent review of all of the repellent studies identified area repellents that the analysts felt had the most potential to keep certain types of animals away from roadways (Kinley et al., Problem statement: Potential to Develop an Area Repellent System to Deter Ungulates from Using Highways, Prepared for the Insurance Cooperation of British Columbia, November 2003). However, as noted above, no studies seem to have documented how effective repellents are in preventing animals from crossing roads. It was also noted in some studies that deer got used to the repellents over time and were not deterred by them if they were sufficiently hungry.

JF:ro