Evaluation of Wildlife Crossing Structures: Their Use and Effectiveness

The preservation of biological diversity has become a focal issue in current conservation strategies. The global community is beginning to recognize the many deleterious impacts our lifestyle thrusts upon the world's flora and fauna. The survival of these species will depend upon ecosystem protection. The preservation of large tracts of viable habitat is being threatened due to land fragmentation caused from human disturbance. Many of the world's endangered and threatened species are at risk of genetic isolation and potential extinction due to this fragmentation. Roads are one of the main causes of land fragmentation and thus a major reason for the decline and isolation of many wildlife species.

The study of road impacts upon wildlife has ignited in the last decade, but only in the past few years have effective efforts been initiated to mitigate some of their impacts. Engineers and biologists are now making a joint effort to design operative wildlife underpasses and overpasses that will lessen the effect roads have upon wildlife. Effective crossing structures will help to reconnect areas of viable habitat that have become isolated due to road construction. This is an arduous task if you consider the wide-ranging species in the world, their diverse biological needs and habitat requirements. This report will briefly outline some of the harmful impacts roads place upon a diverse array of wildlife species, discuss the important features of wildlife crossing structures and species willingness to use them, and analyze the effectiveness of these structures as a true mitigation tool.

IMPACTS OF ROADS:
As the world population grows, so does the construction and impacts of new roads. Currently, at least 3.9 million miles of public roads crisscross the United States(USDOT, 2001). Almost 200 million vehicles use these roads, and research indicates that road density and miles traveled are increasing rapidly (Foreman, 1998). The research that has been conducted to determine impacts of roads upon wildlife is extensive.


Studies show that highways impact wildlife in the following ways:

• Direct loss of habitat
• Degradation of habitat quality
• Habitat fragmentation. Highways dissect contiguous habitat patches resulting in smaller patch sizes and higher edge to interior ratios.
• Road avoidance. Some wildlife avoid areas adjacent to highways due to noise and human activity associated with noise.
• Increased human exploitation. Highways increase human access for hunting and poaching. This may reduce wildlife populations in areas adjacent to roads and highways and contributes to avoidance.
• Road mortality leading to loss of populations
• Reduced access to vital habitats
• Population fragmentation. (Jackson, 1999 )

An estimated 15-20% of the United States is ecologically impacted by roads, making mitigation efforts a focal issue in the plight to preserve biodiversity. (Foreman, 1998).

Roads and vehicles contribute significantly to a decline in wildlife populations through wildlife/vehicle collisions (roadkill) and road avoidance. An estimated one million vertebrates per day are killed on U.S. highways (Foreman, 1998). Roadkill and road avoidance especially impacts threatened and endangered species, as well as large carnivores with low reproductive rates, low population densities, and large home ranges (Rudiger, 1996). Many large carnivores have to cross numerous highways to meet their biological needs. A male grizzly bear, for instance, has a home range of up to 600 square miles and may need to cross many highways to fulfill its spatial requirements. The Florida panther has a home range of up to 500 km2, which is often crisscrossed by many miles of roads. These species need a contiguous system of wild lands to accommodate their spatial needs, without which their existence is imperiled (Land and Lotz, 1996). Road avoidance threatens the survival of large continuous populations because it creates a barrier effect, thus generating separate, smaller, unconnected populations that have less chance for survival. These smaller, isolated populations have a higher probability for extinction because their recolonization process is blocked. Furthermore, genetic exchange is halted, thus escalating the impacts of natural disturbance.

Our society prizes our ability to travel quickly and unimpeded; we want to move freely despite the impact upon wildlife. However, in the last decade, wildlife crossing structures, ranging from amphibian tunnels to large carnivore open span bridges, have been slowly implemented in various areas of the world to combat roadkill and road avoidance. With the passage of a new federal highway bill in the Congress, the Transportation Equity Act (TEA-21), federal support is available for wildlife crossing structures on both new and existing roads in the United States. "Thanks to TEA-21 and expanded "Transportation Enhancements" category, states and communities can get help not only for crossing structures but for habitat connectivity measures" (USDOT, 2001). The European Union passes a similar measure, Cost-341. Both initiatives have heightened the concern for sustainable transportation systems and incorporation of mitigation structures into their road plans (Gloyne and Clevenger, 2001).

Just now gaining nationwide and global support, there is limited information on the efficacy of these structures. Much of the information and research cited here will come from the Proceedings from the International Conference of Wildlife Ecology and Transportation (ICOWET), interviews with biologists and researchers, and preliminary studies. Biologists and engineers are on a learning curve. Data produced from past and ongoing studies are not definitive. Future studies are needed to properly determine what attributes will help to make these structures the most effective for the greatest number of species. However, the preliminary studies that have been conducted in Europe, Canada, and the United States offer important insights concerning significant features/issues that should be considered when building wildlife crossing structures.

WILDLIFE CROSSING STRUCTURES—FEATURES TO CONSIDER:
Wildlife crossing structures come in many sizes and shapes; their features will ultimately depend on the needs of the myriad of species that inhabit the area of consideration. Physical dimensions of these structures such as size, substrate, noise, temperature, light and moisture will be very important to some species but irrelevant to others (Jackson and Curtice, 1998). Each species has different needs. Therefore, when designing these structures it is virtually impossible and impractical to try and accommodate all species in an area. Instead, a more generalized approach should be taken to make highways permeable for as many species as possible. "Today highway planners and land mangers can ill afford the naïve luxury of single-species mitigation structures. Species do not function in isolation but are components of ecological systems that inherently fall into the category of organized complexity. Therefore, any single-species mitigation structure is likely to have cascading effects, some positive and some negative, on non-target species also. If a mitigation structure is to succeed, a multi-species approach is needed to evaluate the efficacy of such mitigation on non-target species as well"(Clevenger and Waltho, 2000). Biologists and engineers need to determine what target species are in the area being studied, then evaluate what features will work the best for the greatest number of species.
Although the different behavioral and habitat requirements make it impossible to design one perfect structure that suits the needs of all species (Clevenger, pers. comm.), the various attributes of these structures, such as light, noise, substrate, natural cover, dimensions, and placement, will be deciding factors in the usage by some animals and therefore should be taken into consideration.

A. Light can play a key role in a species' willingness to use a crossing structure. The cougar, for example, finds any artificial light will be a deterrent. "Lights are especially detrimental in a road undercrossing or in open habitats" (Beier, 1995). Cougars exploring new terrain prefer dark areas and are less likely to habituate to urban lighting and surroundings than would ungulates who easily habituate to human disturbance. Making the crossing structure large enough to allow for natural lighting would help to address this concern for other sensitive species who, like the cougar, are deterred by artificial lighting (Jackson, 1999). Amphibians are another species that benefits from natural lighting. Tunnels designed for this species have slots on the top to allow natural light to enter. Unfortunately, adding slots can increase noise, which can be an added disturbance for other species.

B. Noise can deter some animals from using a structure. Some biologists recommend that the noise in an underpass not exceed 60 db during the expected time of use (Beier, 1995). Placing skylights in underpasses to allow for natural lighting and moisture was seen as a deterrent in an ungulate study of underpass use because they allowed for traffic noise (Reed, 1975). Therefore, considerable effort should be taken to reduce traffic noise when constructing a crossing structure (Clevenger et al., 2001a). Berms, which are raised sections of earth, have been added to overpasses and underpasses on mitigated highways in order to alleviate noise and light pollution (Highway Service Center).

C. Including natural substrate and internal features, such as logs, refuge poles, and dry terrain, can help make the underpasses more welcoming and more appealing for small prey species (Norman et al., 1998 and Mansergh and Scotts, 1989). However, some studies show that although many animals do prefer a natural floor type, others will readily use underpasses made of concrete (box culverts), and culverts constructed of metal (Norman et al., 1998). Wildlife underpasses in the form of bridges along riparian areas or drainage culverts, should provide raised ledges within the structure to offer dry passageways in times of flooding (Clevenger et al., 2001a, Foresman, pers. comm.). These ledges along river/creek beds are usually concrete, or in some cases metal shelving within culverts. These dry crossings are essential for safe passage.

D. Light, substrate, and noise all can play a role in species acceptance and use of these structures; however, two attributes that seem to have more significance are the structure's dimensions and the presence of natural cover. Most studies indicate that the larger the underpass is the better it will accommodate a wide range of species (Reed, 1981, De Santo and Smith, 1993, and Jackson and Curtice, 1998). Evidence indicates that wider overpasses are more effective (Van Wieren and Worm, 2001, Veenbaas and Brandjes, 1999). These studies recommend a minimum width of 40-50 meters for overpasses. Recommendations for underpass dimensions vary between studies; however, most conclude that bigger structures are less constricting and allow wildlife to feel more a part of their natural environment. Species with special habitat needs, like the salamander or badger, will need smaller tunnels to accommodate their temporal and spatial concerns. Ultimately, the biggest crossing structure benefit the largest number of species.

E. Natural cover near the opening of the wildlife crossing structures can play a significant role in overall effectiveness. Regardless of the structure type (large underpass, small culvert, or overpass), natural vegetation near the opening will give both large and small species the security of their preferred environment (Rodriguez et al.1996, Hunt et al., 1992, Clevenger et al. 2001a). Vegetation also can serve as a funneling system, guiding animals to the openings and thus helping to motivate them to use the crossing structure (Yanes et al. 1995). By increasing natural vegetation around the openings, a connection is established between undisturbed vegetation on both sides of the crossing structure (Hunt et al., 1987). By making this connection, species like the black bear, which prefers covered/forested areas, will be less apprehensive about approaching the opening (Rodriguez et al., 1996). The presence of cover around and inside smaller structures can reduce the risk of these structures becoming prey traps (Hunt et al. 1987). On the other hand, it has been concluded in other studies that natural cover can be a hindrance for species that are wary of ambush opportunities (Jackson and Curtice, 1998). These conclusions simply reiterate the point that finding one attribute that will work for the benefit of all species considerations is virtually impossible.

F. Making sure that there is a clear view to the other side of the crossing structure is another important feature to consider when erecting wildlife crossing structures. "Most researchers have concluded that animals using an underpass should have an unobstructed view of the habitat on the far side of the underpass" (Foster and Humphrey, 1995). Biologists feel that a clear view across will aid in decreasing potential threatening feelings that may be caused by small constricted underpasses that do not provide an ample view of the habitat across the crossing structure (Evink, 1996, Beier, 1995). Two overpasses that were erected in Banff National Park are under reevaluation due to their arched design that impedes cross-highway field of view for wildlife. Wildlife species may be apprehensive to climb into the ‘unknown', thus reducing the structures potential effectiveness (Clevenger, 1999).

PLACEMENT:
Proper placement of the crossing structure is the one attribute that serves the needs of almost all species requirements and thus the most important feature in determining success (Jackson and Curtice, 1998, Rodriguez et al., 1996, Land and Lotz, 1996, Singer et al., 1985). Most studies indicate that placing the crossing structure near traditional migration routes will increase effectiveness. Crossing studies conducted in Florida determined that structures placed without regard to traditional paths failed. "We believe that underpass placement based on knowledge of actual travel routes is more important in determining underpass use than other factors such as structural dimensions" (Foster and Humphrey, 1995). Ungulate studies show similar results; placement near existing game paths will lend a hand to the structure's success (Groot Bruinderick and Hazebroek, 1996). In this Florida study, the panther's traditional crossing points were found through radiotracking, a method now utilized in many studies. To determine proper location, biologists and engineers refer to historical roadkill data, known migratory pathways, or crossing points determined by radio telemetry (Scheick and Jones, Land and Lotz, 1996). Other methods that can help determine proper placement are track count surveys, monitoring trails with infrared cameras, and GIS modeling (Scheick and Jones).

In addition to placing the structures close to know migratory pathways, some biologists also have determined that they should be placed away from human disturbance areas. Human activity near the crossing structures correlates negatively with underpass/overpass use. In the studies conducted in Banff National Park, biologists have continuously reported that underpasses may be ineffective if human activity is not controlled (Clevenger and Waltho, 2000, Clevenger, 1998). Grizzly bears, black bears, cougars, and wolves all have shown avoidance of human use areas (Clevenger, 1998). Many of the underpasses in Banff are close to popular hiking trails and other recreational use areas and are not used as frequently by large carnivores. To reduce the risk of species avoidance due to human presence, scientists and researchers recommend that human foot trails be relocated and human use of underpasses be restricted. Taking these precautions could lead to greater permeability and thus increase habitat connectivity (Clevenger and Waltho, 2000). A wildlife passage study in Spain indicated that foxes and wildcats preferred using passages with a low degree of human disturbance (Rodrigues et al. 1996). Crossing rates were just as low in well-placed passages where human presence was high as were passages placed near unsuitable habitat where human presence was low. By placing the structures away from areas that are frequently used by humans and restricting human use of passages, it can be inferred that the structures will be more appealing to large carnivores and other shy wildlife species.

OVERPASS vs. UNDERPASS:
When determining what type of structure to use, cost benefit analysis can play a significant role. Ideally, overpasses should be utilized whenever possible. "They are less confining, quieter, maintain ambient conditions of rainfall, light and temperature, and can serve both as passageways for wildlife and intermediate habitat for small animals such as reptiles, amphibians and small animals" (Jackson and Curtice, 1998). If connected with the habitat on either side of the highway, overpasses can make the road virtually invisible to wildlife. Once again, if not properly placed, overpasses will not be as effective! Their major drawback is their expense; one overpass can cost as much as one to two million dollars.

Due to their minimal cost and ubiquity in existing road corridors, drainage culverts are increasingly considered a promising mitigation tool(Clevenger et al., 2001a). They should be placed as frequently as possible, and their dimensions should vary in order to accommodate animals of various sizes (Clevenger et al., 2001a). Drainage culverts were erected along the newly widened section of highway 93 in Montana. Ramps have been added to the midsection of the culverts to allow for dry crossing during flooding (Foresman, pers. comm.). Studies are now underway to determine the effectiveness of this approach.

Many biologists believe that the placement and frequency of these structures can be more important than the actual physical features. The use of many inexpensive underpasses along a stretch of road may be more effective than just one or two overpasses (Clevenger, pers.comm.). Biologists conclude that, to be effective, one crossing structure should be placed within each species' home range (Clevenger et al., 2001a). Ideally, we would raise all roads for convenient underpass use or place overpasses every mile. The cost to do this would be implausible. Overall, it may be concluded that, although overpasses are the most species friendly, placing inexpensive underpasses frequently has the potential of making the road more permeable for a greater number of species.

FENCING:
Underpasses and overpasses will not be as effective unless they are accompanied by fencing on both sides of the road. Fences were primarily erected to reduce roadkill; however, without a crossing structure, fencing will further reduce the road's permeability. Most studies indicate that fencing should be about eight feet tall (Groot Bruinderink and Hazelbrook, 1996, Sipes and Neff, 2001). Fencing should be built around the crossing structure to guide (funnel) animals to the passageway, thus preventing them from circumventing the system (Jackson and Curtice, 1998). This design may not work for all species, and innovations have been made in recent years. For instance, instead of a fence, a 3 1/2 foot concrete wall with a "lip" (to push off the animals that can scale the wall) has worked to keep toads, frogs, snakes and other amphibians off the highway (USDOT, 2001, Jackson and Curtice, 1998). In a study evaluating desert tortoise use of culverts, fencing proved to be ineffective, due to escape routes made by small gaps created by improperly installed or maintained fencing (Boarman, 1996). Other studies also have concluded that roadkill can increase if fences are not regularly checked and properly maintained.
Some animals can climb over or burrow under the fences. Along the Trans-Canada Highway (TCH), grizzly bears, black bears, and cougars can easily climb over the fencing, and coyotes can burrow under the fence where gaps exist. In certain areas, measures have been taken to mitigate this shortcoming. Burying the fence for burrowing animals or lifting the fence (Groot Bruinderick and Hazelbrook, 1996), and adding barbed wire, outriggers, or 1-m extensions at the top for species like the black bear and mountain lion that can climb over the top can aid in reducing the number of animals able to circumvent the fence (Sipes and Neff, 2001, Gibeau and Heuer, 1996, Clevenger et al., 2001b). The addition of one-way gates also has proven useful in helping trapped animals that have managed to scale the fence to safely escape (Reed et al. 1975).

Regardless of the time and effort spent to prevent passage around the fencing, some will ultimately find a way in. In Banff, three cougars were killed on a mitigated (fenced) stretch of the highway. Deer and antelope can jump a 12-foot fence (Sipes and Neff, 2001), and mischievous bears will surely find a way over if they try hard enough. "This highlights the importance of fence maintenance and the need for a more effective fence design to discourage climbing (or jumping)"(Clevenger et al., 2001b).

The one major drawback to fencing is their potential as prey traps. In Banff National Park, coyotes have been running bighorn sheep into the fence along the Trans-Canada Highway. The bighorn sheep's escape terrain along the cliffs has been removed by the fence. In 1988, 14 sheep were killed along the TCH, and by 1991, the number rose to 47 (Gibeau and Heuer, 1996). Few sheep use the area today. Wolves and cougars also have been observed herding deer up against highway fencing (Foster and Humphrey, 1995). Fences have proven to be successful in reducing wildlife vehicle collisions and funneling animals to crossing structure; however, they are not infallible and steps should be taken to improve their utility.

ARE THESE STRUCTURES EFFECTIVE?
Crossing structures are slowly being incorporated into the road plans of transportation agencies around the globe; however, very little research has been done on their true effectiveness. "This is a field of applied ecology still very much in its infancy" says Parks Canada researcher Tony Clevenger (Wilkinson, 2000). Clevenger also adds that, "virtually nothing is known regarding the effectiveness between overpasses and underpasses, or even between the different types of underpasses being tested" (Clevenger, 1998). We can make inferences about what features are the most important and generalizations and decisions based on the studies that have been conducted so far. However, no study has been able to accurately compare how many animal crossings would have occurred if the roads were not there.

The studies that have attempted to determine effectiveness are arbitrary due to the fact that there is a dearth of data on the population density and behavior of species without the presence of roads. Data is needed on population density and migration patterns before a road is constructed, after it is built, and then again once a mitigation structure is erected (Hardy, pers. comm.). It is important to consider animal communities and their processes and not only a few selected species. Knowledge of the abundance and distribution of populations is essential in developing criteria to measure effectiveness (Clevenger, 1999). In Banff, the studies conducted on the TCH have shown that for the species using the underpasses regularly fragmentation effects may be considerably reduced. "However, to determine whether the TCH is fragmenting wildlife populations in BNP and the Bow Valley we need to know something about crossing and movement prior to highway construction, which unfortunately is impossible (Clevenger, 1998). To try and obtain expected crossing rates, movement patterns, population densities, and life history requirements, years of study is needed. "Answering some of the complex ecological questions around roads and long-lived wildlife like bears may require research timeframes of up to 10-15 years"(Whyte Museum, 2000). Intensive monitoring of the habitat and populations are needed to try and determine expected crossings (Hardy, pers. comm.) To obtain such information, mark-recapture studies, radiotracking, video cameras, tracking counts, will all help to determine ‘expected' crossing rates. However, this takes vast amounts of time and money (Jackson, 1999). Even then, biologists will need to take into consideration that a low crossing rate may be due to a natural fluctuation/decline in wildlife populations.

Low crossing rates may also be attributed to the duration of time it takes certain species to adapt to a crossing structure. If a study on effectiveness is conducted one year after installation crossing rates can be exceedingly different than a study conducted 4 years later when adaptation has occurred. Further studies show that, "ungulates and other mammals ‘remember' the location of such underpasses and adjust movements accordingly"(De Santo and Smith, 1993). In Florida the number of Panther crossings have increased, and may be attributed to acceptance and learning (Land and Lotz, 1996). Effectiveness studies done years after the inception of crossing structures may have more definitive results because enough time has been allotted for adaptation (Boarman, 1996).

When evaluating the efficacy of wildlife structures, biologists, engineers, and land managers need to realize that crossing structures alone are not enough. If the habitat on either side of the crossing structure is not preserved then these crossings are essentially bridges to nowhere (Servheen, personal comm.). "We believe that wildlife underpasses are but one tool, and cannot be viewed as the ultimate panacea in terms of mitigating future adverse effects of roadways on black bears and other wildlife" (Gilbert and Wooding, 1996). Management of surrounding area must be initiated along with these structures (Rodrigues et al., 1996). "Regardless of the type of crossing object, an adequate land use management, aimed at conserving/enhancing the quality of habitat in the surrounding area, will be crucial!"(Kobler and Adamic, 1999).

WHAT DOES EFFECTIVENESS MEAN?
A central question that needs to be addressed is: what does ‘effectiveness' really mean? Different biologists and engineers will have varying points of view on what makes these crossing structures truly effective (Hardy, pers comm., and Clevenger, pers. comm.). If the sole purpose is to reduce roadkill, then, yes, they are effective. Roadkill has been dramatically reduced on most highways that have both fencing and crossing structures. On Interstate 80 in Wyoming, road kills of mule deer have been reduced by 90%(DeSanto and Smith, 1993). There has been a 97% decrease in the number of elk killed in Banff, and other Park species have seen similar declines in mortality (Savage, 2000). But effectiveness involves more than decreasing roadkill.

The issue of effectiveness becomes more complex when discussing road permeability. "One conservation geneticist might define success with one grizzly bear passing per grizzly bear generation (11-13 years), while a conservation biologist might expect more use"(Whyte Museum, 2000). There is never one golden movement or one perfect number (Hardy, pers. comm.). Bruce Leeson feels that the underpasses in Banff National Park have been extremely successful, with approximately "21,000 safe passages across the underpasses and overpasses documented in the last three years" (Savage, 2000), while others like Paul Paquet and Michael Gibeau are skeptical about how well they are working for large carnivores. Wolves are not using the underpasses in Banff on the TCH as much as researchers had hoped. They are showing signs of hesitance and some are turning back midway through the crossing structure. Underpasses may force wolves to modify their travel patterns, resulting in fewer crossings. This in turn may be disrupting pack structure and cohesiveness (Pacquet, 1996). Pacquet believes that, "safe passages are virtually meaningless because there are no data showing how animals used the valley before the highway was built" and that having a few animals cross does not prove success (Savage, 2000). With regard to grizzly bear use, Mike Gibeau, a biologist for the Eastern Slope Grizzly Bear Project, has indicated that no more than 10 percent of the bears are crossing the highway. Bears are reluctant to cross and have also abandoned rich feeding ground that lies close to traffic (Savage, 2000).

On the other hand, Tony Clevenger, a biologist studying the efficacy of the crossing structures in Banff National Park has stated that every year the number of large carnivores crossing the highway is increasing. He claims that about 18 grizzlies in the past year have used the crossings, along with one collared female (Clevenger, pers. comm.). Once again, it is important to compare these numbers with known population demographics in the area of study. "The low use of crossing structures by adult female grizzlies is largely a function of the fact that very few adult females' home ranges include areas in the immediate vicinity of the twinned, mitigated sections of the TCH. If adult female grizzlies tend not to use habitats near the highway, how will they discover and learn to use the crossing structures?"(Whyte Museum, 2000, Clevenger pers. comm.). The more information that becomes available on species distribution, abundance, and ecological needs in the area, the easier it will be to accurately measure mitigation structure effectiveness (Clevenger, 1999). All these issues must play a role in evaluating effectiveness, making any definitive assertion concerning efficacy an arduous task when considering the dearth of scientific data and plethora of conflicting opinions within this field.

CASE STUDIES: Examples of functional and proposed mitigation structures in North America.

A. Banff National Park and Mitigation along the TransCanada Highway (TCH)
A major highway bisects Banff National Park, located in Alberta, Canada. The Trans-Canada Highway (TCH) runs straight through the center of Canada's first and most celebrated national park, with an estimated 9.5 million people entering the Park each year. Some enter the Park for recreation, while others enter simply to find the quickest route from to the Rockies via the TCH. The Tran-Canada highway, which opened in 1962 (Woods, 1996), has often times been called the ‘meateater' due to the plethora of wildlife killed by vehicles. The TCH bisects the Bow River Valley in Banff, which is the home to many species of wildlife due to its rich diversity of habitat. In fact over 70% of the montane habitat in Banff is contained within the Bow Valley (Gloyne and Clevenger, 2000). Many of the Parks 54 mammal species and 280 bird species reside in this area of the Park, and therefore, many perish when trying to cross the TCH to get to habitat on the other side (Gibeau et al. 1998). For wide ranging species like the grizzly bear and cougar who have large home ranges and small reproductive rates this highway is especially detrimental not only because of roadkill but as a result of habitat fragmentation. With almost 14,000 vehicles a day driving through the Park carnage has been high and road permeability extremely low.

To increase habitat connectivity and decrease mortality from wildlife-vehicle collisions 22 underpasses and 2 overpasses (see above photo) have been added to the first 45km of the TCH as the road was being expanded from a 2 to a 4-lane highway. The fist 11 underpasses were completed in 1986, while the later wildlife structures, including the 2 overpasses, were erected in 1997. The types of underpasses constructed varied along the 45km stretch ranging from open-span bridges to metal and box culverts. Wildlife exclusion fencing 2.4 meters in height was added in conjunction to the crossing structures to further reduce wildlife carnage.

Due to the wildlife crossing structures and the mitigation fencing, there have been 80% fewer accidents involving wildlife on the TCH (Clevenger et al. 2001b). Bruce Leeson, environmental advisor in BNP, says that statistics show a 97% decrease in the number of elk killed on the road, and believes that the underpasses and overpasses have helped to maintain the natural travel patterns of wildlife (Savage, 2000). "In the 35 months of monitoring, there have been 15 grizzly bear passes, nine passes from three known radiomarked adult males and six from unknown grizzlies. During the same period black bears have used the crossing structures for cross-highway travel 355 times, cougars 351 times, wolves 256 times and coyotes 3,000 times. Elk have counted for more than 12,000 passes, deer more than 4,000, and sheep 603 passes"(Clevenger, 1999). Although the wildlife crossing structures have resulted in a reduction of roadkill and has increased road permeability for wildlife there is still much controversy revolving around the benefits of these mitigation efforts for large carnivores. Three cougars have been killed recently along the mitigated section showing that the structures are not infallible and the fence in not insurmountable (Gloyne and Clevenger, 2001). Black bears, grizzly bears, and cougars are still climbing over the fencing, and coyotes are finding gaps in the fence to access the right-of-way (Clevenger et.al., 2001b).

There are still improvements to be made, and extensive monitoring to be done. Research on the effectiveness of the mitigation structures began in 1996. As alluded to earlier in this report, the answers concerning how effective the structures are depends on a multitude of factors. Knowing the number of species that cross the underpasses means very little unless the abundance and distribution of that species is evaluated. "For instance, 200 elk crossing at one underpass is not necessarily more important than 2 grizzly bear crossings"(Clevenger, 1999). Studies are still underway in Banff to help determine effectiveness. Biologists and researchers are working on a multi-species approach to determine the efficacy of the 22 underpasses, 2 overpasses, and mitigation fencing. Years of monitoring and research are still needed. Although more research is requisite, the crossing structures have significantly reduced roadkill and have increased road permeability for the diversity of wildlife in the Bow Valley.

B. Florida and I-75
Florida was one of the first states to recognize and react to the detrimental impacts roads have had on our Nation's wildlife. Florida's human population has increased from 1.7 to over 14 million since 1936(Gilbert, 1996). This human increase has led to the development and expansion of roads, greater traffic density, faster speedways, and increased land fragmentation. When trying to evaluate the effects of roads on wildlife in Florida the two target species have been the Florida panther and black bear due to their small population sizes and low reproductive rates. Highway I-75, also known as Alligator Alley, was of particular concern due to the high density of the endangered Florida Panther in the habitat surrounding this highway. Between 1978 and 1994 twenty panther deaths and six injuries were attributed to vehicles on I-75 (Defenders of Wildlife).

To mitigate the effects that the expansion of the highway would have on the endangered Florida panther, 24 underpasses were installed along a 64 km fenced portion of the highway. To determine location for the wildlife crossings, roadkill and radio telemetry data were assessed to find the best place to erect the new underpasses. In addition to the underpasses, a 3.4 meter chain link fence topped with 3 strands of outrigger barbed wire was added to prevent panthers and other local species from crossing the busy highway.

Continuous monitoring and evaluation has shown that the crossing structures have been successful in reducing roadway mortality of both the black bear and panther (Gilbert and Wooding, 1996). Since construction, no panther has been killed on the former Alligator Alley. "The underpasses also reduced fragmentation of animals home ranges encompassing both sides of the highway. However, full evaluation of the latter effect requires analysis of movement patterns"(Foster and Humphrey, 1995). All of the medium and large sized animals in Southern Florida have used the various underpasses on I-75 (Land and Lotz, 1996). Although primarily constructed for the panther, the crossing structures erected in Florida on I-75 have had a cascading effect and have thus been beneficial for a wide range of non-target species such as the bobcat, deer, great blue heron, wild turkey, and alligators (Foreman and Hersperger).

C. Glacier National Park and Highway 2
The construction of goat underpasses in Glacier National Park on U.S. Highway 2 can be seen as a successful mitigation attempt. The mountain goats crossed busy U.S. highway 2 in order to reach a natural salt lick along the Middle Fork of the Flathead River (USDOT, 2000). Increased attraction to the goat lick area by tourists escalated the traffic volume to 34,000 vehicles in 1975 (Singer et al., 1985). The goats were often hesitant to cross the highway, they altered their initial crossing routes, and on occasion nannies were separated from their kids (Singer et al., 1985). The addition of two underpasses in 1980 decreased the potential for mortality and habitat fragmentation. When building the underpasses the location was determined by observing the area where the greatest number of goat crossings had occurred (Singer et al., 1985). To increase security of the underpass, conifer trees were planted near the bridge rail, existing goat trails were obliterated, and new trails guiding the goats to the underpass were added (USDOT, 2000). An 8-foot fence was utilized to deter the goats from crossing U.S. Highway 2. After the first year, 99.4 percent of the goats that were observed used one of the two underpasses, and by 1984 all of the goats used the underpasses (USDOT, 2000). The success can be attributed partly to the location of the underpasses. In addition, the use of conifer covering to make the underpass more shielded and safe helped with their success (Singer, 1985). The use of the fencing, natural cover, and natural goat movement routes, were more important to the success of the underpasses than were the underpasses actual physical features. The road is there to stay and a successful effort was made to mitigate its effects. Now, over twenty years since the erection of the underpasses, goats are still using them to reach the salt lick.

D. Salamander Tunnels:
Two tunnels built specifically to mitigate the effects of a 2-lane street in Amherst Massachusetts on the migration pattern of a local salamander population is a perfect example of a species-specific successful mitigation attempt. Once every spring salamanders must migrate at night to wetland ponds to breed. The salamander population in Amherst Massachusetts had to cross Henry Street during their breeding season to get to the wetlands. Local residents began noticing the many salamanders killed on the roads during this time of year and began carrying the salamanders safely across the road in buckets. This attempt to get the salamanders to their breeding pools via buckets was referred to by volunteers as the "bucket brigade". To eliminate the need for these brigades and to permanently facilitate the salamanders' migration in the spring, two tunnels were built 200 feet apart near the salamanders' normal crossing route. A drift fence was also added to funnel the salamanders to the newly constructed tunnels. The salamanders were marked so that they could be easily tracked for effectiveness studies. The results showed that three quarters of the salamanders were able to use the tunnels and successfully get to their breeding habitat. Adding the fence and making the tunnels species friendly by adding a slotted top to let in light and provide the damp conditions salamanders prefer helped to make these tunnels a successful mitigation attempt. (USDOT, 2000)

E. U.S. Highway 93
U.S. Highway 93, which crosses the Flathead Indian Reservation in western Montana, will soon become a model in the United States illustrating that the combined efforts of local residents, county, local and federal governments can result in an innovative plan to consider wildlife and land ethics when building or reconstructing highways. U.S. Highway 93 from Evaro to Polson is scheduled to be widened over the next decade to improve safety. The area under consideration is approximately 56 miles long. Thanks to the agreement between the Montana Department of Transportation, the Confederated Salish & Kootenai Tribes, and the Federal Highway Commission the highway will be reconstructed with tribal concerns at the forefront. "Traditions teach that land, wildlife, and people are all deeply connected. A highway designed only to connect points on a map can sever some of the most important connections on the land, impacting the lives of many generations who live there…Decisions about the highway are decisions about the land."(Sipes, 2000). The new road will be designed to fit the landscape and enhance scenic vistas. One measure being taken to accomplish this goal is the proposed addition of 42 crossing structures on this 56-mile stretch of highway to aid in restoring/preserving land connectivity. The crossing structures range from small fish culverts to a wide-open span overpass. Fencing will also be added along the mitigated stretch to keep wildlife off the road and to act as an aid in funneling species to the new structures. Most of the crossing structures are being constructed on a multi-species approach in order to accommodate as many species as possible. To determine location the tribal biologists and highway architects have evaluated road-kill data and tracking information. Construction for this project is scheduled to begin in 2003. "Highway 93 is designed to be an eye-opening project, part of a new generation of highways that bring communities together"(Sipes, 2000).

CONCLUSION:
Wildlife crossing structures have had some great successes. They have been useful in decreasing roadkill, and have been successful in enhancing landscape connectivity. However, the dearth of information on effectiveness makes further studies essential if biologists and engineers are expected to make the crossings better for a diverse array of wildlife (Clevenger, 1998). Regardless of how much data is collected, there will never be one perfect structure to suit all specie's needs (Clevenger and Hardy pers. comm.). Therefore, "our first choice would always be to not build a road through wildlife habitat."(Lavendel, 2000). For already existing roads, wildlife passages should be added thoughtfully with all species considerations in mind. The following list provides some salient recommendations to consider when developing or monitoring wildlife crossing structures:

• Take a multi-species approach rather than a single-species focus, remembering that species do not function in isolation but are components of ecological systems
• Know the biology of the species in the area; their distribution, abundance, and ecological and behavioral needs
• Place the structures at know migration routes, away from human disturbance. This can be determined by; roadkill data, infrared cameras, GIS modeling, track-count surveys
• Make the passages wide to accommodate a larger number of species
• Try to build them to allow for natural lighting and low noise levels
• Have a clear view to the other side
• Use fencing designed to reduce wildlife intrusions
• Intensive monitoring before and after construction of the wildlife passages via track count surveys, radio-collaring, mark-recapture studies, etc.
• Share results!

It is imperative that biologists and engineers share their findings and ideas concerning crossing structure effectiveness. Events such as the International Conference on Wildlife Ecology and Transportation have become integral steps in this process. Ideally, the knowledge gained from these meetings will lead to the design of improved crossing structures that will more effectively connect the habitats of a diversity of wildlife.

Literature Cited

Beier, Paul. 1995. Dispersal of Juvenile Cougars in Fragmented Habitat. Journal of Wildlife Management, 59 (2): 228-237.

Boreman, W.I. 1996. Highway Mortality in Desert Tortoises and Small Vertebrates: Success of Barrier Fences and Culverts. in G.L. Evink, P.A. Garrett, D. Zeigler, and J. Berry, eds. Proceedings of the International Conference on Wildlife Ecology and Transportation. June, 1996 Tallahassee, FL.

Clevenger, A.P. 1998. Permeability of the Trans-Canada Highway to Wildlife in Banff National Park: Importance of Crossing Structures and Factors Influencing their Effectiveness. Pp.109-119. in G.L. Evink, P.A. Garrett, D. Zeigler, and J. Berry, eds. Proceedings of the International Conference on Wildlife Ecology and Transportation. Feb. 10-12, 1998 Fort Myers, FL.

Clevenger, A.P. and Waltho, Nigel. 2000. Factors Influencing the Effectiveness of Wildlife Underpasses in Banff National Park, Alberta, Canada. Conservation Biology. 14(1): 47-56.


Clevenger, A.P. 2001. Parks Canada Wildlife Biologist. October 10, 2001. Personal communication.

Clevenger, A.P., Chruszcz, B., and Gunson, K.E. 2001(unpublished report). Drainage Culverts as habitat linkages and factors affecting passage of mammals. Journal of Applied Ecology. In Press.

Clevenger, A.P., Chruszez, B., and Gunson, K.E. 2001. Highway mitigation fencing reduces wildlife-vehicle collisions. Wildlife Society Bulletin, 29(2): 646-653.

Defenders of Wildlife. Habitat and Highways Campaign. http://www.defenders.org

DeSanto, R.S. and Smith, D.G. 1993. Environmental Auditing: An introduction to Issues of Habitat Fragmentation Relative to Transportation Corridors with Special Reference to High-Speed Rail (HSR). Environmental Management. 17(1): 111-114.

Evink, G.L. 1996. Florida Department of Transportation Initiatives Related to Wildlife Mortality. in G.L. Evink, P.A. Garrett, D. Zeigler, and J. Berry, eds. Proceedings of the International Conference on Wildlife Ecology and Transportation. June, 1996 Tallahassee, FL.

Foster, M.L. and Humphrey, S.R. 1995. Use of highway underpasses by Florida panthers and other wildlife. Wildlife Society Bulletin, 23(1): 95-100.

Foreman, R.T. and Alexander, L .E. 1998. Roads and their Major Ecological Effects. Annual Review of Ecological Systems. 29: 207-231.

Foreman, R.T. and Hersperger, A.M. Road Ecology and Road Density in Different Landscapes, with International Planning and Mitigation Solutions. Harvard University, Graduate School of Design, Cambridge, Mass.

Foresman, Kerry. 2001. Professor of Wildlife Biology. October 3, 2001. Personal Communication.

Garrett, L.C. and Conway, G.A. Characteristics of Moose-vehicle Collisions in Anchorage, Alaska, 1991-1995. Journal of Safety Research. 30(4):219-223.

Gibeau, M.L. and Heuer, K. 1996. Effects of Transportation Corridors on Large Carnivores in the Bow River Valley, Alberta. in G.L. Evink, P.A. Garrett, D. Zeigler, and J. Berry, eds. Proceedings of the International Conference on Wildlife Ecology and Transportation. June, 1996 Tallahassee, FL.

Gilbert, T. and Wooding, J. 1996. An Overview of Black Bear Roadkills in Florida 1976-1995. in G.L. Evink, P.A. Garrett, D. Zeigler, and J. Berry, eds. Proceedings of the International Conference on Wildlife Ecology and Transportation. June, 1996 Tallahassee, FL.

Gloyne, C.C. and Clevenger, A.P. 2001. Cougar puma concolor use of wildlife crossing structures on the Trans Canada highway in Banff National Park, Alberta. Wildlife Biology, 7(2): 117-124.

Groot Bruinderink, G.W.T.A. and E. Hazelbrook. 1996 Ungulate Traffic Collisions in Europe. Conservation Biology 10(4):1059-1067.

Hardy, Amanda. 2001. Biologist for the Western Transportation Institute. October 21, 2001. Personal Communication.

Highway Service Center-Parks Canada Agency, Trans-Canada Twinning Project-Banff National Park. http://www.hsctch-twinning.ca/

Hunt, A., Dickens, H.J., and Whelan, R.J. 1987. Movement of mammals through tunnels under railway lines. Australian Zoologist. 24(2) 89-93.

Jackson, S.D. and Curtice, G.R. 1998. Toward a Practical Strategy for Mitigating Highway Impacts on Wildlife. Pp. 17-22 in G.L. Evink, P.A. Garrett, D. Zeigler, and J. Berry, eds. Proceedings of the International Conference on Wildlife Ecology and Transportation. Feb. 10-12, 1998 Fort Myers, FL.

Jackson, S.D. 1999. Overview of Transportation Related Wildlife Problems. Pp.1-4. in G.L. Evink, P. Garrett, and D. Zeigler, eds. Proceedings of the Third International Conference on Wildlife Ecology and Transportation. Sept. 13-16, 1999 Missoula, MT.


Kobler, A. and Adamic, M. 1999. Brown Bears in Slovenia: Identifying Locations for Construction of Wildlife Bridges Across Highways. Pp. 29-28. in G.L. Evink, P. Garrett, and D. Zeigler, eds. Proceedings of the Third International Conference on Wildlife Ecology and Transportation. Sept. 13-16, 1999 Missoula, MT.


Land, D. and Lotz, M. 1996. Wildlife Crossing Designs and Use By Florida Panthers and Other Wildlife In Southwest Florida. in G.L. Evink, P.A. Garrett, D. Zeigler, and J. Berry, eds. Proceedings of the International Conference on Wildlife Ecology and Transportation. June, 1996 Tallahassee, FL.

Lavendel, Brian. 2000. Putting the Breaks on Roadkill. Animals. 133(6):20-23.

Mansergh, I.M., and Scotts, D.J. 1989. Habitat continuity and social organization of the mountain pygmy-possum restored by tunnel. Journal of Wildlife Management. 53(3):701-707.

Norman, T., Finegan, A., and Lean, B. 1998. The Role of Fauna Underpasses in New South Wales. Pp. 195-208. in G.L. Evink, P.A. Garrett, D. Zeigler, and J. Berry, eds. Proceedings of the International Conference on Wildlife Ecology and Transportation. Feb. 10-12, 1998 Fort Myers, FL.
Paquet, P. 1996. Effects of Linear Development on Winter Movements of Gray Wolves in the Bow River Valley of Banff National Park, Alberta. in G.L. Evink, P.A. Garrett, D. Zeigler, and J. Berry, eds. Proceedings of the International Conference on Wildlife Ecology and Transportation. June, 1996 Tallahassee, FL.

Pedevillano, C. and Wright, R.G. 1987. The Influence of Visitors on Mountain Goat Activities in Glacier National Park, Montana. Biological Conservation. 39:1-11.

Reed, D., Woodard, T.N., and Pojar, T.M. 1975. Behavioral Response of Mule Deer to a Highway Underpass. Journal of Wildlife Management. 39(2): 361-367.

Reed, Dale. 1981. Mule Deer Behavior at a Highway Underpass Exit. Journal of Wildlife Management, 45(2):542-543.

Rodriguez, A., Crema, G., and Delibes, M. 1996. Use of non-wildlife passages across a high speed railway by terrestrial vertebrates. Journal of Applied Ecology. 33: 1527-1540.

Rodriguez, A., Crema, G., and Delibes, M. 1997. Factors affecting crossing of red foxes and wildcats through non-wildlife passages across a high-speed railway. Ecography. 20:287-294.

Romin, L.A. and Bissonette, J.A. 1996. Deer-vehicle collisions: status of state monitoriang activities and mitigation efforts. Wildlife Society Bulletin. 24(2):276-283.

Ruediger, B. 1996. The Relationship Between Rare Carnivores and Highways. in G.L. Evink, P.A. Garrett, D. Zeigler, and J. Berry, eds. Proceedings of the International Conference on Wildlife Ecology and Transportation. June, 1996 Tallahassee, FL.

Ruediger, B. 1998. Rare Carnivores and Highways - Moving Into the 21st Century. Pp. 10-16 in G.L. Evink, P.A. Garrett, D. Zeigler, and J. Berry, eds. Proceedings of the International Conference on Wildlife Ecology and Transportation. Feb. 10-12, 1998 Fort Myers, FL.


Savage, Candace. 2000. A Highway Runs Through It. Canadian Geographic. 120(5): 35-42.

Scheick, B.K. and Jones, M. D. Locating Wildlife Underpasses Prior to Expansion of Highway 64 in North Carolina, North Carolina Wildlife Resources Commission, Plymouth and Bridgeton, North Carolina. www.myflorida.com/emo/sched/locate.pdf.

Servheen, C. U.S. Fish and Wildlife Service Grizzly Bear Recovery Coordinator. Nov, 2001. personal Communication.

Singer, F.J., Langlitz, W.L., and Samuelson, E.C. 1985. Design and Construction of Highway Underpasses Used by Mountain Goats. Transportation Research Record. 1016:6-10.

Sipes, J.L. and Neff, J. 2001. Fencing, Wildlife Crossings, and Roads: separating animals and vehicles. Landscape Architecture. 91(6):24-27.

Sipes, J.L. New respect for the land drives highway design: With landscape architects on the team, a new way is found for US 93. http://www.djc.com/news/enviro/11121025.html

U.S. Department of Transportation. 2000. Critter Crossings: Linking Habitats and Reducing Roadkill. Federal Highway Administration.

Van Wieren, S. E. and Worm, P.B. 2001. The use of a motorway wildlife overpass by large mammals. Netherlands Journal of Zoology. 51(1):97-105.

Veenbaas, G and Brandjes, J, 1999. Use of fauna passages along waterways under highways. Pp. 253-259 in G.L. Evink, P. Garrett, and D. Zeigler, eds. Proceedings of the Third International Conference on Wildlife Ecology and Transportation. Sept. 13-16, 1999 Missoula, MT.

Whyte Museum. 2000. Mitigation: Reducing the Impacts of Roads and Railways on Bears. http://www.whyte.org/bears/mitigate.html.

Wilkinson, Todd. Making the road safe - for elk, bears, and wolves. The Christian Science Monitor. 12/09/00. http://www.csmonitor.com/durable/2000/12/19/text/p3sl.html.

Woods, J. 1996. Roads, Rails & the Environment: Wildlife at the Intersection in Canada's Western Mountains. . in G.L. Evink, P.A. Garrett, D. Zeigler, and J. Berry, eds. Proceedings of the International Conference on Wildlife Ecology and Transportation. June, 1996 Tallahassee, FL.

Yanes, M., Velasco, J.M., and Suarez, F. 1995. Permeability of Roads and Railways to Vertebrates: The Importance of Culverts. Biological Conservation. 71:217-222.