Habitat fragmentation, habitat loss, and human caused mortality are the major factors contributing to wildlife decline throughout the world. High-speed, heavily-used highways can divide formerly contiguous blocks of habitat and isolate wildlife populations. Underpasses and culverts have the potential to mitigate the negative impacts of roads on wildlife populations by maintaining connectivity between wildlife populations and decreasing wildlife collisions with vehicles. Using heat and motion sensitive cameras, we monitored seven underpasses and three culverts for ten months along Interstate 90 in western Montana. We documented the type and frequency of wildlife use and compared the level of use to variables associated with the structures. Wildlife use was most frequent at underpasses and limited in culverts. Ungulates were the primary users of underpasses with limited use by medium and large carnivores. We found no significant relationships between wildlife use and structural variables. This is partly due to a small sample size but could be the result of animals using structures opportunistically.

The fragmentation of carnivore habitat in the Rocky Mountains on both sides of the U.S.-Canada border is an ongoing threat to the survival and recovery of these populations. Human developments are the cause of this fragmentation. Major developments causing fragmentation include private land conversion into homesites and highway construction and improvement. If carnivores such as grizzly bears (Ursus arctos horribilis), wolves, (Canis lupus), wolverines (Gulo gulo), lynx (Lynx lynx), and fishers (Martes pennanti) are to survive and recover to healthy population levels in the Rocky Mountains, the issue of fragmentation must be addressed in a proactive and effective manner.

Throughout the world, the effects of highways and railroads on wildlife have been of great concern to scientists, land and wildlife managers, and the public, for over 80 years. Through these years, many researchers have sought to understand and mitigate the negative impacts of roads through theoretical and empirical research. However, to our knowledge, no one has investigated the underlying probability theory that likely governs the extent to which linear transportation features result in wildlife mortality. One reason may be that the number of factors potentially influencing observed patterns of road mortality can be quite large and can quickly become intractable. Our objective here was to suggest that the lethality of linear transportation features to wildlife is governed primarily by two factors: traffic volume and time spent on the roadway. Using a simple Poisson model of expected vehicle arrival times, we estimated the probabilities of animals successfully crossing roads under different traffic volume and animal mobility constraints. We used actual vehicle counts from two study areas as examples, and used a study of grizzly bears along a major railroad and highway to illustrate these concepts. We discuss the usefulness of this approach to conservation problems, and place it in context with other efforts to quantify the occurrence of wildlife mortality due to highways. Our hope is that these ideas will clarify and advance the search for solutions to what previously has been an intractable problem.

Highway construction and expansion through bear habitat can negatively affect brown bear populations. Highway structures can decrease habitat availability through habitat loss and restricted access, roads often displace animals and cause re-direction of natural movements, and highways can act as barriers to decrease gene flow. Lastly, highway traffic can cause direct bear and human mortality through car-animal collisions. We examined the spatial and temporal distribution of brown bear crossings of the Sterling and Seward Highways on the Kenai Peninsula, Alaska. Data were collected between 1995 and 2001 as part of an ongoing population study. We created random walking bears within each bear’s home range and compared the geographic distribution of each bear’s highway crossing locations to the random crossing locations to assess whether the number and pattern of crossing locations were different than expected. An information theoretic approach comparing logistic regression models was used to determine if traffic volume, distance to cover across the highway, road density, and distance to the closest stream crossing were related to locations bears crossed the highway. We conducted a second set of analyses comparing models based on the temporal factors of daylight versus darkness, mean bear movement per hour, and traffic volume. Most bears crossed the highway less frequently than expected. While locations where bears crossed the highway were clustered, none of the spatial models tested strongly explained the observed clustering. Bears were more likely to cross the highway during nighttime. Additional research will be necessary to identify the cues bears use to choose locations to cross the highway.

Successful movement of wildlife across highways to effectively provide population-level wildlife linkage is usually viewed in one dimension - movement either exists or it does not. We believe that there are multiple ways to measure both the existence and value of such movement opportunities to better demonstrate success or failure of these efforts. The use of multiple methods to measure success will provide quantitative and qualitative values that can be used to better judge the effectiveness of wildlife movement across highways and to justify investment in the in¬frastructure to create such movement. Currently, many transportation agency administrators view investments in wild¬life crossing infrastructure as outside their responsibility and as fiscal competition for highway projects of greater value to the traveling public. We believe this is a false paradigm that can be changed by enhanced measures of the values of wildlife highways crossing enhancement. Measures of success should include a wide range of factors that transcend wildlife issues. These factors should include biological impacts; economic impacts on highways, public lands, and private lands; public safety measures; social influences and acceptance; and political factors. Biological measures of success should include wildlife movement, gene flow measures, seasonal range access and dispersal opportunity, potential for re-occupancy of historic habitat, reduction in population isolation, effects on endangered species listing and management, and mortality reduction. Economic successes should include improved project planning efficiency, reductions in project time delays, reduced environmental review and court challenge costs, and improved land values adjacent to linkage areas due to healthy wildlife populations. Public safety measures should include road kill reduc¬tion, reduced probability of accidents and human injury, and improved speed limits. Social measures should include attitude surveys measuring public willingness to invest public funds to reduce wildlife collisions, public acceptance of the concept of linkage zones, and public awareness of the multiple benefits of wildlife population connectivity. Political measures should include measuring the knowledge and understanding of this issue by political interests, their willing¬ness to appropriate funding for such projects, and legislation. We review the application of each of these measures of success to wildlife crossing enhancement and suggest a basic measurement approach to all wildlife crossing efforts. In the long term, successful wildlife linkage efforts associated with highways will require improved public understand¬ing and support, improved agency willingness to accept wildlife crossings as part of their responsibility, and improved understanding of the multiple values and benefits that come from enhancing wildlife movement across highways.

Many human activities affect how bears use habitat. The effects of motorized recreational vehicle use on trails have not been formally assessed previously. Potential effects include displacement from and avoidance of high quality habitat, either temporally or spatially, which could affect reproduction and survival and result in fewer bears. Focusing on displacement, we used hourly locations from four GPS-collared female bears in the Badger-Two Medicine area in the Lewis and Clark National Forest to assess spatial and temporal distributions of bears relative to trail locations and to recreational use on trails. When availability was defined as circles equal to 95 percent of move distances around the previous bear location, all bears used areas near trails less than expected. We iteratively reclassified trail habitat versus non-trail habitat as increasing buffers in 50m increments around trails until we reached a buffer-width at which bears used areas near trails in proportion to availability. Compositional analysis results showed that bears selected against areas within 250 - 900m from ATV trails and within 450 - 600m from single-track trails, which had some motorbike use. The distance from trails at which bear use approximated availability varied by individual bear, by time of day, and by type of trail. Log-ratio differences were used to assess selection. Bears were less likely to spend time near trails with high (~5 trips/day average) motorized use than trails with low motorized use. We used an information-theoretic approach to select between nonlinear regression models with variables that included motorized use estimates, non-motorized use estimates, and trail density.

The Crowsnest Pass in southwestern Alberta, Canada has been highlighted as a critical area for wildlife movement. There are plans to upgrade Highway 3, which cuts through the Pass, to four lanes, with resulting increased traffic volume and speed. Currently, highway traffic volume is between 2,500 to 10,500 vehicles/day. Highway 3 may already be acting as a barrier to large carnivore and ungulate movements patterns, and wildlife mortality from animal/vehicle collisions on Highway 3 is approximately 109 large mammal deaths reported annually for a 46km stretch within the Pass. Detailed wildlife movement information in the Pass is limited.

To assist in understanding wildlife movement patterns along the highway to support decision-making for mitigation, a community based monitoring project was developed. The Alberta research institute Miistakis Institute of the Rockies created Road Watch in the Pass (RW), which allows local citizens to enter their wildlife observations along Highway 3 through an interactive web-based mapping tool. Over 1220 observations have been collected in over sixteen months, including 11 species of ungulates and carnivores.

This innovative approach to data collection would benefit from an analysis to determine whether the citizen reports are accurately representing visible wildlife activity along Highway 3. There are likely biases in citizen reports, based on unequal sampling effort involving location and frequency of travel. To identify and address these biases, this study compares spatial and temporal wildlife observation data from RW to a systematically gathered dataset using various statistical approaches.

We began systematic data collection in May 2006 and will continue through May 2007 to examine spatial and temporal characteristics of large mammal species movement (bighorn sheep, elk, moose, mule deer, white-tailed deer and carnivore species) along the highway. A 46-km stretch of Highway 3 was driven as a strip transect. When we observed an animal along or crossing the highway, UTM location, species, date, time, and other data were recorded. Each hour within the 24-hour period were sampled equally across a full year, allowing temporal analysis. Similar data from RW reports provided by citizens during the same period were extracted from the RW database. From May 2006 to March 2007, 395 transects were driven totaling over 395 hours of data collection and resulting in 681 wildlife observations. Spatial and temporal comparisons will be made between systematically gathered data and concurrent Road Watch data. Analysis will include examination of spatial association between the two data collection processes, comparison of spatial distribution, comparison of hourly and seasonal temporal distribution, effect of any biases on spatial or temporal distribution or species composition, and other analyses.

The Road Watch program is an important use of citizen involvement in transportation science. After analysis of RW’s accuracy in representing visible wildlife activity in the Crowsnest Pass, this study will provide suggestions and stipulations to improve the scientific rigor of this unique citizen-science program.