Twenty-three years after the Endangered Species Act (ESA) listing of the grizzly, intensive research conveys one compelling message: roads kill grizzlies. The mechanisms include: 1) direct mortality; 2) displacement; 3) habituation; and 4) fragmentation of habitat. The effects of access and roads on bears are similar to impacts on other sensitive species such as elk and wolves; but grizzlies are the most vulnerable of all wildlife in the northern Rockies.

Direct Mortality

The risk of grizzly mortality is significantly increased by roads (collisions), and increased encounters between bears and people (potentially lethal for the bear). This relationship is more pronounced on public lands outside Glacier and Yellowstone National Parks, partly due to many hunters travelling the back roads.

On the Rocky Mountain Front in Montana, Keith Aune and Wayne Kasworm (1989) reported that 63% of known human-caused grizzly deaths occurred within 1 kilometer of roads. Another study (Dood et al. 1986) found that from 1967 and 1986, 48% of all known non-hunting mortalities had occurred within 1 mile of roads. The South Fork of the Flathead study (Manley and Mace, 1993) found that in areas of high human use and roaded access, mortality risk for grizzly bears was significant (deaths equaled births). Similarly, on Admiralty and Chichagof Islands, the Alaska Department of Fish and Game (Titus, K., and L.R. Beier, 1992) found "significant positive associations between the autumn brown bear kill and the sum of roads built per year."

Displacement

Roads and related human activities displace grizzlies from preferred habitat. Under-use of areas near roads has been documented in the Greater Yellowstone Ecosystem (GYE) 4-5 kilometers from town sites, and within 2 kilometers of roads (Mattson et al. 1992). Mace and Manley found in the Northern Continental Divide Ecosystem (NCDE) that when total road densities reached 2 miles per square mile, or open road densities exceeded 1 mile per square mile, use by grizzly bears declined significantly. 22% of their study area had road densities greater than 2 miles per square mile, and this habitat was used by radio-collared bears far less than expected.

Bears learn to avoid roads as a result of associated noise, human scent, and hunting or shooting. In the GYE and NCDE, bears consistently avoid areas with high road densities, even after roads are closed. Female bear cubs generally establish home ranges that overlap their mother's. Long-term displacement of a female from part of her home range due to high road densities and human access could cause that area to be effectively lost to female bears, with their offspring having no opportunity to learn feeding opportunities there (Aune and Kasworm 1989, McLellan 1989).

Habituation

Addiction to garbage and the loss of fear by bears to people leads to habituation. In Yellowstone, habituated grizzlies often are found near roads and human developments, while wary grizzlies search elsewhere for high quality foods. Mortality risk among habituated bears is double that of non-habituated or wary bears. Habituation increases encounters that can lead to the bear's death or removal (Mattson and Pease, in press). During years when key food sources fail, sub-adult males and adult females with cubs are much more likely to be found near roads. This could be due in part to those bears being displaced into roaded marginal habitats by male bears dominating quality backcountry habitat.

Habitat Fragmentation

As roads are punched into bear habitat, their populations and habitat fragmenta major mechanism by which grizzly bear populations have been severely reduced. Fragmentation causes populations to become more isolated and vulnerable to extinctionespecially when human-caused mortality continues. This is particularly troublesome to large carnivores such as grizzlies, which require large home ranges (up to 900 square miles for Yellowstone males) and wide vegetative and topographic habitat diversity (Servheen 1986).

Methods for Evaluating Road Density

Rick Mace and Tim Manley's South Fork study applies one of two recognized methodologies that arrive at road density limits for grizzly bears. They predicted that when total road densities reach 2 miles per square mile, or open road densities exceed 1 mile per square mile, use by grizzlies would decline significantly. Their study was based on documentation of displacement, responses of individuals and family groups to roads, and mortality. Researchers concluded from earlier reports (Mace and Manley, 1993) that 68% core areas (roadless habitat) was important to protect grizzlies; however, recent reports conclude that only 60% core (roadless) habitat may be needed (Mace et al. 1996). This method is used widely in grizzly bear habitat, especially the NCDE.

Dave Mattson and Mark Haroldson (1985) arrived at somewhat similar road density recommendations using a more complicated method. They used a grizzly bear foraging radius during a 24-48 hour period, surrounded by a buffer zone. This represents a micro-security area of approximately 7,000 acres that grizzly bears can utilize unimpeded by humans. Expanding this micro security concept, the authors recommended that 7,000 acre areas should be arranged across the landscape, approximately 1.8 kilometers apart. If an entire home range of 884 square kilometers for a grizzly in the GYE was arranged in micro security areas, then approximately 57% of the landscape would be secure. Ideally, these areas would have little or no road access during the grizzly's time of use.

Mattson (1993 ) added to this approach with a new methodology that accounted for the variable influence that roads in cover and closed roads have on grizzly bear behavior and mortality risk. The weighted coefficients are as follows: open roads in cover (.7); open roads in non-cover (1.9); trails or closed roads in cover (.3); trails or closed roads in non-cover (.6). Within cover means the road must travel through a forested stand of trees that is dense enough to make a grizzly bear undetectable from a roadside view.

An example of an application of the method follows. If 1 mile of road exists per square mile and the road is located in a forested stand with roadside cover, the equivalent weighted road density would be .7 miles of road per square mile. Conversely, if 1 mile of open road in 1 square mile has no cover, the weighted road density would be 1.9 miles per square mile.

Using this approach and a system of protecting security habitat, the U.S. Fish and Wildlife Service (USFWS) arrived at the need for road density limitations as low as .26 miles per square mile on parts of the Targhee Plateau Bear Management Unit. When loss of cover, flat topography and absence of secure habitat from past clearcutting and roadbuilding were factored in, the road densities dropped far below Mace/Manley's more generally applicable 1 mile per square mile benchmark. This analysis requires more landscape-specific information than does the Mace/Manley approach.

For either of these methods to be used, motorized vehicle use must be limited to trails or roads. It is impossible to calculate road densities if vehicle use occurs off designated routes; in such cases the road densities essentially would be infinite.

The Biological Opinion written by the USFWS on the Targhee NF's Plateau area, and its companion piece by the Forest Service (Targhee National Forest, 1993) are some of the best documents on the impacts of roads on bears, and the best analyses for a landscape-level restoration plan for bear habitat.

The Interagency Grizzly Bear Committee (IGBC) issued a document (1994) "The Interagency Grizzly Bear Committee Task Force Report on Grizzly Bears/Motorized Access Management". This gave preliminary definitions of roads and trails, as well as procedures for evaluating road densities and identifying existing and potential core security habitat for bears. While it didn't go far enough in addressing grizzly sensitivity to low levels of access or the ineffectiveness of many Forest Service road closures, it did begin to set consistent terms for discussing management of roads and human access.

This access document was first of a two-part series; the second document would have set road limitations and standards for Northern Rockies grizzly bear ecosystems. Unfortunately, the agencies have delayed the second step.

This process of setting roads standards has since been folded into a larger analysis called the "Conservation Strategy" for Grizzly Bears. The Conservation Strategy is slated for release and public comment later this year, along with the revision of the 1993 Grizzly Bear Recovery Plan. This revision was prompted by successul litigation of the Plan brought by thirty eight conservation groups and individuals. A central issue for both documents will be roads and access management.

Conclusion

Grizzly bears are more sensitive to roads and human access than other species. Scientists and managers use them as an indicator of ecosystem health. If the Great Bear is doing well, so are other wide-ranging species sensitive to human disturbances. The future of the bear relies on redoubling current efforts, and expanding habitat protections to include recent information on roads impacts and the importance of protecting secure (roadless) habitat. The grizzly's survival today is testament to the strength of the ESA, the integrity of its watchdogs both within and outside of state and federal agencies, and specific actions based in sound science that reduce human-caused mortality and protect habitat.

Louisa Willcox is the coordinator of the Sierra Club Grizzly Bear Ecosystem Project, based in Bozeman Montana, and on the Board of Advisors of Wildlands CPR. You can contact her at 234 E. Mendenhall, Bozeman, MT. 406-582-8365.

Bibliography

Aune, K., and W. Kasworm. 1989. Final Report East Front Grizzly Bear Study. Montana Department of Fish, Wildlife and Parks. Helena, MT.

Dood, A., R.D. Brannon, R.D. Mace. 1986. Final Programmatic Environmental Impact Statement, the Grizzly Bear in Northwest Montana. Montana Department of Fish, Wildlife and Parks, Helena, MT.

Haroldson, M. and David J. Mattson. 1985. Response of Grizzly Bears to Backcountry Human Use in Yellowstone National Park. U.S. National Park Service, Interagency Grizzly Bear Study Team, Bozeman, MT.

Kasworm W., and T. Manley. 1988. Grizzly Bear and Black Bear Ecology in the Cabinet Mountains of Northwest Montana. Montana Department of Fish, Wildlife and Parks, Helena, MT.

Kasworm, W., and T. Manley. 1989. Road and trail influences on grizzly bears and black bears in northwest Montana. International Conference on Bear Research and Management. 8:79-84

Lyon, L.J. 1983. Road density models describing habitat effectiveness for elk. Journal of Forestry. 81:592-595.

Mace, R.K., and T. Manley. 1993. The Effects of Roads on Grizzly Bears: Scientific Supplement. South Fork Flathead River Grizzly Bear Project: Project Report for 1992, Montana Department of Fish, Wildlife and Parks, Helena, MT.

Mace, R.K. and T. Manley. 1993. South Fork Flathead River Grizzly Bear Project: Progress Report for 1992. Montana Department of Fish, Wildlife and Parks, Helena, MT.

Mace, R.K. and J. Waller. 1997. Final Report: Grizzly Bear Ecology in the Swan Mountains, Montana. Montana Department of Fish, Wildlife and Parks, Helena, MT.

Mace, R.K., J. Waller, T. Manley, L.J. Lyon, and H. Zuring. 1996. Relationships among grizzly bears, roads, and habitat in the Swan Mountains, Montana. Journal of Applied Ecology.

Mattson, D.J., R. Knight, and D.M. Blanchard. 1987. The Effects of development and primary roads on grizzly bears in Yellowstone National Park, Wyoming. International Conference on Bear Research and Management. 7:259-273.

Mattson, D.J., and R. Knight. 1991. Effects of Access on Human-Caused Mortality of Yellowstone Grizzly Bears. USDI National Park Service, Interagency Grizzly Bear Study Team Report, Bozeman, MT.

Mattson, David J. 1992. Conversion Factors for Standardized Calculations of Roads and Trail Densities: Yellowstone Grizzly Bears. Interagency Grizzly Bear Study Team, Forest Sciences Labs, Montana State University, Bozeman, MT.

Mattson, David J. 1992. Microscale Security Areas for Yellowstone Grizzly Bears. Interagency Grizzly Bear Study Team, Forest Sciences Labs, Montana State University, Bozeman, MT.

Mattson, David J. 1993. Grizzly Bear Responses to Human Activities: A Review and Summary. Interagency Grizzly Bear Study Team, Forest Sciences Labs, Montana State University, Bozeman, MT.

Mattson, David J. 1993. Background and Proposed Standards for Managing Grizzly Bear Habitat Security in the Yellowstone Ecosystem. Cooperative Parks Studies Unit, College of Forestry, Wildlife and Range Sciences, University of Idaho, Moscow, Idaho.

Mattson, David J. 1993. Use of Road Density Standards for Management of Yellowstone Grizzly Bear Habitat. Interagency Grizzly Bear Study Team, Forest Sciences Labs, Montana State University, Bozeman, Montana.

McLellan, B.N., and D.M. Shackleton. 1988. Grizzly bears and resource extraction industries: effects of roads on behavior, habitat use, and demography. Journal of Applied Ecology. 25: 451-460.

Meagher, M. and S. Fowler. 1989. The consequences of protecting problem grizzly bears. In M. Bromley, ed. Bear-People Conflicts: Proceedings of a Symposium on Management Strategies. Pp. 141-144. Northwest Territories Department of Renewable Resources. Yellowknife, Northwest Territories, Canada.

Titus, K., and L.R. Beier. 1992. Population and Habitat Ecology of Brown Bears on Admiralty and Chichagof Islands. Alaska Department of Fish and Game. Division of Wildlife Conservation, Federal Aid in Wildlife Restoration. Research Progress Report.

U.S. Department of Agriculture, Forest Service. 1993. Biological Assessment of the Westline Timber Sale and Plateau Bear Management Unit. Targhee National Forest.

Zager, P., and C. Jonkel. 1983. Managing grizzly bear habitat in the northern Rocky Mountains. Journal of Forestry. 81:524-526.