Long-term responses of an ecological community to highway mitigation measures

Abstract

In road mitigation systems characterized by multiple wildlife crossing structures (CS) and multiple-focal species, these species-specific design criteria are important to meeting management goals. CS types and locations are fixed in place and cannot be manipulated experimentally; long term studies may offer the best chance to inform evidence-based designs for new CS projects in the future. Long-term data from Banff National Park are uniquely posed to answer these critical questions. More recently, highway mitigation along US93 in Montana provides an additional case study with which to understand the responses of large animals to different CS designs. The purpose of this study is to identify factors affecting movement of large mammals through CS using data sets from both mitigation projects. Year-round monitoring of CS use was used in an analytical framework to address questions regarding species-specific and community level use of CS; design and habitat factors that best explain species-specific variation; and whether importance of design parameters changes over time. Over the 17 years of the Banff study, and the six years of the Montana study, CS facilitated over 200,000 crossing events at 55 locations. There were significant changes in annual crossing events over time. Variables associated with CS passage rates were species specific, but aligned with a few clusters of preference. With the exception of coyotes, all large carnivore species preferred open span bridges or overpasses to other CS types. In Montana, fencing was positively associated with passage rates for black bears and cougars. We found that wider CS tend to be preferred by most species, irrespective of their location. We also found that wider CS tend to have shorter ‘adaptation’ curves than narrower ones for grizzly bears, coyotes, cougars, and moose. Depending on the heterogeneity of the landscape near the highway, more CS may not create more crossing opportunities if local habitat conditions do not favor animals’ access to the road. At the scale of ecological communities, the flows of mass and energy are likely enough to alter the distribution of ecological processes in the Banff and Montana ecosystems. Our results highlight the value of long-term monitoring for assessing the effectiveness of mitigation measures. Our work confirms the species-specific nature of measure CS performance, leading to our primary recommendation that a diversity of CS designs be considered an essential part of a well-designed mitigation system for the large mammals of western North America. Short-term monitoring efforts may fail to accurately portray the ecological benefits of mitigation for populations and ecological communities. Our results will help to inform design and aid in the establishment of robust, long-term performance measures.