IENE 2016

Roads are widely considered as a major source of disturbance to wildlife. In addition to decreasing the quantity and quality of habitat, they often create a barrier to wildlife movements. However, the most observable effect on a daily basis is the wildlife mortality associated with road crossing attempts. To try to reduce road mortality, mitigation measures such as wildlife passages have been developed. These mitigation measures allow animals to cross the road without having to go across the road pavement. However, most of these mitigation measures were specially designed for large fauna because large mammals represent a potentially deadly danger for drivers. Even though road mortality could also endanger populations of smaller fauna, little work has been done to reduce collisions with them. During the widening of Quebec’s Highway 175 from two to four lanes, 33 wildlife passages designed specifically for small and medium mammals were added under the road. Small fauna fences were also built for 100 meters on each side of every passage to direct animals through the passage entrance. They are among the first wildlife passages built in the province of Quebec. Our study examines the effectiveness of these passages and exclusion fences to reduce road mortality of small and medium mammals while controlling for the potential confounding effects of landscape variables. To do so, we analyse the spatial distribution of road mortalities along Highway 175. Daily mortality surveys were conducted by car during summers 2012 to 2015. During these four years, 892 mortalities were found comprising 13 different species or taxonomic groups. The North American Porcupine was the mammal most often found. We examined the relationship between roadkill locations and distance to various landscape features such as forest cover, and water bodies, and also the presence of mitigation measures. For all species combined, mortalities were significantly higher at the fence ends than within the fenced sections and the unfenced sections. Furthermore, the presence of shrubby vegetation in the median strip separating the two directions of the highway also increased the number of roadkill for species over 1 kg. The general detection probability (p) for all species combined was 0.72. For species less than 1 kg and species over 1 kg, p was 0.17 and 0.82, respectively. Future road mortality studies should be combined with adequate data of the surrounding abundance of faunal populations and take into consideration detection probability of roadkills to get a more accurate estimate of the road effects at the population level of each species.