Multi-tools and multi-species approach to study the impact of linear infrastructures on terrestrial mammals in France

Linear infrastructures can be detrimental to wildlife through barrier effects on animal movement and particularly, on dispersal. Populations may exchange few individuals leading to genetic groups that differentiate with generations and expose them to greater risks of genetic depletion. Landscape researchers and practitioners have been organizing themselves to better understand these impacts and act accordingly to maintain connectivity. In northeastern France, we compared the movement behavior of several terrestrial mammal species around linear infrastructures implementing three types of tools. First, we identified the impact of several highways and navigation canals by analyzing the genetic structure of populations on the Red deer (Cervus elaphus), the Roe deer (Capreolus capreolus), the Red fox (Vulpes vulpes), the European pine marten (Martes martes), and the Hedgehog (Erinaceus europaeus). Second, we aimed to identify the effective crossing points of these infrastructures analyzing GPS trajectories on red deers and pine martens. Third, we quantified the use of 70 potential crossing structures using camera trapping. Genetic analyses revealed that the red deers, the roe deers and the pine martens populations showed significant genetic clusters at the regional level. We also recorded between 4000 and 25000 locations of 9 deers. They used a restricted space and none of them crossed the highways. We monitored 23 pine martens and 98 trajectories were recorded in the vicinity of a canal. Five out of 12 individuals crossed it but at very specific points (locks or grassy borders) whereas all of them regularly crossed the nearby river at any place. Moreover, 9 pine martens were monitored next to a highway with 64 recorded trajectories. Three individuals crossed it using different types of structures: culverts, the borders of large underpasses, and agricultural bridges. Finally, 1500000 photos were analyzed and 5537 crossing events could be recorded, identifying a total of 16 distinct species. A gradient of requirement for the wildness potential of the structures was found between species. Main significant factors were the frequentation by cars and the presence of vegetation. Overpasses were also preferred to underpasses.
Our findings suggest that highways and navigation canals limit the exchange of individuals between populations, not impeding all animals to find some specific points to cross them: low frequented by humans and vegetalized places. Our actual concern is more about the cumulative costs of infrastructures and low quality habitats that surround them. Habitat fragmentation and loss seem to add major difficulties to individuals to handle infrastructures impacts. None individuals were settled in this landscape that seems to be low favorable to reproduction and dispersal. A study of the survival rates would allow to obtain an overview of the situation.