Monitoring the use of a canopy bridge and underpasses by arboreal mammals on a Brazilian coastal road

Roads disrupt the canopy and can affect arboreal animals in different ways, such as reducing canopy connectivity, generating habitat loss and degradation, and increasing direct mortality. Since arboreal animals mainly use the canopy for movement, mitigation measures for these species usually focus on maintaining or restoring canopy connectivity to guarantee safe crossings. Here we present a case study of a Brazilian coastal road (ES-060) for which we described the use of a canopy bridge and multiple underpasses by three arboreal mammal species and compared these data with roadkill records of the same species in the vicinity of the crossing structures. Our study includes a 75 m long steel cable canopy bridge, monitored for 3 years, and clusters of different types of underpasses, monitored for 16 years. The use of the crossing structures was monitored with sand track beds installed at entrances on both sides, and roadkill surveys were conducted daily for 16 years. We considered a crossing to be successful if tracks of the same species were recorded on either side of a structure and showed opposite movement trajectories. The canopy bridge survey resulted in an observed rate of 0.16 crossings/month for Callithrix geoffroyi, 7.79 for Coendou insidiosus, and 0.46 for Didelphis aurita, and all types of underpasses combined demonstrated a rate of 0.33, 1.94, and 8.43 crossings/month for each species, respectively. The roadkill surveys resulted in an observed rate of 1.41, 0.78, 2.94 roadkills/month for Callithrix geoffroyi, Coendou insidiosus, and Didelphis aurita, respectively. Even with mitigation structures confirmed to be used by these three species, roadkill hotspots occurred in the road sections with the crossing structures. Our study demonstrated the use of a canopy bridge and different types of underpasses by arboreal mammal species. The canopy bridge was mostly used by Coendou insidiosus, while the underpasses were mainly used by Didelphis aurita. As roadkill hotspots occurred red in the same segments where mitigation crossing structures were installed, our results indicate that some important improvements are needed to mitigate roadkills of arboreal mammals in this area, mainly preventing that these species access the road. We present recommendations for a research agenda to support mitigation planning for arboreal mammals, namely: (1) testing the efficiency of different canopy bridge designs for multispecies mitigation, (2) testing the use of connecting structures, such as ropes that connect to the surrounding forest, to encourage underpass use by arboreal species, and (3) testing fence adaptations to block the access of arboreal mammals to roads. Keywords