Mammalian fatalities on roads: how sampling errors affect road prioritization and dominant species influence spatiotemporal patterns

Prioritizing roads for mitigation can be a challenge in conservation. The comparison of fatality estimates is an important tool to assist in this process, but the sampling errors in these estimates should be considered. Spatial and temporal patterns of wildlife fatalities vary among different roads, and multispecies evaluations can be biased by the abundance of a dominant fatality species. We compared three roads in southern Brazil regarding their mammalian fatality estimates, assessed temporal and spatial patterns of these records, and evaluated the effect of a dominant fatality species on these observed patterns. We found a discrepancy in road prioritization when using the estimated rather than the observed number of fatalities to the extent that the road with the lowest priority became the one with the highest. Therefore, we demonstrated the importance of considering sampling errors in fatality estimates. Our data evidenced the influence of the dominant fatality species on the temporal patterns on the three roads, changing the distribution of fatalities. We also demonstrated the importance of considering the influence of the dominant fatality species on the spatial patterns due to changes in the location of hotspots, representing hotspots with more non-dominant species. The decision of where to mitigate, given a stated mitigation target, could be improved by a simple cost–benefit approach considering all records or only non-dominant species. Independently of road or mitigation target, mitigating a small fraction of the road extension with effective action can avoid a considerable fraction of the fatalities.