Can Roads Be Used as Transects for Primate Population Surveys?

Line transect distance sampling (LTDS) can be applied to either trails or roads. However, it is likely that sampling along roads might result in biased density estimates. In this paper, we compared the results obtained with LTDS applied on trails and roads for two primate species (Callithrix penicillata and Callicebus nigrifrons) to clarify whether roads are appropriate transects to estimate densities. We performed standard LTDS surveys in two nature reserves in south-eastern Brazil. Effective strip width and population density were different between trails and roads for C. penicillata, but not for C. nigrifrons. The results suggest that roads are not appropriate for use as transects in primate surveys, at least for some species. Further work is required to fully understand this issue, but in the meantime we recommend that researchers avoid using roads as transects or treat roads and trails as covariates when sampling on roads is unavoidable.