Effects of land use on zoonotic host communities: a global correlative analysis

Abstract

BackgroundEnvironmental trade-offs associated with land use—for example, between food security and biodiversity conservation—are crucial dimensions of planetary health. Land use-driven biodiversity change might predictably affect disease risk if reservoir host species are consistently more likely to persist under human disturbance (ie, if ecological communities in modified habitats generally have a higher zoonotic potential than those in unmodified habitats). Such a phenomenon has been observed in specific disease systems, but with substantial change in global land use projected for this century, assessing its global and taxonomic generality would shed light on an important hypothesised driver of environmental synergies or trade-offs between conservation and public health. MethodsWe collated data on hosts of human parasites and pathogens from the published literature, and combined these with the Projecting Responses of Ecological Diversity in Changing Terrestrial Systems (PREDICTS) global database of local ecological communities and associated land use data. We analysed the effects of land use on host richness and abundance across 7330 sites globally, controlling for disease-related research effort and differences in survey methods. FindingsEcological communities in anthropogenic land uses (managed and urban ecosystems) contained a consistently higher richness and abundance of host species than did communities in nearby primary (undisturbed) sites. However, among mammal hosts of zoonotic pathogens, we found considerable taxonomic variation in host responses to land use, with abundances of rodents and bats generally increasing and those of primates and carnivores generally declining in modified landscapes. InterpretationOur results suggest that future change in global land use has the potential to drive overall increasing contact between people and ecological communities with increased shared pathogen potential (ie, more potential hosts and possible opportunities for transmission). However, the differences in host responses to modified landscapes among mammal orders suggests that these changes are unlikely to affect all zoonotic diseases consistently. This result supports arguments that, rather than expecting a consistent relationship between biodiversity change and disease, in practice, policies aiming both to conserve biodiversity and reduce human disease (eg, through establishment of protected areas) must be disease and context specific. FundingThis research was supported by a Graduate Research Scholarship from University College London (to RG), the Natural Environment Research Council (to DWR and KEJ), and a Royal Society University Research Fellowship (to TN).