Aim: Asian forests are becoming increasingly degraded and fragmented by the extensive intensification of anthropogenic activities; these activities threaten wildlife and ecosystem sustainability. Facing a defaunation crisis, managers need more information on species assemblages to guide conservation efforts. We tested the relative influence of natural and anthropogenic factors on mammalian species richness and occupancy in temperate forests in Northeast Asia.

Location: Northeast China.

Methods: Camera-trapping data and multi-species occupancy models were used to estimate the species richness of a terrestrial mammal community in a working landscape and assess community-, group-, and species-specific responses to natural and anthropogenic features while accounting for imperfect detection. Species were grouped based on body size, diet, and activity pattern.

Results: We deployed 138 cameras and photographed 21 mammalian species over 22,976 trap days across the China-Russia border. Both natural and anthropogenic correlates varied in their importance in predicting the presence of different animals. Vegetation cover and cattle were found to have significantly positive and negative influences on community-level mammalian occupancy, respectively. The positive relationship with vegetation cover was most evident for large or diurnal species; the negative relationship with cattle was most evident for diurnal and wild ungulate species. Large species occupancy was also negatively associated with human settlements. The predicted richness across each station varied from 5 to 14 unique species, and species had a mean occupancy probability of 0.45 (95% credible interval = 0.09-0.86). Species richness was generally the lowest in livestock grazing areas and close to human settlements. Human influence is more important than the influences of vegetation and environmental variables.

Main conclusions: Our results highlight that livestock grazing was the primary human disturbance that had a negative impact on species occupancy and richness. Multi-species occupancy models helped to identify drivers of biodiversity declines and will inform conservation strategies in human-dominated landscapes in Northeast Asian forests. We suggest that landscape conservation planning seek to maximize forest protection and ecosystem services such as biodiversity and carbon storage.

detection history with effort 14 days.csv - Our observed data were represented by a summarized detection matrix yij that contains the number of 14-day windows where species i was detected at station j at least once.

site covariates.csv - The covariate for each camera site, in the same order as all of the other data (not standardized).

speciesCode.csv- A vector of the observed species used in our community analysis.