Abstract:

Aims: Urban forests serve as key reservoirs of biodiversity within cities. Yet, the ecological strategies of urban mammals and their responses to human activity remain understudied. This study aimed to: (1) document mammal species composition in the urban forests of Chongqing central urban area; (2) analyze the functional traits structuring mammal communities; and (3) assess how native mammals respond to direct human presence and to free-roaming domestic cats (Felis catus) and dogs (Canis lupus familiaris).

Methods: From October 2019 to April 2024, we deployed 45 infrared camera traps across 29 sites in Chongqing central urban area, accumulating 5,814 effective camera days (mean ± SD: 200.48 ± 20.5 days per site). Functional traits— body size, dispersal, diet, life history, and habitat breadth—were used to construct a two-dimensional functional trait space. We applied kernel density estimation to analyze diel activity patterns and evaluated the temporal overlap between human activity (including domestic cats and dogs) and five wildlife species: wild boar (Sus scrofa), Reeves’ muntjac (Muntiacus reevesi), masked palm civet (Paguma larvata), Siberian weasel (Mustela sibirica), and red-hipped squirrel. We further used Wilcoxon rank-sum tests to assess differences in species richness between sites with and without direct human activity.

Results: We recorded 13 mammal species from 3 orders and 7 families. This includes the Class-I nationally protected animal small Indian civet (Viverricula indica) and the Class-II nationally protected animal leopard cat (Prionailurus bengalensis). Species with moderate body size, intermediate dispersal ability, and omnivorous diets clustered near the center of functional trait space, indicating convergence in adaptive strategies. In contrast, species at the periphery of trait space were more sensitive to human disturbance. For instance, wild boar displayed more diurnal activity in areas without direct human presence, while reeves’ muntjac exhibited compressed activity windows in areas frequented by cats, and temporal shifts in response to dog activity. Notably, domestic cats and dogs showed higher temporal overlap with native mammals than did direct human activity.

Conclusion: Mammalian responses to human activity in urban forests are complex. Indirect anthropogenic pressures— especially those mediated by domestic cats and dogs—may pose greater risks to native wildlife than low-intensity human recreation. These findings provide important insights for urban biodiversity conservation and management.

Key words: urban forest, camera trap, human activity, functional trait, diel activity pattern