LAUSR

Abstract Livestock are a major human-induced threat to wildlife worldwide, especially in forest landscapes where livestock degrade the food and habitat of forest-dwelling wildlife. However, few empirical studies on this topic have been conducted at fine spatiotemporal scales that are crucial for wildlife-livestock interactions, in particular those involving multiple sympatric wildlife species under policy changes. Here, we demonstrate wildlife-livestock interactions through examining the interactions of several sympatric, threatened wildlife species with livestock in Wolong Nature Reserve, China, using data collected from infrared camera traps, DNA analysis of panda fecal samples and panda distribution predictive modeling along with habitat predictors. Camera trapping revealed an increase in livestock after the government implemented an incentive policy to encourage livestock production midway through the study. Three species (giant panda, red panda, and golden snub-nosed monkey) were displaced as more livestock encroached on forest habitat. In contrast, the detection rate of sambar deer was not affected by livestock encroachment, but sambar shifted the timing of visiting water sources (streams) to dusk (when livestock disturbance and other human activities were lower). The number of giant pandas detected via DNA testing of feces was relatively stable, but panda distribution modeling showed that pandas occurred across a wider area after disturbance. Our research shows that with increased livestock, different wildlife species may respond in different ways, which is likely associated with their biological traits (e.g., life history strategy and diet). Our study underscores the need for careful livestock policy making and planning.