Plant community-mediated effects of grazing on plant diseases

Grazing is one of the most important management practices for grasslands. To date, most studies on how grazing affects plant diseases have focused on a single plant species, ignoring plant community characteristics and phylogeny. We used data from a 6-year yak grazing experiment (0, 1, 2, and 3 yak(s) ha^- 1 treatment) in an alpine meadow ecosystem of Qinghai-Tibetan Plateau, from which we tested grazing effects on foliar fungal diseases at both population and community levels. By measuring plant community variables (including richness, evenness, phylogenetic diversity, and composition) and disease severity, we evaluated the relative importance of plant community-mediated effects of yak grazing on community pathogen load with a multi-model inference approach. We found significant differences in pathogen load among different grazing treatments; we recorded the highest and lowest pathogen loads in the 1 yak ha^- 1 treatment and in the 3 yaks ha^- 1 treatment, respectively. Pielou's evenness index and community proneness (i.e., an estimate of the capacity of plant communities to support diseases) best explained variation in pathogen load, indicating that plant community-mediated effects (through evenness and proneness) of yak grazing determined pathogen load. Our study provides empirical evidence that grazing influences foliar fungal disease prevalence through plant community evenness and composition, which demonstrates the necessity of incorporating host plant community characteristics into disease load prediction frameworks.