Tree diversity promotes generalist herbivore community patterns in a young subtropical forest experiment

Jiayong Zhang, Helge Bruelheide, Xufei Chen, David Eichenberg, Wenzel Kröber, Xuwen Xu, Liting Xu, Andreas Schuldt

Stand diversification is considered a promising management approach to increasing the multifunctionality and ecological stability of forests. However, how tree diversity affects higher trophic levels and their role in regulating forest functioning is not well explored particularly for (sub)tropical regions. We analyzed the effects of tree species richness, community composition, and functional diversity on the abundance, species richness, and beta diversity of important functional groups of herbivores and predators in a large-scale forest biodiversity experiment in south-east China. Tree species richness promoted the abundance, but not the species richness, of the dominant, generalist herbivores (especially, adult leaf chewers), probably through diet mixing effects. In contrast, tree richness did not affect the abundance of more specialized herbivores (larval leaf chewers, sap suckers) or predators (web and hunting spiders), and only increased the species richness of larval chewers. Leaf chemical diversity was unrelated to the arthropod data, and leaf morphological diversity only positively affected oligophagous herbivore and hunting spider abundance. However, richness and abundance of all arthropods showed relationships with community-weighted leaf trait means (CWM). The effects of trait diversity and CWMs probably reflect specific nutritional or habitat requirements. This is supported by the strong effects of tree species composition and CWMs on herbivore and spider beta diversity. Although specialized herbivores are generally assumed to determine herbivore effects in species-rich forests, our study suggests that generalist herbivores can be crucial for trophic interactions. Our results indicate that promoting pest control through stand diversification might require a stronger focus on identifying the best-performing tree species mixtures.