Metabarcoding reveals the dietary diversity and food web structure of spider functional guilds in a highly diverse subtropical forest.

Abstract

Predator-prey interactions are key to regulating lower trophic levels and stabilizing ecosystem processes. Therefore, understanding prey selection and prey composition of predators is essential, yet especially the small size and diverse diets of predatory arthropods present a significant challenge to conventional field methods. Here, we employed DNA metabarcoding on gut contents of over 1500 arboreal spiders in a subtropical forest in China to construct a high-resolution spider-prey interaction network. Subsequently, we compared prey diversity, composition, starvation rates, and network metrics (predator niche overlap, generality, and prey vulnerability) across different spider hunting guilds and families. Our results revealed an exceptionally broad diet for these spiders, with spiders themselves constituting a significant proportion of prey (i.e. spider on spider predation), besides a wide range of insects, particularly flies and moths. Although, prey composition was broadly similar at high taxonomic levels and functional groups, distinct dietary partitioning was evident at lower (MOTU) taxonomic level. Interaction networks revealed a broader and more generalized diet spectrum of active-hunters compared to web-builders, which resulted in higher diet overlap within the hunting guild (high prey vulnerability and niche overlap). However, network metrics for active hunters were lower than null expectations, suggesting a structuring role of intraspecific competition, while web-builders displayed greater random associations, possibly reflective of the passive mode of prey capture. Collectively, our study offers a high-resolution overview of the dietary niches of subtropical arboreal spiders, revealing how hunting mode shapes their ecological impact and providing significant implications for leveraging these predators in biological control strategies.