Post-dispersal seed predation reflects tree species occurrence beyond local neighborhoods in a forest biodiversity experiment

FEILONG JI, Leif Magnus Pantléon, Alexandra Erfmeier, Tim Diekötter

Post-dispersal seed predation is a key mechanism underlying Janzen-Connell (JC) effects with seed predation expected to decline with increasing distance from conspecific trees. However, the spatial scale at which seed predators respond to the occurrence of their resource tree species remains poorly understood. In particular, mobile consumers may respond to resource distributions beyond the local neighborhood, potentially weakening distance-dependent patterns at the spatial scales commonly used to quantify biodiversity and neighborhood effects. Here, we tested whether tree species richness, vertebrate access, and conspecific tree occurrence at the plot and site level shaped post-dispersal seed predation in the BEF-China tree diversity experiment. We deployed paired open and vertebrate-exclusion seed cards tracking 41,580 seeds from seven tree species across 297 plots spanning monocultures to 16-species mixtures. Across the seven focal tree species, seed predation declined with increasing tree species richness, was consistently higher on open than on vertebrate-exclusion cards and differed strongly among seed identities. Contrary to the expectation of a local JC-type effect, seed predation did not differ between seeds of focal tree species present and absent from the focal plot, provided that the species belonged to the species pool of the respective site. In contrast, seeds of species absent from the site-level species pool experienced substantially lower predation. This pattern persisted across the tree species richness gradient. Our results provide little evidence that local conspecific occurrence drives post-dispersal seed predation at the scale of individual experimental plots. Thus, spatial patterns in seed predation consistent with processes underlying JC effects may emerge at scales larger than the local neighborhood, particularly when consumers are mobile. More generally, our findings highlight the importance of considering consumer mobility and spatial scale when interpreting seed predation effects on forest regeneration in biodiversity experiments.