Effects of stochastic intraspecific seed dispersal variation on dispersal distance predictions in a temperate forest in Japan

The incorporation of seed dispersal models into conservation practices plays a pivotal role in predicting future ecosystems. Seed dispersal significantly influences plant population dynamics; however, our understanding of its implications at a landscape scale remains limited. We investigated how different seed dispersal processes impact the distributions of individual species within a 20 km × 20 km plot centered around a temperate old-growth forest reserve in Japan. We hypothesized that random variation in seed dispersal (i.e., stochastic processes) and variation mediated by distance (i.e., deterministic processes) would explain landscape-scale species distributions more effectively than deterministic seed dispersal processes alone. We evaluated 16 tree species with different seed dispersal modes by simulating seed dispersal processes. Stochastic simulations predicted higher dispersal distances for the majority of species. The findings of this study suggest that stochastic simulations are a more reliable tool for identifying dispersal restrictions in the context of climate change and biodiversity loss, highlighting the significance of using stochastic simulations for analyzing seed dispersal patterns in temperate forests, rather than more conventional deterministic methods.