Effects of space partitioning in a plant species diversity model

Understanding the mechanisms of species diversity maintenance within plant communities has become a fundamental issue in ecology over the past decades. While some models have tried to explore these mechanisms, few studies have integrated the dynamic interactions with neighbours in a spatially explicit way. The present model uses Voronoi polygons to dynamically partition a landscape patch into areas occupied by individual plants. It thus incorporates neighbourhood competition for space, unlike grid-based models with nearest-neighbour competition. In closed two-species communities, dynamic Voronoi partitioning promoted species coexistence, especially under local dispersal. This suggests that grid-based models overestimate species extinction rates. Likewise, multispecies communities without immigration had substantially greater species richness in the space partitioning model than in the grid-based model but only under distance-limited dispersal. In contrast, richness levels were similar in both models under global dispersal or with immigration from the metacommunity. Trait variation among species reduced species richness, but more so for traits associated with competition for space. This suggests that some traits are more important than others in governing species richness. Overall, our study demonstrates that combining species identity (traits) with partitioning of physical space can improve understanding of diversity regulation.