COS 248-1 - Using networks to understand plant diversity in a variable world

Abstract: Understanding how ecological communities function and respond to change requires an understanding of how the different species within it relate to each other. When two species interact, this can determine whether one will go extinct or if they can both coexist. Such relationships can be represented by networks, where species which have a positive or negative effect on one another are linked. A key benefit of this approach is that we can investigate processes which arise when many species are involved, and visualise how a whole community is connected. Moreover, how species interact with one another will often vary across different environments. Subsequently, certain features of interaction networks are expected to vary across time and space too. Conversely, other aspects of species interactions and their networks are expected to remain consistent across all environmental conditions. It is thought that those key features which are invariant are crucial to maintaining diversity and stability, without which an ecological community would not be able to persist. When it comes to plant networks however, we are still figuring out which is which.
Indeed, while applications of network theory to food webs, plant-pollinator and parasite-host systems have been well studied, applications to plant systems are still in their infancy and plant networks remain rare. We therefore experimented with a newly-developed method to quantify interaction networks and applied it to a diverse annual plant community experiencing varying levels of natural shade (canopy cover) from Western Australia. We explored how species interactions, and the overall networks, varied from one micro-environment to the next. Interaction strengths, directions, and network properties all varied despite close spatial proximity and consistent patterns of local species richness, with most species interacting in different ways under different environmental conditions. In addition, we found that many of the interactions occurring between these plants subverted common expectations: species frequently experienced facilitation under all environmental conditions and networks were intransitive. Self-facilitation was also present and unrelated to abundance. These findings raise important questions about the mechanisms driving biodiversity in plant systems, and illustrate how a network perspective can broaden our understanding of coexistence.