Coexistence between plant species is well known to depend on the outcomes of species
interactions within an environmental context. Few studies, however, incorporate
environmental variation into empirical studies of these processes due to the complex
experiments needed to assess coexistence in the field, and the lack of feasible modelling
approaches for determining how environmental factors alter specific coexistence
mechanisms.
In this paper, we present a simple modelling framework for assessing how variation in
species interactions across environmental gradients impact on niche overlap and fitness
differences, two core determinants of coexistence. We use a novel formulation of an
annual plant population dynamics model that allows for competitive and facilitative species
interactions, and for variation in the strength and direction of these interactions across
environmental gradients. Using this framework, we examine plant-plant interactions and
the relative variation in niche overlap and fitness differences between four commonly co-
occurring annual plant species from Western Australian woodlands; and how those vary
across environmental gradients previously identified as important for structuring diversity
patterns in this system: soil phosphorus, shade and water.
Results/Conclusions
We found facilitation to be a wide-spread phenomenon, and that interactions between
most species pairs shifts between competition and facilitation across multiple
environmental gradients. Environmental conditions also altered the strength, direction and
relative variation of both niche overlap and fitness differences in non-linear and
unpredictable ways. Our findings highlight the importance of the environment in determining the outcome of species interactions and the potential for pairwise coexistence between species. The prevalence of facilitation in our system indicates a need to improve current theoretical frameworks of coexistence to include non-competitive interactions, and ways of translating their effects into explicit predictions of coexistence. Our results also suggest a need for further research into which factors determine consistent responses of niche overlap and fitness differences to environmental
variation. Such information will improve our ability to predict outcomes of coexistence,
invasion events and responses of whole communities to future environmental change.