Species coexistence in resource-limited patterned ecosystems is facilitated by the interplay of spatial self-organisation and intraspecific competition

Abstract

The exploration of mechanisms that enable species coexistence under competition for a sole limiting resource is widespread across ecology. One classical example is the coexistence of herbaceous and woody species in self-organised dryland vegetation patterns. Previous theoretical investigations have explained this phenomenon by making strong assumptions on the differences between grasses and trees (e.g. contrasting dispersal behaviours or different functional responses to soil moisture). Motivated by classical theory on Lotka-Volterra competition models, I argue that the interplay between interspecific and intraspecific competition can explain species coexistence without relying on such assumptions. I use an ecohydrological reaction-advection-diffusion system that captures the interactions of two plant species with an explicitly modelled resource to show that coexistence is facilitated by strong intraspecific competition of the species superior in its colonisation abilities, if its competitor's local average fitness is higher. The inclusion of a spatial self-organisation principle yields significant differences from the nonspatial case in which strong intraspecific competition of the locally superior species enables coexistence. Results presented in this paper also capture the empirically observed spatial species distribution in single bands of vegetation and propose differences in plants' dispersal behaviour as its cause.

Self‐organised pattern formation increases local diversity in metacommunities

“…In studies of savanna ecosystems, the coexistence of grass and tree species has also been linked to self‐organised pattern formation (Baudena & Rietkerk, 2013; Gilad et al, 2007; Nathan et al, 2013). However, these studies usually assume that only one of the plant species is involved in the pattern‐forming process (notable exceptions are Eigentler, 2021; Eigentler & Sherratt, 2020)), which then facilitates the persistence of the other species. The results of the present study suggest that self‐organised pattern formation, especially when it interacts with pre‐existing ‐diversity (or spatial heterogeneity in general), might be an important mechanism complementing the well‐established theories on plant diversity.…”

“…They demonstrated that coexistence of two plant species (grasses and trees) is possible due to a colonisation-competition trade-off. In Eigentler (2021), these results were extended to cover a wider range of environmental conditions by including intraspecific competition. In general, conditions under which pattern formation might facilitate -diversity include less strongly stabilising selection regimes, more flexible shapes of the local trait distributions (e.g.…”

Interpreting how nonlinear diffusion affects the fate of bistable populations using a discrete modelling framework

“…There are also many ways that the current modelling framework can be extended. For example, the discrete model can be extended to consider multiple interacting subpopulations, and the same discrete-to-continuum averaging approach could be used to construct a continuum limit model, which would take the form of a system of coupled partial differential equation models [43,64–68]. We leave this extension for future consideration.…”

Processes and predictions in plant ecological models: logic and causality

“…There is increasing awareness of the limitations of classical community ecological models, e.g. Lotka-Volterra-type competition models, to predict ecosystem effects of a changing environment (Clark et al 2020;Damgaard and Weiner 2021;Eigentler 2021;Martyn et al 2020;Mayfield and Stouffer 2017;Neill et al 2009;Vellend 2010). For example, it is widely appreciated that frequencydependent species interactions, in which relatively rare species are favored over more common species, may play an important role in plant species co-existence and community dynamics in plant communities (Chisholm and Fung 2020;Connell et al 1984).…”