Self‐organised pattern formation increases local diversity in metacommunities

Abstract

Self-organised formation of spatial patterns is known from a variety of different ecosystems, yet little is known about how these patterns affect the diversity of communities. Here, we use a food chain model in which autotroph diversity is described by a continuous distribution of a trait that affects both growth and defence against heterotrophs. On isolated patches, diversity is always lost over time due to stabilising selection, and the local communities settle on one of two alternative stable community states that are characterised by a dominance of either defended or undefended species. In a metacommunity context, dispersal can destabilise these states and complex spatio-temporal patterns in the species' abundances emerge.The resulting biomass-trait feedback increases local diversity by an order of magnitude compared to scenarios without self-organised pattern formation, thereby maintaining the ability of communities to adapt to potential future changes in biotic or abiotic environmental conditions.

Habitat isolation diminishes potential of self-organised pattern formation to promote local diversity in metacommunities

“…While the random geometric graphs that form the basis of our metacommunity networks are not scale-free, we also observe an increase of the pattern amplitude when changing the diffusion coefficient d X in such a way that dynamics are led further into the Turing instability region (Box 1, Figure A1). Similar observations were made in previous studies (Hata et al, 2014; Guill et al, 2021). With increasing size and connectance, the patch degrees increase (Figure 1).…”

“…Between-patch functional diversity is measured as the difference between the mean traits of the local communities, with larger differences between patch means associated with higher β -diversity. In line with Guill et al (2021), the biomass-weighted average local and between-patch ( β ) diversity of a metacommunity are given by where and v i are temporal averages and is the grand mean trait across all patches of a metacommunity. Regional diversity ( γ ) can be calculated as the sum of and β diversity (see supplementary material of Guill et al (2021) for derivation) and is therefore not separately evaluated here.…”

“…Therefore, land use change that alters the spatial structure of a metacommunity might affect self-organised pattern formation, and in consequence threaten the diversity of local communities. In this study we thus expand the trophic metacommunity model of Guill et al (2021) to assess the impact of two structural properties directly affected by anthropogenic changes, the number of habitat patches and dispersal links, i.e., the size and connectance, of a metacommunity, on the functional diversity within and between autotroph communities (Fig. 1).…”

“…Grey areas: pattern formation of undefended autotroph communities, red areas: pattern formation of defended autotroph communities. The arrow denotes the gradient of d N used in this analysis. (Brechtel et al, 2018; Guill et al, 2021). Here J is the Jacobian matrix of the dispersing local food web components ( N, A, H ) evaluated at the homogeneous equilibrium, D is the Jacobian matrix of the emigration dynamics (which in the case of diffusive, i.e., random, dispersal is simply a diagonal matrix of the three diffusion coefficients d X ) and κ i are the eigenvalues of L , their number corresponding to the number of habitat patches ( P ).…”

Modelling approaches for capturing plankton diversity (MODIV), their societal applications and data needs

“…Terseleer et al, 2014;Acevedo-Trejos et al, 2015) and fundamental (e.g. Coutinho et al, 2016;Guill et al, 2021) research questions related to plankton diversity, but this comes at the cost of limited ability to accurately resolve the fitness landscape and inability to capture certain observed diversity distributions.…”