Beyond connectivity: how the structure of dispersal influences metacommunity dynamics

Abstract

Dispersal within metacommunities can play a major role in species persistence by promoting asynchrony between communities. Understanding this role is crucial both for explaining species coexistence and managing landscapes that are increasingly fragmented by human activities. Here, we demonstrate that spatial patterning of dispersal connections can drastically alter both the tendency toward asynchrony and the effect of asynchrony on metacommunity dynamics commonly used to infer the potential for persistence. We also demonstrate that changes in dispersal connections in strictly homogeneous predator-prey metacommunities can generate an extremely rich variety of dynamics even when previously investigated properties of connectivity such as the magnitude and distribution of dispersal among patches are held constant. Furthermore, the dynamics we observe depend strongly on initial conditions. Our results illustrate the effectiveness of measures of spatial structure for predicting asynchrony and its effects on community dynamics, providing a deeper understanding of the relationship between spatial structure and species persistence in metacommunities.

Metacommunity Structures of Macroinvertebrates and Diatoms in High Mountain Streams, Yunnan, China

“…In stream ecosystems, connectivity of dendritic networks and dispersal ability through that network play an important role in structuring metacommunities (Altermatt and Fronhofer, 2017;Hayes and Anderson, 2017;Tonkin et al, 2018). When the connectivity between communities is weakened, dispersal pathways are disrupted and the communities become more isolated (Cañedo-Argüelles et al, 2015).…”

Predicting Pattern Formation in Multilayer Networks

“…We compare the dynamics of our model with a commonly studied and more complex ecological model, the Rosenzweig-MacArthur predator-prey model (Rosenzweig and MacArthur 1963). This model has been used previously to study the effects of structure on dynamics (Holland and Hastings 2008;Hayes and Anderson 2018), though analytical prediction from the model has been challenging due to its complexity. Other work has been successful in extracting analytical predictions of the conditions for and qualities of synchronous and asynchronous states from this model (Goldwyn and Hastings 2008), but only when coupling is too weak to change the amplitude of limit cycles, only the phase of oscillations.…”

“…Specifically, we compare our model with the non-dimensional form of the Rosenzweig-MacArthur model with dispersal between subpopulations of predators and prey, which has previously been used to study the effects of structure on dynamics (Holland and Hastings 2008;Goldwyn and Hastings 2008;Hayes and Anderson 2018):…”

“…2. We use parameters for the ecological model used in previous work (Holland and Hastings 2008;Hayes and Anderson 2018) to represent systems with strong density dependence in prey and high predation, creating high amplitude oscillations which emphasize the role of spatial structure in these systems. Despite the different shape of limit cycle and intrinsic nonlinearity of the ecological model, the asynchronous states it produces and their dynamics resemble those found in our multilayer oscillator model.…”

“…Simulations were run until dynamics were stationary for 5000 timesteps. We used the parameters ω = .5, δ = .1, and ψ = .7 for the multilayer model and K = .3, e = 5, m = 1, and δ = .001 for the Rosenzweig-MacArthur model following (Holland and Hastings 2008;Hayes and Anderson 2018). The two three-oscillator interaction structures were used, a line and a triangle, both with L i j = 1 for all i, j.…”