Abstract

Theoretical studies have shown that coexistence between competitors can be favored in a spatially heterogeneous environment by a number of mechanisms, which ultimately allow the expression of persistent or transitory variation in species competitive abilities, colonization, or reproduction. Four distinctive paradigms to model metacommunities have been identified according to assumptions about the biology of the species and essential aspects of the environment. Missing from these are mechanisms of coexistence that can arise from the dispersal process itself without explicit spatial heterogeneity or biological trade-offs. These mechanisms have only recently received attention, but they may be common in marine communities and other systems in which dispersal is obligatory and modulated by the physical environment. We investigate coexistence in spatially homogeneous metacommunities where there is no partitioning of resources, no competition colonization trade-off, and no possibility of source sink dynamics. Coexistence is shown to be possible through three distinct mechanisms related to the dispersal process itself. Firstly, in a neutral scenario, inclusion of temporal variability in the connectivity matrix, emulating an intrinsic attribute of ocean character and other turbulent environments, can promote the invasion of an equally matched competitor and, in a hierarchical competition scenario, the persistence of an otherwise unviable, inferior competitor (the dispersal variability mechanism). Secondly, a sufficiently large difference in the shape of the time-independent dispersal kernels of the two species, which may result from differences in larval-release timing, buoyancy, or behavior, can produce stable coexistence in the center of their shared range (the dispersal-shape mechanism). Thirdly, asymmetry in the dispersal process due to biased advection renders the metapopulation model reactive,, such that small variations in the upstream abundances can be sufficient for the subordinate species to stably persist (the dispersal-bias mechanism). These results demonstrate that a subordinate species may persist by occupying a dispersal niche that differs sufficiently from that of the dominant species. Further theoretical research is necessary to develop simple empirical tests for these and other dispersal-based coexistence mechanisms.