The effects of dispersal on spatial synchrony in metapopulations differ by timescale.

Understanding the processes that stabilize species populations is a fundamental question in ecology and central to conservation biology. In metapopulations, dispersal can act as a 'double edged' sword for species stability by simultaneously decreasing local population variability (thereby decreasing local extinction risk) while increasing spatial synchrony (thereby increasing landscape‐level extinction risk). These dynamics may operate at different timescales, complicating efforts to assess their relative importance for long‐term stability. Here, we use a simple metapopulation model to understand how dispersal affects population variability and spatial synchrony across timescales. Our model shows that dispersal has contrasting effects at short versus long timescales on the variability and synchrony of populations. For populations that exhibit slow recovery when perturbed (i.e. under‐compensatory growth), dispersal decreases local population variability while increasing spatial synchrony at long timescales. In contrast, at short timescales dispersal increases local population variability while decreasing spatial synchrony. For populations that recover via damped oscillation when perturbed (i.e. over‐compensatory growth), the effects of dispersal are all opposite to those for populations with under‐compensatory growth, at both short and long timescales. The timescale‐dependent effect of dispersal has important implications for empirical studies. Specifically, studies conducted over short periods may only observe population variability increasing and spatial synchrony decreasing with dispersal, whereas the opposite patterns may predominate over longer periods. Our results provide novel insights on the dynamics underlying the role of dispersal and highlight the importance of time series length in empirical studies of metapopulations. [ABSTRACT FROM AUTHOR]