Spatially-separated populations of the same species often flucutate synchronously in nature. Two processes can generate spatial synchrony: spatially-synchronized exogenous environmental fluctuations (the Moran effect), and dispersal between populations. The efficacy of both processes is thought to be mediated by species interactions. In particular, cyclic population dynamics generated by predator-prey interactions are thought to be easily synchronized (‘phase-locked’) by dispersal. In a protist microcosm experiment, we provided the first experimental demonstration of phase-locking of predator-prey cycles by dispersal. Here we build on previous work by asking whether dispersal can generate phase-locking over large spatial domains (linear arrays of patches connected by ‘stepping-stone’ dispersal). We also examined the synchronizing effects of realistic patterns of spatiotemporal environmental fluctuations. In treatments with the Moran effect, nearby patches experiencing highly-correlated environmental fluctuations (=strong Moran effect), while distant patches experiencing uncorrelated environmental fluctuations (=no Moran effect). In control treatments, environmental fluctuations were uncorrelated at all spatial scales.

Results/Conclusions

As in previous work, the protist predator Euplotes patella and its prey Tetrahymena pyriformis exhibited cyclic dynamics within patches (microcosms). Both dispersal and the Moran effect increased spatial synchrony of these cycles. Unexpectedly, synchrony gradually declined with increasing inter-patch distance, even when all patches experienced uncorrelated environmental fluctuations. This indicates that dispersal did not produce perfect phase-locking across all patches. Nor did dispersal divide the spatial arrays into clusters of patches phase-locked with one another but out of sync with other clusters. In studies of natural populations, gradual decay of synchrony at inter-patch distances greater than the typical dispersal distance is routinely attributed to the Moran effect. Our results indicate that short-distance dispersal may provide an alternative explanation for gradual decay of synchrony with increasing inter-patch distance. The simplest theoretical models do not reproduce the results reported here. We discuss how these models might be modified to account for the observed results.