Print PageMetapopulation models are a useful way to describe the population dynamics of species that occupy patchily distributed habitats but they have rarely been applied to migratory species because of the challenge of separating a migratory population into demographically independent local populations. Here, I present an extension to metapopulation theory for seasonally migratory populations. I consider a set of discrete patches that are two different types, breeding and non-breeding. Colonization of one type of habitat only occurs from the opposite type. I present an extension of the classic (Levins) metapopulation model as well as an extension to spatially realistic metapopulation model.
Results/Conclusions
I show that a migratory metapopulation will persist if the product of colonization rates of the two types of patch exceeds the product of extinction rates in the two types of patch. The practical implication of this result is that one season can act as a buffer for a population that would otherwise be unable to persist if it were not migratory and restricted to one set of patches. I also show that in spatially realistic migratory metapopualtions, analogous to non-migratory metapopulations, there exists a landscape metric, the migratory-metapopulation capacity (λMM), that determines persistence of the species. I show that a migratory metapopulation will persist when λMM exceeds a value that is determined by the properties of the species. This work increases the range of applications of the metapopulation concept to include animals that migrate seasonally and occupy two different habitats during the annual cycle and introduces an important tool for management and conservation of migratory species. This extension of metapopulation theory also has a potential application to model the dynamics of parasites that have two obligate hosts.
