Most landscapes are complex mosaics of many types of habitat. From the viewpoint of a particular species living in such a landscape, only some habitat types, called suitable habitat, provide the resources that are necessary for population growth. The remaining landscape, often called the (landscape) matrix, can only be traversed by migrating individuals. Often the suitable habitat occurs in discrete patches (also called habitat fragments). Individual habitat patches may be occupied by a local population of the focal species, but many patches are likely to be unoccupied at a particular point in time, because a local population went extinct in the past or the patch appeared in the landscape only recently due to, for example, successional changes. The currently unoccupied patches may become colonized in the future. The set of local populations inhabiting the network of habitat patches is called a metapopulation. The local populations are typically connected to other populations by some degree of migration, but how strong this coupling is depends on the structure of the landscape (how far apart the habitat patches are located) as well as on the powers of migration of the species.
Mathematical metapopulation models are used to describe, analyze, and predict the dynamics of metapopu-lations in fragmented landscapes. This article presents an overview of different kinds of metapopulation models, with an emphasis on spatially realistic models, which can be applied to real metapopulations for the purposes of research, management, and conservation.
Was this article helpful?