Department of Adaptive Systems

A metapopulation consists of a group of spatially separated populations of the same species partially connected by dispersal. The term metapopulation was coined by Richard Levins in 1969 to describe a model of population dynamics of insect pests in agricultural fields, but the idea has been most broadly applied in conservation ecology, to species in naturally or artificially fragmented habitats. The major drawbacks of the Levins model are its deterministic nature and the lack of consideration of the landscape characteristics. Indeed, it is assumed the existence of infinitely many identical habitat patches (theoretical metapopulation). But for management decisions it is necessary to consider more realistic  models that allow for specific and testable prediction about the importance of habitat destruction and fragmentation. Spatially explicit metapopulation models can consider combined environmental effects and hence, are useful for representing the spatio-temporal population dynamics. In this respect, the Incidence Function Model (IFM) introduced by Hanski in 1994 can be considered as the reference model, for simplicity and spread of application. Despite its wide use, however, the literature provides limited support to ecologists interested in learning how the information obtained from the estimated IFM model could be applied to ecosystem management. As conservation is concerned, a few indexes that could be considered for the purposes of strategy comparison have been introduced. However, their interpretation seems sometimes to conflict with the assumptions of the model and often they can not be used, in practice, because of computational difficulties. To our knowledge, the same indexes did not find application to pest management. In this talk we will discuss the use of the Kullback-Leibler divergence to deal with ranking of (a) conservation options for endangered species and (b) options of pest management. The application and properties of the proposed measure will be illustrated on the basis of two case study. The first is concerned with amphibian metapopulation conservation strategies in the alpine Rhine valley and the second with the evaluation of pest control strategies for the Pine Processionary moth metapopulation. In the latter case, a first attempt to include costs by applying Fuzzy Decision Theory has been made.