Hierarchical Bayesian spatio–temporal models for population spread

Abstract

Abstract There is a long history in the ecological sciences concerning the development of mathematical models for describing the distribution of organisms over time and space. Although such models have sometimes been evaluated by comparison to observations, they have seldom been “fit” to data in a formal statistical sense. Thus, uncertainties regarding data, model, and parameters are not readily accounted for in such analyses. Alternatively, realistic statistical models for spatio–temporal processes in ecology often require that one estimate a very large number of parameters. This is typically not possible given the relatively limited amount of data collected over space and time.