Modelling ecological succession, with particular reference to Markovian models

Abstract

There is a brief review of models of succession: these are classified as verbal or descriptive, simulation, population dynamic, and Markovian. Many facets of the latter class of models are discussed, demonstrating that there are far more disadvantages to their use than apparent advantages. However, Markovian models do appear to have predictive ability, and it is also considered that the patterns of probabilities in Markovian matrices may have a role to play in interpreting opposing views on the mechanisms of succession. Data from the Breckland grasslands (Watt, 1960b) are used as an example.