Vegetation dynamics modelling:: a method for coupling local and space dynamics

Simulation of some biological or ecological phenomena is generally based on differential equations. We propose a new approach based on coupling a local model and a spatial model described by a cellular automaton. The coupling process is referred to a prediction correction scheme. To illustrate our methodological approach we present an example of vegetation dynamics problem. The local model describes the vegetation growth in its ecosystem with two state variables, the biomass and the soil moisture, using Monteith growth approach (Plant Canopies: Their Form and Function 31 (1989) 21). To introduce the spatial variable, we have chosen a cellular automaton to describe the expansion biomass. The vegetation growth model is described by rules related to seed dispersal and soil moisture phenomena which are interacting with neighbouring plants. The coupling process describes the evolution of the geographical layers (with cellular automaton) and biological layers (with local model) of vegetation in the mediterranean climate environment. (C) 2002 Elsevier Science B.V. All rights reserved.