Computer based development of large scale ecological models problems and prospects

Modelling an ecosystem requires knowledge of the system, obtained with experiments, and its abstraction within a mathematical framework. Systems methods can be used effectively in the latter phase of the modelling process. Indeed, when coupled with experimental work, these mathematical methods can help in the development of an ecologically realistic mathematical model. These models formalize hypotheses proposed to describe the ecosystem structure and explain its behavior. For example, a forest is not only made of trees, but also of shrubs, fungi and other flora. Animals contribute to modify and influence the forest, and chemical nutrients cycle between flora, fauna and soil. A mathematical model of this ecosystem would try to conceptualize relationships among these domains. If this model is set in dynamical terms, e.g., a set of differential equations, then the model behavior should be similar to the forest behavior.