Dynamic Modelling of the Annual Cycle

Abstract

A hypothetico-deductive framework for dynamic ecophysiological modelling of the annual cycle of boreal and temperate trees is described and discussed. In the framework used, the modelling is closely associated with experimental and observational empirical studies, so that inductive (`empirical') and deductive (`theoretical') phases alternate in the research. Computer simulations are deductive, and their results therefore contain nothing but implications of the assumptions of the model used in the simulations. Empirical data are used either for inductive formulation of the models or for testing the models after deducing their predictions by means of simulations. In dynamic modelling, the time courses of seasonal ecophysiological processes are simulated by first calculating the momentary rate of development on the basis of the input data of environmental factors, such as air temperature and night length. After that, the time course of the state of development is obtained by mathematical integration of the rate of development with respect to time. A unifying notation is described for the rate and state variables of different aspects of the annual cycle. Diverse model categories, based on major differences in the ecophysiological phenomena addressed, are described and compared. Finally, the realism, accuracy, and generality of the models are discussed. A novel concept of coverage is introduced for use in the assessment of the realism of the models. Biological levels of organisation, vertical reduction, and emergent properties are briefly discussed in relation to the realism of the models. The framework described and discussed in this chapter forms the basis for the subsequent chapters, in which different aspects of the annual cycle in boreal and temperate trees are addressed.