The state-space approach in growth modelling

Abstract

Systems theory can provide a useful conceptual framework for the development of forest growth models. Modelling approaches are examined and compared, based on the state-space description of dynamic systems. This viewpoint is used also to discuss the principles behind height-based site indices and possible extensions, and problems of statistical parameter estimation. Four main kinds of estimation methods for dynamic systems can be distinguished. A multidimensional generalization of the Richards model illustrates many of the ideas.