Framework for modelling tropical forest dynamics

Autor: Young, Allen C.
Rok vydání: 1999
Předmět:
Druh dokumentu: Electronic Thesis or Dissertation
Popis: A framework for modelling the dynamics of tropical forests is described. The framework makes use of simulation models to predict the long term growth and yield of forests under different management regimes. It is designed to have practical application for the sustainable management of forest resources in tropical countries. The framework comprises a suite of simulation models, each of which is appropriate in particular circumstances. Each model uses a disaggregated representation of a forest stand. An individual-based representation is used for trees in the forest stand above a threshold size, while a more aggregate representation (such as a cohort representation) is used for seedlings and saplings. Processes of stand disturbance and recovery from disturbance are captured. Disturbance results from tree falls and from stand harvesting. Recovery from disturbance may involve seed production, seedling establishment, and competition between trees. Local interactions within the stand are captured, as is species-specific behaviour. The content of each model in the framework is represented in a text-based model design. Details of the content are specified using a formal representation language. The language has semantics and syntax for specifying how a set of generic modelling concepts is employed in an individual model. Details specified in this way include the names and kinds of model variables, and the algorithms used in model calculations. A formal-representation (an 'ontology') of the set of generic modelling concepts developed for use in the framework was created using the Ontolingua server. This provides an unambiguous specification of the modelling concepts used. This assists communication and may also make it easier to design rules and procedures for translating the model content into forms compatible with different modelling systems.
Databáze: Networked Digital Library of Theses & Dissertations