Building dynamic spatial environmental models

Abstract

An environmental model is a representation or imitation of complex natural phenomena that can be discerned by human cognitive processes. This thesis deals with the type of environmental models referred to as dynamic spatial environmental models. The word ‘spatial’ refers to the geographic domain which they represent, which is the two- or three-dimensional space, while ‘dynamic’ refers to models simulating changes through time using rules of cause and effect, represented in mathematical equations. Since these equations generally include complex interactions which can only be solved by numerical solution, dynamic spatial models are programmed and run on a computer. The aim of dynamic spatial environmental model building is to find the optimal representation of environmental processes in the numerical equations (and parameters) of a computer program of the model, for a given case study defined by the aim of modelling, the properties of the study site, the field data present, the software en hardware technology available to construct the model, and the researchers involved. Since most of these factors will be different for each case study, a new (or modified) model should be made for each case study, by executing the procedural steps of the model development cycle until the optimal model has been found. This thesis evaluates whether existing technology and/or science provide sufficient means to deal with relevant issues related to the model development cycle.