Understanding Urban Growth System: Theories and Methods

Understanding the urban growth system is a prerequisite for modelling and forecasting future trends of urban land use/cover change and its ecological impacts. As urban growth involves various actors with different patterns of behaviour, we argue that scientific understanding must be based on elaborated complexity theory and a multidisciplinary framework. The theoretical analysis can provide a guideline for selecting modelling methods currently available in complexity modelling and in remote sensing and GIS environments. This paper first proposes a conceptual model for defining urban growth and its complexity, in which spatial, temporal and decision-making complexity are distinguished as separate domains. Second, this paper links the conceptual model with the major current methods of modern urban modelling, such as cellular automata, fractals, neural networks, multi-agent, spatial statistics etc. This confrontation enables the possibilities of various modelling methods to understand urban growth complexity to be indicated. Third, this paper evaluates the operational implementation of representative methods based on criteria such as interpretability, data need and GIS embeddedness.