Modelling urban growth prediction: an application of complexity theory

Abstract

An urban system has been identified as a non-linear complex system which is a recursive one and an amalgamation of various physical and stochastic factors. Urban growth can be defined as a system resulting from the complex dynamic interactions between the developable, the developed and the planned systems. Complexity of urban systems is distinguished by analysing the spatial, temporal and decision-making dimensions. It is linked with the major current methods of modern urban modelling, such as cellular automata, fractal theory, neural networks, multi-agent and spatial statistics, etc. This research concentrates on the complexity in structural and functional changes, temporal comparability, spatial patterns and spatio-temporal processes of urban growth. With remotely sensed imagery and secondary data, this research work presents a methodology for monitoring and evaluating land use changes over time. My methodology primarily comprises of morphology analysis, urban land use structure change and spatial pattern analysis, using fractal and landscape matrices approach. The outcome shows that the integration of multiple spatial indicators can improve the capacity for interpretation and evaluation. This case study reveals temporal variation in the spatial urban growth process. This work presents an innovative method for the temporal measurement of urban growth for comparing urban sprawl. The method comprises temporal mapping, data disaggregation, integration on spatial gravity, and global evaluation. The findings reveal that the macro patterns of urban sprawl can be interpreted and compared from micro urban activities. This research also shows that pattern, process and behaviour must be integrated into a whole towards understanding the complexity in urban growth. My study presents a preliminary multi-scale perspective for understanding spatial patterns based on the theory of spatial hierarchy comprising of planning, analysis and data that are interrelated. This framework is implemented by using exploratory data analysis and spatial logistic regression and the combination is proven to have a strong capacity for interpretation. Project-based cellular VI automata model is developed for interpreting the spatial and temporal logic between various projects forming the whole urban growth. As a nonlinear function of temporal land development, dynamic weighting is able to link spatial processes and temporal patterns. The findings from this research suggest that this method can improve the temporal interpretation and visualization of the dynamic process of urban growth both globally and locally. My research has found that complexity theories are very helpful in understanding a complex system such as urban growth.