An integrated approach of Multi-Criteria Suitability Evaluation and Cellular Automata modeling for urban growth simulation

In this paper, based on the ArcGIS 9.3 platform and python scripts, an approach of integrated CA and Multi-Criteria Suitability Evaluation (MCSE) analysis is developed to simulate the changing land use patterns of urban area in the past decades. The methods are designed to allow slow edge growth around towns and villages, rapid edge growth around cities and permits random growth only within the suitable development area. Take the City of Waterloo as a study area, and base on the understanding of the logistic trend and spatial pattern of urban and region development processes. By implementing the model in raster analysis environment of ArcGIS, this study extracts more precise influencing factors of urban growth, and then uses Multi-Criteria suitability evaluation analysis methods to create a satisfying suitability raster dataset. After the improved accuracy tests with CA automation, especially in the random growth model, the desirable results show the dynamic development of urban areas of Waterloo region between the 1960s and 1990s. Several experimental scenarios of development of Waterloo region under realistic conditions are also presented using animation technology in ArcScene. This study shows that experimental application in the study area which produced realistic results and demonstrated the model was theoretically valid and feasible. In the last part, the analytical thinking on urban stimulation using CA and GIS are also discussed.