Abstract:
The exponential increase of urban areas in Africa during the last decade has become a major concern in the context of local climatic change and the increasing amount of impervious surface. Major Nigeriantowns such as Suleja have undergone rapid urban expansion and sprawl in comparison tothe rest of the country. This research investigated and predicted the urban sprawl pattern in Suleja using the GIS based cellular automata (CA) model up to the year 2030. Three multi-temporal Landsat imageries for the years 1987, 2001 and 2015 were used to map the urban land-use changes in the study area. The study carried out a supervised classification using the support vector machine (SVM) algorithm. Factors that influence urban sprawl were determined and a logistic regression analysis was carried out to check the level of impact each variable factor had. The variables were then used alongside transition files to predict the urban sprawl pattern for 2015. The accuracy of the predicted 2015 image was validate as against the actual 2015 image. A 14% error was recorded which made the model valid. The model was then used to predict the sprawl pattern for year 2030 which indicated more expansion with a decline in the expansion rate due to the proportion of urban land available. The research concluded that cellular automata model represents a worthwhile approach for urban and regional planning as it is able to simulate complex behavior of urban sprawl. The model provided a good guide to the sprawl and expansion pattern of Suleja local government area illustrating area which can be expanded in future
