Characterizing the scale sensitivity of the cellular automata simulated urban growth: A case study of the Seberang Perai Region, Penang State, Malaysia

In modeling the complex behaviour of urban systems using a cellular automata approach, scale is an important concept in representing space since the result obtained at specific scale shall not be expected to be valid at another scale. The selection of scale, however, has often been taken for granted on the basis of information available to be mapped. Although it is desirable to run a model with fine spatial scale datasets, such datasets may well be costly to collect and add computational expenses in processing. This article investigates the effect of changing scale on the performance of a GIS-based cellular automata (CA) model developed to simulate the spatial pattern of urban growth. The sensitivity of scale of the input data used to parameterize the model was undertaken by evaluating the resulting prediction accuracy of the model and the morphology of urban areas. It seemed that the model managed to perform well and produce realistic urban form only up to specific range of scale. Thus, the selection of scale to represent urban system under consideration has to be undertaken with care, in order to ensure that the output produced maintains the overall accuracy of the model and morphology of urban areas.