Simulating urban expansion by coupling a stochastic cellular automata model and socioeconomic indicators

Abstract

Urbanization is one of the most important anthropogenic activities that create extensive environmental implications at both local and global scales. Dynamic urban expansion models are useful tools to understand the urbanization process, project its spatiotemporal dynamics and provide useful information for assessing the environmental implications of urbanization. A hybrid urban expansion model (NNSCA model) was proposed to simulate rapid urban growth in a typical industrial city, Dongying, China, by coupling a artificial-neural-network-based stochastic cellular automata model and several socioeconomic indictors, i.e., the per capita income of the rural population, the per capita income of the urban population, population and gross domestic products of the city. Good conformity between simulated and actual urban patterns suggested that the NNSCA model was able to effectively simulate historic urban growth and to generate realistic urban patterns. A series of scenario analyses suggested that the expanding urban would threaten the ecosystem health of coastal wetlands in the city unless environmental protection actions are taken in the future. The NNSCA model provides abilities to assess future urban growth under various planning and management scenarios, and can be integrated into ecological or environmental process models to evaluate urbanization’s environmental implications.