| description abstract | Urban land-use planning is becoming increasingly complex in China because it requires estimating the future urban extent and balancing its various aspects, e.g., provision of enough housing and employment opportunities and preservation of the environment. This paper presents an integrative approach to the problem of sustainable land-use planning. Specifically, an urban growth simulation tool, a cellular automata model, was used to determine the extent of the urban area, and a neural network approach was used to quantitatively predict the land-use structure in the projected year. A multiobjective optimization genetic algorithm was then developed to search for Pareto-optimal urban land-use plans within the urban extent determined by the urban growth model. To search for a sustainable urban land-use plan, social, economic, and environmental objectives were defined that reflect the multiple objectives of the urban system. In addition to these nonspatial goals, objectives concerning the spatial distribution of land usages were proposed. These are classified as local and global spatial objectives. A case study of the Donghu Lake watershed was conducted. Donghu Lake is one of the largest downtown lakes in central China, and the watershed area is undergoing rapid urbanization and suffering from nonpoint source water pollution. The study was carried out to validate the proposed method, and it demonstrated that the models are appropriate for areas undergoing urban expansion. | |
