Urban Growth Simulation Based On CAS-CA Model

For a long time, influenced by the Western Rationalism, the urban planning scholars in China are indulging in explore the general law of the development of the city, while ignoring the city’s inherent complexity and particularity. Recent years, with the rising of the complexity science, the complexity science paradigm is affecting the urban planning ideas gradually and deeply. Among them, the Cellular automata, as a discrete dynamical model from Below researching on self-organizing system, is getting more and more attention on the study of the spatial growth in urban planning. The paper attempts to introduce the complex adaptive system theory into the CA model, and extending the existing CA models to simulate the urban spatial growth more reasonably.Firstly, the paper differentiates and analyses the behavior pattern of the urban adaptive agents and the characteristics of the urban complex adaptive system, from the perspective of the complex adaptive systems, on the basis of the review of the domestic and foreign relevant literature. Human is the basic adaptive agent in urban complex adaptive system. However, in urban system, people do not play a role with a single person. In the process of urban development, all kinds of seemingly independent body compose different sizes, different properties and different structure social groups (aggregation) through interests (tagging). And they affect urban spatial growth through group decision, considering the different social groups’ interests and environmental benefits, to maximize the benefits to meet their needs (emerge). The urban development is the common result of the interaction of the urban adaptive agents and the interaction between the urban adaptive agents and the urban geospatial environment.Secondly, the paper constructs CAS-CA model based on the complex adaptive systems from the macro level and the micro level based on the cognitive of the urban complex adaptive system. On macro level, we will simulate the harmonious coexistence between the city and the natural environment by the control of the land suitability evaluation and the land scale. On micro level, we will develop the model’s transition rules based on the behavior pattern of the urban adaptive agents. The paper introduces the concept of probability control into the CAS-CA model, and converts public services accessibility and transportation accessibility value into probability value to include these two elements in the calculation of the cellular comprehensive state values. The cellular comprehensive state value is decided by the cellular neighbor effect value, the transportation accessibility value and the public service accessibility value, and chooses the highest comprehensive value cellular to convert its land-use status. Besides, the paper develops the CAS-CA model system by the components integration technology of GIS.Finally, the paper takes Changting County as an example, building a society economic and land-use geospatial database based on GIS, simulating the urban spatial growth of the year 2015 and the year 2020 of the Tingzhou Town of Changting, and comparing with the 2015 land-use map and the 2020 land-use plans. The model simulation result is in line with the reality and the general law of the urban development from the structure. In addition, the paper simulates the evolution process of the commercial service facilities on both sides of the Zhaozheng road from the year 2015 to the year 2020. The simulation result shows the dynamic process of urban development.In current urban planning, the urban plan is an ultimate blueprint plan, lacking the process control and the guidance, and lacking the figurative expression of the dynamic process of the urban development. The paper attempts to build a CA model based on CAS transition rules, and show the urban spatial development process concretely by quantitative analysis, so that the public can have an intuitive understanding of the urban development, and it also provides decision supports for the urban planning.