Nanjing Urban Spatial Extension Characteristics And Simulation Studies

The study of urban space is one of the important aspects of urban studies, and it is the theoretical basis of city planning. So it has become one of the research hotspots in city planning research. As a result of urbanization, urban spatial expansion clearly reflects the urbanization process. At the macro level, it is illustrated by the expansion of city size and city build-up area. At the micro level, it is a process that the different types of non construction land (grassland, farmland and so on) are converted into construction land. How to interpret the processes and changes scientifically, and to visualize these processes and changes qualitatively and quantitatively are very important and are the subjects of this thesis. Along with the continuous development of science and technology, to support our research with the help of RS, GIS and other mathematical models is feasible to analyze the urban expansion rules, to predict the direction and trend of urban development.First, the paper discusses the process of urban space expansion, and elaborates the theories, methods and models. All these works are the theoretical bases of the empirical investigation. In the case study, we use RS, GIS means to present the city spatial expansion process intuitively and scientifically. Secondly, according to the analysis of urban spatial expansion characteristics based on the expansion of the scale, expansion intensity changes in time and space, and the directional anisotropy of the extension, the author comes to know that the urban spatial expansion process of Nanjing is not balanced. Furthermore, the author tries to analysis the extended intrinsic mechanism from the natural geography environment, economic and social conditions, traffic construction and policy planning and other aspects. Thirdly, based on the study of general urban expansion model, the author introduces cellular automata model which has complex’spatial and temporal characteristics, and focuses on the perspective from the model, practical application, expanding the classical CA model by introducing the geographically weighted (GWR) thought when calculates the conversion rules, making a simulation to forecast the urban spatial expansion, and it makes comparisons with the traditional secondary logistic regression analysis. The results show that when the secondary logistic regression is combined with the CA to simulate the transformation of urban element cell state, the outcomes have a high similarity with the actual results and when the geographically weighted regression is combined with the CA to simulate the transformation of urban element cell state, the outcomes also have a high similarity with the actual results, but the latter simulation is more detailed and real, and this shows fully and forcefully that the GWR model has significant improvements than the logistic regression in simulating accuracy and the impact on factors, spatial representation.