Urban Spatial Organization, Multifractals, and Evolutionary Patterns in Large Cities

Abstract

Understanding urban spatial organization and evolutionary patterns is critical to formulating spatial development strategies. Multifractal analysis has been effectively applied to investigate urban spatial organization in a multiscale manner. Without effective approaches to deal with local parameters, however, its ability to identify urban spatial arrangements intuitively and morphologically remains limited. Therefore, in this article, a new method is proposed to characterize local characteristics by introducing a new parameter, the slope coefficient (Formula presented.) This coefficient allows one to define urban spatial structure based on the similarity of local multiscale spatial distributions and explore urban evolutionary patterns. To test its validity, the city of Beijing, China, is selected as a case study. Based on six sets of remote sensing images obtained at six-year intervals from 1988 to 2018, the results show that three types of urban clusters (urban core areas, medium-sized urban settlements, and small villages and towns) dominate the urban spatial organization of Beijing. During the stage of accelerating urbanization, areas of urban growth were dominated by the expansion of urban core areas in the central urban district due to the concentration of population and abundant land resources. During the later process of decelerating urbanization, with declining population growth and limited land resources, the connectivity of urban core areas to the center and periphery, with increasingly scattered medium-sized urban settlement patterns, has become the main mode of urban growth. The findings enable us to reexamine urban spatial organization from a multiscale perspective and provide planning reference for cities experiencing rapid urbanization.