Identification of Urban Spatial Structure of Pearl River Delta Urban Agglomeration Based on Multisource Spatial Data

Identifying urban center structure and urban centers is an important component of urban geography research. This study aimed to propose an urban center identification method, which was constructed by combining nighttime light data, point-of-interest data, and population density data. The method involved constructing the Main Center Score Index and the Mean Subcenter Score Index and using the existing Getis-Ord Gi* model, natural discontinuity method, and other models. This study used this method to identify the urban center structure of the Pearl River Delta urban agglomeration. The results of this study are as follows: A total of 69 urban centers in the Pearl River Delta urban agglomeration were identified in this study, including 9 main centers and 60 subcenters. However, the urban center of the Pearl River Delta urban agglomeration exhibits the phenomenon of uneven spatial distribution, and the urban center is concentrated in the central area of the urban agglomeration. In the urban center structure, Guangzhou, Foshan, Zhaoqing, Huizhou, Dongguan, Shenzhen, Zhongshan, Jiangmen, and Zhuhai were polycentric structures. Among the dominant factors affecting the spatial distribution of urban centers, the dominant factor in Guangzhou and Foshan was the government-led spatial development model, the dominant factor in Dongguan and Zhongshan was the administrative division system, the dominant factors in Shenzhen and Zhuhai were the special economic zone policy and government industry guidance policy, and the dominant factors in Huizhou, Jiangmen, and Zhaoqing were geographical proximity, industrial transfer, and transportation infrastructure. In addition, this study compared the urban center identification results with three data sets and Chinese government urban planning documents to test the accuracy of the method for identifying urban centers. The results showed that the overall identification accuracy of urban centers in this study was 85.48%, among which Foshan exhibited the highest identification accuracy of 90.9%.