Evolution Pattern of Urban Agglomerations Based on Bayesian Networks from the Perspective of Spatial Connection: A Case Study of Guangdong-Hong Kong-Macao Greater Bay Area, China

The study of the formation and development of urban agglomerations is of great significance, and the connection between cities is the critical foundation for shaping these agglomerations. However, the mechanism behind spatial connection between cities in the formation of urban agglomerations remains unclear. Using the Greater Bay Area (GBA) as a case study, we proposed a Bayesian network framework that integrated the spatial connection index and land use intensity. We constructed a dependency network of land use intensity from the perspective of spatial connection, and summarized the spatiotemporal evolution patterns of urban agglomeration combined with social network analysis methods. The results indicate that: (1) From 1980 to 2020, both land use intensity and spatial connection strength in the GBA have significantly increased, though the characteristics of different cities varied noticeably; (2) The spatial connection center of the urban agglomeration has shifted geographically from Hong Kong and Macao to the Pearl River Delta, and then to the east bank of the Pearl River. Hong Kong, Guangzhou, and Shenzhen are the three core cities in the spatial connection network, each with different development trajectories. (3) A dependency network of changes in land use intensity among cities at different stages from the perspective of spatial connection was constructed, identifying the evolving roles of each city in the development of the urban agglomeration. The study discussed a three-stage development model of urban agglomerations from the perspective of spatial connection, providing a new perspective for exploring the formation mechanism of urban agglomerations.