Spatial heterogeneity analysis between street network configurations and various service activities: evidence from the Wuhan metropolitan area

Abstract China’s economic growth is increasingly being driven by the contemporary service industry in the context of a new economy. This study aims to examine the spatial heterogeneous relationship between various service industry activities and street network design configurations by integrating multisource big data and geospatial analysis to provide insightful implications for human-centered design for compact cities by taking the case study of an inland megacity in central China, Wuhan. Street configurations under the walking/driving modes including closeness, betweenness, severance and efficiency, are characterized from the perspective of spatial design network analysis and angular distance to effectively reflect network shapes and subjective perceptions when navigating through the streets. The point-like, point-axis and ring patterns of various service activities are identified using the kernel density estimation (KDE). Then two sets of densities are analyzed to investigate whether various service activities are spatially associated with specific street metrics and whether spatial stratified heterogeneity exists. The results show that severance and efficiency are two promising indicators to represent the human-scale street design besides the conventional street centrality indices. The spatial mismatch is mainly observed between street metrics and the tourism sector whereas spatial clusters are detected in other types of service activities. Diverse service activities have distinct location preferences for street designs under different transport modes. The walking mode values global closeness and betweenness, while the driving mode values severance and efficiency.