Multiscale Associations Between Street Built Environment and Street Vitality in a Polycentric City: Evidence from MGWR Analysis in Chongqing, China

Understanding how the street built environment (SBE) relates to street vitality is critical for promoting livable and sustainable cities, yet its multiscale and spatially heterogeneous patterns remain insufficiently understood, particularly in polycentric urban contexts. Focusing on the core urban area of Chongqing, this study adopts 7951 street segments as the analytical unit to capture street-level spatial processes. A street vitality index was constructed using multi-source data integrating population, social, and economic activities. The SBE was quantified across three dimensions: macroscale street-network structure derived from spatial design network analysis, mesoscale functional characteristics measured using point-of-interest data, and microscale streetscape perception extracted from street-view imagery. The multiscale geographically weighted regression (MGWR) model was employed to examine spatially varying associations between the SBE and street vitality. Results reveal clear spatial non-stationarity in these associations. Closeness, functional density, and functional mix show positive associations with street vitality, whereas connectivity, betweenness, and greenness exhibit mainly negative associations. Transit stop density and enclosure demonstrate bidirectional spatial associations. These findings provide empirical evidence of spatially differentiated associations between the SBE and street vitality in polycentric cities and offer a data-driven basis for differentiated street planning and urban spatial optimization.