A Spatial Typology of University Campus Areas in Cold-Climate Cities: Evidence from Three Cities in Northeast China

Abstract

Campus–city relationships are important for optimising campus areas in cold-climate cities. Existing classifications are largely based on location, openness, or spatial layout, but rarely distinguish combinations of campus–city spatial relationships. From a campus-area perspective, this study develops a spatial typology framework. It examines 53 samples in Harbin, Changchun, and Shenyang, treating the campus proper and its proximate urban area within 1000 m as a unified analytical unit. The study combines multilevel spatial characterisation with correlation, factor, and cluster analyses. The results show that four factors explain 61.25% of the variance before rotation and correspond to four dimensions: spatial interaction support, morphological difference, transport accessibility, and built-up intensity difference. Factor-score clustering identifies four types: interaction-dominant, transport-accessible, connection-imbalanced, and peripherally isolated. The silhouette coefficient averages 0.311, and the adjusted Rand index (ARI) averages 0.869 across resamples. All four occur in each city. Type composition does not differ significantly, but spatial distributions vary. Four winter-relevant spatial indicators load on three of these factors, representing spatial conditions potentially relevant in winter rather than measured climate effects. The typology extends classification from individual attributes to combinations of campus–city spatial relationships, supporting comparative research and type-specific regeneration.