Multiple pathways to urban density: A comparative diachronic analysis of building type transitions across three city centers

Pursuing built density purely as a quantitative target overlooks how similar density levels can be achieved through different building configurations. This insight calls for a more nuanced understanding of densification - not just how much we densify but how we achieve it through distinct building types. Recent typo-morphological studies have advanced quantitative methods for identifying building types, yet remain predominantly synchronic without insight into long-term evolutionary patterns. This paper presents one of the first attempts to operationalize these methods in a diachronic framework to investigate how building types emerge and influence densification processes across various periods. We hypothesize that cities do not simply become denser uniformly but follow distinct densification pathways through various transitions between building types. Through hierarchical cluster analysis of a novel longitudinal dataset spanning the 1800s–2000s across three international city centers –New York, Melbourne, and Barcelona– we identify seven distinct building types and trace their evolution. Our analysis reveals three fundamental pathways to density: (1) compact densification, where cities fill in available space while maintaining building heights; (2) vertical densification, where building heights increase while maintaining ground coverage; and (3) dispersed densification, where taller buildings are constructed with reduced ground coverage, creating more open space between them. By extending built density measurement into diachronic analysis and combining it with machine learning clustering, we contribute to urban morphometrics with a more time-conscious focus. This methodology enriches transnational urban form comparisons and provides insights into achieving density targets while balancing intensification and environmental sustainability.