Patterning in Urban Population Densities: A Spatiotemporal Model Compared with Toronto 1971–2001

We build on the literature on population-density distributions, but translate the consensus cross-sectional progression into a three-dimensional and six-stage geographic information system (GIS) based ‘volcano’ model. Visual comparison and descriptive statistics show Toronto's recent density patterns to be very similar to those suggested by the model: the central density cluster has reversed its decline, while peripheral clusters have developed at increasing distances from downtown. Local autocorrelation (LISA) allowed areas of significant clustering and diversity to be mapped, and strong conformity was found between the model and Toronto's empirical patterns. Overall, density levels throughout the metropolitan area are homogenizing and randomizing, even while inner-city redensification and peripheral densification proceed.