Answers are grounded in this paper and cite their sources.
Intensifying demographic polarization has placed a subset of cities on a trajectory of persistent population decline, accompanied by extensive vacant land and, in some cases, urban perforation. Previous studies have linked population decline to vacant land, but how vacancy varies with the intensity and trajectory of shrinkage remains underexplored. Using shrinking cities in Northeast China, we refined shrinkage typologies with explicit spatio-temporal trajectories, mapped vacant land in central urban areas with a VGG16–U-Net segmentation model, and applied generalized linear mixed models to examine associated factors. Between 2010 and 2020, the region lost more than 10% of its population; 35% of cities experienced severe or extreme shrinkage, and nearly 20% of central urban areas contained vacant land (364.7 km 2 ). By refining shrinkage in both space and time, we find a nonlinear relationship between urban perforation and shrinkage severity. A one-standard-deviation increase in standardized shrinkage is associated with 5% higher odds of perforation. Core districts show lower odds of perforation, while inner suburban and peripheral areas are more heterogeneous. In contrast to the inverted U-shaped relationship reported for increasing urbanization, our study finds a non-monotonic pattern across shrinkage-severity groups. It rises under slight shrinkage, flattens under moderate shrinkage, and resurges under severe shrinkage. Economic vitality most strongly suppresses perforation, whereas building density amplifies it; both peak under moderate shrinkage.
No takes yet. Share an insight, caveat, or question.
Li et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: