Spatial analysis of urban shrinkage: a scoping review of methodological and analytical gaps

Urban shrinkage – the sustained loss of population and economic activity that leaves cities burdened with vacant land, surplus infrastructure, and fiscal strain – has generated a rapid growing body of spatial-analytic research. Yet no synthesis has clarified how such studies operationalize and model the phenomenon at fine spatial scales. This scoping review systematically analyzes 73 peer-reviewed studies restricted to sub-city analyses (parcel, block, neighborhood, or town level), coded by geography, spatial unit, metric family, and method family under a transparent PRISMA protocol. Four methodological blind spots emerge. (i) Geographic bias – nearly all studies are concentrated in a small set of data-rich cities across North America, East Asia, and Europe, leaving most of the Global South unrepresented. (ii) Scalar imbalance – about 90% analyze a single spatial unit, limiting insights into cross-level feedbacks between parcels, neighborhoods, and cities. (iii) Metric divergence – key constructs such as vacancy or land-use change are defined in non-comparable ways across 62 datasets and seven metric families. (iv) Fragmented methodological tiers – roughly two-thirds of studies rely on cross-sectional diagnostics (spatial autocorrelation with regression), while dynamic, multilevel, and agent-based approaches remain scattered and seldom integrated. To advance the field, we recommend three incremental shifts: (1) geographic diversification toward intra-urban ‘perforation’ and Global-South contexts; (2) metric standardization anchored in transparent data-selection rationales; and (3) methodological integration that links diagnostic pattern detection to dynamic or multilevel models. Together, these steps can transform isolated case studies into a cumulative, transferable evidence base for planners and policymakers confronting uneven urban decline.HighlightsReviews sub-city spatial analyses of urban shrinkage across 73 empirical studies.Reveals four structural gaps in geography, scale, measurement, and methodology.Shows dominance of single-scale, snapshot-based spatial analysis approaches.Proposes multi-scale and longitudinal designs to advance shrinkage research.