Spatiotemporal Lock-In of Short-Term Rentals in Dubrovnik’s Historic Core

Platform-mediated short-term rental (STR) markets concentrate intensely in heritage urban cores, yet the temporal stability of this concentration remains poorly understood. This study examines spatial dynamics of STR concentration in Dubrovnik’s UNESCO-listed historic core across five biennial cross-sections (2017–2025) using complete administrative eVisitor registration data and an H3 hexagonal grid at resolution 11. Global and local spatial autocorrelation (Moran’s I, LISA, Getis-Ord Gi*), Emerging Hot Spot Analysis for spatiotemporal typologies, and bivariate LISA to distinguish capacity saturation from fragmentation were applied. Results demonstrate structural persistence: Moran’s I remained highly significant (p < 0.001) across all periods including COVID-19 disruption (range 0.417–0.467, CV = 4.4%), despite 53% supply growth and only 3.7% spatial expansion. Consecutive hotspots dominated typological classification, indicating active consolidation. Capacity analysis revealed concordant High–High patterns (56.8% of significant cells) with zero High–Low associations, confirming saturation not fragmentation. Findings support spatial lock-in in STR markets: concentration persists because locational advantages are properties of place rather than market volume, requiring spatially differentiated regulation rather than aggregate supply controls.