Seasonal surface urban heat island clustering and thermal hotspot mapping: Dual-scale spatial diagnostics and planning implications for Colombo, Sri Lanka

This is the authors' abstract. We don't add key points for this paper.

Urban surface heating has intensified worldwide, increasing heat stress, energy demand, and health risks. In rapidly urbanizing Asian cities, fine-scale, decision-ready evidence linking surface urban heat island (SUHI) patterns to urban form and population exposure remains limited. Translating SUHI spatial patterns into actionable, climate-resilient planning guidance at operational scales remains a persistent challenge. This study integrates geospatial data, spatial analytics, and planning-relevant outputs to address this gap for the Colombo Metropolitan Area, Sri Lanka. Daytime and night-time land surface temperature from MODIS Terra and Aqua were seasonally composited for the HOT (December–April) and WET (May–November) seasons and analyzed using two hexagonal neighborhood grids with nominal spacings of 500 m and 250 m, which served as alternative analytical neighborhood supports for evaluating the robustness of spatial patterns across different aggregation scales. Covariates included built-up and water/wetland classes from ESA WorldCover (10 m), terrain from SRTM (30 m), road networks and hydrography from OpenStreetMap, and gridded population from WorldPop. Spatial diagnostics combined Global Moran's I, false-discovery-rate-controlled Local Indicators of Spatial Association (LISA), and GetisOrd Gi hotspot analysis to identify statistically significant and spatially coherent hotspot neighborhoods beyond simple temperature mapping. Wet-season daytime SUHI exhibited strong positive spatial autocorrelation across both analytical supports. Hotspots concentrated along inland impervious corridors, while cool clusters aligned with wetlands and canals. Stable hotspot footprints supported corridor-based cooling, wetland conservation, street-tree deployment, and prioritization of high-exposure communities, providing reproducible, open-data evidence for climate-resilient urban planning supporting SDGs 3, 11, and 13.