Identifiering av urbana värmeöar och bedömning av flodnära åtgärder : Fallstudie om Coimbra, Portugal

Urban Heat Islands (UHI) pose a growing challenge for climate-sensitive urban planning, particularly in Mediterranean cities vulnerable to heatwaves and drought hazards. This study investigates the spatial distribution of UHI in Coimbra, Portugal through remote sensing and geospatial analysis. Further, it evaluates the cooling potential of blue-green infrastructures (BGI), emphasising proximity to the Mondego River through ten land cover change scenarios. UHI intensity was calculated using both z-score standardisation of land surface temperature (LST) (UHI-z) and urban–rural LST differentials (UHI-r) for comparative analysis. On average, the urban area experienced temperatures 4.31 °C higher than surrounding rural zones. Two predictive models, multiple linear regression (MLR) and Random Forest (RF) were tested to link UHI with land cover variables. RF outperformed MLR (R² = 0.916) and indicates strong predictive accuracy and robustness to complex spatial patterns.Scenario-based modelling tested the cooling effectiveness of various land cover interventions, including the conversion of impervious surfaces (buildings, roads) to vegetated ones (grassland, cropland, tree cover). Ten land cover change scenarios were tested. The largest surface temperature reduction, averaging -1.9 °C in dense urban zones resulted from the scenario converting 25% of built-up areas to irrigated cropland (simulating intensive green roofs). The Mondego River demonstrated a localised cooling effect within a 688 m buffer, while tree cover exhibited a broader influence, reducing temperatures up to 1,023 m from source areas. These findings confirm the synergetic role of water and vegetation in urban climate regulation. The results highlight the importance of spatial context, land use configuration, and targeted BGI placement in mitigating urban heat. The study provides practical insights for urban planners in designing effective cooling strategies and emphasizes the need for further research that incorporates seasonal, nocturnal, and ground-based thermal data to enhance mitigation strategies.