Mapping heat-related risk in three mid-sized European cities

Urban heat risk arises from the interaction of thermal hazard, population distribution, and socioeconomic conditions, yet comparative high-resolution assessments across European cities remain limited. We develop a harmonised 100 m heat-risk framework for three climatically contrasting mid-sized cities, Iași (Romania), Jena (Germany), and Turku (Finland), using ten years of observation-based daytime and nighttime air-temperature data. The framework distinguishes societal heat risk, which reflects the number of residents exposed, from per-capita heat risk, which normalises exposure to reveal inequalities that aggregated metrics may conceal. To our knowledge, this is the first study to provide an integrated heat-risk assessment of this kind for each of the three cities, combining diurnally resolved hazard, exposure, and rent-based vulnerability at high spatial resolution. As a fine-scale and timely proxy for vulnerability, we incorporate apartment rent from geolocated real-estate listings; while not a comprehensive socioeconomic indicator, it captures affordability- and housing-related conditions relevant to heat sensitivity. To characterise environmental context, we quantify local (100 m) and neighbourhood-scale (300 m) land-cover fractions (tree cover, impervious surfaces, grassland) and assess their statistical associations with hazard, exposure, vulnerability, and risk components using Spearman correlations. In addition, we use Lorenz curves and Gini coefficients to quantify spatial inequalities in heat risk and green infrastructure within each city. Across all cities, societal heat risk closely follows the residential population distribution, whereas per-capita risk aligns more strongly with spatial patterns of thermal hazard and adaptive capacity. Nighttime conditions (Tropical Nights) emerge as particularly influential for both risk measures, underscoring the importance of nocturnal heat for recovery and health. Green land cover exhibits predominantly negative, and impervious surfaces predominantly positive, associations with hazard and risk, with effect sizes varying across cities and spatial scales; these relationships are interpreted as correlational rather than causal. Lorenz and Gini analyses further show that societal heat risk and green infrastructure are unevenly distributed, with pronounced inequalities in exposure and nighttime hazard, particularly in Iași and Turku. Overall, societal heat risk is primarily governed by where people live, while per-capita risk reveals how hazard and adaptive capacity shape the burden experienced by individuals. Apartment rent offers a practical, location-specific proxy for vulnerability that can be applied where detailed socioeconomic data are limited. The proposed framework provides a transferable basis for identifying both areas where many residents are exposed and areas where individuals may face disproportionate heat burdens, supporting more context-sensitive and equitable urban heat-adaptation planning.