The global population is increasingly urbanized, leading to greater exposure to extreme temperatures exacerbated by the urban heat island (UHI) effect. UHIs, characterized by warmer urban than rural temperatures, result from factors such as building materials, human activities, and reduced evaporative cooling. While UHIs pose significant heat-related health risks, potentially increasing mortality during heat waves, they also offer protection against cold extremes. The balance between these summer risks and winter benefits remains largely unexplored across diverse cities and climates. Existing literature often focuses on a limited number of cities, lacks granularity in addressing intra-urban temperature variations, and often neglects the economic consequences of UHI-related mortality. This study aims to provide a comprehensive analysis of UHI's impact on mortality risk across 85 European cities, considering both heat and cold temperatures throughout the year. High-resolution urban climate simulations, population density data, and city-level temperature-mortality relationships are used to quantify mortality risks across urbanization gradients. Furthermore, the study monetizes these climate-related risks, enabling comparison with other urban living costs to emphasize the significance of considering UHI's impact on public health and economic planning.

Previous research has documented the adverse effects of extreme temperatures on human health, particularly the significant increase in mortality observed during heat waves like the 2003 European heatwave. Studies show that UHIs exacerbate heat-related risks but also offer protection against cold extremes, although this protective effect is less studied. While many studies examine UHI mitigation strategies focused on thermal comfort, fewer quantify health risks with the necessary granularity to account for intra-urban variations. Economic assessments of UHI's impact, including its effect on mortality, have been limited in scope and scale. Existing literature highlights that UHIs can significantly increase the economic impact of climate change, but comprehensive, city-specific economic valuations are lacking. This gap in the literature motivates this study to provide a city-level analysis of UHI impact, considering seasonal variability and integrating economic assessments of mortality.

This study utilizes hourly near-surface air temperature data from the UrbClim urban boundary layer climate model, provided by the Copernicus Climate Change Service, for 85 European cities from 2015-2017. UrbClim simulates urban climates considering land surface characteristics, local climate conditions, and anthropogenic heat flux. Daily average temperatures, re-gridded to 500 m resolution, are used. Urban and rural areas are classified based on land cover type (CORINE land cover class), considering >80% impermeable land cover for urban areas. Elevation data (MERIT DEM), population density (NASA SEDAC), land imperviousness (Copernicus Land Monitoring Service), and Köppen-Geiger climate classification data are incorporated. Temperature-mortality exposure-response relationships for five age groups are obtained from Masselot et al. (2023), using a two-stage framework involving quasi-Poisson regression with distributed lag nonlinear models (DLNMs) and meta-regression. These relationships provide the average risk for each city and age group at different temperatures. Daily attributable mortality is calculated using the attributable fraction (AF = (RR-1)/RR, where RR is relative risk). The UHI impact on mortality risk is computed by comparing average attributable mortality in urban and rural areas. Heat and cold extremes are defined as the warmest and coldest 2% of days. Cities are classified into Köppen-Geiger climate groups based on the dominant climate within each UrbClim urban domain. Economic impacts are quantified using the value of statistical life (VSL) approach, converting OECD's estimate to 2021 Euros, and considering inflation. The value of a life year (VOLY) approach is also considered. Analyses are performed in R and Python, with code available on GitHub.

The study reveals that UHIs have the greatest impact on mortality risk during heat extreme days, with a median increase of 0.25 additional deaths per 100,000 adult city inhabitants per day (45% increase compared to rural areas). During cold extreme days, UHIs show a protective effect, with a median decrease of 0.05 deaths per 100,000 inhabitants per day (7% decrease). However, the annual net impact of UHIs on temperature-related mortality is generally minimal due to the prevalence of colder temperatures in many European cities. The median annual economic impact of UHI heat-related mortality is €192 per adult inhabitant, while cold-related mortality shows an economic benefit of €-314. There's a weak correlation between the magnitudes of heat and cold-related mortality impacts. Fifteen cities show a net adverse economic impact. The economic impacts of UHI-related mortality are comparable in magnitude to those of air pollution (PM2.5 and ozone) and transport costs. Cities with higher economic impacts from UHI heat-related mortality tend to also have higher impacts from air pollution-related mortality. The magnitude of UHI impact is influenced by population vulnerability to heat and cold, frequency of warm days, and the magnitude of UHI itself. Geographically, northern and eastern European cities, and those with colder average temperatures, exhibit weaker protective effects of UHIs during winter, potentially reflecting better adaptation to cold weather. Southern and eastern European cities experience stronger adverse summer impacts due to more warm days. The analysis considering years of life lost (YLL) demonstrates that younger populations are more vulnerable to heat, potentially leading to adverse effects in more cities using the YLL approach. The study reveals a notable spatial heterogeneity in UHI impact within cities, with more built-up areas experiencing more pronounced effects, either adverse or beneficial. The analysis considering socioeconomic deprivation shows that more deprived neighborhoods in London and Leeds experienced greater UHI impacts.

The findings highlight that UHI mitigation strategies must account for the seasonality of risks. Reducing heat-related risks without sacrificing cold-related benefits is crucial. Strategies such as enhancing evapotranspiration through urban vegetation or implementing seasonally optimized materials like thermochromic roof membranes offer potential solutions. Climate change is expected to increase the proportion of warm days, potentially shifting the annual net impact of UHIs from protective to adverse. The study's economic assessment provides a first-order valuation of UHI's impact on mortality, but the true economic cost likely involves additional factors such as morbidity, reduced productivity, and energy consumption. The study acknowledges limitations in using city-level exposure-response relationships to represent individual variations in exposure and vulnerability, and the need for integrating socioeconomic factors into vulnerability assessments to capture the effect of UHI in diverse settings and populations.

This study demonstrates the complex and seasonal nature of UHI's impact on temperature-related mortality in European cities. The economic valuation of this impact highlights the importance of incorporating health and social costs into urban planning. Future research should focus on improving the granularity of data on social vulnerabilities, individual exposures, and population mobility to create healthier and more equitable cities.

The study uses city-level exposure-response relationships, which might not fully capture individual variations in exposure and vulnerability within cities. Factors such as access to healthcare, and population health, may also influence health outcomes and are not fully reflected in this study. Indoor temperatures and population mobility are not explicitly considered, which could affect exposure estimates. The economic assessment primarily focuses on mortality costs, and other economic impacts (morbidity, productivity losses, energy use) are not fully quantified. The analysis relied on relatively average years for European temperatures. The impacts of exceptional events would likely be considerably greater.