European Union flag

New EEA analysis shows that, within cities, some groups of people are more exposed to air pollution than others. This may lead to these people suffering disproportionately higher health impacts from poor air quality if the problem is left unaddressed.

The recently revised EU Ambient Air Quality Directive requires Member States to develop air quality plans with specific actions to protect sensitive populations and vulnerable groups, as defined by their higher sensitivity and reduced ability to protect themselves (EU, 2024).

This EEA assessment – part of a broader scope of work on environmental inequalities – helps to identify the most critical areas within countries and cities in EU Member States and Norway. It can guide the development of effective national and local measures to reduce environmental inequalities and protect vulnerable people.

Exploring environmental inequalities within European cities

Three-quarters of Europe’s population live in urban areas (Eurostat, 2021). Due to high population density, levels of economic activity and traffic congestion, urban areas often see higher levels of air pollution (Eurostat, 2024b) as well as other environmental concerns such as transport noise and the urban heat island effect. In fact, almost all EU urban citizens (94%) are exposed to a key harmful air pollutant – fine particulate matter (PM2.5) – at concentrations above recommended World Health Organization (WHO) levels (EEA, 2025b).

Air quality varies between European cities, and also within them. The new analysis shows that all EU-27 countries and Norway have air pollution hotspots. These are locations where concentrations are higher for one or more key air pollutants (up to four) than in other areas. These hotspots can be caused by pollution from nearby transport networks, industrial facilities or home heating systems.

In parallel, socio-economic factors such as type of employment, educational attainment and income level also influence both where people live and their vulnerability to environmental pollution.

Vulnerability to environmental pollution

Some people are more vulnerable to the health impacts of air quality than others. This vulnerability can be caused by sensitivity (e.g. due to pre-existing conditions, age) and also by concurrent factors such as occupational risks, limited access to healthcare, limited coping capacity and the inability to relocate to another, less polluted area. These differential sensitivities, coping capacities and possibilities for relocation are influenced by the economic resources of the individuals, their social and cultural circumstances and networks, and levels of pollution awareness, among other variables (EEA, 2018; Diekmann et al., 2023; Brazil, 2022). This means that, depending on their socio-economic status, people are not only unevenly exposed to pollution in their work and living environments, but they also face differences in the risk that pollution poses to their health, as well as their capacity to respond. Because of the diversity of socio-economic factors involved, the manifestation of environmental inequality has been shown to differ between subpopulations, local areas and countries that might be considered fairly similar (Fecht et al., 2015).

While age is a well understood and reported variable at high resolution at the European level, income level and most key economic factors that describe a household’s socio-economic status are currently not available at a high resolution (EEA, 2018; Fairburn et al., 2019). To illustrate this limitation, the EEA’s only current indicator measuring inequalities in air pollution exposure, GDP-related air pollution exposure inequalities between regions in Europe, is reported at the regional level (NUTS3). While air pollution information is available at a 1x1km scale for the entire EU, the highest current resolution for GDP data (and other indicators, such as disposable income) as reported by EUROSTAT is, at best, the level of small regions (NUTS3). Hence, until more detailed datasets on direct economic indicators become available, this assessment and the associated viewer make use of data on age and of two proxy variables: employment and place of birth. These are imperfect proxies for socioeconomic status, but the evidence supports the interlinkages. For example, in 2025, 35% of people living in a given EU country, but born in another country, faced the risk of poverty or social exclusion, compared with 18% of people living and born in the same country (Eurostat, 2025b). Previous work has also shown that people’s place of birth could be related to inequalities in exposure to air quality due to a variety of socio-economic factors (Brazil, 2022; Fong et al., 2022; OECD and EC, 2023; OECD, 2024; Fecht et al., 2015).

In this context, the study identifies air pollution hotspots where there is a coincident occurrence of a higher proportion of a particular group – older people, children, employed or unemployed or foreign-born – without directly exploring possible reasons. Depending on the country, the identified hotspots may both cover large areas and several cities showing regional differences, and more localised ones within cities.

Older populations and children

In 20 countries of the EU-27 and Norway, air pollution hotspots hosted a larger proportion of older people. Similarly, in 16 countries, air pollution hotspots were identified with higher proportions of children. This is noteworthy, given that older people and children are particularly sensitive and vulnerable to air pollution as they tend to be more adversely affected while having limited agency to act (EEA, 2018, 2023). The pollution hotspots with more old people and children all had at least one air pollutant above the WHO annual guidance value, typically fine particulate matter (PM2.5), with pollution levels between 5% and 53% higher than in other areas of the same city or country, indicating increased health risks.

Employment

Whether people are in employment or not and the nature of their employment can also influence their socioeconomic status, concurrent health risks, access to healthcare, and ability and willingness to relocate, for example if the most accessible jobs are found in the highly polluted areas. However, the absence of quantitative income data makes it difficult to discern economic vulnerability among employed people. Potentially due to these uncertainties and limited data, the analysis shows a mixed picture, with 18 countries having hotspots with higher than average levels of employed or unemployed people – people who were therefore disproportionately exposed to air pollution. Additional data are needed to further explore these inequalities.

Place of birth

Place of birth is a standard, harmonised census variable across the EU, encompassing the following standard categories: born in the reporting country; born in another EU Member State; and born outside the EU (see the methodology box). Place of birth does not relate to ethnicity or cultural grouping.

In all countries except Germany, some air pollution hotspots hosted a significantly higher proportion of foreign-born people, while only one country – Luxembourg – had an air pollution hotspot with a higher proportion of people born in the country. This indicates that the areas with a greater proportion of foreign-born populations are disproportionately exposed to air pollution.

Further information

The results for all the EU-27 countries and Norway are available through the European Environment and Heath Atlas (see Figure 1).

Many factors contribute to sensitivity and vulnerability to air pollution. In addition, 1km grids can mask substantial variability in both air quality and demographics within a single square, driven for example by building heights and proximity to roads and industrial sites.

The area-level results presented in the viewer do not imply individual risk. They only highlight the statistically disproportionate exposure to air pollution of areas featuring different proportions of various population groups – they are not a direct indicator of risk or impact. Results may identify a need for further investigation at a local level.

The atlas also allows users to check how the environment affects health and well-being in any given urban area within the EU.

More information on health inequalities in relation to climate change is available through the European Climate and Health Observatory, including data viewers and analysis showing population exposure to potential flood-prone areas and the urban heat island effect.

The EEA is broadening its work on environmental inequalities, with work ongoing on assessing inequalities related to noise, climate-related hazards and access to green areas.

Figure 1. European Environment and Health Atlas extract

Please select a resource that has a preview image available.

Other sources on information on inequalities in EEA assessments

Box 1. Methodology

The analysis explored clusters of urban locations showing similarities in both air pollution (levels of PM2.5, PM10, NO2 and O3) and proportions of certain population categories. The identification of clusters was done for each country separately, considering each country’s urban areas. This allowed the identification of air pollution hotspots – clusters of locations with higher air pollution levels than other urban locations – and their associated demographic characteristics.

The European Commission’s Geographic Information System (GISCO), provided by Eurostat, released a Europe-wide set of demographic data in 2024 (Eurostat, 2024a). These data are based on national censuses carried out in 2021 and are spatially organised into a 1x1km grid. With information on people’s sex, age, employment status, mobility and place of birth, the dataset offers the opportunity to study, at a much finer scale, possible environmental inequalities for different population groups. This analysis focuses on a series of population categories:

  • Age categories:
    • Children: Proportion of the population under 15 years old within each 1km grid square;
    • 15-64 years old: Proportion of the population above 15 and under 65 years old within each 1km grid square;
    • Elderly: Proportion of the population above 65 years old within each 1km grid square;
  • Place of birth categories:
    • Native-born: Proportion of the population born in the reporting country;
    • EU-born: Proportion of the population born outside the reporting country but in an EU country;
    • Non-EU born: Proportion of the population born outside the reporting country and in a non-EU country;
  • Employment category: Employed: Proportion of the population employed.

These categories were selected for their direct relevance to studying inequalities in exposure to air pollution as informed by research studies on the topic.

The census data were analysed in combination with EEA Europe-wide air quality data, produced at the same spatial resolution (EEA, 2025a).

Definition of urban locations

To identify urban locations, the six refined degrees of urbanisation produced by DG REGIO and the Joint Research Centre were used. The degrees are defined according to grid population and population density (Eurostat, 2025a) and the definition of an urban cluster i.e. a cluster of contiguous grid cells of 1 km2 with a population density of at least 300 inhabitants per km² and a minimum population of at least 5,000 inhabitants.

Only urban degrees were used in the study, namely:

  1. Urban centres, defined as urban clusters of contiguous grid cells of 1km2 with a population density of at least 1,500 inhabitants per km² and collectively a minimum population of 50,000 inhabitants;
  2. Towns, defined as:
    1. A dense urban cluster of contiguous grid cells with a population density of at least 1,500 residents per km², but with a total population of between 5,000 and 50,000, or;
    2. A semi-dense urban cluster of contiguous grid cells with a density of at least 300 inhabitants per km2, with a total population of at least 5,000 inhabitants, and located at least 3km away from a dense urban cluster or an urban centre (distance measured between the edges of the clusters);
  3. Suburbs, defined as all the other grid cells that belong to an urban cluster, but which are not part of an urban centre, dense urban cluster or a semi-dense urban cluster.

The remaining degrees of urbanisation not used in this study are rural degrees described as villages, dispersed rural areas and mostly uninhabited areas.

Cluster analysis and definition of air quality and demographic hotspots

The machine learning technique k-mean clustering was used to identify, for each country, data points of similar characteristics in terms of air pollution levels and population groups as outlined above (MacQueen, 1967). The number of clusters was optimised for the highest ratio of between-cluster variance to within-cluster variance (also known as pseudo F-statistic).

For each country, air pollution hotspots with an overrepresentation of one of the population groups were identified among the national clusters and defined as air quality and demographic hotspots if they fulfilled the two following conditions:

  • They have higher averages (mean) of both air pollutant concentrations and proportions of population groups than the median of all national clusters’ averages;
  • The air pollutant concentrations and proportions of population groups are at least 5% higher than in the other clusters.

The WHO guideline levels were used in the assessment of the average yearly levels within the identified clusters, i.e. 5µg/m3 (PM2.5), 15µg/m3 (PM10) and 10µg/m3 (NO2). For O3, the concentrations from the EEA interpolated maps were comparable to the EU target threshold of 120µg/m³. The WHO guidelines refer to long-term exposure (annual). The EU target threshold is based on a one-year average.

Brazil, N., 2022, ‘Environmental inequality in the neighborhood networks of urban mobility in US cities’, Proceedings of the National Academy of Sciences 119(17), p. e2117776119 (DOI: 10.1073/pnas.2117776119).

Diekmann, A., et al., 2023, ‘Environmental Inequality in Four European Cities: A Study Combining Household Survey and Geo-Referenced Data’, European Sociological Review 39(1), pp. 44-66 (DOI: 10.1093/esr/jcac028).

EEA, 2018, Unequal exposure and unequal impacts, Publications Office of the European Union, LU (https://www.eea.europa.eu/publications/unequal-exposure-and-unequal-impacts/) accessed 6 June 2023.

EEA, 2023, ‘Air pollution and children’s health’ (https://www.eea.europa.eu/en/analysis/publications/air-pollution-and-childrens-health) accessed 7 October 2025.

EEA, 2025a, European air quality data (interpolated data)- Series, eea_aq-interpolated_s (https://sdi.eea.europa.eu/catalogue/srv/eng/catalog.search#/metadata/82700fbd-2953-467b-be0a-78a520c3a7ef) accessed 21 July 2025.

EEA, 2025b, ‘Exceedance of air quality standards in Europe’, European Environment Agency (https://www.eea.europa.eu/en/analysis/indicators/exceedance-of-air-quality-standards) accessed 15 April 2025.

EU, 2024, Directive (EU) 2024/2881 of the European Parliament and of the Council of 23 October 2024 on ambient air quality and cleaner air for Europe (recast) (OJ L, 2024/2881).

Eurostat, 2021, ‘Urban-rural Europe - introduction’, Eurostat (https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Urban-rural_Europe_-_introduction) accessed 19 March 2025.

Eurostat, 2024a, Population grids - GISCO - Eurostat, (https://ec.europa.eu/eurostat/web/gisco/geodata/population-distribution/population-grids) accessed 21 July 2025.

Eurostat, 2024b, ‘Urban-rural Europe - economic activity in capital cities and metropolitan regions’, Eurostat (https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Urban-rural_Europe_-_economic_activity_in_capital_cities_and_metropolitan_regions) accessed 19 March 2025.

Eurostat, 2025, Clusters - GISCO - Eurostat, (https://ec.europa.eu/eurostat/web/gisco/geodata/population-distribution/clusters) accessed 29 April 2025.

Eurostat, 2025b, Persons at risk of poverty or social exclusion by group of country of birth (population aged 18 and over), (https://ec.europa.eu/eurostat/databrowser/product/page/ILC_PEPS06N) accessed 29 April 2026, Eurostat.

Fairburn, J., et al., 2019, ‘Social Inequalities in Exposure to Ambient Air Pollution: A Systematic Review in the WHO European Region’, International Journal of Environmental Research and Public Health 16(17), p. 3127 (DOI: 10.3390/ijerph16173127).

Fecht, D., et al., 2015, ‘Associations between air pollution and socioeconomic characteristics, ethnicity and age profile of neighbourhoods in England and the Netherlands’, Environmental Pollution 198, pp. 201-210 (DOI: 10.1016/j.envpol.2014.12.014).

Fong, K. C., et al., 2022, ‘The Intersection of Immigrant and Environmental Health: A Scoping Review of Observational Population Exposure and Epidemiologic Studies’, Environmental Health Perspectives 130(9), p. 096001 (DOI: 10.1289/EHP9855).

MacQueen, J., 1967, ‘Some methods for classification and analysis of multivariate observations’, Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Volume 1: Statistics 5.1, pp. 281-298.

OECD and EC, 2023, Indicators of Immigrant Integration 2023: Settling In, OECD Publishing, Paris (https://www.oecd.org/en/publications/indicators-of-immigrant-integration-2023_1d5020a6-en/full-report/component-8.html) accessed 4 December 2024.

OECD, 2024, ‘Environmental Justice: Context, Challenges and National Approaches’, OECD (https://www.oecd.org/en/publications/environmental-justice_57616eb4-en.html) accessed 7 April 2026.