Which countries have the highest PM2.5 exposure?
Across the 28 European countries covered (EU-27 plus the United Kingdom), the highest population-weighted PM2.5 annual mean in 2024 was in Poland at 16 µg/m³, followed by Bulgaria (15.6 µg/m³) and Slovakia (15.2 µg/m³). The lowest exposure was in Ireland (1 µg/m³), followed by Estonia (4.6 µg/m³) and Sweden (5.1 µg/m³). The cross-country arithmetic mean was 10.37 µg/m³ - well above the WHO 2021 guideline. Cumulative premature deaths attributable to PM2.5 across the 28 countries totalled approximately 238,000.
The geographic pattern is consistent with the long-established "north-clean, south-and-east-polluted" gradient: Nordic and Atlantic-fringe countries dominate the low end (low population density, dominant westerly winds, near-elimination of coal-based residential heating); the highest end is dominated by countries where residential solid-fuel heating (Bulgaria, Poland, Romania) or basin topography (the Po Valley in Italy) traps emissions.
Ranking
| Rank | Country | PM2.5 (µg/m³) | × WHO 2021 | NO2 (µg/m³) | Stations | Deaths (PM2.5) |
|---|---|---|---|---|---|---|
| 1 | Poland | 16 | 3.2× | 22.8 | 2 | 34,155 |
| 2 | Bulgaria | 15.6 | 3.1× | 21.6 | 1 | 8,297 |
| 3 | Slovakia | 15.2 | 3.0× | 19.6 | 1 | 3,623 |
| 4 | Cyprus | 14.4 | 2.9× | 17.8 | 1 | 669 |
| 5 | Croatia | 14.4 | 2.9× | 19.2 | 1 | 3,690 |
| 6 | Slovenia | 14 | 2.8× | 17.8 | 1 | 1,277 |
| 7 | Romania | 14 | 2.8× | 22.4 | 1 | 17,903 |
| 8 | Italy | 13.9 | 2.8× | 23.8 | 3 | 51,000 |
| 9 | Hungary | 13.9 | 2.8× | 22.6 | 1 | 8,503 |
| 10 | Greece | 13.3 | 2.7× | 22.1 | 1 | 10,409 |
| 11 | Czech Republic | 12.2 | 2.4× | 20.5 | 1 | 6,826 |
| 12 | Spain | 11.1 | 2.2× | 18.9 | 2 | 18,315 |
| 13 | Malta | 10.9 | 2.2× | 18.4 | 1 | 169 |
| 14 | Latvia | 9.9 | 2.0× | 13.6 | 1 | 804 |
| 15 | Austria | 9.9 | 2.0× | 18.9 | 1 | 3,332 |
| 16 | United Kingdom | 9.8 | 2.0× | 21.4 | 3 | - |
| 17 | Belgium | 9.8 | 2.0× | 22.3 | 1 | 4,105 |
| 18 | Germany | 9.5 | 1.9× | 19.4 | 3 | 32,339 |
| 19 | France | 9.3 | 1.9× | 20.6 | 3 | 20,286 |
| 20 | Netherlands | 8.7 | 1.7× | 21.7 | 2 | 5,171 |
| 21 | Portugal | 7.5 | 1.5× | 18.6 | 1 | 3,659 |
| 22 | Lithuania | 7.3 | 1.5× | 14.2 | 1 | 1,435 |
| 23 | Denmark | 7 | 1.4× | 16.8 | 1 | 1,183 |
| 24 | Luxembourg | 6.7 | 1.3× | 19.8 | 1 | 74 |
| 25 | Finland | 5.3 | 1.1× | 13.5 | 1 | 70 |
| 26 | Sweden | 5.1 | 1.0× | 12.4 | 1 | 478 |
| 27 | Estonia | 4.6 | 0.9× | 11.2 | 1 | 89 |
| 28 | Ireland | 1 | 0.2× | 14.8 | 1 | 517 |
Methodology
The ranking above was computed by querying the countries table for
pm25_annual_avg, which holds the EEA Average Exposure Indicator for PM2.5
(Air Quality Status Report 2025 Table 3, three-year urban-background mean for 2021-2023),
transcribed in August 2026. Sister columns for context include editorial-composite
no2_annual_avg where present, station_count, and
premature_deaths_pm25_2023 from the EEA Burden of Disease (2022) series.
Population weighting for the AEI is computed by the EEA (not by this portal) using a 1 km gridded population overlay (GHS-POP) and related reanalysis inputs. We restate the published AEI table; we do not download CAMS grids or Dataflow E1a hourly series ourselves.
See our methodology page for what is wired today, what remains editorial composite, and known limitations.
Limitations
Country-level ranking masks within-country variation. The PM2.5 annual mean in central Madrid differs substantially from that in coastal Galicia even though both contribute to the Spain figure. The EEA population-weighted metric we use is the appropriate national-level summary, but for individual exposure assessment the relevant figure is the nearest representative station to the residential address, see country-detail pages for station-level data. The ranking also reflects a single year (2024); inter-annual variability driven by wildfire seasons (notably the 2017, 2021, 2022, 2023 Mediterranean seasons), Saharan dust intrusions, and weather patterns can shift a country by 1–3 µg/m³ year on year.
Source
Underlying data: European Environment Agency, Air Quality Status Report 2025 AEI Table 3 (2021-2023), transcribed August 2026. Mortality estimates: EEA Burden of Disease / HRAPIE methodology, see EEA in-depth air pollution topic page.
Figures on PlainAirQuality are rendered from this site's database rather than typed into the page. Country values restate the EEA's published series and are held in our dataset pending an automated re-import; station-level values are modelled from those national means, not measured at the station. Figures on this page come from the EEA's published series (population-weighted country exposure and Burden of Disease mortality); anything shown per station is a modelled estimate derived from those national means, never an independent reading. See our editorial standards & corrections policy, the methodology behind these numbers, or report a data error.