Geographic sources of ozone air pollution and mortality burden in Europe

Tropospheric ozone (O3), produced from nitrogen oxides (NOx) and volatile organic compounds (VOCs) under sunlight, peaks in the warm season and is linked with adverse respiratory outcomes and mortality. In Europe, >95% of the population is exposed to O3 exceeding WHO air-quality guidelines (daily maximum 8-hour average of 100 µg m−3), with impacts exacerbated by warming temperatures. Because O3 and its precursors are transported over long distances, local concentrations are heavily influenced by remote sources. A continent-wide quantification of O3 pollution by geographic source and its health burden has been lacking. This study aims to quantify national versus imported (transboundary and hemispheric) contributions to O3 and their associated mortality across Europe, focusing on warm seasons (May–September) of 2015–2017.

Prior work has documented the health hazards of O3 and its short-term association with all-cause mortality across many countries. EEA reports indicate widespread exceedance of WHO O3 guidelines in Europe. Studies have shown substantial contributions of imported O3 to surface levels in southwestern Europe (46–68%), underscoring transboundary transport. While several assessments estimated O3-attributable mortality in various contexts, a comprehensive European, source-specific attribution of O3 and related mortality has been missing. Intercontinental transport of O3 and precursors has been established, motivating the need for coordinated, cross-border mitigation.

Design: Europe-wide health-impact assessment integrating air-quality source apportionment and exposure–response modeling. Scope: 813 regions in 35 European countries; warm seasons (May–September) 2015–2017. Health data: Weekly all-cause mortality (weeks 18–39) and annual populations from Eurostat (2015–2017), covering ~530 million people and 6,291,008 death counts. Air-quality modeling: CALIOPE system at ~18 km resolution integrating WRF-ARW (v3.6), HERMESv3 anthropogenic emissions (CAMS-REG-AP v4.2), MEGAN v2.0.4 biogenic emissions, and CMAQ v5.0.2 chemistry. Boundary conditions from ERA-Interim and CAMS global analysis. Domain includes Europe and surrounding areas. Source apportionment: CMAQ-integrated tagging approach attributing O3 formation to NOx or VOC-limited regimes using H2O2:HNO3 ratio threshold (0.35). Tagged contributions include: (1) national, (2) other 34 European countries, (3) other countries inside domain (non-EU-35), (4) ocean and sea (maritime), and (5) outside-domain (hemispheric) sources. Both O3 and its precursors (NOx, VOCs) traced through transport, chemistry, and deposition. Health-impact assessment: Exposure–response from a multicountry study (β=0.00018 per 1 µg m−3 increase in daily max 8-h O3; 95% CI 0.00012–0.00024). For each grid cell and day, computed attributable fraction (AF) = 1−exp(−β×O3). Weekly AF averaged per grid and population-weighted to regions using GHSL 2015 population grid (1-km). Weekly attributable numbers (AN) = observed deaths N(r,w) × AF(r,w). Summed across weeks to obtain seasonal AN; mortality rates per million derived by dividing by population. 95% empirical CIs from β uncertainty. Primary analysis used full exposure range; sensitivity analysis applied a 70 µg m−3 threshold (excluding lower days and centering at 70 µg m−3). Model evaluation: Compared modeled daily max 8-h O3 with EEA rural background stations (<1,000 m). Performance: Pearson r=0.66±0.11; normalized mean bias 5.49±9.25%; nRMSE 20.46±5.7. Met FAIRMODE MQI ≤1 at 100% of considered stations, satisfying model quality objectives.

- Average O3 concentration across countries (2015–2017 warm seasons): 101.9 µg m−3 (range: 76.7 in Finland to 130.1 in Malta). - Total O3-attributable deaths (warm seasons, 2015–2017): 114,447 (95% eCI 76,539–152,108), corresponding to 72.0 (95% eCI 48.1–95.6) annual deaths per 1 million inhabitants. - Spatial patterns: Higher O3 and mortality rates in southern/southeastern Europe; highest national burdens in populous countries (Germany, Italy, France, UK, Spain, Poland); highest rates in southeastern countries (Bulgaria, Serbia, Croatia, Hungary, Greece, Romania). - Source apportionment of mortality: Imported (non-national) O3 accounted for 88.3% of all O3-attributable deaths (range 83.0–100%). Hemispheric sources (outside domain) contributed 56.7% (range 42.5–87.2%). Contributions from the 34 other European countries averaged 20.9% (range 5.1–40.0%). Maritime sources averaged 7.2% (range 0–24.1%), notable in Malta (24.1%) and Cyprus (14.0%). - Country-to-country impacts: Major exporters of transboundary O3 mortality included France (4,003 deaths caused abroad, warm seasons 2015–2017) and Germany (3,260). Notable bilateral effects: France → Luxembourg (32.3%), Switzerland (29.3%), Belgium (24.4%), Liechtenstein (20.2%), Spain (16.8%), Germany (16.3%); Germany → Luxembourg (24.2%), Czechia (23.3%), Netherlands (21.5%), Denmark (20.3%), Austria (19.9%), Belgium (17.8%), Poland (17.2%). - National vs imported: Southwestern countries showed higher shares of national O3-attributable deaths and lower imported:exported ratios: Spain (53.7% national), France (47.1%), Portugal (46.2%). - Sensitivity analysis (70 µg m−3 threshold): Attributable deaths reduced to 36,523 (95% CI 24,382–48,633), equivalent to 23.0 (95% CI 15.3–30.6) annual deaths per 1 million. The country apportionment of sources changed minimally.

Findings show that only 11.7% of O3-attributable deaths stem from national sources, with the majority due to hemispheric transport (56.7%) and other European countries (20.9%). This underscores that national or regional policies alone cannot achieve WHO air-quality targets; coordinated EU-wide and global strategies are necessary. Maritime contributions are substantial for some Mediterranean countries, suggesting benefits from implementing nitrogen emission control areas in the Mediterranean akin to the North and Baltic Seas. Despite large non-national shares, local actions remain crucial during high-O3 episodes to reduce exceedances and to curb exported pollution. Differences with EEA estimates arise from methodological choices (season, exposure range, thresholds, baseline mortality handling), indicating potential underestimation when assuming safe thresholds below 70 µg m−3. Climate warming will likely increase O3 formation and biogenic VOC emissions, potentially offsetting gains from anthropogenic emission reductions, reinforcing the need to integrate climate policy with air-quality management. The study also notes the role of international trade in embedding emissions and health impacts across borders, calling for broader accounting frameworks.

A comprehensive European source-apportionment of O3 and associated mortality reveals that most of the health burden is imported, predominantly from hemispheric sources. Achieving WHO guidelines and preventing O3-related mortality will require coordinated national, EU-level, and global strategies. Quantitative attribution linking local health impacts to geographic O3 sources is essential for targeted mitigation. Future work should refine sector-specific contributions (energy, industry, transport, residential, agriculture) by country, assess economic costs, incorporate years of life lost, and further integrate climate change considerations into long-term policies.

- Only acute (short-term) mortality effects were considered; potential chronic effects remain uncertain and were not included. - A fixed exposure–response function was applied uniformly across countries due to lack of region-specific estimates, which may bias results given population heterogeneity. - Years of life lost (YLL) were not estimated due to lack of age-specific mortality data and limited European evidence on short-term O3–YLL associations. - Economic costs of O3-attributable deaths were not quantified. - Uncertainty from air pollution models was not propagated into health impact estimates. - Source-apportionment (tagging) methods carry inherent uncertainties despite advantages over brute-force sensitivity approaches.