A hotter world could have cleaner air, but deadlier heatwaves

By Pooja Toshniwal PahariaReviewed by Susha Cheriyedath, M.Sc.Nov 2 2026

What would happen if the extreme heat of 2018 and 2019 occurred in a world nearly 3°C warmer? Researchers examine how shifting heatwaves, ozone and fine particle pollution could alter health risks across continents.

Study: Global health impacts of the 2018–2019 heatwaves and air pollution events under warmer climates. Image Credit: 418 House / Shutterstock

In a recent study published in the journal npj Clean Air, researchers explored how extreme temperatures affect global air quality and the health impacts of heatwaves (HWs), ozone (O₃), and fine particulate matter (PM_2.5) under warmer climate conditions.

Fewer heatwave events in some regions could reflect a shift toward longer, more continuous episodes. Warmer conditions may also reduce O₃ concentrations across many regions and PM_2.5 levels in certain areas, particularly under a scenario in which global average temperatures are 2.75°C above pre-industrial levels.

Lower pollutant levels may not necessarily reduce the overall health burden. If heatwaves become longer and more severe, deaths linked to extreme heat could rise more than the separately estimated reductions in pollution-related deaths.

Both outdoor air pollution and extreme heat already pose serious health risks worldwide. As temperatures rise, changes in weather patterns and natural environments are also contributing to more frequent heatwaves and wildfires.

Higher temperatures and shifting weather conditions can alter air pollutant concentrations, partly through changes in biogenic volatile organic compound (VOC) emissions and the temperature-dependent partitioning of ammonium nitrate. Regional differences in these effects also need to be considered when planning responses to a warmer climate.

About the study

In the present study, researchers assessed changes in O₃, PM_2.5, HWs, and their associated mortality burdens under global warming scenarios of 2°C and 2.75°C, relative to pre-industrial conditions, with the present-day climate representing approximately 1.1°C of warming.

The researchers worked at Forschungszentrum Jülich, the University of Wuppertal and the Max Planck Institute for Meteorology in Germany, the University of Reading in the United Kingdom, Shandong University in China, and Monash University in Australia.

The team used the unusually hot conditions recorded in 2018–2019 as a reference for modeling warmer climates. They retained the same atmospheric circulation and human-caused emission levels across the scenarios.

Population size, underlying death rates, vulnerability, adaptive capacity, and the relationships between current pollution exposure and health outcomes were also kept unchanged. These controlled comparisons were designed to isolate the effects of warming rather than forecast future deaths.

Temperature and air pollutant levels were simulated with the ECHAM/MESSy global atmospheric chemistry model (EMAC). Heatwave mortality estimates drew on the Multi-Country Multi-City (MCC) dataset covering 750 locations in 43 countries or regions, while short-term O₃ and PM_2.5 risk estimates came from published meta-analyses. The team also assessed four additional pollutants: PM₁₀, carbon monoxide (CO), sulfur dioxide (SO₂), and nitrogen dioxide (NO₂).

Overall air quality and associated health risks were evaluated using the Air Quality Index (AQI) and Health-risk-based AQI (HAQI), with pollutant thresholds adapted from the World Health Organization (WHO) 2021 air quality guidelines. The AQI reflects the highest individual pollutant index, while the HAQI accounts for the combined health risks associated with multiple pollutants.

The researchers adjusted modeled PM2.5 concentrations using a statistical correction method informed by machine-learning estimates and comparisons with Copernicus atmospheric reanalysis data.

The researchers evaluated O3 simulations against Tropospheric Ozone Assessment Report data without bias correction, and the daily modeled means approximated the eight-hour exposure measure. Heatwaves were identified using bias-corrected EMAC temperatures and an ERA5 reanalysis baseline from 1990 to 2019, originally available at 0.25° resolution. The AQI calculations followed the United States Environmental Protection Agency method, with pollutant-specific category thresholds based on the WHO interim targets.

Results

The analysis revealed large regional differences in pollution- and heat-related mortality. In 2018–2019, PM_2.5 accounted for the largest attributable death burden in Asia and Africa, averaging 100 and 57.7 deaths per million people, respectively. Heatwaves were the leading contributor in Europe, with 148.5 deaths per million, and in parts of North America and Australia.

Heatwave-attributable mortality increased across all six regions in the 2.75°C scenario. South America recorded the steepest rise, at over 95%, while the increase was about 82% in Africa, 67% in Australia, 59% in North America, 45% in Asia, and 37% in Europe. Some regions experienced fewer distinct heatwave events, but episodes lasted 5–10 days longer in parts of North Africa, the Arabian Peninsula and the southern United States.

Air pollution showed a different response. Under 2.75°C warming, ozone concentrations and associated mortality generally declined across many regions, while the 2°C scenario produced increases in several locations.

Deaths linked to PM_2.5 exposure also declined in Asia, Africa, Europe and North America, while South America showed an increase. Changes in atmospheric chemistry, ozone removal, ammonium nitrate evaporation, natural emissions and dust production may help explain these regional differences, though the contribution of individual aerosol components was not quantified.

Cleaner air did not necessarily mean safer living conditions. The air quality indices indicated that approximately 10% of the global population lived in areas meeting the study's most stringent WHO-based index category, Target III, under the modeled 2018–2019 conditions.

In China and India, many people remained exposed to pollution above the less stringent WHO-based index thresholds. The mortality estimates for heatwaves, ozone and PM_2.5 were calculated separately and cannot be added or subtracted to establish a net death toll.

Conclusion

Based on the findings, climate-related health risks varied considerably from one region to another. Heatwaves were the leading contributor to attributable mortality in Europe and Australia, while PM_2.5 pollution accounted for the largest burden in Asia and Africa.

Heatwaves and ozone exposure were both linked to deaths in North America. Further warming could improve air quality in some regions, but benefits would vary, and substantial pollution exposure would persist.

Responses will need to differ by region, addressing heat exposure alongside pollution. The estimates also leave some factors unaccounted for, including uncertainties in climate models and future changes in people's vulnerability and ability to adapt. The reported confidence intervals reflect uncertainty in exposure-related mortality risks, not the entire modeling process.

Further research should examine how warming affects individual aerosol components and whether heat exposure increases susceptibility to air pollution.