I know it seems like forever ago, but think back to the early days of covid-19 when lockdowns first halted travel and industry in huge swaths of the world. It was devastating to the economy and was a symptom of a huge public health problem, but reports showed there was at least one good thing to come out of it: As levels of activity in many sectors fell, so did global air pollution. Well, a new study shows that we may have overestimated those reductions in air pollution.
The new study, published in Science Advances on Wednesday, examines the changes in atmospheric concentrations of toxic air pollutants, including fine particulate matter (PM2.5), nitrogen dioxide, and ozone, in 11 cities around the world that saw severe pandemic restrictions. The data was quite granular, including hourly pollution levels from December 2015 through May 2020 obtained from local and national environmental agencies or accredited third parties. And the three pollutants researchers looked at are a huge drag on public health.
“Globally air pollution, primary PM2.5, NO2 and O3 are associated with about 7 million premature deaths,” Zongbo Shi, professor of Atmospheric Biogeochemistry at the University of Birmingham and the study’s lead author, wrote in an email.
The authors found that recent studies on reductions in air pollution amid covid-19 did not always sufficiently isolate the effects of the lockdowns themselves from weather changes. For instance, cooler temperatures can slow the reactions that cause pollutants to form, and both rain and strong winds can cause pollutants to disperse more quickly. The new study attempts to remedy this by creating a new machine learning model to isolate and remove the impacts of weather on pollution levels.
Since some previous analyses merely compared pollution levels over a one-year period—for instance, comparing March 2019 to March 2020—the researchers also feared those could have failed to account for changes in pollution over a longer time. For the new analysis, they used a statistical model to determine what concentrations would have been, based on those pollution reductions without lockdowns. They then compared those with the real-life data to determine what difference citywide covid-19 precautions made.
“By looking at the difference between ‘business as usual’ and deweathered concentrations during the lockdown, we calculated the real changes in air pollutant concentration attributed to lockdown,” Shi said.
The analysis suggest that nitrogen dioxide concentrations did sharply decrease in all 11 cities during coronavirus restrictions. But while previous studies showed declines on the order of 60% on average, the new study found that only a 30% decrease was attributable to the lockdowns on average.
The findings on ozone and PM2.5 diverged even further from other findings. When controlling for other factors, the study found all 11 cities saw increases in levels of ozone. That could offset some of the health benefits of the nitrogen dioxide dip, according to the study.
When it came to PM2.5, Wuhan and Delhi—the most polluted of the cities examined—both saw significant declines. However, the study found “no clear changes” in other cities. In fact, when the authors isolated the impacts of covid-19, two cities—London and Paris—saw slightly higher levels of PM2.5 on average.
These discrepancies may be due to how different pollution sources were impacted by coronavirus restrictions. Much of the world’s nitrogen dioxide production comes from cars, trucks, buses, and planes, all of which spent less time on the roads during lockdowns. But other key sources of ozone and PM2.5 pollution, like power plants, industrial boilers, and refineries, didn’t see as steep reductions in activity. The study’s findings show that we need comprehensive policies to lower various forms of air pollution.
“A systematic approach, considering all major air pollutants is needed to deliver the greatest health benefits,” Shi said.
This has implications for how to align climate action with pollution-tackling policies in the future.
“Aggressive actions to reduce carbon emissions, including the phasing out of diesel and petrol vehicles, will deliver immediate reduction in NO2 concentrations,” Shi said. “But air quality improvement is likely to be more challenging than we believed.”