A new study reveals that the COVID-19 induced lockdown did bring the air pollution levels down, however, not as significantly as earlier predicted.
The researchers developed new corrections for the impact of weather and seasonal trends. They analysed reduced NO2 emissions from winter to summer.
The authors of the study further evaluated changes in ambient NO2, O3, and fine particle (PM2.5) concentrations arising from lockdown emission changes in 11 global cities. These include Beijing, Wuhan, Milan, Rome, Madrid, London, Paris, Berlin, New York, Los Angeles, and Delhi.
Led by experts at the University of Birmingham, the international team of scientists discovered that the beneficial reductions in NO2 due to the lockdowns were smaller than expected, after removing the effects of weather. In parallel, the lockdowns caused (weather-corrected) concentrations of ozone in cities to increase.
Also read: Towards a fossil-free India
NO2 is a key air pollutant from traffic emissions, associated with respiratory problems, while ozone is also harmful to health, and damages crops.
The team revealed that concentrations of PM2.5, which can worsen medical conditions such as asthma and heart disease, decreased in all cities studied except London and Paris.
Lead-author Zongbo Shi, Professor of Atmospheric Biogeochemistry at the University of Birmingham, commented: “Rapid, unprecedented reduction in the economic activity provided a unique opportunity to study the impact of interventions on air quality. Emission changes associated with the early lockdown restrictions led to abrupt changes in air pollutant levels but their impacts on air quality were more complex than we thought and smaller than we expected.”
Also read: ‘Pests, climate change and yield key worries’
He added: “Weather changes can mask changes in emissions on air quality. Importantly, our study has provided a new framework for assessing air pollution interventions, by separating the effects of weather and season from the effects of emission changes.”
For the study, the researchers used machine learning to strip out weather impacts and seasonal trends before analysing the data - site-specific hourly concentrations of key pollutants from December 2015 to May 2020.
The findings of the study were published in the journal Science Advances.