In the early weeks of the coronavirus pandemic, there were signs that virus-related restrictions such as lockdowns and social distancing measures, meant to slow the spread of the virus, had a silver lining: air quality improvements.
Air quality initially did improve in many cities of Latin America and the Caribbean, as measured both by air quality monitoring stations and satellite data. However, over time it has become clear that the improvements are not as large and persistent as hoped for and that they vary widely across locations and type of pollutant. This shows the importance of local weather conditions, as well as emission sources other than those directly related to daily human activities, such as driving and industrial activities, for air quality levels.
To analyze the magnitude and persistence of the effect of coronavirus-related restrictions on air quality in the region, we looked at several pollutants in three Latin American cities: Bogotá, Mexico City, and Santiago. The analysis suggests that substantial reductions observed in mobility alone are not going to bring us very far down the path toward good air quality all year long. Other substantial systemic changes would be required to achieve that goal.
We started by looking at nitrogen dioxide (NO2) concentrations. NO2 is emitted through the burning of fuel by cars, trucks and buses, power plants, and off-road equipment. Given these sources of NO2, one would expect mobility restrictions to impact its concentrations. We compared measures of NO2 using data from ground station monitoring networks and satellite data collected in the IDB and IDB Invest Coronavirus Impact Dashboard. These two sources of data provide different but complementary information. Ground monitoring stations provide accurate measurements of the volume of pollutant gases/particles in the air and the level of exposure for people near the stations, but the measurements rely on having a dense geographic network of stations. Satellite data are available across the globe and allow for greater geographic coverage, but satellites cannot take into account different weather conditions happening near the surface such as wind, rain or clouds, which can lead to inaccurate assessment of air quality observations for any given location.
We used the average measures between 2018 and 2019 for monitoring stations and the measure in 2019 for satellite data as a benchmark to ensure that seasonality, environmental conditions, and meteorological variability that can influence the concentration of pollutants between different months of the year do not influence the analysis. We began the analysis in the first week of January 2020 to confirm that 2020 air pollution levels were similar to those observed in previous years and prior to the coronavirus pandemic.
Using both ground monitoring station and satellite data, Figures 1 and 2 show consistent trends: coronavirus-related restrictions initially contributed to reductions in NO2 concentrations, but improvements are fading. As Figure 1 shows, immediately following the declaration of a pandemic by the World Health Organization (WHO) on March 11, 2020, and the introduction of lockdowns (to different degrees in different cities), the 2020 NO2 levels declined compared to those of 2018—2019 for all cities. The gap is particularly evident in the monitoring station data, but it is also detected with satellite data. However, over time, the NO2 reductions become smaller and reach levels of concentration similar to those seen in previous years.
Figure 1. Ground Monitoring Station and Satellite Data for Nitrogen Dioxide Concentration Levels, 2018–2019 versus 2020
Note: NO2: nitrogen dioxide. “WHO pandemic” marks the date when the World Health Organization declared the coronavirus a pandemic.
To have a better idea of the magnitude of the effect, Figure 2 reports percentage changes compared to the baseline years (2018–2019). For example, in Bogotá, for most of the time period after the declaration of the pandemic by the WHO in March 2020, NO2 levels were approximately 40 percent lower than levels observed for the same dates in 2018–2019. By the time Colombia declared a national lockdown on March 25, concentration levels were far below those in the same week of prior years. But by the final weeks of May, concentration levels began moving toward levels similar to previous years and the earlier gains had been nearly erased. As opposed, the changes observed in Santiago have remained below those observed in 2018-2019.
Figure 2. Change in Nitrogen Dioxide Concentration Levels in 2020 versus 2018–2019
Note: “WHO pandemic” marks the date when the World Health Organization declared the coronavirus a pandemic.
To see the full picture of air quality changes over this period, we also analyzed other common air pollutants using data from the ground station monitoring network. In general, the levels of most common air pollutants have decreased relative to levels in 2018–2019, but in some cases the decrease has been small and not persistent. For example, in Bogota, after dipping to levels around 70 percent below those of 2018–2019 in midApril 2020, PM10 and PM2.5 rebounded to levels comparable to prior years (Figure 3). The trend is similar in Mexico City and Santiago, although the reduction was smaller at the beginning (around 20 percent). Particulate matter is one of the pollutants with the strongest evidence of negative health effects. Although one source of PM is combustion engines (both diesel and petrol), many sources are not necessarily affected by the decrease in mobility caused by the pandemic. These include solid fuel (coal, lignite, heavy oil and biomass) combustion for energy production for household and industrial consumption.
Figure 3. Change in Particulate Matter Concentration Levels in 2020 relative to 2018–2019
Note: PM: particulate matter (the numbers 2.5 and 10 refer to the diameter of the particulate matter measured in micrometers). “WHO pandemic” marks the date when the World Health Organization declared the coronavirus a pandemic.
In terms of levels of other pollutants, Bogotá has seen substantial declines in sulfur dioxide (SO2) and carbon monoxide (CO) since mid-March, but Santiago has actually seen a substantial increase in SO2 levels, and its CO levels are similar to prior years (Figure 4). Sulfur dioxide is primarily produced from the burning of fossil fuels (coal and oil) and the smelting of mineral ores that contain sulfur. Carbon monoxide is primarily emitted by motor vehicles and machinery that burns fossil fuels. Levels of SO2 and CO below those of comparable dates in 2018–2019 could indicate a slowdown in economic activity, particularly in industrial and manufacturing processes. The differences across the cities could be related to different levels of stringency and enforcement of the coronavirus-related restrictions.
In terms of ozone levels (O3), the data show levels in 2020 that are usually above those observed for prior years—by more than 100 percent during some weeks in Bogotá and Santiago. Ozone creation is associated with volatile organic compounds that are usually contained in a variety of cleaning products, disinfectants, aerosols, air fresheners, paints, and solvents. The analysis supports the hypothesis that people might be doing more cleaning and home repairs due to the pandemic and the lockdowns.
Figure 4. Change in Other Pollutant Concentration Levels in 2020 relative to 2018–2019
Note: NO: nitrogen monoxide ; SO2: sulfur dioxide; O3: ozone; CO: carbon monoxide.
In sum, while the residents of Bogotá, Santiago, and Mexico City have experienced some improvements in air quality over the past few months, the evidence suggests that we should not expect these improvements to last. Further analysis will be needed to evaluate the effects of different types of lockdowns in cities and the gradual return to pre-lockdown conditions. Despite this, the unprecedented circumstances created by COVID-19 provide scientists and policymakers with insights into how reductions in certain air pollutants could be achieved through major improvements in sustainable mobility, for example, through increased adoption of mass transit and lower-emission vehicles. Also, changes in businesses and households’ practices will be key to contribute to improvements in air quality. Both the public and private sectors will have to work together to transform that fleeting “silver lining” of coronavirus-related restrictions into a sustainable outcome over the long term in Latin America and the Caribbean.