A preliminary analysis of satellite data on air pollution shows a decrease in the concentrations of pollutants over South Africa during the national lockdown caused by the outbreak of the Coronavirus in the country.

This is according to a team of researchers at the Council for Scientific and Industrial Research (CSIR) who are working in collaboration with Dr Eloise Marais from the University of Leicester, in the United Kingdom, to understand the impact of the lockdown on air quality in South Africa. These air quality experts say some parts of the country have been showing a decrease in the concentration of pollutants during this period.

Using satellites data, the team can explore the impacts of the lockdown on the concentrations of the pollutants in the atmosphere. Using ground-based measurement data, the team can explore the impacts of the lockdown on ground-level concentrations that have a direct impact on health, as they are measured at the level at which people breathe.

The preliminary analysis of the satellite data shows that there has been a decrease in the column of tropospheric nitrogen dioxide (NO₂) and sulfur dioxide (SO₂). This data was obtained from the Earth Observing TROPOMI instrument on the Sentinel-5P satellite (see attached Figure 1). Over the Highveld, NO₂ concentrations decreased by 23% after the lockdown (27 March to 20 April 2020) compared to before (10-26 March 2020), and SO₂ concentrations are 47% lower. The SO₂ data only extends to 17 April 2020.

Spatially, there is a larger decrease in NO₂ in and around Gauteng, as would be expected from a decrease in vehicle use. There is also a substantial decline in NO₂ over the industrialised Highveld region. Preliminary results from ground-based monitoring stations in the Highveld domain are less conclusive, as these show large variability in changes of NO₂ before and during the lockdown.

“Assessing the impacts of the lockdown on ground-level concentrations is complex. However, it is critical in order to understand how pollution affects people’s health. One aspect that we have to consider is that ground-level pollution concentrations generally increase over this period in the run-up to winter. In order to account for that, we are looking at comparing measurements from January 2020 until now to average measurements from previous years. An example of this analysis is highlighted by our preliminary look at the Kliprivier monitoring station in Gauteng (see attached Figure 2). When comparing concentrations before lockdown to those seen in previous years, the concentration levels were similar. However, when comparing the NO₂ concentrations from during lockdown to previous years, we saw a decrease in the 24-hour average NO₂ concentrations. We did not see such a decrease in the SO₂ 24-hour average concentrations. This could be because the monitoring station is impacted by emissions from traffic, that would affect NO₂ concentrations more than SO₂, but it is not strongly impacted by coal-fired power plants that are large sources of SO₂. However, analysis is continuing to quantify reasons for these changes”, says CSIR principal researcher, Prof. Rebecca Garland.

South Africa’s large sources of nitrogen oxides (NO_X; NO_X = NO + NO₂) emissions are vehicles, industrial activities and energy generation that mostly comes from coal-fired power plants. There are also natural sources from lightning, bacteria in soils, as well as semi-natural sources, like biomass burning. These natural sources tend to be seasonal with the burning season in South Africa occurring in late winter through spring. The largest sources of sulfur emissions in South Africa are coal-fired power plants and industrial activity that uses or processes materials containing sulfur, such as coal.

Prof. Garland adds that there are numerous aspects that could affect the pollutant concentrations. “It would be premature to attribute this change solely to the decreases in emissions from the lockdown. Confirmation with additional surface observations and a model are required. This work is on-going.” Quantifying the change in air quality is not trivial, as it will vary in space and time, and will be affected by the fact that we would normally expect pollution levels to be increasing slowly towards colder months. “Attributing changes in air pollution to specific actions is not simple, as it can be influenced by many factors, including a change in temperature, rainfall, wind speed and direction, as well as changes in close and distant emission sources,” says Garland.

Dr Marais, Associate Professor in Earth Observations at the University of Leicester, says the decreases in emissions of pollutants from lockdowns and travel restrictions leading to changes in air quality were documented extensively in countries like China, Italy, India, Spain and the UK, which have been in lockdown mode longer than South Africa. Many of these reports used data from the space-based TROPOMI instrument. It is not surprising then that South Africa is starting to see the same effect of the lockdown on air quality.

“This change will likely be temporary, as many of the emissions reductions due to the lockdown will reverse when we go back to our normal day-to-day activities that make large contributions to air pollution. This period does provide us with a unique opportunity to learn about the influence of different sources on air quality, whether natural or anthropogenic. This information is crucial for developing effective air quality policies,” Says Marais.

Other team members are CSIR air quality experts, Ms Seneca Naidoo, Mr Mogesh Naidoo and Mrs Juanette John.

NB: Please see charts below and click here for the Report.

ENDS

Figure 1: Concentrations of the air pollutant NO₂ (top) and SO₂ (bottom) before and after the lockdown due to COVID-19. Data are median values from the TROPOMI NO₂ and SO₂ instrument on the Sentinel-5P satellite for the whole country (top of each figure) and the Highveld (bottom on each figure) before (left) and after (right) the lockdown. The blue box is the Highveld sampling domain, where the decreases were quantified (in text). Green circles indicate the locations of power plants. Black grid-squares in the NO₂ data show missing data due to the presence of clouds. The location of the Kliprivier station is shown by the yellow star.

Figure 2: Preliminary analysis of daily averages for ambient NO₂ (top) and SO₂ (bottom) concentrations at the Kliprivier site in Gauteng. The black lines show the 24-hour average values pre-lockdown (1 January to 16 April 2020), the light orange vertical lines show the start of lockdown (27 March 2020), and the blue lines show the 24-hour average values after lockdown (27 March 2020 to 16 April 2020). The light gray solid lines show the long-term median values of the 24-hour averages for 1 January to 16 April for 2007-2017. The grey shading extends to the 25^th percentile (on the bottom) and the 75^th percentile values (on the top) of the long-term 24-hour averages. This information on the long-term concentrations highlights the “expected” range of pollution levels during this time. This is a preliminary analysis to start to understand if changes are seen in pollution concentrations during lockdown. Air quality data were downloaded from http://saaqis.environment.gov.za/.

Enquires:

David Mandaha, CSIR Media Relations Manager
Email: dmandaha@csir.co.za
Mobile: 072 126 8910