This guest blog from Dr Douglas Finch & Prof. Paul Palmer explores the COVID-19 lockdown and its effect on air quality in Scotland’s urban areas. As with all guest blogs, what follows are the views of the author, not those of SPICe or indeed the Scottish Parliament.
The COVID-19 lockdown has seen fewer cars on roads that are typically packed with daily traffic. With fewer cars comes fewer engines emitting pollutants into our atmosphere. So what has lockdown meant for air quality in Scotland’s urban areas?
This blog describes changes in air pollution most associated with traffic for five weeks from the start of the lockdown (23 March – 26 April). Although other sectors (e.g. industry) also affect air quality, the largest change in emissions in cities is assumed to be from traffic reductions and this lockdown gives a unique opportunity to study the impact traffic has on air quality.
Pollution and data
Combustion engines emit nitrogen dioxide (NO2), volatile organic compounds (VOCs) and particulate matter (PM) among others. The mix of NO2 and VOCs leads to the formation of ozone (O3) which although it protects us from harmful UV radiation high in the atmosphere, is poisonous and therefore dangerous at the surface. All of these pollutants are known to have adverse effect on the respiratory system, including reduced immunity to lung infections, which is particularly relevant to the COVID-19 virus.
We use air pollution measurements from the Department for Environment, Food and Rural Affairs’ Automatic Urban and Rural Network (AURN). There are 19 sites operating in Scotland, scattered among urban and rural locations. As we are limited by which pollutants are measured and where the measurements are taken, this analysis focuses on NO2, O3 and PM2.5 (particulate matter smaller than 2.5 microns) in urban locations.
You can explore more AURN data though an online interactive tool developed as part of our work at the University of Edinburgh.
Changes seen during the lockdown
Air pollution levels after the lockdown started (23 March) reflect reductions in traffic emissions across the whole of Scotland. The figure below shows the mean average values for nitrogen dioxide expected at this time of year and the mean value observed during the lockdown. Measurement sites located on busy roads (Glasgow, Aberdeen and Greenock) show the largest reductions in nitrogen dioxide (>50%), whereas the other sites measure atmosphere further from the roadside where pollutants have had time to mix and dilute and so have lower levels to begin with.
If we look at a typical week of nitrogen dioxide concentrations in Glasgow (see figure below), we can see the lockdown has led to a stark decrease in emissions. During a normal, pre-lockdown week, variations in nitrogen dioxide concentrations reflect traffic patterns, with a daily cycle peaking at morning rush hour and a smaller peak for the evening rush hour. There are lower concentrations at the weekend due to less traffic on the roads. During the lockdown, there is still a morning rush hour but nitrogen dioxide levels are substantially lower for every day of the week, lower than values we normally expect on a Sunday.
A similar pattern of nitrogen dioxide can be seen across cities in Scotland as well as the rest of the UK. If we look at the change seen in other pollutants however the situation is less straight-forward. The figure below shows a heat map of the differences between daily nitrogen dioxide, ozone and PM2.5 during the lockdown (as well the preceding two weeks) from the mean of the same weeks for the previous five years. This shows the impact of the lockdown, even after taking into account expected year-to-year variations in pollutants that are influenced by changes in weather. We show some of the larger urban areas in Scotland as well as Manchester and Birmingham for comparison.
As we’ve seen previously, nitrogen dioxide concentrations have reduced, however, the heatmap shows that this was the case before the lockdown, meaning either lower levels of traffic prior to the lockdown (when social distancing was advised) or traffic might not be solely responsible. As the concentrations have remained low for such a long duration, it indicated the reason is traffic rather than other factors such as changes in weather.
Ozone concentrations have increased in Glasgow and Edinburgh as they have for most cities across the UK. This is expected to happen in cities where high levels of nitrogen dioxide can start to destroy ozone instead of help create it. When there is less nitrogen dioxide, less ozone gets destroyed and the atmospheric concentrations of ozone increases. This pattern doesn’t hold true for Aberdeen however and to some extent for Fort William. The reason behind this is unclear and will be the subject of further investigation.
Unlike nitrogen dioxide and ozone, PM2.5 does not show a clear directional change throughout the lockdown. Early in the lockdown, concentrations of PM2.5 increased. One potential explanation for this is the influence wind has on particulate matter. A new weather system arrived over the UK at the same time as the lockdown came into force and caused a change in wind direction over much of the country. This meant air was being brought from the east (from continental Europe) that may have been more polluted than air blown in from the west.
Another potential explanation for this higher PM2.5 is an increase in barbecues and garden bonfires occurring due to the warmer weather and people staying at home. This is difficult to confirm and the number of fires needed to see this change over the whole country is extremely uncertain. Either way, this indicates that PM2.5 is not as regulated by traffic levels as nitrogen dioxide and ozone.
Heatmap showing the change in pollution concentration (nitrogen dioxide [top], ozone [middle] & PM2.5 [bottom]) compared to the mean of the corresponding days for the previous five years. Blue shows a reduction, red shows an increase. The intensity of colour shows the magnitude of change. Manchester and Birmingham are included for comparison.
Implications and the future
This period of enforced traffic reduction has given us an opportunity to see what our cities would be like with reduced emissions. How air quality changes over Scotland and the rest of the world as the lockdown continues will be extremely interesting and no doubt the subject of many research projects.
This brief analysis shows that traffic is not the only consideration when cleaning up our air. Variations in PM2.5 concentrations have highlighted that changes in local emissions do not exclusively control local pollution levels. As emissions from newer vehicle fleets get progressively cleaner we will likely need to shift our focus onto other pollutants. For instance particulate matter smaller than one millionth of a metre that can penetrate further into the human body.
By adopting cleaner travel options, such as more electrically powered transport and more cycling and walking infrastructure, then we would see lower levels of nitrogen dioxide. Strategies to lower VOCs emissions would also reduce ozone levels and therefore improve respiratory health across the country.
Dr Douglas Finch & Prof. Paul Palmer
School of Geosciences, University of Edinburgh