Making sense of our sensor data

In 2019 the University of Southampton installed 11 air quality sensor boxes for the Breathing Spaces project, in and around St Denys. They monitor particulate matter (PM10 and PM2.5), temperature and humidity. Real-time information on local air quality in St Denys can be found here.

The locations of the 11 sensors in and around St Denys

We have been collecting local air quality data since February 2019. Here are some of the things we’ve found out:

  • The whole city needs to work towards cleaner air for all
  • High levels of air pollution are still present on quiet streets
  • Peak-time traffic means peak air pollution
  • Car-free days can mean cleaner air
  • Weekends don’t provide any respite from bad air
  • Clear, sunny skies can mean higher levels of pollution

There is no safe level of air pollution

Even though we have to treat the data from the Breathing Spaces sensors with caution, we can look at the overall trends and show that people are exposed to elevated levels of air pollution almost every day in the city. High pollution episodes are hidden when you talk about averaged data and whether legal limits have been met in any particular year.

Research has shown that no amount of air pollution can be considered ‘safe’, especially for children, the elderly, those with existing health problems and residents of more deprived neighbourhoods, as they are more vulnerable to the effects of dirty air. This is why the WHO guideline limit of 10ug/m3 for PM2.5 should be enshrined in law (the current UK annual limit is 25ug/m3) and our cities re-designed for sustainable and active travel.

Quiet city streets still have dirty air

There are two graphs below. Both are from the same period (1st-30th May 2019) but one is from a sensor located on the busy main road through St Denys and the other from a sensor where there is little through traffic. If you compare the patterns of the two they are really quite similar aren’t they? This shows that, in terms of fine particles (PM2.5), people are exposed to the same levels of pollution, regardless of whether they are on a quiet or busy street. This is because those tiny airborne particles disperse easily over a wide area – they aren’t confined to busy roads.

Data from St Denys Road/A3035 sensor, May 2019
Data from Horseshoe Bridge sensor, May 2019

Rush hour is bad for your health

We know instinctively that lots of traffic means high levels of pollution. The real-time graphs above and the multi-coloured graph below also paint this picture – that there are definite peaks in air pollution at certain times of the day – rush hours. The evening one happens later than expected, showing that dirty air builds up throughout the afternoon, into the night.

Average daily trends in air pollution

Car-free days can help us breathe better

There are currently only two opportunities each year for the residents of Southampton to enjoy car-free streets across large areas of the city and both of those are sporting events, rather than Open Streets where people can roam freely. However, removing vehicles from lots of city roads does seem to have a positive impact on the air that we breathe. The graph below (data from Government sensors this time) shows that the Half-Marathon (blue line) & Let’s Ride (orange line) had lower levels of pollution than many other Sundays (grey lines) between May and September.

Comparison of data from Sundays with and without car-free events

Pray for rain!

Perhaps slightly counter-intuitive, but it is well known that higher levels of PM2.5 are associated with ‘good weather’ i.e. clear skies and minimal precipitation. On the other hand, cloudy skies and rain are associated with lower air pollution.

Our sensor data have also shown that when the weather isn’t so good, pollution levels are lower. We will also be using the data to investigate the seasonal differences in air quality in St Denys; but it is well known that PM2.5 emissions are higher in the winter months, in part due to domestic coal & wood burning.

You can’t breathe easy at the weekends

The colourful graph below is a way of showing average trends in air quality on different days of the week, over a 4 month period. Saturdays and Sundays show similar patterns to mid-week days, so there isn’t really any respite at the weekend. (Please note however that averages hide the daily fluctuations in readings.)

City-wide collaboration for cleaner air

As toxic air does not respect geographical boundaries, everyone needs to work together to clean up our air. Designating ‘quiet routes’ to school or work only protects people from exposure to Nitrogen Dioxide (the other main pollutant in our air) and ignores the effect of fine particles which have a much more pervasive effect on our health.

We are really pleased that the Breathing Spaces project in St Denys is empowering more people to make changes for a more liveable neighbourhood and healthier streets, but local action is not enough. There needs to be a concerted effort from policymakers, businesses, communities and external stakeholders to work together to make significant improvements to the city’s air quality. We all deserve to breathe clean air.


DISCLAIMER: The data collected by the Breathing Spaces sensor boxes has not been recorded using legally validated reference equipment and should therefore be treated with caution. In other words, please don’t place too much emphasis on the numbers presented in the graphs as we cannot guarantee that a particular legal limit or air quality threshold has been exceeded. If you’re interested in finding out more about the methods used, the University of Southampton team has recently published a paper in Nature.

However, as you can see by the graph below, the general trends match the data provided by the Government automatic monitoring station in the city centre, so they provide a fairly reliable indication of local air pollution.

Average readings from the Breathing Spaces sensors plotted against readings from the city centre AURN for April 2020

The graphs & map on this page were produced by Robin Wilson.

%d bloggers like this: