Colchester Air Quality Sensor Project

At the time the project was conceived (2020), Colchester City Council (CCC) had declared two Air Quality Management Areas (AQMAs) in central Colchester due to exceedances of the annual mean nitrogen dioxide (NO₂) air quality objective (AQO). A key contributor to these exceedances was traffic congestion, which can occur during peak times, resulting in increased road traffic emissions.

CCC, Essex County Council (ECC) and Essex Highways therefore worked in partnership to obtain funding from the 2020/2021 Defra Air Quality Grant to implement an innovative air quality and traffic sensor network in Colchester. The aim of this project was to collect evidence of the causes of elevated NO₂ concentrations in central Colchester to inform the development of traffic management measures.

The aim of this project was to collect evidence of the causes of elevated NO₂ concentrations in central Colchester to inform the development of traffic management measures and to then assess the impact of these measures on air quality in the real-world.  Find out more about the Air Quality Sensor Project on Essex Air website.

Optimum sensor locations were identified collaboratively to obtain the most useful data on air quality and traffic. Installing the lamppost mounted sensor network was more difficult than originally planned, however, especially due to the impact of COVID-19 on resource capacity within the ECC street lighting team. Working together in a flexible manner to surmount these issues enabled the sensor network to be implemented in late 2021 and go live in early 2022. 

Overall, a particularly innovative approach was taken to this study. The traffic sensors selected for this study were (at the time) a new approach to monitoring road traffic, using machine learning to identify motorised road users, pedestrians and cyclists from video imagery, thereby allowing trends in active travel to be observed. The data were combined with pollutant concentration data from the air quality sensors that were paired with them, and both sets of data analysed (along with other inputs), allowing the causes of high pollutant concentrations to be whittled down to specific vehicle classes (e.g. cars travelling in a certain direction). 

The detailed analysis of NO₂ concentrations across the monitoring sites found that each location is impacted by its unique environment. However, there are overarching trends impacting multiple sites in similar ways across Colchester, which are summarised below:

• All sites demonstrate a diurnal profile that correlates with traffic flows where the morning peak is lower compared to the afternoon peak. Other seasonal and weekday similarities also exist, with concentrations typically higher in spring and autumn months, than summer months, as well as on weekdays relative to weekend days.
• Evidence of increased concentrations associated with road gradient as a result of excess emissions caused by vehicles travelling uphill.
• Evidence of street canyon impacts at all sites, particularly those with the highest concentrations. This is indicated by the relationship between wind direction and pollutant concentrations, which suggests a lack of dispersion and the re-circulation of pollutants within the canyon under certain meteorological conditions,
• Traffic variables were the most important variable in accounting for the variability in NO₂ concentrations. Bus flows were most important for more central locations, whereas car flows were most important for outer locations.
• In terms of traffic flows by direction, the nearest lane of traffic to the sensor was not always found to have the greatest influence on measured concentrations, nor was traffic speed shown to have a substantial influence.

An indirect success of the project was to demonstrate the capabilities of the traffic sensors, resulting in a large number of additional sensors since being installed in Colchester and more widely across Essex. These are being used to measure the impact of proposed active travel measures as part of the Active Travel Fund, to replace existing and broken methods of traffic monitoring (e.g inductive-loop traffic detectors), and to monitor traffic and active travel in locations that would previously have been deemed too expensive to monitor, all of which are resulting in wider benefits.

One unexpected challenge on this project, was to deal with the noise levels from the sensors. Prior to installation, it was not known that the pumps within the sensors generate a constant tonal hum measured at 59 dBA. Three of these sensors were located in close proximity to first floor windows of residential dwellings and consequently, these residents complained about the noise. The sensors were temporarily switched off whilst the Essex Highways team worked with the equipment suppliers to identify and implement a solution, which in the end was a modification to the exhaust outlet, which reduced the output noise level to 38 dBA. This modification was applied to all sensors, which resolved the issue at all but one sensor, which was particularly close to a resident’s single glazed bedroom window. Unfortunately, this site had to be decommissioned.

Key lessons learned on this project, which others should consider include:

• Liaise with the Street Lighting and Asset Management teams when selecting locations for monitoring.
• Consider noise levels when selecting monitoring equipment and locations (e.g. avoid positioning equipment in close proximity to resident’s windows and/or consider the type of sensor being selected).
• Consider the level of analysis to be applied to the data. There are powerful resources available, but they require access to and the ability to use packages like R.

Overall, this project can be considered a success. All the VivaCity sensors have been passed over to the Transport Planning team at ECC and will be maintained by them, owing to the amount and quality of data provided. A selection of the AQS1 sensors will remain at key locations across Colchester, with the others being redeployed to new locations across Essex to allow for similar studies to be undertaken with the aim of identifying the causes of high pollutant emissions and developing mitigation measures.

David Wright 

Essex Highways 

[email protected]