This use case is from South London Partnership – InnOvaTe “IoT” Project.
Outcome
This trial consisted of three approaches:
*The ANPR was only used for vehicle classification as it was deemed more acceptable, and gave us the same real data i.e. X cars, Y vans and Z trucks.
This will hopefully lead to enable local decision making to improve the environment and avoid the adverse impacts on health.
Sectors
Highways.
Key Stakeholders
Schools and Highways.
Overview
All five of the SLP boroughs (Croydon, Kingston, Merton, Richmond and Sutton) were keen to run at least one air quality monitoring IoT trial in their borough. The purpose was to either ascertain whether school street schemes are effective in reducing air pollution around schools at key times of the day; to monitor air quality more generally across the boroughs; to combine data from air quality and sound sensors with *ANPR (automatic number plate recognition) sensors to build a picture of activity at three waste transfer industrial sites in Merton.
*The ANPR was only used for vehicle classification as it was deemed more acceptable, and gave us the same real data i.e. X cars, Y vans and Z trucks.
About the Author
The InnOvaTe Programme is an Internet of Things initiative by South London Partnership (SLP) to “pilot and research” IoT across the 5 London boroughs of Croydon, Merton, Richmond upon Thames, Sutton, and the Royal Borough of Kingston upon Thames. The programme looks at ways to generate economic growth, support local businesses, help people live better, healthier lives and assist with addressing the climate emergency. The project assessed 150 IoT ideas for the councils concerned, implementing 48 of them successfully over 18 months. The programme was formally completed in March 2023.
Case Study Challenge
A number of school streets schemes have been trialled that aim to improve the safety of school children travelling to school, specifically the area around the schools. Many parents use their cars to transport their child and often leave their car idling. Our IoT trial would like to monitor the various effects of the schemes, namely air quality. The air quality sensors would be based near the entrance to the schools and monitor real time data. We would also need to monitor traffic flows including vehicles, pedestrians, cycles and scooters. It is expected that this data will help to build the case for implementation and retention of school streets and potentially of other traffic management schemes.
Secondly, information around travel and transport and its impact upon air quality is vital to understanding this complex issue, and to deliver the best outcomes in line with our statutory requirements. Detailed information of the links between transport and local air quality will better influence policy making, prioritising actions and interventions as well as enable local decision making to improve the environment and avoid the adverse impacts on health. This project will provide a detailed picture of transport and associated pollution in the borough, as well as that caused by vehicles passing through the borough.
Last but not least, one trial worked with businesses in the London Borough of Merton, specifically to reduce air quality emissions by using data, its management and new monitoring technology. Merton has a number of waste transfer stations that help deliver the South London Waste Plan which are located in a specific area of the borough. With businesses of this nature there is an environmental impact which includes, noise, dust (particulates), NO2 and associated vehicle movements in and around the sites. Merton worked with 3 companies by introducing technology to assist with measuring air quality, noise and vehicle movement around these sites, and as they enter the industrial estate. Real-time data allowed these businesses to better self-regulate, enable flagging of incidents of dust or noise and allow these to be identified and addressed immediately by the business. It also allowed for better regulation and understanding of the impact of individual sites on air quality in the area. Further, at the time the trial started there was active legal action being taken by one of the traveller sites, and local residents had also got the Environment Agency involved. Merton was firmly stuck in the middle with little solid data on which to respond.
Use Case Design Objectives
SLP designed these trials to monitor air quality in key locations across the boroughs in order to inform decisions about whether and where potential interventions may be needed. In addition, by recording baseline data before changes are implemented, the effects of any new schemes can be measured. The sensors provide several air quality indicators in real-time. Some of the trials utilised Breathe London devices, which combine data from a huge network of sensors with that from its reference sites, which are used to correct and calibrate the data. Reference units are technical equipment, which collect detailed air quality hourly data. This data is then fed into a statistical model that is used to assess historical and recent data to determine air quality and whether the data is likely to be representative. Over a hundred air quality sensors were installed across the SLP by the end of 2021, demonstrating a clear commitment to tackling air quality issues.
The last trial worked in a different way. It used EMSOL equipment to deploy ANPR (automatic number plate recognition) and air quality sensors to generate a picture of current dust and noise pollution on three waste transfer sites. The ANPR data would be useful to determine whether some types of vehicles exacerbate the dust problem more than others.
Commissioning (budget/procurement)
A selection of relevant suppliers listed on the Crown Commercial Services (CSS) Spark dynamic purchasing systems (DPS) were invited to tender a competitive process managed by the Sutton Procurement team.
Contracts were awarded to:
The budget range for the trials varied as the Breathe London sensors were very expensive (£1,600 each), so it was dependent on the number used. The EMSOL trial cost between £50k and £100k. All trials ran for 18 to 24 months.
Deployment (what / who / where / how long)
Vivacity Labs’ traffic sensors were mounted to lamp posts or similar infrastructure with a power source. The sensors use AI to identify how many people and vehicles pass the sensor site and record relevant information such as time, mode of transport, direction of travel, etc. This constant monitoring generates a detailed picture of traffic at the sensor site, which can then be mined for insights and trends developing over times of day, week vs. weekend, month to month etc. In the case of schools, term-time and school holidays might also be interesting comparison points. Breathe London air quality sensors are co-located with the traffic sensors and log detailed readings of several air quality indicators in real-time.
Sensors were installed as follows:
In Merton, for many years there have been ongoing concerns around waste vehicles and site pollution in Weir Road. These sites play a vital role in the waste management plans for the South London Waste Partnership. A number of the sites converge in Weir Road, leading to a disproportionate number of skip and waste vehicles in the local area. From a public perspective this can lead to concerns over pollution, load shedding and safety. A number of residential properties surround this area and we see ongoing complaints regarding re-suspended road surface dust and site specific waste. Additionally, the area is being made over with a number of new, tall and expensive apartments that will now be overlooking the sites adding further pressure to the situation. The trial was to work with three key sites and use an approach that has developed to cover the environmental impact of construction and apply this to industrial waste management. The project consisted of a number of activities, including the installation of pollution monitoring at each site and at the site entrances of three key waste management sites. Additionally, traffic movement will be required to identify the makeup of the traffic in the area and where possible tagging (with the permission of businesses) specific site equipment. The type of data to be monitored will be both noise and dust in 6 locations (2 on each site).
Sensors were installed as follows:
Technology Implemented
Suppliers of the sensors were:
Results / Key Findings
Benefits / Usefulness of Data
The creation of the Tracsis dashboard showing traffic data and air quality data together was helpful. The granularity of Vivacity data is a benefit here as, for example, officers can now discern how many pedestrians are affected by pollution at a particular time in a particular location, which is believed to be an innovative feature. Richmond officers, in particular, recognised that the air quality data available in this trial was much more sophisticated than what they had access to before, i.e. PM2.5 readings, which indicate levels of particles of less than 2.5 micrometres in diameter (< 2.5μm). This fine particulate matter is the most likely to cause health problems so is a metric of special interest.
Richmond is motivated to learn about what is happening to air quality in the borough in real-time and potentially linking this to an updated health messaging service. This is a good example of how data generated by these trials can be drawn on for diverse uses.
At least one officer on the Comms and Engagement team has used the Tracsis dashboard data to demonstrate the efficacy of a school streets scheme. The data showed reduced car numbers and improved air quality readings at Cheam Common Junior Academy in Kingston (see Table 5 below).
Table 5 – Tracsis Dashboard data showing reduction in care numbers and lower air pollution readings after school streets implementation at Cheam Common Junior Academy, Sept to Nov 2021
Interestingly, the InnOvaTe programme has approached both Breathe London and the GLA to seek feedback on the combined traffic and air quality datasets, to see if broader benefits are being realised beyond the SLP. We believe this is an angle that should be more fully explored.
For the Merton industrial site trial, in keeping with officer expectations, initial data from the one fully installed site suggests that it is well within the limits for construction site air quality measures. Early data shows that the sites can be quite dirty even outside of working hours, though we do not yet know why. Interestingly, on one occasion some nearby Breathe London air quality sensors picked up a spike in air pollution that was captured on site. This indicated that the data from those may also be somewhat useful in conjunction with the data captured specifically for this trial.
EMSOL proactively tailored their alert system to share timely updates with officers. For instance, they have decreased the dust sensor alert range from a change in air quality over three minutes to only when there are spikes lasting for at least one hour after initial detection. This new alert range is a much better measure of actual meaningful air pollution given the complexities of measuring air quality.
EMSOL have shared detailed information with officers on how alerts are triggered and how thresholds have been set, as well as proposals for adjusting them. When there is a pollution spike, the dashboard provides a snapshot photo of the area. Meanwhile, data from the ANPR cameras list all vehicles in the location for ten minutes prior to the breach, so that officers can begin to ascertain whether their presence affects the breach rate.
From the beginning, the officers handling the waste site trials worked hard to coordinate activities between the SLP, the sensor/data providers, and the industrial waste facilities.
Lessons Learned
With the Breathe London sensors, air quality recordings must be calibrated in order to establish baseline air quality levels against which to compare changes before it can be fully ratified. In addition, as the sensors are solar powered, some struggled to function fully through the winter as sunlight is weak and the days are short. This resulted in power management controls to reduce functionality to conserve energy. This caused some initial confusion for officers as the dashboard showed some sensors as greyed out, which meant no index data could be generated for that particular hour, although underlying data was still accessible.
This innovation was not fully recognised by some officers involved in the pilots or communicated to colleagues with sufficient impact. Some officers involved in the scheme did not use the Tracsis dashboard at all, and programme management had to push some councils to provide information on which sensors should be added to the dashboard for their borough. Even where councils were engaging, the message was not getting through, although it was not possible to ascertain whether this was due to low internal appetite or a simple lack of awareness. Council efforts were not being fully publicised and therefore will not be recognised by residents.
The main problem with this trial was the lack of engagement with colleagues. This is essential for the programme to work. The Breathe London sensors are very expensive. At the time they were the best available, but shopping around would be suggested.
For the waste site trial, officers reported spending a great deal of time building trust among the three groups. They assured all parties that the sensors were meant to serve as objective data collection mechanisms, rather than as monitoring devices implemented with preconceived expectations.
Data results and insights were shared with all partners. Findings revealed that the major sources of air and noise pollution were vehicles on the adjacent roads, rather than the waste facilities. The facilities also learned of a few basic ways to improve their practices, which they swiftly implemented.
Conclusion
On the whole, the air quality trials were a success.
For the schools air quality trial, the AQ colleagues didn’t wish to engage with the schools and build a real programme. However, the units subsequently proved very useful for Highways colleagues, and Councillors loved them. In Richmond, the data fed in at the right time to focus on this area. It has subsequently become embedded and part of BAU to get baseline data. Data was proven to be very important, although the link to the school, children and education was missed, so Highways took it over. With better officer and CEO engagement the data was attended to. This is absolutely key to generate baseline information and assist in making supportive decisions based on facts.
The Merton waste site trial was a big success! It was perhaps our most interesting trial. The three waste transfer sites were literally being targeted by all sides, as new flats had now been built, two traveller camps overlook the sites and in short they could never appear to be treated fairly. The introduction of sensors on the sites and outside, allowed facts to be obtained covering noise, vehicle movements and air quality. Whilst the sites clearly were silly at times e.g. not turning on sprinklers during dry and windy days, they were nowhere near as bad as portrayed. Also, by allowing live data to be fed to admin staff they could immediately get involved if noisy or dirty vehicles came on site. In the end the data showed a 28-40% improvement in air quality and noise at nil cost to the council, and based on WHO averages, the three sites were compliant. Ultimately, this led not only to the environment agency referring to this scheme as a poster child, but also the suspension of their interest in the sites as they were now deemed clean and within all approved limits.
Photos from the use case
Contact
For further information, please contact the service leads involved in this project, listed below.
Paul Garside
Sustainable Transport Officer, Kingston and Sutton Shared Environment Service
London Borough of Sutton and Royal Borough of Kingston
paul.garside@sutton.gov.uk
Peter Bond
Environmental Protection Officer, Kingston & Sutton Shared Environment Service
London Borough of Sutton and Royal Borough of Kingston
peter.bond@sutton.gov.uk
Jason Andrews
EH Pollution Manager (Air Quality)
London Boroughs of Merton, Richmond upon Thames and Wandsworth
Jason.Andrews@merton.gov.uk
Pierre Venter
IoT Delivery Manager
Royal Borough of Kingston and London Borough of Sutton
pierre.venter@sutton.gov.uk
