COMPAIR results can be grouped into five categories.
1) Digital tools: Four custom-built apps for analysing the impact of urban mobility policies (Policy Monitoring Dashboard), measuring carbon footprint and simulating abatement strategies (CO2 Calculator), visualising air pollution in Augmented Reality (Dynamic Exposure Visualisation App), and understanding air pollution along the route (Dynamic Exposure Visualisation Dashboard). PMD was used in two Flemish cities; in Herzele it helped to assess the effects of a school street, in Sint-Niklaas to measure the effects of a new circulation plan and a Vijfstraten bicycle bridge. CO2 Calculator was used in Athens, Plovdiv and Sofia, mostly by senior citizens and high-school students as part of a campaign to promote green lifestyles. DEVA was published on Google Play and integrated with Google's BreezoMeter to display modelled air quality data. DEV-D was tested in Flanders and Berlin for over a year, during which time more than 750 km worth of trips were recorded on foot and by bike.
2) Sensors and data on traffic and air quality collected by citizen scientists: Three sensors were significantly improved during the project. Telraam became more robust, accurate, and easier to use. NitroSense was upgraded to work in new locations (outside the Netherlands) and contexts (policy use cases). bcMeter benefited from improved technical design and WiFi protocols to support data transfers. All original sensor data is managed by the data framework following FAIR principles and is shared using the OGC SensorThingsAPI.
3) Calibration: Data from air quality sensors was calibrated with data from reference stations to account for influencing factors that affect low-cost sensor (LCS) accuracy and precision. Initially, a distant calibration approach was used. However, given the limited density of reference stations and the fact that many LCS were stored indoors, the team decided to try another method: auto-calibration. It uses electrochemical sensor readings instead of reference data. Preliminary tests conducted in Ghent with NO2 devices prove auto-calibration to be a viable alternative.
4) Communities & policies: Networks of citizen scientists and other stakeholders (public authorities, NGOs, schools) that contributed to and were influenced by local policies, as well as by COMPAIR in terms of improved knowledge, skills, motivation, social network, lifestyle changes. In Athens, 80% of air quality data was collected by senior citizens who are identified as a priority at-risk group in the city's climate-mitigation strategy. In Berlin, two campaigns happened, one focusing on redevelopments in the Graefikiez Kiezblock, another on dynamic exposure during cycling in which more than 100 Berliners took part. In Flanders, besides urban mobility use cases (school street, circulation plan, bicycle bridge), considerable effort went into promoting STEM education and behavioural change through citizen science. After one such experiment, results showed a 20% decrease in car use and a 7% increase in the use of active and collective transport modes among students. Boosting STEM was also the focus in Plovdiv, where hundreds of students from 3 schools participated in citizen science. And in Sofia, the main activities focused on promoting green mobility habits through a school bus service. It was found that 60% of students from grades 1-4 who started using the service were previously driven by car.
5) Project knowledge base: All of the above are described in project reports, publications, webinars and a recently published online course "Better Together: Citizen Science and Digital Tools to Improve Air Quality Policy". Interested adopters can use this knowledge to learn, get inspired and reproduce COMPAIR in whole or in part, in the same or different settings and cities.
