Pollution Explorers

Pollution Explorers is a participatory project in which people make sense of the quality of air in their environment through their innate subjective perception using wearable technology and machine learning, with the aim to harness collective actions to improve air quality in neighbourhoods.

Workshop at: Tower Hamlets(London, UK), Macclesfield(UK), Koekelberg(Brussels) and Sablon(Brussels)
Presented at: Centre Pompidou 2019 and STRP Festival, Eindhoven, Netherlands, 2019

Through the engagement, participants make a statement on the air quality with their own physical actions of recording perceptual data using the wearable tools, help to fill in missing ‘gaps’ of air quality data in order to make sense of the quality of air in their neighbourhood and devise collective pledges that they can commit to helping tackle air quality issues in their neighbourhood for a period of time.

Over the past 6 months, we worked with a total of 90 participants (children, young adults, parents, activists and city officials) to pollution explore in their neighbourhoods in Borough of Tower Hamlets(London,UK), Macclesfield(UK) and Brussels(Belgium) using wearable technology and machine learning processes to emphasise personal agency and responsibility in tackling complex issues surrounding air pollution.

The project has proven that people’s ability to assess the quality of air compared to the digital sensors is fairly high in some instances and a high percentage of participants have shown dedication in committing to an action to tackle air quality for a period of time in their everyday life. These are important findings from which participating residents and other cities can benefit greatly.

In each workshop, participants are encouraged to make sense of air quality issues by collectively describing, discussing, and pollution exploring in their vicinity using technological tools. Depending on the length of each workshop, participants went through the various key activities:

  • An open discussion among participants regarding air quality in their neighbourhood
  • Carry out Pollution Explorers walk in specific locations around the workshop space. Each location was selected for it’s mixed built up so that participants can look for the different cues in the environment that might inform them about the quality of air  (e.g park facing the highway, streets with wind tunnel effect). At each location, participants were asked to:
    • describe the quality of air using 2 written keywords
    • record their perception of air quality using the custom made wearable suits, the data were uploaded in real time to an online database. 
  • A debrief session where participants collectively look at similarities and differences between each other’s description of air quality, and the correlation between their recorded perceptual data (from wearables) and the air quality data (from air quality sensors).
  • Each workshop was ended with participants discussing ways they can tackle air quality issues in their everyday life and choosing one action to commit to for a period of 7 or 21 days. The actions were tracked through:
    • prepaid postcard given to each participant for them to track their actions for 7 days before sending it back to us.
    • custom-made badges and poster that enable students in a school to track their actions over 21 days

The outcomes from the workshops have been revealing. Through data analysis and observations, we have found that:

  • People are not necessarily precise in determining the quality of air but they are accurate in terms of comparison between air quality i.e they are good at telling if air quality is better or worse than a location they were at before.
  • People are especially good at making correct perceptions in areas with variability in the environment and that is badly polluted.
  • Perception is completely relative. There are differences from location to location and when the base air quality is good, people feel a lot stronger about changes or worse AQ because of that relativity
  • Children are very sensitive to momentary changes in the air around them, such as a truck or car driving by, and adults tend to be more holistic about how they perceive the air around them. People are shown to have an emotional connection to certain smells (e.g perfume, barbecue or food smell) that contribute to bad air quality This can sometimes cause them to make an erroneous judgement about the air quality even if they knew the air was bad.
  • When it comes to describing air quality using words that are not related to just good or bad, people are shown to have difficulty describing air quality in spaces with complex environmental build ups (e.g a park situated near the highway). An evaluation on the list of words contributed by all participants has shown that the combined words description are in some instance comparable to what an air quality sensor has captured.
  • Children are more committed to long-term actions to tackle air quality issues (e.g planting or growing more plants and reducing electrical consumption) while adults have shown to prefer short-term actions (e.g reduce unnecessary car journeys, walk to work) but have shown to be less committed to keeping up with the actions for an extended period of time.
  • People who were more sensitive to air quality changes in their environment were shown to be more committed to actions to tackle air quality in their neighbourhood.

Through the learnings, we aim to improve on our tools and work with more communities on pollution exploring and tackling air quality issues in their neighbourhoods and cities.

Pollution Explorers is a project led by Ling Tan of Umbrellium and funded by Vertigo STARTS, with the support of the STARTS Residencies Project as part of the STARTS initiative of the European Commission.