Blog 9 August 2023
It’s impossible to tackle air pollution without first measuring and mapping its spread and intensity. But air quality monitoring stations and sensors can be expensive and difficult to scale.
Cutting-edge technology is transforming what’s possible for the clean air movement. Data models using artificial intelligence (AI) can now predict air pollution to street level and identify breaches to legal limits. Wearable sensors democratise the collection of air quality data, making it possible for ordinary citizens to understand their personal exposure to air pollution and use this data to campaign for environmental justice.
Read on to explore the latest tech trends transforming the global clean air movement’s action.
AI helps scientists to accurately measure air pollution
Measuring PM2.5 pollution at a local level in New York
Scientists used to rely on models that used an extraordinary amount of data points to predict air pollution levels. These models were complex and inaccessible to non-experts.
Engineers at Cornell University have invented a simplified model which is easy to use. They use artificial intelligence to provide a detailed view of air pollution at street level. Their model focuses on fine particulate pollution (PM2.5), which comes from traffic exhaust and harms human health. The new model combines traffic data, topology and meteorology in an AI algorithm to learn simulations for traffic-related air pollution concentrations.
City planners and policymakers can use the machine learning model to gain a more accurate understanding of air pollution at a hyperlocal level. This data enables policymakers to design projects that reduce citizens’ air pollution exposure at a hyperlocal level. When considering new transport and infrastructure projects, governments can use the hyperlocal data to understand a project’s potential health impacts for smarter planning.
Poor air quality is known to cause millions of premature deaths globally. High resolution air quality data can help in effective air quality management and save peoples lives, however there is a paucity of data due to resource constraints. Careful use of AI can provide hyperlocal, high frequency accurate information and predictions that can be used for effective air quality management especially in regions with dense population and complex air pollution sources.”
Sachchida Nand Tripathi, Professor at Indian Institute of Technology Kanpur
Identifying excessive air pollution in Barcelona
99% of the world’s population breathes air which breaches the World Health Organization’s air pollution guidelines. AI can pinpoint when and where the level of air pollution is likely to exceed legal limits.
Scientists in Barcelona pioneered an AI model that uses machine learning to predict the likelihood of urban areas breaching legal nitrogen dioxide (NO2) limits. The scientists from the Barcelona Supercomputing Centre combine data from several sources:
- the CALLIOPE-Urban model that predicts air pollution at high resolutions, at varying heights, and at any location in the city
- official air quality stations and low-cost sensors
- details on building density and other geospatial data
- meteorological information
The output is NO2 concentration maps at street level every hour. Policymakers and citizens can see when there’s a breach in legal thresholds for nitrogen dioxide pollution. The group hopes decision makers will use these maps to improve air quality management in cities.
Plugging data gaps in Africa
Air pollution is universal, but its sources are highly specific to different geographies. Machine learning has the potential to revolutionise air pollution monitoring in countries with less air quality monitoring resources and technical expertise.
Rapid urbanisation and industrialization in major cities in Africa has worsened air quality in recent years, causing increasing mortality rates and health budgets associated with respiratory illnesses. Ambitious efforts by different countries to combat air pollution have put data collection and research at the forefront. Data and tools have had a high barrier of entry in Africa due to the cost and technical complexity of handling the tools. Machine learning seeks to bridge the data gap. It also enables cutting edge technology to be used in areas that have been formally excluded in data coverage and as an error minimization tool in research related to air quality in Africa.”
Christine Muthee, Air Quality Analyst, WRI Africa
Wearable tech advances campaigns for environmental justice
The invention of low-cost, lightweight air quality sensors has democratised access to air quality data. People can now wear sensors to record and understand the levels of pollution they experience in their daily lives. Wearable sensors like the AirBeam, created by HabitatMap, have been a game changer for environmental justice and local campaigns.
Cityzens4CleanAir used AirBeam sensors to run a clean air campaign in Accra, Cape Town and Lagos. Levels of air pollution vary greatly among and within the three cities, none of which have adequate air quality measurement. The NGO equipped young people with wearable sensors to collect data as they ran through their cities. A bespoke app captured the data on the sources of air pollution they encountered while running.
The young people reported that they felt empowered by the ability to actively monitor and access air quality data themselves. They used the evidence they had collected to advocate for clean air during COP27.
Wearables, such as the AirBeam, an air quality sensor, are increasingly accessible through community based organisations and libraries in response to the public’s growing desire to monitor their personal environment. By the end of next year, based on trends we are seeing at SciStarter and Arizona State University, we anticipate that between 500 and 1,000 libraries across the country will loan wearables and provide additional resources to support public participation in scientific research.”
Darlene Cavalier, Professor of Practice, School for the Future of Innovation in Society, Arizona State University
Supporting in innovative air quality monitoring technology
Innovations in air quality technology help generate data that’s accessible, usable and actionable. At Clean Air Fund, we work with partners to pilot innovative technologies in the air quality field.
Find out how we’re improving air quality data.
Main photo: AirBeam, wearable sensor. Credit: HabitatMap
