Researchers adapted methods from experimental particle physics to build AI_r, a system that monitors air quality in real time. AI_r links a network of low-cost sensors in pollution hotspots with Internet of Things technology and artificial intelligence. The team says this approach avoids heavy investment in traditional monitoring infrastructure and can give faster, local data.
Air pollution has large health and economic costs. The World Health Organization estimates almost 7 million deaths every year, with 89 per cent of those deaths in low- and middle-income countries. The World Bank finds that poor health from outdoor air pollution costs up to US$6 trillion each year. Data from other organisations underline gaps in monitoring: a 2024 report from OpenAQ found over a third of countries are not monitoring air pollution at all, and nearly 1 billion people live in a country without any programme to monitor air pollution. Vumile Senene, country lead for South Africa at the Clean Air Fund, said forty per cent of South Africans live more than 25 kilometres from the nearest monitoring station and argued that low-cost sensor networks can measure air in underserved communities and support campaigns and evidence-based policy.
The first AI_r pilot began over a year ago in Soweto, where sensors in schools showed how pollution spikes can move across the city. Mellado, who heads the Institute for Collider Particle Physics at the University of the Witwatersrand and is lead researcher for the South African Consortium of Air Quality Monitoring, said a single burn can affect air quality up to ten kilometres away. Over the next year the project will place 500 sensors across the Sedibeng district, home to around 1 million people; Mellado described this as the largest network in Africa in one area. The project has support from the governments of South Africa, the UK and Canada, as well as CERN and iThemba Labs. The team received the Global South eHealth Observatory (ODESS) prize from the Pierre Fabre Foundation to be presented in October. Researchers say they hope to expand sensor networks across Africa and beyond, but they note that further funding is needed to develop a continental approach.
- Supporters: South Africa, UK, Canada
- Partners: CERN, iThemba Labs
- Prize: ODESS from Pierre Fabre Foundation
Difficult words
- adapt — change methods to suit a new useadapted
- monitor — watch something regularly to check its conditionmonitors, monitoring
- hotspot — area with unusually high levels of pollutionhotspots
- infrastructure — basic systems and services needed for a society
- underserved — not receiving enough services or public support
- consortium — group of organisations working together on a project
- continental — relating to a continent or very large region
- sensor — device that detects or measures physical conditionssensors
Tip: hover, focus or tap highlighted words in the article to see quick definitions while you read or listen.
Discussion questions
- What are the advantages and possible limitations of using low-cost sensor networks for air quality monitoring?
- How could governments use faster, local data from systems like AI_r when making environmental policy decisions?
- What practical challenges might teams face when trying to expand sensor networks across an entire continent?
Related articles
New models rethink the insides of Uranus and Neptune
A University of Zurich team created new interior models for Uranus and Neptune. The work shows the planets could be rock-rich or water-rich, helps explain their odd magnetic fields, and says we need dedicated missions to learn more.
Virtual diet and exercise program helps people with lymphoma
A study tested a virtual diet and exercise program for people getting chemotherapy for lymphoma. The program improved symptoms, attendance was high, and participants had better strength and physical performance during treatment.
Africa uses AI to strengthen health systems and self-reliance
At the CPHIA conference in Durban, Africa CDC said AI and digital tools can help protect 1.4 billion people, improve surveillance and support primary health care. Data governance, infrastructure and domestic financing are key concerns.
Clean water reduces child stunting in Mozambique
A University of Notre Dame study using national survey data finds that improved access to safe drinking water lowers the odds of stunting in Mozambican children by about 20%. Improved sanitation showed no independent effect on stunting.