I remember driving into Johannesburg over the wintertime and seeing the smog overhanging the city. I could see that something wasn't right.
Tapiwa Chiwewe, Research Engineer at IBM.

On the outskirts of Cairo, the pyramids shimmer in a brownish haze. In gridlocked Nairobi, elderly minibus taxis belch black smoke into the fume-filled air. Driving into Johannesburg, the skyline is often encased in a dome of smog.

As Africa urbanises and industrialises, its economic growth is powered by coal, its vehicles by dirty fuels, and many still cook over paraffin and three-stone fires, burning their rubbish in the street. Researchers believe that more than 700,000 Africans die each year as a result of air pollution – and the number continues to grow.

Few African cities even measure air pollution, but those that do routinely rank among the dirtiest on the planet. Nigeria is home to four of the worst, including Onitsha, which holds the dubious distinction of the Earth's most toxic air. The health and lifestyle consequences of living amid pollution are vast, not least in Johannesburg, a five million-plus city that's set to become a megacity within a generation.

Air pollution

Breathing becomes a pain

From the bedroom of her suburban Johannesburg home, Hilary Pace, a model agent, can see towering piles of ash, residue from the nearby coal-powered Kelvin Power Station. Every morning she sweeps a fine layer of black dust off her patio. In the 10 years she has lived in the area she has suffered more chest and sinus infections than she can count; her stepson was diagnosed with asthma, a condition that disappeared when he moved away.

“Depending which way the wind blows, the sulphur smell can be quite asphyxiating,” she says.

In the slums and shantytowns where Johannesburg's toxic load of mining dust and vehicle fumes mingle with the grime from cooking, heating and rubbish fires, the burden lies even heavier on some of the nation's poorest citizens.

Time for change

Tapiwa Chiwewe, a Zimbabwe-born research engineer at IBM, believes harnessing data could help solve the pollution crisis.

“When I joined IBM, I used to live in Pretoria,” Chiwewe recalls. “I remember driving into Johannesburg over the wintertime and seeing the smog overhanging the city. I could see that something wasn't right.”

Air pollution

Inspired by the problem, Chiwewe got to work on a solution: an air quality forecasting platform, developed in collaboration with South Africa's Council for Scientific and Industrial Research, that's undergoing trials in Johannesburg.

The platform reports, analyses and predicts the intensity and location of key pollutants: ozone, nitrogen dioxide and microscopic particles known as PM2.5 and PM10. PM is short for particulate matter, the tiny particles that make up dust, dirt, soot and smoke: PM10 particles are 10 micrometres or smaller, just a fraction of the diameter of a human hair, while PM2.5 are even tinier, no larger than 2.5 millionths of a metre.

Data that drives change

To create his platform, Chiwewe drew on IBM's Green Horizons technology. The system harnesses massive slabs of historical and real-time data about weather, air quality and topography, through to traffic reports and social media, to deliver granular, high-accuracy forecasting. The results could help shape policy, planning and even law enforcement across Johannesburg.

With advance warning of adverse pollution events, city authorities could issue public health alerts and suspend polluting activities, such as scaling back industry or diverting traffic from a specific road. With an accurate understanding of pollution patterns, the city could identify –and prosecute – major polluters, plan the location of future roads and settlements, and tailor intervention strategies.

Intervention strategies, explains Lebo Molefe, director of Air Quality and Climate Change for the City of Johannesburg, play a major role in air quality management. “These are projects like the roll-out of electricity to unelectrified settlements, surfacing of road infrastructure, rehabilitation of mines,” she says. “And provision of proper housing linked to renewable energy sources.”

Artificial intelligence steps in

Behind the platform, and managing wildly varying types of data, are sophisticated, self-improving machine learning algorithms, a type of artificial intelligence (AI).

“In its current form it's a cognitive platform. We get real-time data from traffic, from air quality and from social media,” Chiwewe says. “It can actually improve itself. It can understand, reason and use evidence.”

Chiwewe plans to roll out an application programming interface (API) that developers can use to build apps based on his platform. He is keen to see a solution that will help people who suffer from respiratory problems and enable them to plan their lives more easily. Molefe, too, sees protection of vulnerable citizens as a key potential benefit of the system.

Chiwewe would like to see the platform rolled out to other cities in Africa. But there is a problem. While Beijing has a network of more than 30 air quality monitoring stations, and Johannesburg has eight, many African cities have no stations to measure air pollution at all. “Even though the mechanism for Kenya is the same,” Chiwewe says. “We don't have data for Kenya.”

Self-improving technology

The issue is challenging but Chiwewe sees a way through. The next stage of the platform’s evolution, still at the blue-sky thinking stage, would be a virtual monitoring system. “The virtual monitoring system based on computer models is a way of sensing using software, rather than sensing using hardware,” he says.

Rather than use on-site sensors to test for pollutants, Chiwewe's system would exploit other options such as remote sensing and climate chemistry. “We can combine satellite data, weather data and climate chemistry models using machine learning and AI," Chiwewe says. "The result would be a virtual monitoring station that indicates the level of pollution in a particular area."

Like the current incarnation of the platform, the future version will be self-improving. And many hope it will not just measure and predict pollution, but actually drive enforcement to address the problem as African cities identify the pollutants that are turning their blue skies grey.

The Green Horizons initiative from IBM is using data analytics to measure air pollution in big African cities. It is also working with various government bodies to improve air quality and increase the use of renewable energy.

Building a sustainable future with data

The Green Horizons initiative from IBM is using data analytics to measure air pollution in big African cities. It is also working with various government bodies to improve air quality and increase the use of renewable energy.

Discover more