New AI-powered early warning system to combat climate-induced disasters in KZN
In an era where climate change poses an increasing threat to communities worldwide, a groundbreaking initiative is taking shape in South Africa.
The collaborative effort involved the University of the West of Scotland (UWS), the University of KwaZulu-Natal (UKZN), the Royal College of Surgeons in Ireland (RCSI) Faculty of Nursing and Midwifery and the University of Portsmouth.
UKZN DVC Research and Innovation Professor Anil Chuturgoon said: “This initiative known as the ‘weather’ project, which stands for warning system for extreme weather events, awareness, technology for healthcare, equitable delivery and resilience represent a transformative step in safeguarding lives and strengthening our communities against growing threat of climate-induced disasters.
“This grant will enable us to develop a cutting edge early warning system tailored for the unique needs of vulnerable communities in eThekwini and Ugu districts.
“This initiative is more than just a response to climate change. It is a commitment to a safer, healthier and more resilient future. The weather project is a beacon of hope, a demonstration of how science and technology, when combined with community engagement, can mitigate for the impacts of our changing climate.”
The project has been awarded a pivotal grant of more than £2 million from the National Institute for Health and Care Research (NIHR) to develop the system.
University of Portsmouth professor of wireless communication systems and School of Electrical and Mechanical Engineering head Professor David Ndzi said: “This early warning system would be to focus and predict rain and also to predict flooding. One other key aspect is to predict disease symptoms outbreak.”
He said the aim of the project is really to have a system that helps the population to know well in advance when it is going to rain and whether that rain is going to cause flooding.
“We are taking a slightly different approach, because in rain prediction, you often focus on past data... How is that going to evolve in the future? We are taking a slightly different approach, and that is to use artificial intelligence to predict when it is going to rain and how heavy that rain will be, and that can be information that is available to the different communities, because there is an increase, overpopulation in flood plains, and also to indicate when there is going to be floods, and if there is flooding, what sort of health issues the population will have,” Ndzi explained.
“One of the key things within our prediction system is wind and wind speed. We will be able to predict the wind speed and see how that is changing, because that does drive other factors. A key thing just to emphasise, is that these systems have to be adapted to the geography of the area.”
“We are going to work with the different communities, work with the municipality, and also work with the South African Weather Service,” Ndzi said on how to ensure the systems are reliable.
“The predictive system is to give that advance warning for people to really take precautionary measures and minimise harm, either to property or to the loss of life.”
Ndzi said the approach they are taking is that there will be prediction at specific locations, and you will be able to download an app that you can zoom in to see exactly where you have either heavy rainfall or flooding.
He said the fundamental premise of this project is climate change.
“AI does one thing when we implement that system, it will be doing the prediction, but it will also be learning and adapting as the pattern of rainfall changes… It will adapt and change with time, but that will be running in the background.
“The users, you would be able have on your mobile or on the app, what the situation, what the conditions are now, maybe in six hours, in 18 hours, in two days and so on. And it will continually improve that accuracy,” Ndzi further explained.
thobeka.ngema@inl.co.za