EXTREME weather conditions in KwaZulu-Natal which caused many deaths and billions of rands in damage has resulted in a new project to develop an early warning system that will assist vulnerable communities.
It will focus on two vulnerable districts in KZN, eThekwini and Ugu and researchers will use cutting-edge technology and artificial intelligence to develop a comprehensive warning system tailored to the needs of affected communities.
The project known as ‘Warning system for Extreme weather events, Awareness Technology for Healthcare, Equitable delivery, and Resilience (WEATHER) is a multi-partner collaboration between the University of the West of Scotland (UWS), University of KwaZulu-Natal (UKZN), the Royal College of Surgeons in Ireland (RCSI) Faculty of Nursing and Midwifery and the University of Portsmouth.
Professor Saloshni Naidoo, Head of Public Health Medicine at UKZN says the collaboration arose following a call between her department and Professor Mary Lynch (UWS/RCSI).
“Following the discussion about the call put out by the UK NIHR RIGHT (Research on Interventions for Global Health Transformation) for proposals on climate and health, it was agreed that we would jointly collaborate on a submission. The call was made a few months after the mid-year floods we experienced in KZN in 2022,” said Naidoo.
The NIHR has set aside about R46 million (£2 million) to develop the system which aims to empower vulnerable communities and healthcare systems in response to the devastating impacts of flooding and associated health risks.
Naidoo said they were currently addressing administrative matters concerning ethics and once it was approved they would start with the fieldwork for the project which runs until 2027.
“The first year of the project will focus on analysing historical weather data and collecting new climate data to develop the predictive models needed to develop the alert system. We will also assess the resilience of the existing health system in KZN to strengthen the health system and ensure it is responsive to adverse events,” she said.
Representatives from national and provincial departments of health, the South African Weather Service and eThekwini Municipality form part of the project's advisory committee and would make input related to the strategic direction of the project.
“KZN has experienced several adverse weather events in the past two years and it is appropriate to base the study in this province. Furthermore, we would like to use our expertise to help our province and the people of KZN,” said Naidoo.
She said a successful WEATHER project would also serve as a valuable model for other regions facing similar climate-related risks, potentially contributing to broader improvements across South Africa and in other low- and middle-income countries.
Professor Fiona Henriquez, co-investigator on the WEATHER project and Associate Dean of Research and Innovation in the School of Health and Life Sciences at UWS said, "This is an important project which will support areas in eThekwini, South Africa grappling with increasingly unpredictable weather patterns due to climate change, leaving communities exposed to floods, waterborne diseases, and disrupted healthcare services.”
The researchers say that geospatial technologies will collate weather and climate data while AI algorithms will analyse weather, climate and health data to identify patterns and trends which allow for more accurate predictions of flooding and associated health risks. In addition, AI-powered tools would continuously monitor and assess flood risks in real-time, enabling swift interventions to minimise disease outbreaks.
Warnings would be shared through a mobile app providing more detailed information, maps, and resources tailored to specific locations and needs and through SMS text messages, ensuring widespread reach even in areas with limited internet access.
Professor Michael Gebreslasie, Associate Professor at UKZN said: “Geospatial technology offers perspective on evolving weather patterns by visualising historic data and overlaying it with current records and helps to identify significant changes in precipitation and flood risks to human health. “Furthermore, it supports the development of predictive models that can forecast future climate scenarios. Informative visualisation shows these developments in ways scientists and non-experts can understand the risk.”
Researchers say the system will use the existing mobile phone infrastructure in the communities and warnings will be disseminated through SMS text messages ensuring widespread reach even in areas with limited internet access.
There will also be mobile applications which will provide more detailed information, maps, and resources tailored to specific locations and needs.
According to Professor Mohammed Zeeshan Shakir, School of Computing, Engineering and Physical Sciences at UWS,“We can't change the weather, but we can change how we respond to it.
“By integrating AI into our warning system, we're empowering healthcare and communities with real-time predictive information, giving them the time they need to prepare and reduce the harm from extreme events.”