AI Predicts Extreme Weather to Save Lives in Africa

Extreme weather events such as floods, droughts, and wildfires are becoming more frequent and severe due to climate change, impacting countries globally. In the Horn of Africa, heavy rains often lead to devastating localized flooding, affecting livelihoods and causing fatalities.
To address this, scientists are employing artificial intelligence (AI) to predict extreme weather in the region, potentially saving lives. The United Nations World Food Programme is collaborating with researchers from the University of Oxford’s physics department to develop an AI-based weather forecasting system using machine learning.
This hybrid approach leverages AI to fill gaps in physical models, representing complex processes through data-driven methods. By analyzing satellite data of cloud formations and temperatures, the AI makes predictions about upcoming extreme weather events. The system is trained on historical observational data, constantly improving as it rewards accurate predictions and penalizes inaccurate ones.
Unlike costly supercomputers used in countries like the United Kingdom, the AI forecasting code from Oxford can run on a laptop, making it accessible for developing nations. Currently, the pilot program is operating in Kenya and Ethiopia, with plans to expand across the region. If successful, this technology could be deployed in other parts of the world severely affected by climate change.
With a 48-hour warning from the AI forecasts, those at risk can be alerted via text messages, emails, and broadcasts, providing crucial time to prepare and potentially save lives.