TL;DR
The Alan Turing Institute is leading Project Cumulus, a new initiative funded by the Gates Foundation and the UK’s Foreign, Commonwealth & Development Office. The project will develop AI-based weather forecasting systems specifically designed for West African conditions, helping farmers make better decisions about planting and harvesting.
Addressing a Critical Gap
Weather forecasting in West Africa currently presents unique challenges. As a region increasingly vulnerable to climate change effects, unpredictable weather patterns directly impact food security and economic stability. Farmers often lack access to adequate forecasts, forcing them to make critical decisions about crops without the information they need.
Global forecasting systems like the Integrated Forecasting System have achieved remarkable accuracy overall, but in sub-Saharan Africa—where weather systems behave differently and local observations are limited—forecasts can still be unreliable. The way Earth rotates and solar heating drives the atmosphere means African weather behaves differently from mid-latitudes like Europe and the US, where most forecasting methods were designed.
AI-Powered Solutions
Project Cumulus will draw on emerging technologies including Aardvark Weather and the Aurora Earth System Foundation Model to develop new AI-based forecasting methods tailored to African conditions.
Unlike traditional physics-based weather prediction systems, Aardvark is fully AI-driven, combining satellite imagery, ground observations and existing forecast data to create a clearer atmospheric picture. It can learn from data-rich regions to improve predictions where data is scarce.
Professor Richard Turner from the University of Cambridge noted: “AI is both revolutionising and democratising weather prediction. Tasks that once required a supercomputer can now run on a laptop, producing accurate forecasts in a fraction of the time and cost.”
International Partnership
The initiative brings together UK institutions—the Alan Turing Institute, University of Cambridge and University of Leeds—with Kwame Nkrumah University of Science and Technology in Ghana, University Cheikh Anta Diop in Senegal, and both countries’ meteorological agencies.
Crucially, these systems are designed to be affordable and adaptable, enabling West African partners to produce their own forecasts, build expertise and drive local innovation.
Looking Forward
Dr Scott Hosking, Mission Director for Environmental Forecasting at the Turing Institute, emphasised the project’s collaborative approach: “To protect lives and livelihoods in these regions, we cannot rely on off-the-shelf AI solutions. This partnership focuses on developing AI weather models specifically designed for conditions in West Africa.”
The project represents a promising model for how AI can be developed collaboratively to address real-world challenges in regions that have historically been underserved by technology.