How AI Revolutionizes Weather Forecasting in Malawi

By: Mai-Linn Finstad Svehagen, MET Norway a nd Lucy Stone, ECMWF.

Climate change intensifies extreme weather events and the need for accurate and accessible forecasts is crucial. This year, El Niño is particularly strong, and statistically, the likelihood of extreme weather increases during such periods. This is concerning for a country like Malawi, which is highly dependent on the agricultural sector, which employs 80% of the population. Droughts, floods, and tropical cyclones threaten food security and public health, making early warnings about extreme weather vital. 

Through a collaboration between the Norwegian Meteorological Institute (MET Norway) and the European Centre for Medium-Range Weather Forecasts (ECMWF), AI-Based Weather Prediction (AIWP) can pave the way for early warning systems.

A pilot project in Malawi is now testing how Artificial Intelligence (AI) can help facilitate weather forecasts and warnings to the public. 

Closing the Gap in Weather Warning Systems

About 70% of countries globally now report having early warning systems capability according to the UN's Early Warnings for All (EW4All) initiative. This represents a significant increase since the initiative was launched in 2022. Malawi is amongst the countries preparing to implement an early warning system by the end of 2027. 

Since 2019, the Norwegian Meteorological Institute has been collaborating with the Department of Climate Change and Meteorological Services (DCCMS) in Malawi. Meteorologists in Malawi have gained access to and training in meteorological analysis systems and have developed a weather app based on data from Yr, a digital public good developed in Norway. 

MET Norway now provides Malawi with “Forecast-in-a-Box”, an all-in-one weather forecasting application developed by ECMWF. The application makes it possible to run AI-based weather forecasting models and can be deployed on a range of computing infrastructure. With this technology, Malawi can now run a version of the Bris weather forecasting model, tuned for Africa.

–Providing a portable forecasting system such as Forecast in a Box literally puts forecasting into the hands of those who need it. The system puts users in control, strengthens their sovereignty and enables them to generate the products they need. Our collaboration with meteorological services such as MET Norway and working closely all the way through with Malawi’s Department of Climate Change and Meteorological Services (DCCMS) has made this possible, and helped us steer the development of Forecast-in-a-Box, states Florian Pappenberger, Director-General of ECMWF.

AI-Powered Tools: A Game Changer for Malawi's Climate Resilience

AI has the potential to transform meteorology, but is not yet fully accessible to some of the regions most affected by climate change. This project aims to democratize forecasting by enabling DCCMS Malawi to run AI weather prediction systems locally and independently. These recent efforts include adapting the AI-based weather model Bris, developed by MET Norway, to Malawi’s needs, such as expanding geographic coverage to monitor tropical cyclones and generating high-resolution forecasts. 

In addition to Malawi, the same technology is being made available to several countries in Africa, and MET Norway is collaborating with Ethiopia, Mozambique, and Tanzania in addition to Malawi. The long-term goal is for all countries to gain access over time. The tools  and data introduced in the pilot project are free and open to facilitate sustainability and local governance.

–We have made significant progress in using AI-based models and tools for weather forecasting. The question is when these models and tools will become operational, and what services they will provide. MET Norway shows together with ECMWF and DCCMS Malawi how AI can support the UN's Early Warnings for All program and help countries with limited resources for traditional forecasting systems, states Roar Skålin, Director General of MET Norway. 

Empowering Forecasters, Empowering People 

Using advanced tools like the Bris AI-based weather prediction model from MET Norway and ECMWF’s Forecast-in-a-Box, the project equips DCCMS Malawi’s with AI-based forecasting capabilities. The goal is to improve early warning systems, enhance forecast accuracy, and empower local meteorologists through training and technology. 

Forecast-in-a-Box pilots will enable more accurate, timely and accessible forecasts that can better support decision-making in resource-limited and vulnerable regions.

– With the latest improvements we think the model is good enough for this pilot. We know there will be adjustments to be made, but a pilot gives us valuable feedback to further improve the AI weather model Bris, says Lene Østvand, researcher and project manager for METs capacity building project in Malawi. 

From Training to Independence: Building Local Forecasting Capacity

The road to full implementation involves continued capacity building, refining AI models, and integrating local data. The goal is to reach implementation within a year's time. The project will also focus on developing educational materials and ensuring sustainable governance to maintain public trust and operational continuity. 

– We are proud to be piloting and testing AI in weather forecasting in Malawi, and we look forward to implementing the Forecast-in-a-Box and Bris forecasting model to operationalize and integrate it into our meteorology and climate services. This collaboration, which ensures our involvement throughout the entire process, means that the capacity and knowledge remain with us, and we greatly appreciate that,” explains Director General Lucy Mtilatila.

This progress not only strengthens Malawi’s resilience to climate change, but also serves as a model for other countries. By making cutting-edge tools accessible, the EW4All initiative is bringing the vision of universal early warning systems closer to reality – saving lives and protecting livelihoods in the process.