Delivery of Services towards the Strengthening Early Warning Systems for Anticipatory Action using Machine Learning Model

Region Africa, East Asia
Years 2026–2028
Partners World Food Programme
Delivery of Services towards the Strengthening Early Warning Systems for Anticipatory Action using Machine Learning Model

This initiative strengthens early warning systems across Africa and East Asia by integrating AI and machine learning with traditional climate models to improve forecasting, support anticipatory humanitarian action, and build resilience among vulnerable communities.

As a core implementing partner, we are advancing AI and ML-enabled climate and weather intelligence by integrating physics-based and data-driven models to strengthen early warning systems, risk assessment, and anticipatory humanitarian action across Africa and Eastern Asia.

Data Integration

Integrate climate, satellite, hazard, geospatial, and socio-economic datasets with physics-based models to enhance representation of extreme events and enable accurate, impact-based forecasting.

Numerical Weather Prediction (NWP) Modelling and Multi-model Integration:

Combine NMHS models with geospatial and socio-economic data to deliver localised, impact-based early warnings through a modular, API-driven system integrated with national and WFP platforms.

Model Training & Calibration

Train and calibrate hybrid AI/ML models using multi-decadal datasets and physics-based outputs to generate reliable forecasts with validation and uncertainty analysis, improving predictive accuracy to minimise forecast uncertainty and associated Loss and Damage risks.

Documentation & Training

Provide user-friendly documentation and capacity-building support to ensure effective adoption, integration, and long-term sustainability of the system by NMHSs and regional partners, strengthening institutional capacity for preparedness, response, and management of residual Loss and Damage.

Impact Metrics

USD 1.5 Bn

in potential Loss and Damage costs that could be averted

100 Mn

people in climate-sensitive areas expected to benefit from enhanced climate resilience and early warning coverage

Multi-country early warning coverage

across Africa and Eastern Asia through AI - ML enabled forecasting systems