Suitable Measures for the Potential Operational Utility of AI NWP Rainfall Forecasts Over Africa

2026-09-24

AI Summary

This study addresses the forecasting challenges posed by limited numerical weather prediction infrastructure and intensifying extreme rainfall in Africa by conducting the first systematic evaluation of AI models, such as GraphCast, against the Integrated Forecasting System (IFS) for precipitation prediction over complex African terrain. Methodologically, isotonic distributional regression is employed for post-processing calibration, complemented by a multidimensional assessment utilizing the Continuous Ranked Probability Score (CRPS) and multi-source satellite data. Results indicate that AI models achieve approximately 5% higher median accuracy than IFS, demonstrating superior performance in humid regions, whereas the physics-based model retains advantages in arid, high-elevation areas. These findings confirm the operational potential of low-cost, highly efficient AI-driven rainfall forecasting across Africa.

Abstract

Artificial intelligence (AI)-based weather prediction is approaching the skill of physical numerical weather prediction (NWP) systems at a fraction of the computational cost. This is particularly promising for Africa, where rainfall extremes are intensifying and many forecasting centres lack the infrastructure to run physical models at extended lead times. We present a calibrated comparison of GraphCast, GenCast and the Functional Generative Network (FGN) against the physical NWP model IFS for rainfall prediction across Africa. Deterministic and probabilistic forecasts are postprocessed using Isotonic Distributional Regression and evaluated with the Continuous Ranked Probability Score against IMERG, RFEv2 and CHIRPS across seasons, wet and dry regimes, elevation zones and lead times. All models retain skill beyond climatology across most seasons and at extended lead times. AI models generally outperform IFS in wet regions, whereas IFS performs better in dry, high-elevation areas, where its finer resolution better represents orographic controls on rainfall. Across observational datasets and seasons, AI models achieve a median improvement of approximately 5% over IFS. GraphCast achieves calibrated skill comparable to the ensemble-based FGN, although FGN provides greater significant skill at longer lead times. These results highlight the potential of calibrated AI weather prediction to provide accessible and computationally efficient rainfall forecasts across Africa, while demonstrating the continuing importance of spatial resolution, ensemble design and regional characteristics.

Problem

• AI weather prediction
• rainfall forecasting
• Africa
• numerical weather prediction
• operational utility

Innovation

• AI weather prediction
• Isotonic Distributional Regression
• rainfall forecasting
• calibrated comparison
• Africa