Predicting East Africa's rainfall extremes with calibrated, hybrid physical and AI systems
Research Square
Download
Abstract: Accurate and reliable rainfall forecasts are crucial for Early Warning Systems (EWS). This study presents a low-cost, fine-tuning method that calibrates ensemble forecasts to probabilities of exceedances for rainfall extremes, applied to physical and hybrid artificial intelligence rainfall forecasts over East Africa. Case-study results demonstrate encouraging skill improvement from hybrid AI systems at higher rainfall thresholds, with sub-regional analysis showing hybrid AI systems better reflecting observed chances of extremes at longer lead times. The research decomposes forecast skill across rainfall and probability thresholds to highlight ranges where actionable improvement arises, whether using satellite or rain-gauge data as reference. This tailored evaluation provides guidance for forecast adoption in Early Warning Systems and Anticipatory Action, particularly relevant for resource-constrained settings across vulnerable East African regions affected by rainfall extremes and related hazards.