The deluge was driven by a slow-moving cut-off low pressure system that remained anchored over the region, repeatedly drawing in moisture and triggering intense downpours, according to Ishani Mistry and Ollie Lewis. The flood warning was raised to the highest level as roads were washed away and rain forced the evacuation of Kruger National Park.

The meteorological phenomenon behind this extreme weather event, a cut-off low pressure system, is characterized by its detachment from the main westerly air current. This isolation allows it to linger over a specific area, drawing in moisture and leading to prolonged periods of heavy rainfall. The predictability of these events is an area of ongoing research in climate science, with AI playing an increasingly important role. Machine learning algorithms are being used to analyze vast datasets of historical weather patterns to improve the accuracy of forecasting models. These models can identify subtle indicators that might be missed by traditional forecasting methods, potentially providing earlier and more precise warnings about impending extreme weather events.

The implications of such extreme weather events extend beyond immediate damage and displacement. Prolonged flooding can contaminate water supplies, leading to outbreaks of waterborne diseases. It can also disrupt agricultural production, leading to food shortages and economic hardship. The ability to accurately predict and prepare for these events is therefore crucial for mitigating their impact on society.

Looking ahead, further heavy rainfall is expected on Friday and over the weekend. Maputo, Mozambique's capital, could expect daily rainfall totals to exceed 200mm by the end of Friday, according to forecasts. Authorities are urging residents to take precautions and evacuate from low-lying areas. The situation remains critical, and ongoing monitoring and response efforts are essential to minimize the impact of this extreme weather event.