‘Current weather forecast models can’t capture the intensity of the most extreme rainfall events,’ says Niklas Boers from the Potsdam Institute for Climate Impact Research, ‘yet these events are the most dangerous and can have severe impacts such as major floods or even landslides.’
Boers and his colleagues used 50,000 high-resolution satellite weather pictures gathered over 15 years and tracked weather patterns from Buenos Aires to the mountains. The result was a model that successfully predicted 90 per cent of extreme weather events in the Andes during the El Niño phenomenon.
Winter monsoons in the Andes hit the mountain range, turn south and encounter cold, dry air sweeping up from the south. The result is high rainfall in the foothills, which are densely populated areas in Argentina and Bolivia.
Comparing the weather events was simple, but the team also devised a mathematical tool to identify the connections that led to weather extremes. ‘The data was there, but nobody had joined the dots like this before,’ said researcher Jürgen Kurths. ‘The method provides a general framework that could now be applied to forecast extreme changes over time in a series of other complex systems. This could be financial markets, brain activity, or even earthquakes.’