Researchers eye AI revolution in natural disaster forecasts
A revolution in forecasting natural disasters is underway, say researchers in Switzerland who are training AI models using vast troves of NASA climate data to produce potentially lifesaving data at lightning speed.
Researchers in Switzerland are harnessing AI models, trained on massive NASA climate data, to revolutionize natural disaster forecasting. By feeding petabytes of climate data into one of the world's most powerful supercomputers, artificial intelligence can rapidly and comprehensively analyze weather patterns that would otherwise go unnoticed.
This development holds significant potential for early warning systems, potentially saving lives in events like Nepal's catastrophic flood last month, which left thousands dead or missing. However, the training of such complex AI models necessitates substantial high-quality climate and Earth observation data, along with immense computing power.
At Switzerland's Federal Institute of Technology Zurich (ETH), around 100 petabytes of NASA data have been copied onto servers adjacent to the Alps supercomputer. This immense dataset, equivalent to 20 million feature-length films, enables scientists to develop AI models capable of generating multi-day global forecasts in just a minute, a drastic improvement from the traditional process that once took hours.
Written by urgent.news from Phys.org's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
