GenCast surpasses the accuracy of the European Center for Medium-Range Weather Forecasts by more than 97%, Google assures,
An artificial intelligence developed by Google is capable of making 15-day weather forecasts with unparalleled accuracy, saving lives as climate disasters multiply, the US Internet giant announced Wednesday.
The GenCast model devised by DeepMind, Google’s artificial intelligence research lab based in London, “showed better predictive abilities” than the current reference model, according to Google. The European Center for Medium-Range Meteorological Predictions currently produces forecasts for 35 countries and is considered the global benchmark for meteorological accuracy. But now GenCast, Google claims, outperforms the Centre’s predictions in more than 97% of 1,320 climate disasters recorded in 2019, on which the two models were tested.
Google’s model, developed over four decades from 1979 to 2018 based on certain data – temperature, wind speed and atmospheric pressure – can produce a 15-day forecast in just eight minutes, compared to a time of hours, valid today. “GenCast provides better forecasts of everyday weather and extreme events… up to 15 days in advance,” Google said in a statement.
According to Google, GenCast “systematically outperformed” the current reference model for predicting extreme heat, extreme noise and high wind speed. “More accurate predictions of the risks of extreme weather conditions can thus help authorities protect more lives, avoid significant property damage and save money,” Google underlines.
