Text size
ADDS reax from Manchester University professor
A new weather model based on artificial intelligence can provide 15-day forecasts with unparalleled accuracy and speed, a Google lab said, with potentially lifesaving applications as climate change accelerates.
GenCast, invented by London-based artificial intelligence research lab Google DeepMind, “demonstrated better forecasting capabilities” than the current world-leading model, the company announced Wednesday.
The European Center for Medium-Range Weather Forecasts (ECMWF) produces forecasts for 35 countries and is considered the global benchmark for weather accuracy.
But DeepMind said GenCast exceeded the center’s prediction accuracy in more than 97% of the 1,320 real-world 2019 scenarios they were both tested on.
Details of his findings were published in Nature, a leading scientific journal.
ECMWF head Florence Rabier told AFP that the project was a “first step” toward integrating AI into weather forecasting, but that it “was indeed a step in progress.” Before”.
At this point, it can be used to supplement their current models, she said.
“We are progressing year by year,” she added. “Any new methods that can improve and accelerate this progress are extremely welcome in the context of the extreme societal pressures of climate change.”
The model was trained on four decades of temperature, wind speed and air pressure data, from 1979 to 2018, and can produce a 15-day forecast in just eight minutes, compared to the hours it currently takes.
“GenCast provides better forecasts of daily weather and extreme events than the main operational system… up to 15 days in advance,” a DeepMind statement said.
DeepMind said GenCast “consistently outperformed” the current leading forecasting model in predicting extreme heat, extreme cold and high wind speeds.
“More accurate predictions of severe weather risks can help authorities protect more lives, avoid damage and save money,” DeepMind said.
Extreme weather conditions are becoming more frequent and more severe due to human-caused climate change.
In August 2023, a series of wildfires in Hawaii killed around 100 people. Authorities were criticized by residents who said they were not warned of the impending fire.
This summer, a sudden heat wave in Morocco killed at least 21 people over a 24-hour period. And in September, Hurricane Helene killed 237 people in Florida and other southeastern US states.
“I am confident that AI-based weather forecasting systems will continue to gradually improve in the future, including better forecasting of extreme events and their intensity, for which there is much room for improvement,” said David Schultz, professor of synoptic meteorology at the University of Manchester, who was not involved in the research.
But he added that these forecasting systems depend on weather forecasting models already in progress, such as the one operated by ECMWF.
jts/klm/bjt