In the 1960s, weather scientists found that the chaotic nature of Earth’s atmosphere would put a limit on how far into the future their forecasts might peer. Two weeks seemed to be the limit. Still, by the early 2000s, the great difficulty of the undertaking kept reliable forecasts restricted to about a week.
Now, a new artificial intelligence tool from DeepMind, a Google company in London that develops AI applications, has smashed through the old barriers and achieved what its makers call unmatched skill and speed in devising 15-day weather forecasts. They report in the journal Nature on Wednesday that their new model can, among other things, outperform the world’s best forecasts meant to track deadly storms and save lives.
“It’s a big deal,” said Kerry Emanuel, a professor emeritus of atmospheric science at the Massachusetts Institute of Technology who was not involved in the DeepMind research. “It’s an important step forward.”
In 2019, Emanuel and six other experts, writing in the Journal of the Atmospheric Sciences, argued that advancing the development of reliable forecasts to a length of 15 days from 10 days would have “enormous socioeconomic benefits” by helping the public avoid the worst effects of extreme weather.
Ilan Price, the new paper’s lead author and a senior research scientist at DeepMind, described the new AI agent, which the team calls GenCast, as much faster than traditional methods. “And it’s more accurate,” he added.
He and his colleagues found that GenCast ran circles around DeepMind’s previous AI weather program, which debuted in late 2023 with reliable 10-day forecasts. Rémi Lam, the lead scientist on that project and one of a dozen co-authors on the new paper, described the company’s weather team as having made surprisingly fast progress.
Related
Can AI spot the next bomb cyclone far in advance? Microsoft hopes so
More
“I’m a little bit reluctant to say it, but it’s like we’ve made decades worth of improvements in one year,” he said in an interview. “We’re seeing really, really rapid progress.”
The world leader in atmospheric prediction is the European Center for Medium-Range Weather Forecasts. Comparative tests regularly show that its projections exceed all others in accuracy.
DeepMind tested its new AI program against the center’s Ensemble Prediction System — a service that 35 nations rely on to produce their own weather forecasts. The team compared how the 15-day forecasts of both systems performed in predicting a designated set of 1,320 global wind speeds, temperatures and other atmospheric features.
The Nature report said the new agent outdid the center’s forecasts 97.2% of time. The AI achievement, the authors wrote, “helps open the next chapter in operational weather forecasting.”
Matthew Chantry, an AI specialist at the European Center for Medium-Range Weather Forecasts, said his agency was already adopting some of its features.
“That’s how highly we think of it,” he said. Machine learning in general, Chantry added, was accelerating human bids to outmaneuver some of nature’s deadliest threats.
DeepMind’s weather advance comes two months after other AI researchers in the company shared the Nobel Prize for chemistry. The scientific news forms a bright counterpoint to public fears of AI stealing jobs and driving humans to the edge of obsolescence.
The natural chaos in Earth’s atmosphere means that all weather forecasts, including the two-week variety, grow less reliable as they peer further into the future. Even so, AccuWeather offers 90-day forecasts. And the Old Farmer’s Almanac says it can gaze ahead 60 days.
DeepMind backs its 15-day declaration with pages of evidence laid out in one of the world’s leading science journals, Nature. So too, Google posted an online blog that details the AI advance.
The new GenCast agent takes a radically different approach from mainstream forecasting, which uses room-size supercomputers that turn millions of global observations and calculations into predictions. Instead, the DeepMind agent runs on smaller machines and studies the atmospheric patterns of the past to learn the subtle dynamics that result in the planet’s weather.
The DeepMind team trained GenCast on a massive archive of weather data curated by the European center. The training period went from 1979 to 2018, or 40 years. The team then tested how well the agent could predict 2019’s weather.
Such training empowers all types of generative AI — the kind that’s creative. Mimicking how humans learn, it spots patterns in mountains of data and then makes new, original material that has similar characteristics.
Lam of DeepMind noted that GenCast’s generative skills were rooted in factual data gathered from nature rather than the internet, notorious for its confusing mix of facts, biases and fallacies. “We have a ground truth,” he said of its dependence on natural phenomena. “We have a reality check.”
The new agent’s forecasts are probabilistic — like those on the weather apps of smartphones. For instance, GenCast can give a range of percentages for the likelihood of rain in a specific region on a given day.
In contrast, its DeepMind predecessor, GraphCast, offers a single forecast for a particular time and location. Known as deterministic, its method is essentially a best guess that gives no indication of the prediction’s uncertainty.
Probabilistic forecasts are considered more nuanced and sophisticated than the deterministic kind, and are more difficult to create. Typically, a GenCast forecast draws from a set of 50 or more predictions that produce its range of probabilities.
Despite all the effort that goes into those calculations, Price of DeepMind said, the new agent can generate a 15-day forecast in minutes compared with hours for a supercomputer. That can make its projections much timelier — an advantage in tracking fast-moving storms.
GenCast, the team says, can predict with great accuracy the paths of hurricanes, which annually can take thousands of lives and rack up hundreds of billions of dollars in property damage. The Nature paper said comparative testing showed that its hurricane track predictions consistently outdid those of the European center.
Emanuel of MIT said the DeepMind team failed to mention that its new agent provides little information about hurricane intensity.
Price, the paper’s lead author, concurred. He said the problem lay in training data limitations on hurricane wind speed. The weather team, he added, was confident it could devise a solution.
GenCast will most likely complement current methods rather than replace them, Emanuel argued. Each type, he said, has its own strengths and weaknesses in predicting the riot of variable phenomena that constitute the weather.
“The status quo isn’t going to disappear,” Emanuel said. “Perhaps the two of them working together will prove to be the best way forward.”
For its part, the DeepMind team acknowledged its heavy reliance on the conventional world of weather readings — noting, for instance, how its AI training data comes from the giant European weather archive. Its computations also start with a snapshot of the world’s current weather, what the team calls initial conditions.
The team hopes that other weather experts will test its new technology. Price said that the DeepMind team would share online its AI agent and underlying computer code.
He added that GenCast’s weather predictions would soon be posted publicly on Google’s Earth Engine and Big Query, giving scientists access to the new forecasts.
“We’re excited for the community to use and build on our research,” Price said.
Chantry of the European center said Google and DeepMind might have hidden their AI advance behind a wall of corporate secrecy, using it “to make a better weather forecast for their own apps and telling no one how they did it.”
Instead, he added, the emerging field has embraced a public openness that’s helping “lots and lots of people engage in this revolution.”
This story was originally published at nytimes.com. Read it here.
The Seattle Times does not append comment threads to stories from wire services such as the Associated Press, The New York Times, The Washington Post or Bloomberg News. Rather, we focus on discussions related to local stories by our own staff. You can read more about our community policies here.