The forecast looks good for this new weather program.
Google DeepMind, the search engine’s AI-focused lab, has introduced a revolutionary weather forecasting model that is said to deliver 10-day predictions in less than a minute, “with unprecedented accuracy.”
GraphCast Outperforms traditional and high-tech weather models by at least 90% as it emphasizes historical weather data, according to a research paper about the AI model published on Tuesday.
Google touts over a million global grid points for hyperlocal data. GraphCast requires two pieces of information: the current weather and six hours ahead, as it predicts weather six hours into the future and up to 10 days ahead.
“We believe this marks a turning point in weather forecasting,” the Science article reads, “helping to open new avenues for strengthening the breadth of climate-dependent decision-making by individuals and industries.” , by making cheap predictions more accurate, more accessible, and suitable for specific applications.”
The model was tested with the European Center for Medium-Range Weather Forecasts’ gold standard system, called high-resolution forecasting (HRES).
GraphCast was found to be 99.7% more accurate in some cases.
The ECMRWF is I’m already experimenting with the tool.thanks to open source coding that anyone can use.
Google also touts the system as innovative in detecting extreme weather events before they arrive.
During training, it was able to more accurately predict the movement of cyclones than HRES. In September, he also predicted that Hurricane Lee would reach Nova Scotia nine days before making landfall.
Traditional models took three days longer, Google said.
Additionally, GraphCast excels at flood prediction through atmospheric data on rainfall patterns.
What differentiates this interface from traditional approaches is that it does not run algorithms or mega-equations like those on supercomputers. Instead, deep learning is used.
“Deep learning offers a different approach: using data instead of physical equations to create a weather forecasting system,” wrote Remi Lam, a member of the GraphCast team. “GraphCast draws on decades of historical weather data to learn a model of the cause-and-effect relationships that govern how Earth’s climate evolves, from the present into the future.”
Meanwhile, Google recently launched a new, separate 24-hour forecast model. called MetNet-3 that also offers incredible precision.
“Pioneering the use of AI in weather forecasting will benefit billions of people in their daily lives,” Lam added. “But our broader research is not just about anticipating the weather, it’s about understanding the broader patterns of our climate.”
The use of AI in weather forecasting is something that experts have called a “quiet revolution” that is substantially changing the industry. Science.org reported.
“It’s very, very exciting to know that we can generate global predictions that are skillful and really economical,” María Molina, a research meteorologist focused on artificial intelligence at the University of Maryland, told the outlet on Tuesday.
And this is just the beginning, says Christopher Bretherton, an atmospheric scientist at the Allen AI Institute.
The institute’s 40 years of training data could be instrumental in catalyzing new models on which the AI of the future would be trained, Bretherton says, “and then running them 100 times faster.”
