The federal government’s premier weather-focused science agency has launched an array of new artificial intelligence-powered weather models that the agency said marks the first-ever deployment of a “hybrid physical-AI ensemble system” for weather forecasting.
The Commerce Department’s National Oceanic and Atmospheric Administration (NOAA) component unveiled the new suite of global models in a Dec. 17 announcement.
The new models, the agency said, offer “a significant advancement in forecast speed, efficiency, and accuracy.”
“The models will provide forecasters with faster delivery of more accurate guidance, while using a fraction of computational resources,” NOAA said.
“NOAA’s strategic application of AI is a significant leap forward in American weather model innovation,” said NOAA Administrator Neil Jacobs.
“These AI models reflect a new paradigm for NOAA in providing improved accuracy for large-scale weather and tropical tracks, and faster delivery of forecast products to meteorologists and the public at a lower cost through drastically reduced computational expenses,” Jacobs said.
NOAA pointed to three applications for the new suite of AI weather models.
The first is the Artificial Intelligence Global Forecast System (AIGFS), which it said “implements AI to deliver improved weather forecasts more quickly and efficiently (using up to 99.7% less computing resources) than its traditional counterpart.”
The second is the Artificial Intelligence Global Ensemble Forecast System (AIGEFS). That system, the agency said, “provides a range of probable forecast outcomes to meteorologists and decision-makers” that in testing has shown “improved performance over the traditional GEFS, extending forecast skill by an additional 18 to 24 hours.”
Finally, NOAA pointed to the Hybrid-GEFS (HGEFS) model, which it called a “pioneering, hybrid ‘grand ensemble’ that combines the new AI-based AIGEFS with NOAA’s flagship ensemble model, the Global Ensemble Forecast System.”
“Initial testing shows that this model, a first-of-its kind approach for an operational weather center, consistently outperforms both the AI-only and physics-only ensemble systems,” the agency said.