A new paper on artificial intelligence in forecasting, authored by the NOAA National Severe Storms Laboratory (NSSL) and the Cooperative Institute for Severe and High-Impact Weather Research and Operations (CIWRO), was selected by the American Geophysical Union (AGU) and the editors of Geophysical Research Letters to be featured as a Research Spotlight on EOS.org.
Right: Actual real-world thunderstorm reflectivity
NSSL and CIWRO’s submission discusses a new artificial intelligence model, WoFSCast, that can accurately predict how storms will evolve up to two hours in advance. The process of generating a prediction takes less than a minute, making it significantly faster than current systems.
“The cheaper and faster forecasts enabled by WoFSCast will further extend forecast and warning lead times based on the Warn-on-Forecast concept,” said Adam Clark, acting NSSL Branch Director.
By using an AI-enhanced system, the National Weather Service may be able to issue severe weather warnings earlier and reduce the harm caused by these extreme weather events.
AGU Research Spotlights summarize the research and findings of the best accepted articles for the broad Earth and space science community.
