A new artificial intelligence model born out of a collaboration between the National Oceanic and Atmospheric Administration (NOAA), Google DeepMind, and Google Research is aiming to improve tropical weather forecasting.
The effort is led by NOAA’s National Hurricane Center (NHC) and will accelerate how quickly NOAA can evaluate, integrate, and improve AI-generated tropical cyclone forecasts with the goal of increasing public safety and preparedness, according to the July 2 announcement.
“The pace of weather modeling innovation is increasing and Google is a stellar partner in AI weather model development,” said Michael Brennan, director of NHC. “This collaboration between NOAA and Google will ensure that NOAA’s National Hurricane Center is able to quickly evaluate new tropical cyclone forecasting technology as it arises.”
The AI model will provide near-real-time AI tropical cyclone forecasts to NHC, while Google said it will host and maintain a visualization platform called “Weather Lab” to view the AI outputs. NOAA will routinely evaluate the model and said it will share progress and results with the public and scientific community.
“The partnership will put new models in the hands of the NHC forecasters, who directly communicate hazard risk to the public and partners to enable life saving actions,” reads the announcement.
“We are honored to collaborate with NOAA, bringing the power of AI models to the U.S. National Weather Service agency,” said Ferran Alet, research scientist at Google DeepMind. “We look forward to working together on the agency’s smooth integration of AI models into NHC operations to enhance their track and intensity forecasts.”
The announcement came the same week that flash floods hit central Texas, resulting in at least 90 confirmed deaths. Reports appear to blame insufficient and inaccurate forecasts, while some say staffing cuts and funding shortages at the National Weather Service by the Trump administration are to blame.