Hurricane model developed by Google boosted with artificial intelligence showing promise

An artificial intelligence model could give hurricane forecasters another day’s information about where a storm is going, how strong it may become, and how far its damaging winds may extend.

Google scientists have led the effort to build the model that can forecast storm tracks, intensity, and wind-field as far as 15 days ahead.

Tests found its three-day forecasts were about as accurate as two-day forecasts from top current top forecasters, who were involved in the effort.

That extra day could influence Gulf Coast evacuation decisions and emergency preparations, particularly when a hurricane strengthens rapidly.

The model can also produce as many as 1,000 possible forecasts, compared with a typical 50 now. Such a huge dataset could help meteorologists identify dangerous outcomes, however unlikely, that smaller groups may miss.

The research, led by Google DeepMind, the National Hurricane Center in Miami, and Colorado State University, was published Aug. 6 in Nature.

Environmental reporting for WGCU is funded in part by VoLo Foundation, a nonprofit with a mission to accelerate change and global impact by supporting science-based climate solutions, enhancing education, and improving health.

WGCU is your trusted source for news and information in Southwest Florida. We are a nonprofit public service, and your support is more critical than ever. Keep public media strong and donate now. Thank you.