New AI-Powered Hurricane Prediction Model Developed
Developed in collaboration with Google DeepMind, the National Hurricane Center, and Colorado State University, a new artificial intelligence model can predict hurricane tracks, intensity, and wind fields up to 15 days in advance.
Google scientists have developed a new artificial intelligence-powered model capable of predicting the track, intensity, and wind field of hurricanes up to 15 days in advance. This innovative system is expected to give meteorologists approximately one extra day of lead time in their forecasts.
The Role of Artificial Intelligence in Hurricane Forecasting
Led by Google DeepMind, the study designed an advanced artificial intelligence model capable of determining the path and intensity of hurricanes in advance. Thanks to this model, forecasting experts can access more detailed information regarding the paths that storms will follow.
Forecast Accuracy and Extra Time Advantage
Tests showed that the three-day forecasts generated by the developed artificial intelligence model have nearly the same accuracy as the two-day forecasts of the currently best forecasters. This gained extra day is anticipated to play a critical role in emergency preparations.
Big Data Sets and Risk Analysis
The new model can generate 1,000 different possible forecasts, compared to an average of 50 in existing systems. This massive dataset enables meteorologists to more easily identify low-probability yet hazardous scenarios.
Interinstitutional Collaboration and Publication Process
Conducted in partnership with Google DeepMind, the National Hurricane Center in Miami, and Colorado State University, this significant research was published in the journal Nature on August 6.