An artificial intelligence system developed at the University of Miami is giving hurricane forecasters a powerful new tool to track tropical weather patterns, now officially in use at the National Hurricane Center (NHC) for the 2025 Atlantic hurricane season.
The technology, designed by Ph.D. student Will Downs at the Rosenstiel School of Marine, Atmospheric, and Earth Science, marks the first automated method to distinguish between tropical easterly waves (TEWs) the clusters of clouds and winds that often evolve into hurricanes and two other major tropical wind patterns: the Intertropical Convergence Zone (ITCZ) and the monsoon trough (MT).
“With this wave tracking tool, we have a new way to detect different patterns, and the types of systems that can grow into hurricanes,” Downs said. “It’s one important step toward improving forecasts and giving communities more time to prepare.”
How the AI model works
To build the system Downs and his collaborators used four decades of weather data, spanning 1981 to 2023, training convolutional neural networks (CNNs) to detect and separate TEWs, the ITCZ, and the MT. The model merges historical records from the NHC’s Tropical Analysis and Forecast Branch with reanalysis data of past climate conditions, allowing forecasters to track these systems in real time.
The AI tool has already proven reliable. “It has captured the waves where they seem to be going, with remarkable accuracy so far,” said Professor Sharan Majumdar, Downs’ advisor at the Rosenstiel School. “The robust dataset it produces will help researchers more effectively study the behavior of these waves from weak clusters of clouds to developing tropical cyclones.”
New insights into tropical systems
Early findings from the system show that tropical waves in the Caribbean are typically weaker than those in the open Atlantic, though still trackable with AI. It also identified a westward expansion of the monsoon trough in the Atlantic over recent decades, along with noticeable shifts in the Pacific during strong El Niño events.
The project benefited from collaboration with fellow Ph.D. student Aidan Mahoney, who interns at the NHC. “What started as a quick question about tropical wave analysis turned into many long discussions about the complexities of tropical wave dynamics,” Mahoney said. “Will developed an expert understanding of the training data, which allowed him to create the best possible version of the tracker.”
For Downs, the work is personal. His fascination with storms began during Hurricane Katrina, when his family evacuated from New Orleans to rural Louisiana. After Hurricane Isaac struck in 2012, he began tracking storms online, a passion that eventually led him to pursue a doctorate focused on cyclogenesis the process by which hurricanes form and intensify.
With the AI tracker now in the hands of NHC forecasters, scientists say the tool could become a game-changer in helping predict and prepare for powerful storms like Hurricane Beryl, which originated from a tropical wave off the African coast in June 2024 before rapidly intensifying.