Using weather satellite observations and AI to study hurricanes - Department of Atmospheric Science | Colorado State University
Using weather satellite observations and AI to study hurricanes - The Department of Atmospheric Science at Colorado State University is a leading global institution focusing on graduate education, cutting-edge research, and public service.
July 22, 2026Alum Muhammad Naufal Razin uses machine learning to identify precipitating cloud types from weather satellite observations of hurricanes, or tropical cyclones. “The work provides a tool that could help us study the relationship between precipitating cloud types and tropical cyclone intensity and structure,” said Razin. Razin is a Research Scientist at Colorado State University’s Cooperative Institute for Research in the Atmosphere. Working with CSU Professor Michael Bell and NOAA NESDIS Physical Scientist Chris Slocum, Razin recentlypublished the work in the journal Artificial Intelligence for the Earth Systems. The research team developed a machine learning algorithm to identify precipitating cloud types, such as strong localized convective and widespread stratiform rain, from satellite passive microwave observations of tropical cyclones. Passive microwave observations capture the vertical structure of water vapor and precipitation. Previous studies show that the precipitating cloud type is related to tropical cyclone intensity and structure. The team hopes to use the algorithm developed in this work to perform a comprehensive study on the relationship between the precipitating cloud types and tropical cyclone intensity and structure. In the paper, the team also highlights variables and observing frequencies that are important for a skillful classification of precipitating cloud types, providing insights for the design and development of future satellite passive microwave sensors. Read more onCIRA’s science news page. Image: Satellite passive microwave brightness temperature of Hurricane Beryl, 2024