A new artificial intelligence model has the potential to provide an extra day of warning for tropical cyclones, according to research led by Florent Alet. The model improves upon existing forecasting methods by more accurately predicting cyclone tracks.
Alet, and colleagues, developed the AI model to enhance the prediction of tropical cyclone movement. Current forecasting methods often struggle to accurately predict cyclone paths beyond a certain timeframe. This new model aims to extend the reliable forecast range by approximately 24 hours.
The research builds upon previous studies by K.R. Knapp, M.C. Kruk, D.H. Levinson, H.J. Diamond, and C.J. Neumann (2010); Zhang, Z. et al. (2023); and others. The model’s development also incorporates insights from work done by Mooney, K.R. et al. (2026), Gomez, M., Poulain-Auzéau, L., Berne, A. & Beucler, T. (2026), and Selz, T. & Craig, G.C. (2026). Further research by Sun, Y.Q. et al. (2025) and Knutson, T. et al. (2020) also informed the project. McGovern, A. et al. (2024) contributed to the understanding of relevant meteorological phenomena. The authors declare no competing interests.
閱讀原始報道
💬 评论
📜 评论政策