Huang Cheng, Mu Pan, Zhang Jinglin, Chan Sixian, Zhang Shiqi, Yan Hanting, Chen Shengyong, Bai Cong
College of Computer Science, Zhejiang University of Technology, Hangzhou, China.
School of Control Science and Engineering, Shangdong University, Jinan, China.
Nat Commun. 2025 Jul 1;16(1):5923. doi: 10.1038/s41467-025-61087-4.
Accurate tropical cyclone (TC) forecasting is critical for disaster prevention. While deep learning shows promise in weather prediction, existing approaches demonstrate limited accuracy in TC track and intensity forecasting, hindered by the lack of open multimodal datasets and insufficient integration of meteorological knowledge. Here we propose TropiCycloneNet containing TCN - a open multimodal TC dataset spanning six major ocean basins with 70 years of multi-source data, and TCN - an AI-meteorology integrated prediction model including multiple modules such as Generator Chooser Network and Environment-Time Net. Comprehensive evaluations demonstrate that TCN outperforms both existing deep learning methods and official meteorological forecasts across multiple metrics. This advancement stems from synergistic optimization of our meteorologically-informed architecture and the dataset's comprehensive spatiotemporal coverage. The released resources and method can attract more researchers to the field, thereby accelerating data-driven tropical cyclone prediction research.
准确的热带气旋（TC）预报对于防灾至关重要。虽然深度学习在天气预报中显示出前景，但现有的方法在TC路径和强度预报中显示出有限的准确性，受到缺乏开放的多模态数据集以及气象知识整合不足的阻碍。在此，我们提出了热带气旋网络（TropiCycloneNet），它包含TCN——一个跨越六个主要海洋盆地、拥有70年多源数据的开放多模态TC数据集，以及TCN——一个人工智能与气象学集成的预测模型，包括生成器选择网络和环境-时间网络等多个模块。综合评估表明，TCN在多个指标上均优于现有的深度学习方法和官方气象预报。这一进展源于我们基于气象知识的架构的协同优化以及数据集全面的时空覆盖。所发布的资源和方法可以吸引更多研究人员进入该领域，从而加速数据驱动的热带气旋预测研究。
