复旦大学“大禹”辐射-云-降水分析系统

“大禹”辐射-云-降水分析系统（DaYu Radiation-Cloud-Precipitation Analysis Systems, DaYu-RCPAS）是由复旦大学大气与海洋科学系自主研发的辐射-云-降水特性综合分析平台，专注于辐射模拟计算，云和降水的实时监测、精准反演及短临预报，为天气气候研究和业务应用提供高精度数据支持。

该系统集成了团队“大禹”辐射传输模式（DaYu Radiation Transfer Model, DaYu-RTM）、“大禹”云分析系统（DaYu CLoud Analysis System, DaYu-CLAS）、“大禹”降水分析系统（DaYu PRecipitation Analysis System, DaYu‐PRAS）和“大禹”云图预报大模型技术。其中DaYu-RTM主要用于辐射模拟计算（W. Li et al., 2023）；DaYu-CLAS主要包括用于云物理特性反演的Cloud-ResUNet（Tong et al., 2023; Zhao et al., 2024），Cloud-SmaAtUNet（J. Li et al., 2023），DaYu-RTM与深度学习的混合算法（W. Li et al., 2022, 2024），CloudDiff（Xiao et al., 2025a）和Overlap‐CloudDiff（J. Li et al., 2026）模型，以及用于云物理特性短临预测的Cloud-FNO模型（Zhang et al., 2026）；DaYu-PRAS主要包括用于降水监测和短临预测的TPWDiff‐CB（Xiao et al., 2025b）和RePPIC‐Net （Yang et al., 2026）模型；“大禹”云图预报大模型主要用于亮温云图的短临预报。

目前，DaYu-RCPAS提供了中国新一代静止气象卫星风云四号AGRI成像仪多通道亮温、全天时云物理特性等实时监测和预报产品。该系统依托复旦大学在大气遥感领域的学科优势，已应用于东亚区域云辐射相互作用研究及极端天气诊断。其中，DaYu-CLAS中的静止卫星云物理特性全天时反演算法已在中国气象局人工影响天气中心完成业务化部署，并在人影中心“天工”指挥平台成功上线，在2024年西南林火和上海进博会期间、2025年“春雨”行动和宁夏六盘山增雨试验中为无人机、飞机作业方案设计和跟踪指挥提供了重要的数据支撑。未来，团队将持续优化算法并拓展卫星数据应用范围。

主要参考文献（按年份排序）:

1. “大禹”辐射传输模式（DaYu-RTM）

• Li, W., Zhang, F., Shi, Y. N., Iwabuchi, H., Zhu, M., Li, J., ... & Ishimoto, H. (2020). Efficient radiative transfer model for thermal infrared brightness temperature simulation in cloudy atmospheres. Optics Express, 28(18), 25730-25749.
• Li, W., Zhang, F., Lu, C., Jin, J., Shi, Y. N., Cai, Y., ... & Han, W. (2023). Integrated efficient radiative transfer model named Dayu for simulating the imager measurements in cloudy atmospheres. Optics Express, 31(10), 15256-15288.
• Cai, Y., Zhang, F., Lin, H., Li, J., Zhang, H., Li, W., & Hu, S. (2023). Optimized alternate mapping correlated k‐distribution method for atmospheric longwave radiative transfer. Journal of Advances in Modeling Earth Systems, 15(5), e2022MS003419.
• Cai, Y., Zhang, F., Li, J., Wu, K., & Yang, Q. (2025). An accurate shortwave gaseous transmittance scheme using modified alternate mapping correlated K‐distribution method. Journal of Geophysical Research: Atmospheres, 130(10), e2024JD041921.

2. “大禹”云分析系统（DaYu-CLAS）

• Li, W., Zhang, F., Lin, H., Chen, X., Li, J., & Han, W. (2022). Cloud detection and classification algorithms for Himawari-8 imager measurements based on deep learning. IEEE Transactions on Geoscience and Remote Sensing, 60, 1-17.
• Zhao, Z., Zhang, F., Wu, Q., Li, Z., Tong, X., Li, J., & Han, W. (2023). Cloud identification and properties retrieval of the Fengyun-4A satellite using a ResUnet model. IEEE Transactions on Geoscience and Remote Sensing, 61, 1-18.
• Tong, X., Li, J., Zhang, F., Li, W., Pan, B., Li, J., & Letu, H. (2023). The Deep-Learning-Based fast efficient nighttime retrieval of thermodynamic phase from Himawari-8 AHI measurements. Geophysical Research Letters, 50(11), e2022GL100901.
• Li, J., Zhang, F., Li, W., Tong, X., Pan, B., Li, J., ... & Mustafa, F. (2023). Transfer-learning-based approach to retrieve the cloud properties using diverse remote sensing datasets. IEEE Transactions on Geoscience and Remote Sensing, 61, 1-10.
• Guo, B., Zhang, F., Li, W., & Zhao, Z. (2024). Cloud classification by machine learning for geostationary radiation imager. IEEE Transactions on Geoscience and Remote Sensing, 62, 1-14.
• Zhao, Z., Zhang, F., Li, W., & Li, J. (2024). Image-based retrieval of all-day cloud physical parameters for FY4A/AGRI and its application over the Tibetan Plateau. Journal of Geophysical Research: Atmospheres, 129(18), e2024JD041032.
• Li, W., Zhang, F., Guo, B., Fu, H., & Letu, H. (2024). Physics-driven machine learning algorithm facilitates multilayer cloud property retrievals from geostationary passive imager measurements. IEEE Transactions on Geoscience and Remote Sensing, 62, 1-18.
• Guo, B., Zhang, F., Zhao, Z., Guo, J., & Li, W. (2024). Retrieval of cloud macro-physical properties using the FY-4A advanced geostationary radiation imager (AGRI) and the geostationary interferometric infrared sounder (GIIRS). Geophysical Research Letters, 51(24), e2024GL109772.
• Yang, Z., Zhao, Z., Zhou, T., Fu, H., & Li, J. (2025). All-day retrieval of cloud physical properties from Meteosat second generation satellite. IEEE Transactions on Geoscience and Remote Sensing, 63, 1-12.
• Liu, C., Zhang, F., Ouyang, H., Li, W., & Zhao, Z. (2025). Diurnal variation of cloud physical properties for tropical cyclones over North Atlantic in 2019–2023. Geophysical Research Letters, 52(13), e2025GL115566.
• Zhao, Z., Zhang, F., Li, W., Yang, B., He, Q., Fu, H., & Cai, M. (2025). Identification and tracking of deep convection systems over the Tibetan Plateau and its surrounding areas in summer using all-day cloud physical properties. Geophysical Research Letters, 52, e2025GL118433.
• Xiao, H., Zhang, F., Wang, L., Pan, B., Zhu, Y., Wang, M., ... & Li, J. (2025a). High-resolution ensemble retrieval of cloud properties for all-day based on geostationary satellite. npj Climate and Atmospheric Science, 8(1), 386.
• Li, J., Pan, B., Zhang, F., Guo, B., Li, W., Jiang, G. M., ... & Wang, Q. (2026). Probabilistic Retrieval of All-Day Overlapping Cloud Microphysical Properties. Advances in Atmospheric Sciences, 1-14.
• Zhang, F., Hong, X., Zhao, Z., Gan, Z., Ouyang, P., Xiao, H., ... & Lu, F. (2026). Short‐term forecasting of cloud physical properties based on Fourier neural operator method. Geophysical Research Letters, 53(8), e2025GL119553.

3. “大禹”降水分析系统（DaYu-PRAS）

• Xiao, H., Zhang, F., Zhang, R., Lu, F., Cai, M., & Wang, L. (2025b). Retrieval of total precipitable water under all‐weather conditions from Himawari‐8/AHI observations using the generative diffusion model. Geophysical Research Letters, 52(15), e2025GL117075.
• Yang, C., Li, H., Zhu, R., Wang, Y., Zhang, F., Gu, M., ... & Tang, X. (2026). Snow or rain? hybrid AI deciphers surface precipitation phase from satellite observations. Nature Communications, 17, 2813.

4. “大禹”云图预报大模型

• Wei, X., Zhang, F., Zhang, R., Li, W., Liu, C., Guo, B., ... & Tang, X. (2024). DaYu: Data-Driven Model for Geostationary Satellite Observed Cloud Images Forecasting. arXiv preprint arXiv:2411.10144.