| Citation: |
ZHUANG Xiaoran, LIU Mei, CAI Ninghao, et al. 2025. Evaluation of deep learning-based precipitation nowcasting methods during the 2023 flood season in Jiangsu Province [J]. Torrential Rain and Disasters,44(1):1−8. DOI: 10.12406/byzh.2023-206
|
Compared to the traditional nowcasting methods based on extrapolation, deep learning-based precipitation nowcasting methods can effectively forecast the trigger, development, and dissipation of heavy rainfall events. Currently, there are three widely used deep learning methods, which are PhyDNet, PredRNN-v2, and GAN. The first two are spatiotemporal convolutional neural networks under the deep learning framework, while the third is the generative adversarial network with PhyDNet as the generator and multilayer convolution as the discriminator. This study evaluates the application of the above three deep learning-based precipitation nowcasting methods during the flood season of 2023 (April to July) in Jiangsu region. The applicability of these methods in different precipitation types was also discussed through the representative case studies. The results are as follows. (1) From the entire evaluation period, PhyDNet and PredRNN-v2 show better TS performance than GAN. However, GAN performs the best in terms of BIAS for the main precipitation bands, which helps eliminate the "blurry" issue observed in the first two methods with the increase of forecast lead time. (2) During the systematic heavy precipitation periods, the evaluation results for all three methods are consistent with the overall period assessment. However, during the localized heavy precipitation periods, GAN outperforms PhyDNet and PredRNN-v2 in both TS and BIAS. (3) The results of the typical case analysis show that all three methods can capture the evolution of precipitation systems. In the case of systematic heavy rainfall events, PredRNN-v2 exhibits better performance in precipitation intensification compared to the other two methods. In localized heavy rainfall events, GAN not only overcomes the issue of precipitation overestimation caused by "blurriness" but also provides a better representation of the intensity and locations of the heavy precipitation centers. The above results suggest that each of the three forecasting methods has its advantages for different scenarios in heavy precipitation forecasting and warning. PhyDNet performs the best overall, PredRNN-v2 can be used for forecasts with longer lead time, while GAN is suitable for localized heavy rainfall events. Therefore, in practical operational applications, the choice of method should be based on their respective strengths and characteristics.
|
程丛兰,陈明轩,王建捷,等.2013.基于雷达外推临近预报和中尺度数值预报融合技术的短时定量降水预报试验[J].气象学报,73(3):397−415. doi: 10.11676/qxxb2013.041
Cheng C, Chen M, Wang J, et al. 2013. Short-term quantitative precipitation forecast experiments based on blending of nowcasting with numerical weather prediction [J]. Acta Meteorologica Sinica,73(3):397−415 (in Chinese). doi: 10.11676/qxxb2013.041
|
|
顾建峰,周国兵,刘伯骏,聂磊,张亚萍,张勇,吴胜刚.2020.人工智能技术在重庆临近预报业务中的初步研究与应用[J].气象,46(10):1286−1296.
GU Jianfeng, ZHOU Guobing, LIU Bojun, NIE Lei, ZHANG Yaping, ZHANG Yong, WU Shenggang. 2020. Study on Artificial Intelligence Technology and Its Application to Chongqing Operational Nowcasting [J]. Meteor Mon,46(10):1286−1296 (in Chinese).
|
|
韩丰,唐文苑,周楚炫,等.2023.基于SWAN系统的降水临近预报算法改进和应用评估[J].气象学报,81(2):304−315. doi: 10.11676/qxxb2023.20220066
Han F, Tang W, Zhou C, et al. 2023. Improving a precipitation nowcasting algorithm based on the SWAN system and related application assessment [J]. Acta Meteorologica Sinica,81(2):304−315 (in Chinese). doi: 10.11676/qxxb2023.20220066
|
|
田刚,陈良华,魏凡,等.2021.基于光流法雷达外推的2020年长江致洪降水临近预报检验评估[J].暴雨灾害,40(3):316−325. doi: 10.3969/j.issn.1004-9045.2021.03.010
Tian G, Chen L, Wei F, et al. 2021. Evaluation of flood-producing rainfall nowcasting based on radar extrapolation with the variational optical flow method in the Yangtze River Basin in 2020 [J]. Torrential Rain and Disasters,40(3):316−325 (in Chinese). doi:10.3969/j.issn.1004−9045.2021.03.010
|
|
解小寒,王勇,郭倩.2018.具有复杂地形适应能力的inca短临预报系统介绍[J].气象科技进展,8(3):70−76. doi: 10.3969/j.issn.2095-1973.2018.03.006
Xie X, Wang Y, Guo Q. 2018. Introduction to the INCA nowcasting system with complex terrain adaptability [J]. Advances in Meteorological Science and Technology,8(3):70−76 (in Chinese). doi:10.3969/j.issn.2095−1973.2018.03.006
|
|
曾康,闵锦忠,庄潇然,等.2024.基于生成对抗网络的强对流临近预报方法及其在中国东部地区的应用评估[J].大气科学,48(X):1−13. doi: 10.3878/j.issn.1006-9895.2310.23094
Zeng K, Min J Z, Zhuang X R, et al. 2024. Severe convection nowcasting method based on a generative adversarial network and its application evaluation in Eastern China [J]. Chinese Journal of Atmospheric Sciences,48(X):1−13. doi:10.3878/j.issn.1006−9895.2310.23094
|
|
张勇,刘慧,郑颖菲,等.2023.人工智能模型的分类临近预报产品效果检验与分析[J].沙漠与绿洲气象,17(1):115−121. doi: 10.12057/j.issn.1002-0799.2023.01.017
Zhang Y, Liu H, Zheng Y F, et al. 2023. Effect verification of classified products outputted by artificial intelligent nowcasting model [J]. Desert and Oasis Meteorology,17(1):115−121 (in Chinese). doi:10.12057/j.issn.1002−0799.2023.01.017
|
|
庄潇然,郑玉,王亚强,等.2023.基于深度学习的融合降水临近预报方法及其在中国东部地区的应用研究[J].气象学报,81(2):286−303. doi: 10.11676/qxxb2023.20220081
Zhuang X, Zheng Y, Wang Y, et al. 2023. A deep learning-based precipitation nowcast model and its application over East China [J]. Acta Meteorologica Sinica,81(2):286−303 (in Chinese). doi: 10.11676/qxxb2023.20220081
|
|
Bowler N E, Pierce C E, See A W. 2007. STEPS: A probabilistic precipitation forecasting scheme which merges an extrapolation nowcast with downscaled NWP [J]. Quarterly Journal of the Royal Meteorological Society,132(620):2127−2155. doi: 10.1256/qj.04.100
|
|
Espeholt L, Agrawal S, Sønderby C, et al. 2022. Deep learning for twelve hour precipitation forecasts [J]. Nature Communications,13:5145. doi:10.1038/s41467−022−32483−x
|
|
Le Guen V, Thome N. 2020. Disentangling physical dynamics from unknown factors for unsupervised video prediction [C]//Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle: IEEE, 11471-11481
|
|
Ko J, Lee K, Hwang H, et al. 2023. Deep-learning-Based precipitation nowcasting with ground weather station data and radar data [J]. arXiv: 2210.12853[physics. ao-ph]. doi: 10.48550/arXiv.2210.12853
|
|
Mandapaka P V, Germann U, Panziera L, et al. 2012. Can Lagrangian extrapolation of radar fields be used for precipitation nowcasting over complex alpine orography? [J] Weather and Forecasting, 27(1): 28-49. doi: 10.1175/WAF−D−11−00050.1
|
|
Pulkkinen S, Neini D, Hortal A A, et al. 2019. Pysteps: an open-source Python library for probabilistic precipitation nowcasting (v1.0) [J]. Geoscientific Model Development,12:4185−4219. doi:10.5194/gmd−12−4185−2019
|
|
Ravuri S, Lenc K, Willson M, et al. 2021. Skilful precipitation nowcasting using deep generative models of radar [J]. Nature 597, 672–677. doi: https://doi.org/10.1038/s41586-021-03854-z
|
|
Sønderby C K, Espeholt L, Heek J, et al. 2020. MetNet: A neural weather model for precipitation forecasting [J]. arXiv: 2003.12140v2. doi: 10.48550/arXiv.2003.12140
|
|
Wang Y, Coning E D, Harou A, et al. 2017. Guidelines for nowcasting techniques [M]. World Meteorological Organization: Geneva, Swizerland
|
|
Wang Y, Long M, Wang J, et al. 2017. Predrnn: recurrent neural networks for predictive learning using spatiotemporal lstms [C] // Proceedings of the 31st International Conference on Neural Information Processing Systems. Long Beach: Curran Associates Inc. , 879-888
|
|
Wang Y, Wu H, Zhang J, et al. 2022. PredRNN: A recurrent neural network spatiotemporal predictive learning [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,45(2):2208−2225. doi: 10.1109/TPAMI.2022.3165153
|
|
Zhang Y, Long M, Chen K, et al. 2023. Skillful nowcasting of extreme precipitation with NowcastNet [J]. Nature,619:526−546. doi:10.1038/s41586−023−06184−4
|
Supported by: Beijing Renhe Information Technology Co., Ltd. 
DownLoad: