黄渤海站点强阵风的预报检验与订正初探

作者：韩永清1,2, 戈瑶1,2, 刘畅1,2, 李建明1,3, 姚凯4

单位：1. 山东省气象防灾减灾重点实验室, 山东 济南 250031; 2. 山东省气象台, 山东 济南 250031; 3. 东营市气象局, 山东 东营 257091; 4. 吉林省气象台, 吉林 长春 130062

关键词：强阵风;检验;订正;初探

分类号：P732.4

出版年·卷·期（页码）：2026·43·第4期(107-117)

摘要：

基于2022—2024年欧洲中期天气预报中心（ECMWF）模式10 m阵风预报产品，检验分析其对黄渤海强阵风的预报性能，采用频率匹配与百分位阈值相结合的方法对站点强阵风进行预报订正试验。结果表明：ECMWF模式对黄渤海强阵风预报强度偏弱、范围偏小，而且漏报高、离散度高和强度预报严重偏弱等因素造成模式对10级及以上阵风的预报TS评分较低。经频率匹配方法订正后的强阵风预报发生频率与实况接近，可有效改善7～9级阵风的预报漏报，但订正后对10级及以上阵风的空报率过高。在频率匹配订正预报的基础上，采用百分位阈值方法进行二次订正，可提升10级及以上阵风的预报TS评分，且以99.5%分位阈值的订正试验综合结果最优，与ECMWF模式相比，10级及以上、11级及以上阵风预报的TS评分分别提升27.3%和10倍，12级及以上阵风的预报TS评分由0提升到0.10。

Based on the 10 m gust forecast products of the European Centre for Medium-Range Weather Forecasts(ECMWF) model from 2022 to 2024, this study analyzes the forecast performance of strong gusts in the Yellow Sea and Bohai Sea, and conducts a forecast correction experiment for strong gusts by adopting the frequency matching method with percentile threshold. The results show that: The ECMWF model underestimates the intensity and coverage of strong gusts in the Yellow Sea and Bohai Sea. Factors such as high missing rate,high dispersion, and severe underestimation of intensity lead to the low TS score for the model in predicting gusts of level 10 and above. After correction using the frequency matching method, the forecast frequency of strong gusts is close to the observation. The correction method effectively reduces the missed alarms of forecasts from level 7 to level 9, but leads to higher false alarms of forecasts of level 10 and above. Applying the percentile threshold method for a secondary correction improves the forecast TS score for gusts of level 10 and above, and the 99.5th percentile threshold gives the best performance. Compared with the ECMWF model, the improvement rates of the forecast TS score for level 10 and level 11 gusts reach 27.3% and 10 times, respectively, while the forecast TS score for level 12 gusts increases from 0 to 0.10.

参考文献：

[1] 王彬,李文博,李锐,等.EC模式和GFS模式对黄渤海10 m风速预报性能对比分析[J].海洋预报, 2023, 40(4):64-71, doi:10.11737/j.issn.1003-0239.2023.04.007.Wang Bin, Li Wenbo, Li Rui, et al.Comparative analysis on 10 m wind forecasts in the Yellow Sea and Bohai Sea from the ECMWF and GFS models[J].Marine Forecasts, 2023, 40(4):64-71, doi:10.11737/j.issn.1003-0239.2023.04.007. [2] 隋玉秀,杨景泰,常慧琳,等.大连地区ECMWF细网格风向与实况的对比检验[J].海洋预报, 2022, 39(1):21-31, doi:10.11737/j.issn.1003-0239.2022.01.003.Sui Yuxiu, Yang Jingtai, Chang Huilin, et al.Comparison and validation of ECMWF fine-grid wind direction and observation data in Dalian area[J].Marine Forecasts, 2022, 39(1):21-31, doi:10.11737/j.issn.1003-0239.2022.01.003. [3] 连喜虎,刘桂艳,高松,等.EC细网格10 m风场产品在渤、黄海海域的预报能力检验与评估[J].海洋预报, 2020, 37(6):74-82, doi:10.11737/j.issn.1003-0239.2020.06.008.Lian Xihu, Liu Guiyan, Gao Song, et al.Evaluation of the ECMWF forecast skills for the 10 m wind field in the Bohai Sea and the Yellow Sea[J].Marine Forecasts, 2020, 37(6):74-82, doi:10.11737/j.issn.1003-0239.2020.06.008. [4] 全继萍,李青春,仲跻芹,等.“CMA北京模式”中三种不同阵风诊断方案在北京地区大风预报中的评估[J].气象学报, 2022, 80(1):108-123.Quan Jiping, Li Qingchun, Zhong Jiqin, et al.Evaluation of three different gust diagnostic schemes in the CMA-BJ for gale forecasting over Beijing[J].Acta Meteorologica Sinica, 2022, 80(1):108-123. [5] 董双林.中国的阵风极值及其统计研究[J].气象学报, 2001, 59(3):327-333.Dong Shuanglin.Gust extremes in China and its statistical study[J].Acta Meteorologica Sinica, 2001, 59(3):327-333. [6] 杨璐,宋林烨,荆浩,等.复杂地形下高精度风场融合预报订正技术在冬奥会赛区风速预报中的应用研究[J].气象, 2022, 48(2):162-176.Yang Lu, Song Linye, Jing Hao, et al.Fusion prediction and correction technique for high-resolution wind field in Winter Olympic Games area under complex terrain[J].Meteorological Monthly, 2022, 48(2):162-176. [7] 胡海川,钱传海,渠鸿宇.黄渤海及其邻近地区阵风估测改进[J].应用气象学报, 2023, 34(6):668-680.Hu Haichuan, Qian Chuanhai, Qu Hongyu.Correction of gust estimation in the Yellow and Bohai seas and adjacent areas[J].Journal of Applied Meteorological Science, 2023, 34(6):668-680. [8] 胡海川,刘珺,林建.基于预报方程的我国近海阵风预报[J].气象, 2022, 48(3):334-344.Hu Haichuan, Liu Jun, Lin Jian.Application of prediction equation to gust forecasting for Chinese offshore seas[J].Meteorological Monthly, 2022, 48(3):334-344. [9] 苗青,董春卿,孙颖姝,等.CMA-MESO 3 km模式10 m风预报产品在山西区域的定量检验及客观订正[J].气象科技, 2025, 53(1):22-34.Miao Qing, Dong Chunqing, Sun Yingshu, et al.Quantitative test and objective correction of CMA-MESO 3 km model 10 m wind forecast products in Shanxi region[J].Meteorological Science and Technology, 2025, 53(1):22-34. [10] 苏翔,李超.基于最优TS评分和频率匹配的江苏近海风速订正[J].海洋通报, 2020, 39(5):548-557.Su Xiang, Li Chao.Jiangsu offshore wind speed calibration based on optimal threat score and frequency matching[J].Marine Science Bulletin, 2020, 39(5):548-557. [11] 钱磊,邱学兴,郑淋淋.基于概率密度匹配方法的WRF模式阵风风速误差订正[J].气象科技, 2019, 47(6):916-926.Qian Lei, Qiu Xuexing, Zheng Linlin.Error correction of WRF model gust speed based on probability density function matching method[J].Meteorological Science and Technology, 2019, 47(6):916-926. [12] 钟有亮,李勋,陈静,等.ECMWF细网格10 m极大风速预报在海南岛的评估与订正[J].热带农业科学, 2024, 44(10):140-147.Zhong Youliang, Li Xun, Chen Jing, et al.Verification and correction of the ECMWF Ten-meter wind forecast for Hainan Island[J].Chinese Journal of Tropical Agriculture, 2024, 44(10):140-147. [13] 梁利,赵华生,吴玉霜.基于Tabnet的日极大风风速订正预报模型[J].气象科技, 2024, 52(5):714-722.Liang Li, Zhao Huasheng, Wu Yushuang.Research on daily extreme wind speed correction forecast based on Tabnet[J].Meteorological Science and Technology, 2024, 52(5):714-722. [14] 胡海川,代刊.我国近海阵风预报研究[J].气象, 2024, 50(6):711-722.Hu Haichuan, Dai Kan.Research on gust forecasting in China's offshore[J].Meteorological Monthly, 2024, 50(6):711-722. [15] 胡海川,曹勇.基于全连接神经网络与分位数匹配相结合的阵风预报[J].大气科学学报, 2025, 48(3):417-428.Hu Haichuan, Cao Yong.Offshore gust forecasting based on combined fully connected neural network and quantile matching approach[J].Transactions of Atmospheric Sciences, 2025, 48(3):417-428. [16] 杨凡,刘志丰,任兆鹏,等.基于深度学习的大风订正预报研究[J].海洋预报, 2024, 41(6):23-31.Yang Fan, Liu Zhifeng, Ren Zhaopeng, et al.Research on wind speed forecasting correction based on deep learning[J].Marine Forecasts, 2024, 41(6):23-31. [17] 郝雨辰,杨勤胜,霍雪松,等.基于Resnet50的江苏近海海面风场预报订正方法研究[J].海洋预报, 2024, 41(4):57-65.Hao Yuchen, Yang Qinsheng, Huo Xuesong, et al.Correction method for surface wind forecasts in Jiangsu offshore area based on Resnet50[J].Marine Forecasts, 2024, 41(4):57-65. [18] 李俊,杜钧,陈超君.降水偏差订正的频率(或面积)匹配方法介绍和分析[J].气象, 2014, 40(5):580-588.Li Jun, Du Jun, Chen Chaojun.Introduction and analysis to frequency or area matching method applied to precipitation forecast bias correction[J].Meteorological Monthly, 2014, 40(5):580-588. [19] 苏翔,袁慧玲,朱跃建.四种定量降水预报客观订正方法对比研究[J].气象学报, 2021, 79(1):132-149.Su Xiang, Yuan Huiling, Zhu Yuejian.A comparative study of four objective quantitative precipitation forecast calibration methods[J].Acta Meteorologica Sinica, 2021, 79(1):132-149. [20] 唐红玉,吴遥,董新宁,等.华西秋雨期极端降水事件及其环流特征[J].气象, 2025, 51(6):711-723.Tang Hongyu, Wu Yao, Dong Xinning, et al.Extreme precipitation events and their circulation characteristics during the autumn rain period in West China[J].Meteorological Monthly,2025, 51(6):711-723. [21] 李昱薇,杨成芳,傅刚,等.2016-2021年山东省大风时空分布特征分析[J].中国海洋大学学报, 2024, 54(8):23-32.Li Yuwei, Yang Chengfang, Fu Gang, et al.Spatial and temporal characteristics of strong winds in Shandong Province from 2016to 2021[J].Periodical of Ocean University of China, 2024, 54(8):23-32. [22] 张秋锦,黄一民,郝丽婷,等.基于百分位阈值法的洞庭湖流域高温热浪时空特征[J].气象与环境学报, 2025, 41(1):66-73.Zhang Qiujin, Huang Yimin, Hao Liting, et al.Spatiotemporal characteristics of high temperature and heat wave in Dongting Lake Basin based on percentile threshold method[J].Journal of Meteorology and Environment, 2025, 41(1):66-73. [23] Sheridan P.Current gust forecasting techniques, developments and challenges[J].Advances in Science and Research, 2018, 15:159-172.

服务与反馈：

【文章下载】【发表评论】【查看评论】【加入收藏】