EC细网格10 m风场产品在青岛港区的预报检验和随机森林订正-海洋预报

EC细网格10 m风场产品在青岛港区的预报检验和随机森林订正

单位：1. 青岛市气象服务中心, 山东青岛 266003;
2. 青岛市气象灾害防御工程技术研究中心, 山东青岛 266003;
3. 青岛港国际股份有限公司, 山东青岛 266000

分类号：P457.5

出版年·卷·期（页码）：2024·41·第三期（110-119）

摘要：

利用2021年3月—2022年2月EC细网格10 m风场预报产品，对其在青岛港区的风场预报能力进行检验和订正。结果表明：对于大港码头、董家口港自动站和董家口港浮标站，EC细网格前 48 h的 10 m 风速预报偏差离散程度相对较小，预报偏差中位数均大于 0，表明模式风速预报较实况有系统性偏强的特征。EC细网格24 h预报的10 m风速与实况风速相关性较好，3个站点的相关系数分别为0.76、0.73、0.82。EC细网格对大港码头各风向的风速预报均偏大，在实况风向为东北风和偏东风时，EC细网格风速预报的均方根误差和平均误差最大；对于董家口港自动站和董家口港浮标站，不同风向下 EC细网格的风速预报均方根误差差别不大。EC细网格对 3个站点的风向预报偏差主要集中在-45°~45°，实况风速越小，风向的预报偏差离散程度越大。利用随机森林算法对青岛港区EC细网格预报风速进行订正，预报精度均得到提高。

The 10 m wind forecasts of the EC fine-grid model in Qingdao Port from March 2021 to February 2022 are validated and corrected. The results show that: In comparison with the observations from Dagang Wharf, Dongjiakou Port Automatic Station and Dongjiakou Port Buoy Station, the dispersion of biases in 10 m wind speed forecasts at lead time of 48 h is relatively small, and the median of biases in the forecasts is larger than 0, indicating that the wind speed forecasts are systematically larger than the observations. The 10 m wind speed forecasts at lead time of 24 h have a good correlation with the observations, and the correlation coefficients for the three stations are 0.76, 0.73, and 0.82, respectively. The wind speed forecasts for each wind direction are relatively large in Dagang Wharf. When the observed wind direction is northeast and east, the root mean square error and average deviation of wind speed forecasts in Dagang Wharf are the largest. For Dongjiakou Port Automatic Station and Dongjiakou Port Buoy Station, small differences in the root mean square error of wind speed forecasts exist under different wind directions. The wind direction biases for the three stations mainly concentrate between -45 and 45°. Smaller actual wind speed is always associated with larger dispersion of bias in the wind direction forecasts. Using the Random Forest to correct the wind speed forecasts improves the accuracy of forecasts in Qingdao Port.

参考文献：

[1] 刘倩,孙小淇.青岛港潮汐的调和分析与预报[J].海洋气象学报, 2022, 42(2):99-106. LIU Q, SUN X Q. Harmonic analysis and forecast of tides in Qingdao Port[J]. Journal of Marine Meteorology, 2022, 42(2):99-106.
[2] 于文金,苏荣,谢涛.极端灾害天气对港口锚地的影响研究--以河北秦皇岛港为例[J].灾害学, 2016, 31(4):17-21. YU W J, SU R, XIE T. Study on the Influence of extreme weather disasters on harbor anchorages--taking Hebei Qinhuangdao Port as an example[J]. Journal of Catastrophology, 2016, 31(4):17-21.
[3] 魏晓雯,陈亮,赵蕾,等.南海港口大风的环流特征及其预报指标--以海南洋浦港为例[J].气象科技, 2021, 49(5):754-761. WEI X W, CHEN L, ZHAO L, et al. Circulation characteristics and forecasting indicators of high wind in a port of South China Sea:taking Yangpu Port as an example[J]. Meteorological Science and Technology, 2021, 49(5):754-761.
[4] 孔扬,赵昶昱,王科,等.宁波舟山港气象灾害风险及服务效益评估[J].热带气象学报, 2021, 37(1):82-90. KONG Y, ZHAO C Y, WANG K, et al. Evaluation of risks of meteorological disasters and benefits from meteorological services for Ningbo-Zhoushan Port[J]. Journal of Tropical Meteorology, 2021, 37(1):82-90.
[5] 李敏,王辉,金啟华.中国近海海面风场预报方法综述[J].海洋预报, 2009, 26(3):114-120. LI M, WANG H, JIN Q H. A review on the forecast method of China offshore wind[J]. Marine Forecasts, 2009, 26(3):114-120.
[6] 郭丽娜,马艳,于慧珍.青岛沿海地区大风特征及其预警评估[J].海洋气象学报, 2022, 42(2):90-98. GUO L N, MA Y, YU H Z. Characteristics of gale and assessment of gale warning in Qingdao coastal areas[J]. Journal of Marine Meteorology, 2022, 42(2):90-98.
[7] 董全,张峰,宗志平.基于ECMWF集合预报产品的降水相态客观预报方法[J].应用气象学报, 2020, 31(5):527-542. DONG Q, ZHANG F, ZONG Z P. Objective precipitation type forecast based on ECMWF ensemble prediction product[J]. Journal of Applied Meteorological Science, 2020, 31(5):527-542.
[8] 涂小萍,姚日升. ECMWF数值产品在热带气旋(TC)路径预报中的应用分析[J].热带气象学报, 2010, 26(6):759-764. TU X P, YAO R S. Analysis on ECMWF NWP products in TC track forecasting[J]. Journal of Tropical Meteorology, 2010, 26(6):759-764.
[9] PINSON P, HAGEDORN R. Verification of the ECMWF ensemble forecasts of wind speed against analyses and observations[J]. Meteorological Applications, 2012, 19(4):484-500.
[10] ZHANG K, MU M, WANG Q. Identifying the sensitive area in adaptive observation for predicting the upstream Kuroshio transport variation in a 3-D ocean model[J]. Science China Earth Sciences, 2017, 60(5):866-875.
[11] 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. SUI Y X, YANG J T, CHANG H L, 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.
[12] 吴俞,冯文,李勋,等. ECMWF细网格10m风场产品在南海海域的预报检验[J].西南师范大学学报(自然科学版), 2015, 40(9):204-212. WU Y, FENG W, LI X, et al. On forecast test of ECMWF 10 m wind field products in South China Sea[J]. Journal of Southwest China Normal University (Natural Science Edition), 2015, 40(9):204-212.
[13] 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. LIAN X H, LIU G Y, GAO S, 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.
[14] 申华羽,方艳莹,涂小萍,等. ECMWF细网格10 m风预报在浙江沿海的评估与订正[J].气象, 2020, 46(11):1485-1494. SHEN H Y, FANG Y Y, TU X P, et al. Verification of ECMWF 10 m wind forecast for coastal Zhejiang Province[J]. Meteorological Monthly, 2020, 46(11):1485-1494.
[15] 孙全德,焦瑞莉,夏江江,等.基于机器学习的数值天气预报风速订正研究[J].气象, 2019, 45(3):426-436. SUN Q D, JIAO R L, XIA J J, et al. Adjusting wind speed prediction of numerical weather forecast model based on machine learning methods[J]. Meteorological Monthly, 2019, 45(3):426-436.
[16] 任萍,陈明轩,曹伟华,等.基于机器学习的复杂地形下短期数值天气预报误差分析与订正[J].气象学报, 2020, 78(6):1002-1020. REN P, CHEN M X, CAO W H, et al. Error analysis and correction of short-term numerical weather prediction under complex terrain based on machine learning[J]. Acta Meteorologica Sinica, 2020, 78(6):1002-1020.
[17] BREIMAN L. Random forests[J]. Machine Learning, 2001, 45(1):5-32.
[18] 耿丹,刘婷婷,李超.结合FY-4A卫星及随机森林的日间沿海海雾识别模型的研究[J].海洋预报, 2022, 39(3):83-93. GENG D, LIU T T, LI C. Research on a daytime sea fog identification model based on FY-4A satellite data and random forest algorithm[J]. Marine Forecasts, 2022, 39(3):83-93.
[19] 许立兵,孔扬,周峥,等.基于机器学习的风场预报订正方法研究[J].陕西气象, 2023(1):15-20. XU L B, KONG Y, ZHOU Z, et al. Research on wind prediction correction method based on machine learning[J]. Journal of Shaanxi Meteorology, 2023(1):15-20.
[20] 王雪洁,施国萍,周子钦,等.基于随机森林算法对ERA5太阳辐射产品的订正[J].自然资源遥感, 2022, 34(2):105-111. WANG X J, SHI G P, ZHOU Z Q, et al. Revision of solar radiation product ERA5 based on random forest algorithm[J]. Remote Sensing for Natural Resources, 2022, 34(2):105-111.
[21] 赵华生,金龙,黄小燕,等.基于CNN和RF算法的ECMWF降水分级订正预报方法[J].气象科技, 2021, 49(3):419-426. ZHAO H S, JIN L, HUANG X Y, et al. ECMWF precipitation grading correction forecast method based on CNN and RF algorithm[J]. Meteorological Science and Technology, 2021, 49(3):419-426.
[22] 门晓磊,焦瑞莉,王鼎,等.基于机器学习的华北气温多模式集合预报的订正方法[J].气候与环境研究, 2019, 24(1):116-124. MEN X L, JIAO R L, WANG D, et al. A temperature correction method for multi-model ensemble forecast in North China based on machine learning[J]. Climatic and Environmental Research, 2019, 24(1):116-124.
[23] PENSIERI S, BOZZANO R, SCHIANO M E. Comparison between QuikSCAT and buoy wind data in the Ligurian Sea[J]. Journal of Marine Systems, 2010, 81(4):286-296.

服务与反馈：

【文章下载】【发表评论】【查看评论】【加入收藏】