EC模式和GFS模式对黄渤海10m风速预报性能对比分析-海洋预报

EC模式和GFS模式对黄渤海10m风速预报性能对比分析

作者：王彬¹ 2 李文博¹ 2 李锐¹ 2 侯乔琨¹ 2 孙雅文¹ 2 刘桂艳¹ 2

单位：1. 自然资源部北海预报减灾中心, 山东青岛 266100;
2. 山东省海洋生态环境与防灾减灾重点实验室, 山东青岛 266100

分类号：P732.1

出版年·卷·期（页码）：2023·40·第四期（64-71）

摘要：

采用黄渤海10个海上浮标和5个近岸海洋站的风速观测资料作为实况，对2021年欧洲中期天气预报中心（EC）和美国国家环境预报中心全球预报系统（GFS）的10 m风速预报产品进行检验评估和对比分析。结果表明：EC模式的整体预报效果优于GFS模式，其中EC模式24～72 h的均方根误差比GFS模式减小了6.8%～8.0%，EC模式的预报偏差中位数为0.19～0.25 m/s，而GFS模式同类中位数为0.33～0.41 m/s，两种预报产品风速的预报结果整体略偏大，但EC模式的偏离程度更小。随着预报时效的增长，预报误差逐渐增大。在不同风速条件下，EC模式预报对5级及以下风的预报效果优于GFS模式预报，但后者对6级及以上风的预报效果更好。空间方面，两种模式对小长山站的预报效果最差，对黄海中部MF03007站的预报效果最好，GFS模式在渤海的预报效果优于EC模式，其他海区则EC模式的预报效果更好；在时间方面，EC模式风速产品在4—8月的预报效果优于GFS模式，其他月份则GFS模式的预报效果更好。在强天气过程方面，随着预报时效的增长，GFS模式的气旋预报出现误差急剧增大的现象。GFS模式对冷空气以及冷空气和气旋配合过程的预报效果好于EC模式，EC模式对渤海气旋的预报效果好于GFS模式，而对江淮气旋预报来说，不同过程下两种模式各有优劣。

Using wind speed observation data collected by 10 buoys and 5 off-shore stations in the Yellow Sea and Bohai Sea, this study evaluates two products of 10m wind speed forecasts in 2021 with daily forecasting onset at 20 o'clock, one from the European Centre for Medium-Range Weather Forecasts (EC) and the other from Global Forecasting System (GFS). The results show that overall forecasting performance of the EC is better than the GFS. Root mean squared error of 24～72 h wind speed forecasts of the EC is reduced by 6.8%～8% compared to the GFS. Median bias of the EC forecasts for 24～72 h ranges between 0.19 m/s and 0.25 m/s, while that of the GFS ranges between 0.33 m/s and 0.41 m/s. The wind speed forecasts of the two products are slightly larger than the observation, in which the EC bias is relatively smaller. Specifically, the forecast error increases gradually along with the prolong of the forecast leading time. The EC forecast is better than the GFS forecast at Beaufort force 5 and below, while at Beaufort force 6 and above the latter is better. The two products perform worst at Xiaochangshan station, while best at MF03007 station in the central Yellow Sea. The GFS forecast is better than the EC in the Bohai Sea, while the EC performs better in other seas. The EC forecast is better than the GFS during April—August, while the GFS performs better in other months. The prediction error of cyclones of the GFS increases sharply along with the increase of forecast leading time. The GFS forecast is better than the EC on cold air, as well as its encountering process to cyclones, while the EC performs better on cyclones in the Bohai Sea. The two products have their own pros and cons on cyclones in Jianghuai area.

参考文献：

[1] 张新玲, 吴增茂. 渤海海上测风与沿岸实测风的对比分析[J]. 海洋预报, 1998, 15(4): 24-31. ZHANG X L, WU Z M. Comparing analyses of the observations of sea and coastal wind of Bohai[J]. Marine Forecasts, 1998, 15(4): 24-31.
[2] 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.
[3] 周伟隆, 陈往溪, 肖巍. 粤东海面冷空气强风的统计分析与预报[J]. 广东气象, 2005(4): 20-22. ZHOU W L, CHEN W X, XIAO W. Statistical analysis and forecast of strong sea surface cold air wind in eastern Guangdong province[J]. Guangdong Meteorology, 2005(4): 20-22.
[4] FRNDA J, DURICA M, ROZHON J, et al. ECMWF short-term prediction accuracy improvement by deep learning[J]. Scientific Reports, 2022, 12(1): 7898.
[5] 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 10m wind field products in south China Sea[J]. Journal of Southwest China Normal University (Natural Science Edition), 2015, 40(9): 204-212.
[6] 方艳莹, 申华羽, 涂小萍, 等. ECMWF细网格对浙江沿海10 m风预报性能评估[J]. 中国农学通报, 2019, 35(13): 119-125. FANG Y Y, SHEN H Y, TU X P, et al. ECMWF fine mesh: forecast performance evaluation of 10 m wind in Zhejiang coastal area[J]. Chinese Agricultural Science Bulletin, 2019, 35(13): 119-125.
[7] 于鑫, 于明洋, 蔡景就. ECMWF细网格10m风产品在南海北部的检验与评估[J]. 广东气象, 2021, 43(5): 36-40. YU X, YU M Y, CAI J J. Inspection and evaluation of ECMWF fine grid 10m wind products in the northern South China Sea[J]. Guangdong Meteorology, 2021, 43(5): 36-40.
[8] 连喜虎, 刘桂艳, 高松, 等. EC细网格10m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 10m wind field in the Bohai Sea and the Yellow Sea[J]. Marine Forecasts, 2020, 37(6): 74-82.
[9] DE CHIARA G, JANSSEN P, HERSBACH H, et al. Assimilation of scatterometer winds at ECMWF[R/OL]. [2023-07-11]. https://www-cdn.eumetsat.int/files/2020-04/pdf_conf_p60_s4_05_dechiara_v.pdf.
[10] HAIDEN T, JANOUSEK M, VITART F, et al. Evaluation of ECMWF forecasts, including the 2021 upgrade[R/OL]. [2023-07- 11]. https://www.ecmwf.int/sites/default/files/elibrary/2021/20142-evaluation-ecmwf-forecasts-including-2021-upgrade.pdf.
[11] 周荣卫, 何晓凤, 孙逸涵. 3种全球环流预报场近地层风速突变预报能力评估[J]. 资源科学, 2013, 35(10): 2121-2129. ZHOU R W, HE X F, SUN Y H. Forecast performance verification of surface wind ramp for three global circulation forecast fields[J]. Resources Science, 2013, 35(10): 2121-2129.
[12] 吴息, 白龙, 崔方, 等. 海面与海岸陆面风速廓线特征[J]. 大气科学学报, 2014, 37(2): 138-145. WU X, BAI L, CUI F, et al. Wind profile features of surface layer over ocean and coast[J]. Transactions of Atmospheric Sciences, 2014, 37(2): 138-145.
[13] 刘焕彬, 董旭光. 不同下垫面条件下风随高度的变化特征[J]. 自然科学, 2021, 9(1): 1-9. LIU H B, DONG X G. Variation characteristics of wind with height under different underlying surface conditions[J]. Open Journal of Nature Science, 2021, 9(1): 1-9.
[14] 朱兵. 箱线图及其在JCR网络版中的应用[J]. 农业图书情报学刊, 2011, 23(8): 15-18. ZHU B. Box plot and its application in JCR web[J]. Journal of Library and Information Sciences in Agriculture, 2011, 23(8): 15- 18.
[15] TAYLOR K E. Summarizing multiple aspects of model performance in a single diagram[J]. Journal of Geophysical Research: Atmospheres, 2001, 106(D7): 7183-7192.
[16] 李培顺, 曹丛华. 渤海风暴潮与灾害[M]. 北京: 海洋出版社, 2019. LI P S, CAO C H. Storm surges and disasters in the Bohai Sea [M]. Beijing: China Ocean Press, 2019.

服务与反馈：

【文章下载】【发表评论】【查看评论】【加入收藏】