智能网格10 m阵风产品对海南洋浦港适用性评估分析  

Applicability Assessment Analysis of the intelligent grid 10 m Gust of Wind Field Products in Yangpu Port of Hainan Province

作　　者：赵蕾 陈亮[1,2] 刘康 张天圣 ZHAO Lei;CHEN Liang;LIU Kang;ZHANG Tiansheng(Key Laboratory of South China Sea Metcorological Disaster Prevention and Mitigation of Hainan Province,Haikou,Hainan 570203,China;Hainan Provincial Meteorological Service Center,Haikou,Hainan 570203,China;Suining Branch of Civil Aviation Flight University in China,Suining,Sichuan 629000,China)

机构地区：[1]海南省南海气象防灾减灾重点实验室,海南海口570203 [2]海南省气象服务中心,海南海口570203 [3]中国民用航空飞行学院遂宁分院,四川遂宁629000

出　　处：《热带农业科学》2022年第10期120-124,共5页Chinese Journal of Tropical Agriculture

基　　金：中国气象局公服气象服务中心创新基金项目(No.M2021013);基于机器学习的港口大风预报服务产品研发。

摘　　要：为了更好地开展海南洋浦港预报气象服务,利用智能网格10 m风场预报产品对2018年3月8日至2020年3月8日,每日12时(世界时)起报的智能网格10 m阵风产品逐3、6、12、24、48、72 h在洋浦港的适用性进行评估,通过最临近插值法将格点预报插值到洋浦港海岛自动站,并采用不同误差分析方法,将格点预报与同期自动站观测实况数据进行对比.结果表明:不同时效、不同风力条件下,智能网格对6级及以下阵风预报效果最好;风速大于6级时,由于样本数量较少,准确率呈递减趋势,风力等级越大漏报率越高,可信度需进一步验证.不同天气型、不同时效,预报值与自动站实况数据相比,有明显的差异;台风影响下,风力等级预报准确率最高,冷空气影响次之,热对流天气很难捕捉,且随着时效增大,准确率越低,很难预测.另外,随着阵风等级增大,预报值存在较多的漏报,尤其是7级及以上大风有低估的效果,其中很难预测热对流天气造成的大风.The paper aims to improve meteorological services in Yangpu port of Hainan province.The intelligent grid 10m gust of wind field product was used to evaluate the applicability in Yangpu port from March 8 of 2018 to March 8 of 2020 at 12:00 daily for 3,6,12,24,48 and 72 hours.The grid prediction were interpolated to the automatic station of Yangpu port island by nearest interpolation method.Different error analysis methods were used to compare the grid prediction with the observation data of the automatic station in the same period.Under the different time effects and different wind conditions,the smart grid forecasts gusts of level 6 and below was the best.When the wind speed was greater than level 6,due to the limited number of samples,the accuracy rate showed a decreasing trend.The higher the wind level was,the omission rate was the higher.The reliability needs to be further verified.There were obvious differences between actual situation of the station and prediction in the different weather types and different time effects.The accuracy rate of wind forecast was the highest under the influence of typhoon and followed by cold air.Hot convective weather was difficult to catch.With the increasing of exposure time,the lower the accuracy was,the harder it was to predict.In addition,with the increasing of wind level,there were many missed gust values.It was an underestimated effect,especially strong winds above 7 level.It was difficult to predict the strong wind caused by hot convective weather.

关 键 词：智能网格 洋浦港 最临近插值法 误差分析方法 

分 类 号：P732[天文地球—海洋科学]