ERA-Interim和NCEP/NCAR再分析资料在我国东南近海适用性分析  

The applicability of ERA-Interim and NCEP/NCAR reanalysis datasets in the adjacent waters of the southeast China Sea

作　　者：申华羽 吕劲文 涂小萍 方艳莹 SHEN Hua-yu;LV Jin-wen;TU Xiao-ping;FANG Yan-ying(Ningbo Meteorological Service Center,Ningbo 315012 China;Ningbo Meteorological Observatory Ningbo 315012 China)

机构地区：[1]宁波市气象服务中心,浙江宁波315012 [2]宁波市气象台,浙江宁波315012

出　　处：《海洋预报》2019年第2期9-20,共12页Marine Forecasts

基　　金：浙江省自然科学基金(LY15D050001);浙江省气象局科技计划项目(2016YB03)

摘　　要：利用我国东南近海5个浮标站观测资料,对2012—2016年ERA-Interim和NCEP/NCAR再分析资料10 m风、2 m气温、海平面气压的适用性进行了评估。结果表明:NCEP/NCAR的再分析10 m风适用性更好,ERA-Interim的2 m气温适用性更好,海平面气压两者差异不大。风速再分析值与观测值具有较好的一致性,相关系数达0.8～0.9,但再分析风速总体上有偏小的趋势,平均偏差在-1.3～0 m/s之间,均方根误差在1.5～3 m/s。再分析资料的平均风向有顺时针偏差的趋势,温州浮标偏右达14°以上,均方根误差大多在40°～50°。不管风速还是风向,5个浮标站中均以舟山浮标的再分析值与观测值最为接近;分析还表明,再分析资料的冬季风代表性相对较差,这是造成风速和风向系统性偏差的主要原因。再分析资料与观测2 m气温相关系数均在0.95以上,且有偏高的趋势,NCEP偏高更为明显,有4个浮标站平均偏差达1～2℃,而ERA-I仅1个浮标站偏差1～2℃,4个在1℃以内。春季和冬季气温偏高最为明显,春季升温过程存在异常偏高的可能,秋季气温与观测值最为接近。海平面气压适用性较好,总体优于10 m风和2 m气温,且季节间差异也不大。The applicability of 10 m wind, 2 m temperature, and sea level pressure of ERA-Interim and NCEP/ NCAR reanalysis datasets from 2012 to 2016 is evaluated based on observations of five buoys in the adjacent waters of the southeast China sea. The results show that the NCEP/NCAR 10 m wind and ERA-Interim 2 m temperature reveal better applicability, while the sea level temperature of the two datasets shows similar statistical error. The two reanalysis datasets shows good consistency with buoy observations in 10m wind speed with a correlation coefficients between 0.8 and 0.9. However, the value of reanalysis is slightly underestimated compared to buoy observations with a bias below 1.3 0 m/s and a root mean square errors (RMSE) of 1.5~3 m/s. The wind direction of reanalysis has a bias tendency toward the right, which reaches 14℃ with RMSE between 40 and 50° in Wenzhou buoy station. Among the five buoy stations, Zhoushan buoy reveals best consistency with the reanalysis datasets in terms of both wind speed and wind direction. We also find that the wind of reanalysis datasets are least accurate in winter, which causes the systematic bias of wind speed and wind direction. The correlation coefficients of 2 m temperature between the two reanalysis datasets and observations are both above 0.95. Nevertheless, the reanalysis temperature tend to be higher than observations. Especially, the deviation of NCEP/NCAR temperature in four buoy stations is 1~2℃ on average, The reanalysis 2 m temperature shows best applicability in autumn, while it is most significantly higher in spring and winter.. Sea level pressure of the two reanalysis datasets reveals better applicability with little seasonal difference.

关 键 词：再分析资料 浮标站 东南近海 评估 

分 类 号：P732[天文地球—海洋科学]