ERA5海面气压和风速再分析资料在渤海和北黄海适用性分析

利用渤海和北黄海9个站位浮标的海面10 m气压和风速观测资料对ERA5再分析资料的适用性进行了初步评估。结果表明：海面气压和风速再分析资料与观测资料具有良好的相关性，不同时间尺度的统计结果具有一定差异，累年逐月的相关性优于日均和日极值，极端天气下的相关系数最低。不同时间尺度的统计结果显示，气压再分析资料与观测值的偏差总体为负值，即高压再分析资料较观测值偏弱、低压再分析资料较观测值偏强，二者存在一定的系统性偏差；风速统计结果显示，偏差与风力级别有关，大风速时，风速再分析资料普遍小于观测值。累年逐月再分析资料反映了研究区气压和风速的季节变化特征，偏差也呈现出季节性变化，且冬季的适用性优于夏季；极端天气情况下，气压和风速再分析资料的适用性较差，需要开展进一步的订正处理。

Applicability analysis of the sea surface pressure and wind speed of ERA5 reanalysis data in the Bohai Sea and the Northern Huanghai Seas

A preliminarily assessment of the applicability of the sea surface pressure and wind speed of ERA5 reanalysis data is carried out using the observation data at 10m height observation data of 9 buoys in the Bohai Sea and the Northern Huanghai Sea. The results show that: the sea surface pressure and wind speed of ERA5 reanalysis data has high correlation coefficients with the observation data, the correlation between sea surface pressure and wind speed is different in different time scales. The correlation of monthly average is better than that of daily average and daily extreme value, and the correlation coefficient is the lowest in extreme weather. In generally, the deviation between statistical products of the ERA5 and the observed products is negative. It means that the high pressure is weaker than the observed data, and the low pressure is stronger than the observed data, and there is some systematic deviation between ERA5 reanalysis data and the observed data. The deviation varies with the different wind speed level, when the wind is high, the reanalysis wind speed is generally less than the measured. By analyzing the monthly average data, the reanalysis data reveal the seasonal variation of sea surface pressure in the study area, and the deviation from the observed data also show seasonal variation characteristics, the applicability in winter is better than in summer. The error of reanalysis data of sea surface pressure and wind speed is large under extreme weather conditions, especially the typhoon process, further evaluation and revision of the data are needed