基于CORA资料的中国海海表面温度季节和年际变化特征分析

基于1986-2008年的中国近海及邻近海域再分析产品(CORA),采用经验正交函数分解方法(EOF)分析了海表面温度(SST)的季节及年际变化特征,并用相应的SODA、AVHRR以及Levitus资料对CORA做了对比评估。相比于AVHRR而言,CORA资料SST的偏差和均方根误差均小于SODA,相比Levitus资料而言CORA资料温度盐度的均方根误差随深度的变化皆小于SODA。 CORA与SODA资料相比,两者前3个模态的时空分布大体一致,区别在于CORA资料能更好地反映参量的一些细微特征。结果表明,CORA资料能很好的刻画中国近海SST的季节、年际变化特征,尤其是黑潮流经区域SST的局地变化特征。季节EOF第二模态显示的是SST对由风引起的潜热释放的响应特征。第三模态刻画了冬夏转换季的分布特征,主要揭示了东北-西南走向的锋面特征。SST年际变化与ENSO密切相关,区域平均的南海SSTA与Nino指数的吻合程度CORA优于SODA。

Analysis of seasonal and interannual variability of sea surface temperature for China seas based on CORA dataset

The sea surface temperature (SST) of China Ocean Reanalysis (CORA) from 1986 to 2008 is validated using Simple Ocean Data Assimilation (SODA),Levitus and Advanced Very High Resolution Radiometer (AVHRR) dataset. Comparing with AVHRR,the bias and root mean square error(RMSE) of CORA are less than that of SODA. Comparing with Levitus,the RMSE of COAR is also less than that of SODA. The results of Empirical Orthogonal Function (EOF) analysis show that the first three modes of CORA have similar temporal and spatial variations with those of SODA dataset,and the main differences are that CORA can represent more subtle features than SODA. The seasonal and interannual variability of SST in the coastal China seas can be well presented by CORA data. The second mode of EOF reveals the SST response to wind-induced latent heat. The third mode presents the northeast-southwestern distribution of SST front,which is coincident to the transitionally seasonal characters between summer and winter. The interannual variability of SST is closely related to ENSO.The regional average of SSTA of CORA in the South China Sea has more coincident correlation to Nino indexes than that of SODA.