集合最优插值方法在北印度洋海浪同化中的应用  

Application of Ensemble Optimal Interpolation Method in Wave Assimilation in North Indian Ocean

作　　者：曹蕾[1,2,3] 齐鹏[1,2] 

机构地区：[1]中国科学院海洋研究所,山东青岛266071 [2]中国科学院海洋环流与波动重点实验室,山东青岛266071 [3]中国科学院大学,北京100049

出　　处：《海洋科学进展》2015年第1期45-55,共11页Advances in Marine Science

基　　金：海洋公益性行业科研专项子任务--印度洋海域海浪资料同化研究(201005033-5);海洋公益性行业科研专项--南海数值预报业务化系统研制(201105002);国家高技术研究发展计划--船载气象监测预报系统研发(2012AA091801)

摘　　要：基于第三代海浪模式WaveWatch Ⅲ,采用集合最优插值（EnOI）方法对北印度洋海浪进行同化数值实验研究。在集合样本选取方案上,针对不同的实验分别选取有效波高（SWH）的历史后报场（样本A）、24h变化（样本B）以及以同一时刻72h预报时效和24h预报时效的差异（样本C）用于估计背景误差协方差。样本A和样本B是为海浪模拟而设计,样本C是为海浪预报而设计;通过与由高度计数据确定的模式背景误差进行比较,认为样本B优于样本A。采用样本B对2011年北印度洋海浪场进行同化模拟,结果表明2011-03-11相对误差改进都在5%及以上,其中7月份改进效果最佳。采用样本C对2013-07的有效波高进行0~72h预报,发现同化使0~24h预报改进最明显：均方根误差改进0.12m,相对误差改进5%。浮标检验结果支持上述结论。Ensemble Optimal Interpolation （EnOI） assimilation of satellite altimetry data into WaveWatch Ⅲ v3.14 is applied in wave modeling and forecast in North Indian Ocean. Different ensembles are designed as the approximation of the background error. For wave simulation, we design Ensemble-A from historical significant wave height （SWH） fields and Ensemble-B from differences between the SWH fields with 24 h in- terval. For wave forecast, we design Ensemble-C from differences between 72 h and 24 h forecasts verif- ying at the same time. Ensemble-B gives a better approximation of the background error covariance than A. Two assimilation experiments in north Indian Ocean, wave simulation with EnOI and Ensemble-B for the whole year of 2011 and 0~72 h forecast with EnOI and Ensemble-C in July 2013, are conducted. The assimilation impacts on wave simulation and forecast are evaluated. In the simulation experiment, the ab- solute improvements of relative error between Envisat altimeter SWH and modeled SWH are more than 5% from March to November, with the best improvement in July. In the0~72 hforecast experiment, da- ta assimilation improves the forecast period of 0~24 h most significantly, which shows 0.12 m reduction in the RMSE and 5 % reduction in the relative error. Validations with buoy data also agrees with the above results.

关 键 词：集合最优插值 WAVEWATCH Ⅲ 海浪预报 海浪模拟 

分 类 号：P731.33[天文地球—海洋科学]