集合卡尔曼滤波同化探空资料的数值试验

Numerical Test of an Ensemble Kalman Filter for Sounding Data Assimilation

摘要:

应用集合卡尔曼滤波(Ensemble Kalman Filter;EnKF)方法,同化了2005年7月一次暴雨过程的探空观测资料,并用非静力中尺度模式MM5进行数值模拟试验。结果表明:在理想模式的假设下,即假设真实模拟和所产生的集合用的是同一个模式并有相同的初始误差,EnKF方法同化的分析结果较好。如果不运用EnKF方法同化探空观测资料,则集合预报结果和不加扰动的单个数值预报结果都没有EnKF方法同化过的好。

Abstract:

Ensemble Kalman filter(EnKF) is used to assimilate the sounding data of a torrential rain process in July 2005 into the mesoscal model MM5.The experiment results show that under the assumption of perfect model,the analysis of EnKF assimilation is closer to the real observations,and the emsemble prediction with the EnKF assimilation is better than that without the EnKF assimilation and the single numerical prediction without adding perturbations.