利用整层模式大气统计求解GRAPES-3DVAR动力平衡约束的数值试验

NUMERICAL EXPERIMENTS ON STATISTICAL ESTIMATION OF DYNAMIC BALANCE CONSTRAINTS IN GRAPES-3DVAR WITH WHOLE LAYERS OF MODEL ATMOSPHERE

针对采用单层模式大气中统计求解风压场平衡关系的不足,采用改进方案,在表达动力平衡约束时考虑不同层次上变量之间的相关,利用整层模式大气中的旋转风场统计求解每一层上质量场的平衡部分。改进方案通过类岭回归方法减少多重共线性以及小尺度噪音对统计求解的不利影响,提高了统计结果的稳健程度。数值试验结果表明,相对于原有方案,采用整层模式大气统计求解的方案能更好地保证质量场在垂直方向分析的合理性,减少数字滤波初始化对分析增量的调整,提高位温的分析效果。在冬、夏两个代表月同化预报循环的对比试验中,改进方案对温度场的预报效果有明显提高,对风压场预报也有一定程度的正效果。

This paper further explores the statistical methods of estimating and expressing dynamical balance constraints in GRAPES-3DVAR （Version GM）.To overcome the shortcomings in a single-level statistical estimation scheme,the correlations among variables are taken into account for different levels and the rotational wind at all model levels is used to calculate the balanced mass field at each layer.In order to avoid the multi-collinearity problem and reduce the noise caused by unbalance mesoscale disturbances,reformed ridge regression is used in the new regression procedure.The results of numerical experiments show that,by using the new constraints scheme,the system can get more reasonable vertical mass fields,reduce the magnitude of the adjustment by the initialization,and improve the analysis of potential temperature.Furthermore,the results of forecast verification for January （winter） and July （summer） both confirm that the new statistical constraints scheme can significantly improve the temperature forecast accuracy and bring slightly positive effects to the pressure and wind forecast.