摘要
本文用σ-坐标原始方程全球谱模式,对1979年1月23日和6月14日两个个例,采用3种不同的垂直分层方案,进行了敏感性试验.通过对高度场均方根误差的分析,发现15层模式对北半球冬季预报效果的改进最明显.当其降低模式最高层高度和减少垂直分辨率时,预报误差首先从模式上层出现,然后影响到对流层中、下层.并指出南、北半球,冬、夏季不同层次上的预报对模式垂直分层的敏感性存在有差异,对南、北半球,冬、夏季的预报可采用不同的垂直分层模式. 南、北半球,冬、夏季预报敏感性差异的产生同行星尺度波动的水平感热通量的垂直分布、行星尺度波动能量的垂直输送特征有关.
In this paper, the Global Spectral Models in coordinate (T21L5 , T21L9 , T21L15) of IAP are used. The effect of the upper boundary and vertical resolution on numerical forecasts is studied by comparing the results of the 15-layer model with the top at 33.3hPa, to those of 9- and 5 -layer models with the top at 55. 5hPa, and at 100 hPa, respectively. Two cases for 14 June and 23 January 1979 from FGGE data have been chosen as initial fields. Five-day predictions have been made.The results show that the tropospheric forecasts are more sensitive to changes of the top level altitude and the vertical resolution in winter, while the forecasts in the Northern Hemisphere winter case for 23 January are more sensitive than those in the Southern Hemisphere winter (case for 14 June).Above differences are mainly due to the following reasons. The vertical distribution of the zonal mean eddy heat flux (v ' T' )for wavenumber 1 - 3 and the vertical propagation of planetary waves in the Northern Hemisphere winter differ from those in the Southern Hemisphere winter and in summer. The zonal mean northward eddy heat flux (v' T') for wavenumber 1 - 3 in the Northern Hemisphere winter is stronger than that in the Southern Hemisphere winter.
出处
《大气科学》
CSCD
北大核心
1993年第5期563-575,共13页
Chinese Journal of Atmospheric Sciences
基金
国家"七五"科技攻关项目
关键词
全球谱模式
敏感性试验
数值预报
Global spectral
Sensitivity experiment
Root-mean-square error in geopotential height.