中尺度模式数值预报中初值的比较

本文利用Anthes-Warwer中尺度十层湿模式讨论了初始资料垂直分辨率的不同对次天气尺度和中尺度预报的影响对比试验结果表明,两组初值都能定性地把次天气尺度和中尺度过程的主要特点预报出来,都能揭示低层950毫巴正涡度中心与暴雨中心中对应关系以及两区上空、边界层顶附近湿暖盖的存在,但增加特性层温、湿和测风资料后,在降水量和暴雨中心位置的预报上更接近于实况 上述资料的增加使得与降水量和暴雨中心有直接关系的初始低层风场和水汽辐合场有了较大改进.

A Comparison of Different Initial Conditions in Mesoscale Numerical Weather Prediction

In this paper, AW 10-layer moist mesoscale model is used to test the impact of different vertical resolutions of initial data on subsynoptic-scale and mesoscale forecasts. The initial data includes or excludes significant-level radiosonde and balloon wind data. The results show that either with or without the reference to the significant-level radiosonde data, the main features of subsynoptic-scale and mesoscale process are predicted qualititatively, such as the moving and intensifying of LLJ, the shifting of warm shear line to the north, the existence of a warm and moist lid near the top of PBL. However, the initial conditions with significant-level data, the foreeast of precipitation and the position of maximum heavy rainfall are more realistic. Besides, the results reveal the corresponding relationship between the maximum heavy rainfall and the maximum positive vorticity at the 950 mb. The results also give evidence of the existence and higher intensity of a warm and moist lid near the top of PBL over rainfall region. Futhermore, the distributious of thelower-level wind field and water vapor convergence are rather different. It causes the differences in the vertical distribution of some physical quantities and directly results in the differences in precipitation intensity and the position of maximum heavy rainfall. These acconnts for the importance of the meteorological elements in the lower troposphere to the heavy rainfall. The simulations give evidence that significant-level data should be used for limited area weather forecast.