两种短期降水集合预报方案的对比试验

将MM5V3作为试验模式，利用初值扰动和物理参数化方案，分别对东南沿海地区2004年7—9月的短期降水进行了模拟，并运用均方根误差法对模拟试验效果进行了评估检验，以比较两种方案的集合预报效果。结果表明，对于汛期短期降水预报而言，两种方案的预报效果都有一定的改进。相对于初值误差而言，积云对流过程在汛期降水过程中起着更为重要的作用，因而物理参数化方案扰动的集合预报效果，总体优于基于初值扰动的集合预报效果。研究还发现，模式物理参数化方案扰动的集合预报效果在9月出现了明显的下滑，这主要由于7—8月对流活动旺盛，而9月对流活动随热力因素减弱而减弱的原因所致。

Comparative Trial of Two Short-term Precipitation Ensemble Prediction Schemes

With the initial perturbation and physical parameterization schemes, the MM5 V3 was employed as experimental model to simulate the short-term precipitations during July to September of 2004 in the southeast coast area, and the effect was evaluated by the root-mean-square error method to compare this two kinds of ensemble schemes. The results indicated that to the flood season short-term precipitation, both the schemes can improve the forecast result to a certain extent. Compared to the initial perturbation, the cumulus cloud convection process played a more vital role in the flood season precipitation, thus the ensemble prediction effect of physical parameterization scheme perturbation was better than that of initial perturbation. The research also discovered that the ensemble prediction effect of physical parameterization scheme perturbation exhibited signs of decline in September. The reason was that the convective activities were significant during July to August, and weakened in September due to less thermal energy.