中文摘要：

目前,卫星资料在数值天气预报中的同化应用显著改善了数值预报效果,尤其是在全球数值预报中更为明显。而在区域数值预报中,卫星资料同化应用的影响还需要开展进一步的研究与分析。采用基于GRAPES-3DVar同化系统和中尺度数值模式WRF构造的区域数值预报流程,引入NOAA16卫星资料分别对T213和T511不同背景场进行变分同化,分析检验了卫星资料直接同化应用对降水和台风天气系统个例效果的影响。结果表明,卫星资料的应用能较好地改善降水的区域和强度,强降水中心位置更接近实况。台风路径预报初期12—18 h同化效果不明显,预报路径甚至偏离实况,而在后期同化效果明显,预报路径更接近实况。

英文摘要：

Satellite observation data are already used in the numerical weather forecast and have improved the prediction accuracy greatly, which play a more important role in global numerical forecast. While for the regional numerical prediction, the affect of satellite data assimilation need to be researched and analyzed more. In this paper, the regional numerical forecast system is built by using the GRAPES-3Dvar data assimilation system and regional mesoscale model WRF. NOAA16 satellite observations are introduced to assimilate different background fields of T213 and TSll data. The effect examination of satellite data assimilation is carried out to simulate the rainfall and typhoon cases. The results show that the assimilation of satellite data may improve the region and intensity of rainfall, the centre of precipitation is close to the actual situation. For the forecast of typhoon path, the effect of assimilation is not clear in the early 12--18 h, but in later,the effect is positive.