Abstract: By the daily, 5-day, 10-day, monthly and seasonal mean images from hourly GMS-5 infrared images, an analysis of cloud characters of severe storm rainfall during the flood of Yangtze river in 1998 was made. The results show that the mean cloud images disclose the mean features of the severe storm systems. These mean images reveal not only the synoptic cloud systems but also the mesoscale convective cloud cluster. The long-term high water level of mid-Yangtze river and the frequent flood peak of the upper Yangtze river from the middle of June to the middle of August in 1998 was caused by the high frequent activity of convection. The 5-day mean and daily mean images show that the long severe rainfall of Yangtze river was produced by the meso-α scale cloud clusters in the end of July. Additionally, there is somewhat possible correlation between the daily mean cloud image and the daily precipitation. The future research is possibly expected to estimate the flux of water by means of the T_BB of the mean cloud images.

Key words: flood, rainfall, satellite image

摘要： 利用GMS卫星的逐时红外云图做成了日、候、旬、月和季平均云图，在此基础上对1998年夏季长江中、上游洪水期间暴雨的云图特征进行了分析。结果表明，多日平均云图可以概括地揭示降水系统的主要特征。它不仅可展示天气尺度的暴雨云带，而且在日和候平均云图上也可以相当清楚地展示出云带内所包含的中尺度对流云团。旬、月和季平均云图表明，长江上游从6月中旬开始到8月中旬长达两个多月对流的频繁发生是“98洪水”长江上游洪峰多，中游高水位维持时间长的重要原因。候和日平均云图表明，降雨云带上频繁产生中α尺度暴雨云团造成了7月下旬长江中游连续出现特大暴雨。用每小时一次的红外云图作出的日平均云图与日雨量分布之间存在一定程度的对应关系显示出用日平均云顶黑体辐射温度(T_BB)来估算径流量的可能性。

关键词: 洪水, 暴雨, 卫星云图