我国大陆大气水汽含量的计算

AN ESTIMATION OF TOTAL ATMOSPHERIC WATER VAPOR IN THE MAINLAND OF CHINA

本文利用高空探测站点的逐日、逐月和多年平均资料,研究了不同时间尺度大气中水汽含量的间接推算方法。结果表明,利用统一公式借助地面水汽压推算的累年月平均值,具有较高的可信度,其平均相对误差一般不超过10%,历年月平均推算值要差一些,其相对误差在13—15%之间,而逐日值则较难利用地面湿度资料进行推算,尤其是南方的夏季。

The total atmospheric water vapor is a neccesary data for not only the quantitative forecast of cloud and rainfall,but also radiance computation. The distribution of the average total precip-itable water in China has been studied by many scientists. They all used the data from upper-air stations, and discussed the mean values of years. The upper-air station are not many. When people want to get total atmospheric water vapor in the aeras where there are no upper-air stations,they usually use an indirect estimating method. However,this kind of method hasn' t been established in China. We generally use the following Hanns, formila: W = Ce? where W is total precipitable water,e is surface vapor pressure,and parameters C,d are undetermined coefficients. On the basis of the recent research abroad,these parameters have rather large temporal and spatial variability. If we indiscriminately imitate those formula abroad,the error will be great. This study shows that the estimation of the monthly mean value of annual total precipitable water based on the surface water pressure feasible. A good linear correlation appears between total atmosperic water vapor and surface vapor pressure in China, which is as follows: W: annual average monthly mean total precipitable water e-annual average monthly mean surface vapor pressure correlation coefficient R=0. 99 Da-ta for this study are collected from forty-eight main upper-air stations( 1960-1969). The total precipitable Water W in the formula above is revised properly, and the accuracy is high. The aver-age percentage of errors is less than 10%. Owing to a limit of calculating load,only ten representative upper-air stations are selected from main climatic regions in China,and their annual monthly mean values are analysed from 1983 to 1986 in this research. The results are as follows; (l)The linear correlation is very good between total atmospheric water vapor and surface vapor pressure,and the annual correlation co-efficients are more than 0. 98 nin different areas; (2)The correlation coefficients vary a little in different years,but the annual variation is not obvious. (3)By means of testing the variation in different areas is not obvious too. Therefore, a commonly used computing formula could be ob-tained by synthesis in China. Of course,the error will be increased correspondingly. (4)The aver-age percentage of errors in most of these stations is less than 10% ,and those stations at which er-rors are than 10 % are located in the northern part of China.Utilizing the daily data in January and July for 10 upper-air stations from 1960 to 1969(10 years) ,the correlation between daily total precipitable water and surface vapor pressur is analysed . The results show that the variation of correlation coefficients is very great in different years. The correlation coefficients in January are higher than those in July in most areas except the western part of China(Xin Jiang and Tibet). In the southern part of China,it is impossible to find out this kind of correlation in Summer. Up to now,no method which can estimate daily total atmospheric water vapor by using daily surface humidity information has been found.