A Compare Of The Thermal Infrared Signal From FY-2C And TRMM Satellites And The Investigation On Rain Detection By The Thermal Infrared Signal

To obtain an accurate precipitation forecast is a goal which many meteorological scientists want to pursue for many years. However, there are lots of uncertainties in our forecast up to now. There are many reasons for this, but the inaccurate space and time measurement of precipitation is a key problem to this.The FY-2C geostationary satellite, has been already on its orbit, equips VISSR, and scans the earth every half an hour. The precipitation retrieval method mainly depends on the data form the scanning radiometer and the data from ground precipitation measuring stations. Because of the missing of the real precipitation data upon the sea surface, this method may have problems in estimating the precipitation upon the sea surface. In order to solve this problem, this study uses the products of the TRMM/PR and TRMM/VIRS to analyze the infrared radiation characteristics of clear sky, no-rain clouds and rain clouds in different weather conditions. Furthermore, 7 typical heavy precipitation examples are selected, and the data from PR and VIRS are used to define the threshold of detect heavy rain in East Asia in summer. In order to apply these results to the data form FY-2C, the relationship of corresponding channels from the VISSR onboarded on FY-2C and VIRS are analyzed and the conclusions are list below:Generally, the brightness temperature of the no-rain pixels in channel 4 (10.8μm) is usually larger than 280K and that of the rain pixel is below 260K in mid and low latitude of the East Asia in September. The BTD45 of the no-rain pixels is between 1.5K and 4K ,and its BTD34 is below 5K. The cloud is thicker in rain pixels than in no-rain pixel, and the BTD45 of rain pixels is usually lower than that of no-rain pixels. However, BTD34 of most of the no-rain pixels is larger than that of rain pixels, but the difference is just a little.7 typical heavy precipitation examples are analyzed. The result shows that single channel threshold (260K in Channel 4) can detect the most of the rain pixels, but it will take some no-rain pixels as rain pixels. The detection ability of the no-rain pixels is greatly improved by introducing the threshold of BTD45 (equal to 3K), although that of the rain pixels is little reduced. However, the BTD43 has no use in the detection of the deep convective precipitation in summer.Finally, in order to apply the result above to the data from the geostationary satellites, the differences of corresponding channels of TRMM/VIRS and FY-2C/VISSR were investigated. Above 70% of the two kinds of the scanners'infrared brightness temperature difference is lower than 5K; they are consistent in thermal infrared channels. However, the brightness temperature of VIRS is usually lower than that of VISSR in most pixels in channel 3 (3.75μm). The difference between the thermal infrared of the TRMM/VIR and FY-2C/VISSR is larger in daytime than in the night, the difference between them is about 5%; however, the difference between the channel 3 (3.75μm) of the TRMM/VIR and FY-2C/VISSR is lower in daylight than in the night, the difference between them is about 4%. The statistical results show that the differences of the corresponding three infrared channels between the TRMM/VIRS and FY-2C/VISSR are good in clear sky but become worse while the height of cloud top increases. Furthermore, this study analyzes the difference between the cloud parameter ( optical thickness, cloud particle scale, cloud top height ) retrieved by thermal infrared channels using the relationship between the difference of the corresponding thermal infrared channels between the TRMM/VIR and FY-2C/VISSR and a single infrared channel.