Application Of GMS-5 Satellite Data For MM5 Mesoscale Numerical Prediction Model

The application of the satellite data has enlarged to every meteorology science. It applies useful data for the weather forecast, global climate change, ocean condition, crop growth and forest fireproofing; especially it improves the capability of the numerical forecast. This paper used the GMS-5 Satellite images to distinguish cloud or clear sky and classify high cloud and low cloud and determine semitransparent or fractional cloud and middle cloud. The result was used to adjust relative humidity and to enhance the ability of MM5 mesoscale modeling system to produce accurate forecast of precipitation.We define the air condition includes 5 kinds: the clear sky, semitransparent or fractional cloud, high cloud and low cloud and middle cloud .in this process, we present the method development for the generation of cloud based on GMS-5 images. MM5 (fifth-generation Perm.State/Near Mesoscale Model) output will be extensively used for the off-line computation of dynamic changeable mutispectral thresholds in order to adapt to variable weather using statistical regressive relations produced by optimal regressive analysis. So we through the cloud types from GMS-5 images can get the cloud-top and cloud-base altitudes for each grid box .in addition, at each grid box we counted cloud fraction (n). At grid point where the satellite cloud fraction is nonzero we adjust the relative humidity based on MM5 cloud fraction equation. Then the initial field of the relative humidity was adjusted, and was used in the MM5, we made a few examinations to improve the forecast result.