An Improved Algorithm For Extracting Atmospheric Motion Signals In"Cloud-free Region" From GEO Satellite And Its Applications In Case Study

The technique for atmospheric motion vectors (AMVs) derivation for cloud-free regions from satellite infrared images is to analysis of the sensitivity to infrared radiation and extract the possible water vapor or dust aerosol tracers from it. But infrared radiation in atmospheric window region is largely impressed by the surface emission infrared radiation, so quadratic difference method is raised in "clear sky region" from FY-2E infrared channel to weaken the surface temperature interference and help trace the weak signal in "clear sky region" by means of split window and temporal difference calculation from infrared cloud mask pictures.With the MODTRAN4radiative transfer model, the total increase in brightness temperature (BT) in FY-2E IR window imagery corresponding to a1K increase in air temperature over half an hour, which contributions to the total increase from each of the10layers has been analyzed under3typical conditions:mid-latitude moist summer, dry summer and dust winter atmospheres. The variation in brightness temperature corresponding to1K surface temperature,30%increase in water vapor density and50%increase in aerosol extinction coefficient over half an hour, the contributions to the total variation has been analyzed under2typical conditions: mid-latitude moist summer and dust winter atmospheres. Validation proves the quadratic difference method can further reduce the disturbance of surface temperature change in "clear sky region" from FY-2E infrared channel and help get water vapor or aerosol cloud texture and mobile information.Examples of low-level winds for cloud-free regions derived from FY-2E IR images using the quadratic difference algorithm have been conducted. Results show that this method can make up for the lacking wind field data in clear sky regions compared to the obvious split window and temporal difference calculation, finally a comparison between the FY-2E IR AMVs using this technique for cloud-free regions and that obtained from the NCEP reanalysis data wind observations shows a good relative accuracy and agreement.