Cloud Detection And Atmospheric Correction For A Routine Mapping Of Vegetation Index With The AVHRR Of Noaa11

AYRRR Daytime AVHRR data from NOAA-11 are routinely processed in CMS in Lannion for different meteorological purposes. This paper presents the use of this automatic scheme to achieve vegetation index mapping on a limited area. The first step is a fully automated cloud detection applied to all the AVHRR pixels of the orbit. The AVHRR image is then located by automatic landmarks recognition and then remapped on stereographic projection at 4km resolution. The visible and near-infrared channels are then corrected from atmospheric effects (except from aerosol) with the water vapor content derived from the radiosonde measurements and the vegetation index is then computed.