This study presents the development and results of an automated tool for the estimation of lower-tropospheric wind speed and direction based on cloud motion. This tool uses visible and near infrared (3.9µm) images from the geostationary GEOS-12 satellite. The visible band's high spatial resolution and contrast between low clouds and land surface were highly beneficial from tracking cloud motions. For nocturnal detection of low clouds, the near infrared channel performed better than both the infrared window and water vapor channels. The estimation of wind-related variables was obtained successfully over oceanic surface. However, over land, given surface heterogeneity, wind retrieval was not successful and only a few wind vectors were retrieved. The study presents the methodology used to assess cloud-base height, quality-control techniques, and validation whit NCEP data.
wind; atmospheric circulation; meteorological satellites; clouds