Use and Impact of Satellite Atmospheric Motion Winds on ECMWF Analyses and Forecasts

Abstract

Recent developments in the extraction technique of wind vectors from geostationary satellite images have led to a significant increase in the data coverage of this product. This is mainly due to the use of visible and water vapor imagery in addition to the infrared. To assess the impact of these additional data on the European Centre for Medium-Range Weather Forecasts analysis and forecast system, data assimilation experiments have been performed. Winds from the water vapor channel images of the two U.S. satellites, GOES-8 and -9, and from the high-resolution visible channel of the European satellite, Meteosat-5, were included in the assimilation process in addition to the other operational observations. The results showed consistent improvements in the quality of the model first-guess wind field and showed how the new winds have a significant positive impact on the short-range forecast of the circulation in the Tropics and Southern Hemisphere. The results of a global observing system experiment withholding all satellite winds from the observing system are also discussed. The relative impact of other data types and the question of forecast verification in data-sparse regions are addressed.