Further Improvements of Cloud Motion Wind Extraction Techniques

The displacements of clouds in.successive satellite images reflects atmospheric motions at various scales. The main application of the satellite derived cloud motion vectors at a scale of about 100 200 km, is their use as winds in the data analysis for numerical weather prediction. This paper focusses on improvements to existing techniques for deriving cloud motion winds (CMW) from the present generation of satellites. Taking recent improvements of the operational CMWs from METEOSAT as an example, the impact of different changes on the quality will be discussed. It is shown that better height assignment of a wind vector and radiance filtering techniques preceding the cloud tracking have ameliorated the errors in METEOSAT winds significantly. For the foreseeable future CMWs will be used as single level winds in the data assimilation schemes and improvements to existing CMW retrieval techniques have to be made accordingly. Areas for improvement are seen in the height assignment, automatic quality control, and quality flags that would allow a more effective use of CMWs in the data assimilation.