Level Assignment in the Assimilation of Satellite Cloud-Tracked Winds.

摘要

Heights assigned to wind vectors derived from geostationary satellite data cannot be determined exactly. When these vectors are assimilated into numerical models or analyses which define finite layers in the vertical, a compounding of errors may result. In this study, the method of determining the insertion level within an assimilation system is varied in an attempt to minimize these errors and study their effect on the resultant analyses. Analyses from the four methods of level determination were compared to each other and to the control experiment. Results show that slight differences in the level of insertion can produce meaningful differences in the resulting analyses, and also confirm the value of cloud winds in data-sparse regions. Advantages and disadvantages of the wind insertion methods are discussed.