Abstract
The impact of satellite-derived cloud motion vectors (CMVs) on analysts of winds measured by rawinsondes during the 1979 SESAME Experiment is studied in two case studies (10 April and 9 May 1979). Cloud motion vectors are both arbitrarily assigned and vertically interpolated to typical “low” levels of 825 mb and &sigma = 0.9 before being combined with the rawinsonde-measured winds at these levels. Magnitudes of vector differences between the combined winds and rawinsonde-measured winds are computed to show the effect of single-level assignment of CMVs on the rawinsonde-measured wind fields. The impact of the existence of horizontal and vertical gradients of wind and moisture on these results is also examined. In addition, divergence and relative vorticity fields are derived to determine whether the addition of CMVs increase the amount of useful information to the kinematic computations.
The results show that the standard method of arbitrarily assigning wind vectors to a “low-level” coordinate surface yields systematic differences between the rawinsonde-measured winds and combined wind fields. Arbitrary assignment of cloud motions to the 0.9 sigma surface produces smaller magnitudes of vector differences than assignment to the 825-mb pressure surface. Additionally, systematic differences in the wind occur near moisture discontinuities and in regions of horizontal and vertical wind shears. If forced to make arbitrary assignments, the use of the terrain-following sigma surface yields more consistent results than arbitrary assignment to a pressure surface in the lower troposphere. The differences between the combined and rawinsonde-measured wind fields are reduced by vertical interpolation to either a pressure or sigma surface. However, the accuracy of these interpolated fields depends to a large extent on the methods used to determine cloud-base levels and the vertical wind shear.
