Cloud retrieval from satellite data in the regions of snow and ice cover is generally acknowledged to be difficult with the present generation of meteorological satellites. Despite potential advances to be made in this field (e.g., 1.6 μm sensor to be operated on NOAA satellites) large-scale cloud analysis techniques are likely to require information on the location of the margin of seasonal snow cover. In this paper we investigate the accuracy of one snow-cover dataset currently utilized by a global nephanalysis model, the United States Air Force (USAF) RT Nephanalysis, and investigate the effect of inaccurates in the snow cover information on the derived cloud field. There are found to be situations where the snow model is inaccurate because of the advanced gate of the spring melt with respect to climatology which causes errors in the derived cloud amount. The USAF policy of including surface cloud observations in the RT Nephanalysis leads to correction of the erroneous cloud amount in regions where surface observations are available.

This content is only available as a PDF.