Relationship between Arctic clouds and synoptic-scale variability

The main goal of this study is to examine possible relations between Arctic cloudiness and underlying synoptic large scale environment responsible for its generation. As a first step in this analysis we identify those geographical areas under the influence of similar weather conditions. In an attempt to determine them, a statistical procedure consisting of principal component analysis (PCA) and a two-stage cluster analysis was carried over the region north of 50 degrees N. Based upon a 10 year period of NCEP reanalysis data, the classification scheme uses the geopotential height, air temperature, RH and wind magnitude at 17 levels to characterize the weather at each point. PCA is used to eliminate the mutual collinearity of the original variables and as a result, the meteorological conditions at each location are described by a smaller number of orthogonal (uncorrelated ?) principal components. The NCEP data are then clustered according to the similarity of their principal component loadings using a two-stage clustering procedure, thus producing a number of homogeneous regions experiencing similar weather variability. Furthermore, we will also present an analysis of the correlations between Arctic cloudiness and cloud type and each synoptic regime.