An Objective Analysis of the Boundary-Layer Thermodynamic Structure During GATE. Part I: Method

Abstract

A method for the analysis and objective classification of meteorological data is presented and illustrated. The first step in this approach is the application of an efficient method to describe the data both qualitatively and quantitatively. This approach, based on the singular decomposition theorem, extracts the important features from the data and is similar to principal component or empirical orthogonal function analysis. The extracted features are then combined in the classification step to form new features, each of which describes a subset of the data. The mathematical details are given and computational considerations are discussed. The method is illustrated with a set of boundary-layer potential temperature profiles derived from structure sonde data taken during the GARP Atlantic Tropical Experiment (GATE). This set of profiles, taken at 3 h intervals over a 20-day period, was objectively classified into two distinct groups. Time, series of surface divergence and radar-estimated precipitation rates are used to demonstrate that one group of profiles is associated with a convectively modified boundary layer, while the other group of profiles is more typical of undisturbed situations. Other uses and limitations of the objective classification method are discussed.