A LOADING CORRELATION MODEL FOR CLIMATIC CLASSIFICATION IN TERMS OF SYNOPTIC CLIMATOLOGY

By using the orthogonal transformations between principal components a nd original variates, a loading correlation model is proposed as a tec hnique for climatic classification derived from synoptic climatology. The criterion for a loading correlation model for a climate region is that all the first loading correlations are positive and significant. The climate region includes all the stations chosen in the model. The first positive significant loading correlations and other significant ones in the model, fitted to the climate region, can be identified as the average circulation type and the average circulation features, res pectively, by using the decomposition of the original variates. Thus, the regional climate can be described in terms of synoptic climatology . The mathematical structure of the loading correlation model is based on empirical orthogonal functions with two additional parameters, the standard deviation of the principal components and the original varia tes.