Resumen
Large-scale modes of variations in sea surface temperatures (SSTs) and the tropospheric circulation are examined with major principal components utilizing-monthly mean fields of the global SSTs and Northern Hemisphere geo-potential height (Z) at 700, 500 and 300 mb levels. The data period covered is from 1955 to 1992. It is found that the heterogeneity of SST data due to availability of satellite observations and difference of analysis schemes may result in a large systematic bias in the dataset. However, the bias may be effectively corrected through elimination of an appropriate principal component. The first three components of monthly SST and Z fields during the 38-year period are presented. The El Niño-Southern Oscillation (ENSO) mode of variations is observed in both 1st and 2nd components of SSTs. The inter-annual viariations of principal components of monthly SST and Z fields are utilized to probe the association of SST components and Z components. Further, cross-correlation patterns of principal components of SSTs in reference to the tropospheric circulation are studied with 500 and 300 mb fields. It is hown that the tropospheric response to SST anomalies are consistent at 500 and 300 mb levels, and the seasonal-range predictability of the tropospheric circulation is recognized with SST anomaly fields.