EMPIRICAL ORTHOGONAL FUNCTIONS ASSOCIATED WITH PARAMETERS USED IN LONG RANGE FORECASTING OF INDIAN SUMMER MONSOON

Abstract

EEmpirical Orthogonal Functions (EOF),. associated with the; parameters for long range forecasting of Indian summer monsoon onset and seasonal. rainfall have been discussed. It was found that the percentage of variance explained was 77 and 67 respectively through the first four EOF. The highest correlation coefficient with the onset date was found for the first function which showed the maximum influence of Cobar (Australia) and Darwin (Australia) zonal winds on the onset date. It was interesting to note that for rainfall prediction predominant effect on the first EOF was noticed of 50 hPa ridge over northern hemisphere, Eurasian snow cover, Argentina pressure (negatively correlated) and 500 hpa ridge, 10 hPa Balboa wind, north, central India and east coast  minimum temperatures, and northern hemisphere temperature. However, the Influence of EI-Nino, equatorial pressure and Darwin pressure (Including Tahiti minus Darwin) and Himalayan snow cover was almost negligible. The eigen index for the onset date suggests a complementary method for its application In long range prediction of summer monsoon onset date.