Banerjee et al. showed for the first time that the number of Indian subdivisions with normal or above-normal monsoon rainfall is related to the location of the April 500-mb ridge along 75°E. Thapliyal brought out the relationship between monsoon rainfall of peninsular India and this ridge.
A detailed investigation of the relationship between all-India (India taken as one unit) monsoon rainfall, as well as monsoon rainfall of the subdivisions of India, and the location of the April 500-mb ridge along 75°E, and of the stability and consistency of the relationship has been made in this study, which is based on data for 1939–84. The relationship between all-India monsoon rainfall and the ridge location is positive (correlation coefficient = 0.71) and is significant at the 0.1% level and that between the subdivisional monsoon rainfall and the ridge is significant at 5% or above for all the subdivisions lying west of 84°E and north of 12°N. During the years when all-India monsoon rainfall is deficient, the ridge position is south of its normal position and the contribution by these years to the covariance is very high.
The stability and consistency of the relationship between all-India rainfall and the ridge location have been examined over sliding 10-, 15-, and 20-year periods, and it is found that the relationship is significant at the 1% level for all 20-yeat periods. Thus the relationship is characterized by high stability and consistency for periods of 20 years or more.
The contingency table for the two parameters shows that when the ridge is south of the mean position by more than one standard deviation, all-India rainfall is deficient on 80% of such occasions and that when the ridge is similarly located to the north of the mean, all-India rainfall is excessive on two-thirds of such occasions. The regression equation between all-India monsoon rainfall (y, in cm) and the ridge location (x, in degrees), based on data for the period 1939–80, is y = 38.02 + 3.10x. This relationship explains about 53% of the total variance. So far, the ridge location appears to be the only single parameter that explains such a high percentage of the variance of all-India monsoon rainfall. Estimates from this relationship for the independent years 1981–84 are found to be quite good, except for 1983, the year of excessive rainfall.
The southernmost ridge location was 11°N and the northernmost, about 18°N. It is intriguing to find that a variation of 7° in the latitude of the ridge location over western peninsular India during April makes a large difference in all-India monsoon rainfall (from deficient to excessive). Possibly, the ridge locations with relatively high departures from the normal in April have a tendency to persist, resulting in early or delayed transition to the tropospheric conditions from premonsoon season to the monsoon season.