Patterns of India monsoon rainfall anomalies

Abstract

Pattern characteristics of annual rainfall anomalies over India during 1900—72 are studied from precipitation records of 31 regional subdivisions and series of annual cyclone frequency and number of break monsoon days. Methods include spatial correlation, pattern stratification, and principal component and spectral analyses.

Pattern stratification with respect to an all-India rainfall index yields largest concordant departures in the Central and Western portions of the country. Pattern stratification with respect to the largest annual frequency of cyclonic storms in the Bay of Bengal and Arabian Sea show large positive rainfall departures over Eastern. Southeastern and Western India, and negative departures in the North and a limited area of the Southwest: for the collective of years with smallest storm frequency a partly inverse rainfall anomaly pattern is obtained. Pattern stratification with respect to the years of most numerous break monsoon days yields deficient rainfall in a large, broadly zonally oriented band across Central India, and positive departures in most other parts of ther country. Stratification with respect to the least numerous break monsoon days yields a broadly inverse pattern.

The first eight principal component patterns account for 74% of the total variance. The first component resembles the pattern obtained by stratification with respect to the all-India rainfall index, the two time series being highly positively correlated. The third component is approximately inverse to the pattern constructed from stratification with respect to the most numerous break monsoon days, the two time series being negatively correlated. The seventh component is similar to the pattern derived from stratification with respect to the largest frequency of cyclonic storms: this is consistent with the positive correlation between the two time series. No physical significance is apparent for the other principal components. The combination of the various analysis techniques thus serves to identify spatial and temporal pattern characteristics of possible physical significance.