Distribution and Prediction of Monsoon Rainfall in Homogeneous Regions of India: A Stochastic Approach

The economy of India is closely and vitally linked with the rainfall during the southwest monsoon season, which occurs from June to September. Here, a comprehensive analysis has been carried out to understand the variations and distribution of southwest monsoon rainfall over different meteorologically homogeneous regions in India using the long-term monsoon rainfall data from 1951 to 2020. Analysis of means (ANOM), a graphical procedure, has been used for this study, and an attempt has been made to characterize the distribution of monsoon rainfall for different regions using univariate probability distributions based on goodness-of-fit tests. Analysis reveals that Northwest India (NWI) and South Peninsular India (SPI) undergo a downward trend, with NWI receiving less monsoon rainfall than SPI. In contrast, the East and Northeast India (ENEI) region shows an upward trend and receives the most significant monsoon rainfall. At the same time, Central India (CI) shows an upward trend and receives consistent monsoon rainfall. The Weibull distribution is a suitable probability function for characterizing rainfall distribution over all homogeneous regions. The Gumbel distribution is also suitable for NWI. Structural time series modeling (STSM), with the hidden components of deterministic trend, deterministic cycle, and stochastic irregular component, is highly effective for modeling and predicting the amount of monsoon rainfall in various meteorological regions of India. A decrease in the amount of monsoon rainfall is predicted for ENEI and CI, whereas the NWI and SPI regions will receive slightly increased monsoon rainfall. A significant decreasing trend in monsoon rainfall is observed for CI. This study is helpful for assessing flood or drought situations and undertaking necessary planning.