Spatio-temporal trends and variability of rainfall in Maharashtra, India: Analysis of 118 years

Abstract

This study examined the spatio-temporal trends and variability of seasonal and annual rainfall for 36 districts of Maharashtra, India. For this purpose, 118 years (1901 to 2018) gridded rainfall data of India Meteorological Department (IMD) were analyzed using Mann-Kendall (MK), modified Mann-Kendall (MMK), Sen’s slope (SS), Spearman’s rank correlation (SRC), simple linear regression (SLR), and innovative trend analysis (ITA). Auto-correlation coefficient was calculated at lag-1 and tested at 5% level of significance. Rainfall variability was examined using the coefficient of variation (CV). The analysis revealed significantly decreasing trends for winter and pre-monsoon rainfall in districts of Maharashtra. Monsoon, post-monsoon, and annual rainfall had both increasing and decreasing trends. Out of 185 time series analyzed, ITA detected trends in 168 (90.8%) time series. All the trends detected by MK/MMK, SRC, and SLR were captured by ITA, along with trends in additional 103 (55.6%) time series which were not captured by any of the aforesaid methods. Rainfall variability was very high in all the districts for winter, pre-monsoon, and post-monsoon seasons. The trends and variability analysis of rainfall in the state along with their maps would be useful for the local stakeholders for planning efficient use of water resources.