An analysis of precipitation climatology over Indian urban agglomeration

While urban areas in India are rapidly expanding, the analysis of how the precipitation regimes are changing is very limited. In the present study, an attempt has been made to explore the trends in rainfall pattern over 20 most populated urban agglomeration (cities) of India using high resolution gridded, daily rainfall product (0.25 degrees x 0.25 degrees) obtained from India Meteorological Department (IMD). The analysis comprised of annual and monsoonal rainfall and rainy days; 1- and 10-day annual maximum rainfall; 95 and 99 percentile extreme threshold rainfall magnitude in 30 years moving window; and number of events above 95 and 99 percentile threshold in each year during 1901-2015. The Mann-Kendall (Modified Mann-Kendall) test at 0.1 significance level to detect trends, and the Thiel-Sen's approach to estimate% change over mean have been used. Annual and monsoonal rainfall are found to be increasing with shrinking rainy days during the same period that infers rise in heavy rainfall events. Besides, the majority of the cities show increasing trends in 1-and 10-day annual maximum rainfall, number of extreme threshold events, and magnitude of extreme threshold rainfall in 30-year moving window. The Pettitt-Mann-Whitney (PMW) test was used to detect change point year. Cities with significant change point at probability level >= 90% were selected to analyze pre-and post-change point year trend in time series. In addition, the trend in precipitation variables for multiple climatic windows of 30 years with an advancement of 5 years in each window (1901-1930, 1906-1935 ..., 1986-2015) show that unlike a unique long-term trend, different climatic windows have varying trend direction over the period of time. Overall, the analysis reveals the rising trend in extreme rainfall events over majority of the cities in long term (1901-2015); however, mixed pattern of trends are observed in moving climatic windows.