Characterizing long-term rainfall trends in a sub-catchment of Lower Mahanadi River Basin, India

Global warming and consequent climate change have altered the spatio-temporal variability of hydro-climatic variables. It is the primary cause of high-intensity and frequent extreme events and modified seasonality in several parts of the world. In this context, adequate and reliable monsoon and non-monsoon rainfalls are of paramount importance for the proper management of cropping patterns and irrigation water for agriculture-dependent countries like India. The concerns of hydrologists and meteorologists for identifying gradual and abrupt changes in the rainfall time series have resulted in the development of advanced graphical trend analysis methods in recent years. The present study was conceived to analyze temporal variability in the long-term (1951-2014) monthly rainfall time series of the Lower Mahanadi River basin using non-parametric trend analysis methods such as classical Mann-Kendall (MK) test, Innovative Trend Analysis (ITA), and Innovative Polygon Trend Analysis (IPTA). The results of ITA and IPTA revealed a significantly decreasing trend (α = 1%) of early phase (June: - 0.81 mm/year) and late phase (September: -0.95 mm/year) monsoon rainfalls for about 63% of the rainfall stations and significant increasing trends for the remaining stations. On the other hand, the rainfalls in July and August exhibit a significantly increasing trend (July: 1.86 mm/year; August: 2.51 mm/year) for about 90% of the stations. The rainfalls during the post-monsoon and winter seasons show a significant decreasing trend for most of the rainfall stations situated in the central portion of the study area. Furthermore, the MK test identifies similar trends for the monthly rainfall time series, but very few stations have statistically significant trends. It has been observed that the ITA and IPTA trend methods along with significance tests are more sensitive in identifying trends in the monthly rainfall time series as compared to the classical MK test. Particularly, the IPTA method provides an improved approach for visualising the 'slope of the trend'. The IPTA method indicated that August is the trend transition point for most stations with an average trend slope of 2.47 mm/month.