Indian Journal of Geo Marine Sciences Vol. 49 (11), November 2020, pp. 1772-1778 Study of trend of rainfall data: A case study of Koraiyar basin, Tamil Nadu N Surendar* ,† & R Nisha $ Department of Civil Engineering, National Institute of Technology Tiruchirappalli, Tiruchirappalli – 620 015, India [E-mail: † surendarnatarajan86@gmail.com; $ nisha@nitt.edu] Received 20 June 2019; revised 16 September 2020 Rainfall is an vital hydro-climatic element at the helm of that heads the hydrological cycle and water availability in a specific region. In current years, significant deviances have been observed in rainfall patterns on a global as well as local scale. Several analyses have established that any study of hydro-climatic variables must be handled and executed on different micro-scales rather than regional scale because the trends and their effects may differ widely from station to station. In the current investigative study, the urban rainfall variation in the Koraiyar basin between the years 1975-2010 is examined to study the trend of precipitation. The non-parametric test is used in this study because the data outliers can be avoided easily. The Mann-Kendall (M-K) method is adopted to analyze the trend, and Sen’s slope is used to estimate the linear trend. To detect the change in trends, the CUSUM test is applied. The coefficient of variance (CV) is applied for variability analysis. From the analysis, it is noted that there is an unpredictable increasing and decreasing trend in the yearly and periodic rainfall patterns in the basin. [Keywords: CUSUM test, Spear Man-Rho test (SR test), Precipitation, Sen’s slope estimator test, Trend] Introduction Precipitation is an essential element responsible for the availability of freshwater resources. The amount of water available in a specific place depends upon quantum of precipitation and ground collection. Precipitation trend varies with time and place. Forecasting trends using seasonal rainfall time series data is much more complicated than forecasting ambient temperature trends. In recent hydrological studies, top priority has been given to rainfall trend analysis. The trend analysis of precipitation is generally conducted on a macro scale, and the gathered data has helped identify a decreasing trend in the mean yearly rainfall in countries such as India, Russia, Northeast, and North China and along the arid plains of Pakistan. An increasing trend in the mean yearly rainfall has been witnessed in the Chang Jiang valley in China, and Bangladesh in summer season 1 . In earlier studies, global trend analysis was always done with historical rainfall data, which is less useful for local-scale planning of water resources; some analyses showed no significant variation in trend 2,3 . The study of rainfall variability and its trends is a useful tool for policymakers for agricultural planning and hazard mapping 4,5 . The rainfall trends are generally observed to vary from one locality to another. The study of rainfall variability in India is carried on a macro scale. It shows that in the case of peak monsoon season, rainfall variability is high and mostly shows a negative trend, which results in increased frequency and magnitude of monsoon rainfall. The reported yearly rainfall in peninsular India shows no major variation in trends 6,8 . The rainfall data for a period of 40 years from 1971 to 2010 in the river basin of Orissa was used to determine the precipitation trend and to identify the pattern of increasing or decreasing cycles for narrowing attributes 9 . The study of rainfall trends is required for applications like flood frequency analysis, monitoring of drought, and for sustainable management of water resources 10-13 . The oscillating variation in rainfall trends has become a challenging issue for the planning of current and future developmental activities 14 . The exact trends and identification should be made for regional-scale analysis for a particular place. There are different statistical methods to detect a trend in hydrology; these are classified as parametric and non-parametric tests. The parametric test is a more powerful technique that requires data to be distributed both normally and independently which is infrequently seen for hydrological period series data. In the non-parametric method, all data should be independent. The outliers in this method are restored. The frequently used non-parametric tests are the M-K