RETURN PERIOD ANALYSIS OF RAINFALL EROSIVITY

Abstract:

The rainfall erosivity is one of the most important and fluctuating factors of soil erosion considering climate change. The Rainfall erosivity factor commonly known as R factor of the famous Revised Universal Soil Loss Equation (RUSLE), is well accepted for expressing the erosive power of rainfall. The R factor is the annual average of summation of EI30 parameter (a parameter expressing the effect of kinetic energy and intensity of rainfall events) of all the erosive rainfall events. Hence, the soil loss calculated by RUSLE is also an average value. However, consideration of the average value of soil loss in designing soil conservation structure may become unsafe sometime. In an area of high variability of annual rainfall, it may result in overestimation of soil loss in dry years and underestimation of soil loss in wet years. However, in studies of soil erosion, this problem is not often appreciated. So, in this study, an effort is made to suggest annual R factor values corresponding to different return periods. This study aims to help in designing soil conservation structure and various land management policies by providing the R factor corresponding to different return periods. To conduct the study 33 years daily rainfall data of Guwahati city, situated in the North-Eastern region of India, is collected. The annual R factors are calculated by a regional model developed by Das and Sarma, 2017 and using the daily rainfall data. Nine widely used theoretical probability distributions are fitted to the series of annual R factor. The goodness of fit of the distributions are tested and the return level values are calculated with the help of best fitted distribution. The Generalized Extreme Value (GEV) distribution was found to be the best-fitted distribution as per the Anderson Darling goodness of fit test. The return levels of Annual R Factor corresponding to 2, 5, 10, 25, 50 and 100 year return period are calculated from fitted GEV distribution with a 95% confidence interval. It was observed that the average annual R factor value falls between 2 and 5 years, and the maximum annual R factor value falls between 50 and 100 years return period value. Level of underestimation of erosivity can be easily understood from the above two observations. Thus, the inclusion of return period analysis can improve the safety of a soil conservation structure.