SCS-CN AND GAMMA DISTRIBUTION BASED RUNOFF GENERATION FOR AN INDUSTRIAL WATERSHED

Abstract

The present study critically reviews the synthetic unit hydrograph (SUH) methods available in hydrologic literature and evaluates popular methods (such as Snyder, SCS Curve Number and probability distribution functions based) using the data of an industrial watershed (Area=200 ha) (subdivided into 3 sub-watersheds) located in the district of Angul, Odisha. During recent decades, use of probability distribution functions (pdfs) in developing SUH has received much attention because of its similarity with unit hydrograph properties. In this study the analysis was carried out using Snyder SUH and SCS dimensionless UH using the catchment properties. Analysis was also done using the statistical distributions such as two parameter gamma distribution, Chi-square distribution, Weibull distribution, and three-parameter beta distribution using observed rainfall runoff data. The unit hydrographs derived were compared with those derived from observed rainfall-runoff data. The unit hydrographs due to the first two methods such as Snyder and SCS deviated far from the observed ones for all the three sub-watersheds, exhibiting their non-suitability to small industrial watersheds. The pdf-based distributions were chosen for further study using the data of observed rainfall and runoff, for the reason that it obviated the manual adjustments for area to be unity, and their behaviour was observed. The hydrograph by Gamma distribution approach shows symmetric relationship to the observed unit hydrograph and also shows efficiency index of 0.8, 0.8, and 0.7 for watershed 1, . watershed 2, and watershed 3 respectively, which are much closer to unity (perfect fit). The area under study consists of industrial area which is quite complex and different from the usual watersheds. So the response of the catchment was studied by observing the unit hydrographs by above methods. Then the adequacy of the existing drainage system of all the three sub-watersheds for four return period (25-yr, 20-yr, 15-yr, and 10-yr) rainfall values was evaluated and it was found to be inadequate. A possibility was further explored for changing the hydrologic condition of the whole watershed to the extent possible and it was found that in none of the cases the capacity- of drains was adequate, implying that the system required a complete replacement/enhancement of the existing one.