Singh, J.K.Pravendra Kumar2016-09-162016-09-162007-12http://krishikosh.egranth.ac.in/handle/1/76994Present study has been carried out for estimation of storm runoff for the Karkara watershed with an area of 17.51 km2 of Tilaiya dam catchment in Upper Damodar Valley using the concept of unit-step and transfer functions. The conceptual models (discrete linear input-output models) having two parameters namely; Lag and Route model (LAR), Variable Storage Coefficient Model with cascade of two linear reservoirs in series (VSC-I) and Variable Storage Coefficient Model with cascade of three linear reservoirs in series (VSC-II) and Muskingum model for single reach (MSR) considering the system as lumped, linear and time-invariant have been developed for the study watershed. Twelve single peaked and isolated storm events have been analyzed in this study, out of which nine storm events were selected for calibration and remaining three events for the prediction purposes. The parameters of LAR model were estimated by the method of cumulants. The parameters of VSC-I, VSC-II and MSR models were estimated by methods as suggested by Sabol (1988) and Nash (1957). The autocorrelation functions of the residuals (observed – predicted) lie within the lines of confidence, that is, 2 SE lines confirming adequacy of the models for the study watershed. The visual comparison between observed and predicted direct runoff hydrographs showed that the MSR-II model (Channel routing with X = 0.10) resulted in the closest agreement for the study watershed. Based on quantitative performance evaluation indices also, the Muskingum single reach model, MSR-II was found to be the best performing model for the estimation of direct runoff hydrograph in the study watershed. The lowest average value of percentage absolute deviation in peak runoff rate of 4.629 further confirms the applicability of MSR-II model for the study watershed. However, the higher values of coefficient of efficiency of 0.90 and above and coefficient of determination and lower values of integral square error, relative squared error, percentage absolute deviation in peak runoff rates and direct runoff volumes indicate higher degree of goodness of fit of all the models for the study watershed. Based on the performance, it is concluded that the developed models can also be used with acceptable degree of accuracy for ungauged watersheds situated in areas having hydro-meteorological and geographical conditions similar to Karkara watershed.enhydrological factors, watershed management, watersheds, mathematical models, water transfer, hydrology, damsThesis