HYDROLOGICAL MODELLING OF WEST RAPTI RIVER BASIN OF NEPAL USING SWAT

Abstract

West Rapti River basin is a middle-class river basin (of Nepal) originating from northern mid hill of Western Nepal. It drains south to Karnali River (Ghaghara in India) which is the major left bank tributary of Ganges River. It is a transboundary river between Nepal and India and also a river of River linking project of Indian government. In this study, Hydrological modelling of the West Rapti river basin (Area 5281 km2) using Semi distributed SWTA model for assessment of water availability and Sediment yield has been envisaged. The key objective of this study is to develop a suitable model to represent hydrology of West Rapti river basin for estimation of water balance components and sediment yield and to study the effect of different best management practices on sediment yield. SWAT model was simulation was carried out for discharge and sediment on monthly basis for 14 years (2000 to 2013). The SWAT model was calibrated and validated for the years 2003-2006 and 2007-2009 respectively, considering observed stream flow and rating curve generated sediment data using SWAT CUP model and sequential uncertainty fitting (SUFI2) technique. Calibration of the SWAT model was carried out using: (i) p-factor which is the percentage of data lying within the 95% prediction uncertainty, and (ii) r-factor, which is the ratio of the average thickness of the 95PPU band and the standard deviation of the observed value of discharge or sediment yield. Furthermore, Nash-Sutcliffe Efficiency (NSE), Coefficient of determination (R2), Percentage bias (PBIAS), and ratio of root mean square error to the standard deviation of measured data (RSR) were also used to evaluate the performance of model. For monthly flow, the value of p-factor and the r-factor for calibration was found to be 0.83 and 0.40 and the value of p-factor and the r-factor for validation was found to be 0.67 and 0.42 respectively. Likewise, for monthly sediment yield, the value of p-factor and the r-factor during calibration was found to be 0.96 and 0.64 and the value of p-factor and the r-factor during validation was found to be 0.86 and 1.12 respectively. The results obtained from the model calibration and validation showed reliable estimate of monthly stream flow (R2 = 0.96, NSE =0.95, PBIAS=4.7 and RSR=0.22) and sediment yield (R2 = 0.71, NSE =0.68, PBIAS=15.10 and RSR=0.57) for calibration period. However, for the validation period, model performance was low as compared to the calibration period with parameter for flow (R2 =0.78, NSE =0.78, PBIAS=5.3 and RSR=0.47) and sediment yield (R2 = 0.69, NSE =0.69, PBIAS=-9.70, and RSR=0.56). Based on the statistical results v obtained from the SWAT simulation, it is seen that the performance of the SWAT model in the West Rapti river basin is very good. Furthermore, the water balance study of the basin showed that 48.60 % of the average annual rainfall of the basin contribute to evapotranspiration. The annual volume of water available at the basin outlet is 4.5 BCM. The average annual sediment yield of the basin is 16.67 t/ha/year and lies under high erosion class. Further, the calibrated SWAT model was used for assessment of Best Management Practices (BMPs) under different scenarios for reducing the sediment yield and recommendation of the most effective BMP for its implementation. This study would be useful for assessment of possibility of storage type project in the basin in terms of water availability and sediment yield. This study will also help in integrated water resources management and sustainable development of the West Rapti river basin, Nepal