Assessing water resources availability in Indrawati river basin in Nepal : application of SWAT model for hydrologic simulation

The SWAT2005 model was applied for the assessment of water resources availability in the Indrawati River Basin of catchment area 1228 km2. The main objective of this study was to test the performance of the SWAT model for streamflow simulation as well as to apply the same model for the hydrological impact assessment of various water transfer/use scenarios. The SWAT model was set up for the period of 1971 to 2008 in daily time resolution. However, considering the quality of available data, the model simulation period was confined for 1981 to 2008. The SRTM 90 m resolution DEM, FAO based global landuse map of about 400 m resolution, soil map from SOTER database for Nepal and the weather data collected from about 13 weather stations for a period of about 30-40 years were used to set up the model. The model was calibrated and validated at two different gauging stations: one at basin outlet and another at the outlet of upstream Melamchi tributary using SWAT-CUP/SUFI-2 and manual calibration. The NSE obtained were from 0.69 to 0.86 for both calibration and validation periods. The calibrated model was applied to simulate the streamflow generation at 12 numbers of subbasin outlets across the Indrawati basin. Similarly environmental flow assessment was also carried out for all subbasin outlets using Tennant's Montana hydrological method. The calibrated model was further applied to evaluate the hydrological impact of three major water transfer/use scenarios at subbasin and basin levels which are: transferring water from upper three tributaries to the city of Kathmandu (S1), upgrading existing rainfed agriculture land into irrigated land (S2) and combination of the two (S3). The results of these scenarios were compared to the present baseline scenario without any development intervention. Analysis for scenario S1, on a daily time resolution, depicted the annual deficit in available streamflow to divert the water at full diversion capacity from the upper three subcatchments for about 3-8% time in a year. This annual streamflow deficit period would be further escalated (6-12% days) when the downstream minimum environmental flow requirement need to be maintained (10% of the monthly average flow). Scenarios S2 and S3 tested in this study reflected that the water transfer and consumptive use activities appeared within the limit of the available annual water resources development potential in basin level. However, under each scenario, monthly flows from January to May would be severely reduced. The situation is most alarming for the upper three subbasins for water transfer scenario and one of the subbasins near the basin outlet for upgradation of rainfed to irrigated agriculture scenario