Estimation of sediment delivery ratio at sub-watershed level using ArcSWAT Model

An attempt was made to estimate average annual sediment yield and sediment delivery ratio at sub-watershed level using ArcSWAT Model. Digital Elevation Model of 20 m resolution was prepared by using Toposheet F43N10 using Topo to Raster interpolation method in ArcGIS interface. The land use/land cover map was prepared from Landsat 8 image by supervised classification technique using ERDAS image processing software. The precipitation statistical parameters and average dew point temperature were derived respectively using pcpSTAT and dewfebruary.exe program while temperature, solar radiation and wind velocity parameters were derived using Pivot table properties of Microsoft eExcel. The model was set and run for the desired period of 30 years (1986–2015) and calibrated for USLE_P values using manual calibration helper. The gross sediment loading and sediment yield were simulated as 34.61 and 22.78 t ha⁻¹ yr⁻¹, respectively The sediment samples for 12 rainfall event during 3 years (2013–15) were collected from the study area outlet using indigenously designed sediment sampler. In order to validate the calibrated SWAT model, four statistical parameters were adopted to judge the model performance for the study area. The SWAT model was validated as values of co-efficient of determination (R²), Nash-Sutcliffe efficiency (NSE), RMSE-observations standard deviation ratio (RSR) and per cent bias (PBIAS) were computed as 0.82, 0.53, 0.69 and 2.76, respectively and these values were within acceptable limits. The sediment delivery ratio (0.65) indicates that 65 per cent of eroded soil goes out of the watershed and reaches Karjan reservoir thus eroding the soil of watershed, depleting fertility of top soil layer of watershed and consequently reducing the capacity of water storage of Karjan reservoir each year.