Performance assessment of SWAT and HEC-HMS model for runoff simulation of Toba watershed, Ethiopia

SWAT and HEC-HMS models are widely used hydrological models for runoff simulation. The performance of these models varies based on spatial and hydro-metrological characteristics of the watershed; These models are not applied on Toba watershed for daily runoff simulation. Therefore, the ultimate objective of this research is to check the performance of SWAT and HEC-HMS models in daily runoff simulation of Toba watershed which is located in Abbay River Basin, Ethiopia using the same input data set and methods. Land Use/Land cover (LU/LC), Soil, Digital Elevation model (DEM) and daily hydro-metrological data were the input data for both models. Therefore, the daily observed stream flow-data were used for both model calibration and validation. The statistical metrics used to measure the performance of both models were, R-2, NSE, PBIAS, and RMSE. The respective values for HEC-HMS were 0.85, 0.805, 33.7%, 22.7 m(3)/s during calibration, and 0.87, 0.84, 2.71%, 20.7 m(3)/s during validation. For SWAT model, the respective values were 0.69, 0.61, -26.1%, 31.85 m(3)/s, and 0.62, 0.54, -18.3, 35.24 m(3)/s during calibration and validation period. These values are at an acceptable level of accuracy. Therefore, both SWAT and HEC-HMS models well performed in the runoff simulation of Toba watershed. The overestimation of peak discharge and average runoff for dominantly cultivated land gives insight about the LU/LC characteristics of the watershed. The study can be used as a baseline scenario for future assessment of LU/LC and runoff potential.