Evaluation of the SWAT Model in Hydrologic Modeling of a Large Watershed in Nigeria

H. Xie, E. Nkonya, and B. Wielgosz (USA)


Watershed, Soil and Water Assessment Tool (SWAT), calibration, and uncertainty


The Soil and Water Assessment Tool (SWAT) is a comprehensive river basin model and has gained international recognition as a valuable tool for watershed analysis. An application of the SWAT model to a large river basin in Nigeria under data-scarce conditions is presented in this paper. The initial setup of the SWAT model in this study relied on the use of data from global databases with easy accessibility. In hydrologic simulation, the model parameters were first optimized with the genetic algorithm, and the uncertainty in the calibration was further analyzed using the generalized likelihood uncertainty estimation (GLUE) method. The major findings from this study are: (1) the calibrated model achieved a model fit that was reasonably good, with Nash–Sutcliffe Efficiency (NSE) coefficient=0.54; and (2) the estimation of SWAT model hydrologic parameters were subject to great uncertainty, but two key parameters that determine the rate of surface runoff and soil water retention capacity exhibited better identifiability. The developed SWAT model was applied to evaluate the water-related soil erosion status of the study river basin. The findings from the hydrologic calibration and uncertainty analysis helped make a robust assessment on the surface runoff and the resulting soil erosion level.

Important Links:

Go Back