Comparison of SWAT and HEC-HMS model performance in simulating catchment runoff

Abstract

Globally, surface water sources are important sources of drinking water and knowledge on their availability and sustainability is important for their protection. Such studies in data-poor regions are very limited. This paper compares the performance of SWAT and HEC-HMS in the event and continuous modeling of rainfall–runoff in two tropical catchments, a low laying and a mountainous one. Model calibration and validation were done using observed streamflow data at Busiu station for Manafwa and at Sezibwa falls for Sezibwa for the period 2000–2013. The results were compared based on objective functions and also the t-test was used to test the statistical significance of the difference in performance. The results show that HEC-HMS performed better than SWAT in Manafwa catchment (p = .003 and p = .000 during calibration and validation, respectively) while in Sezibwa the difference in model performance was not statistically significant (p = .63) although, during calibration, HEC-HMS performed better (p = .01).