Evaluation of uncertainties using swat versus hbv hydrological models in climate change studies the case study of Kilombero river sub-basin

Evaluation of uncertainties using SWAT versus HBV hydrological models was conducted in the Kilombero River Sub-basin as the case study. Recent studies have paid much attention to uncertainty issues in hydrological modelling due to their great effects on the prediction and decision making. The main objective of this study was therefore to undertake the comparative assessment of uncertainties between a complex physical model (SWAT) and a conceptual hydrological model (HBV) in climate change studies. The sensitivity analysis of modeling inputs like precipitation, temperature, and parameters was examined. The results showed that climatic inputs especially rainfall is the most sensitive model input. The uncertainty analysis showed that, SWAT has higher parameter uncertainty (r-factor= 1.78, p-factor= 0.85) than HBV (r-factor=0.84 and p-factor=0.7). In addressing the future climate change prediction by using the two models, precipitation and temperature future data from (2046- 2064) for Kilombero River Sub-basin were used. Using the future climate data (precipitation and temperature) in the calibrated HBV and SWAT models, based on the resulting FDCs analysis, both models predict the increase of runoff in future with SWAT giving a bit higher values compared to HBV. Using the 95% CI of the daily average observed flow, the results showed that both flows generated by SWAT and HBV were within the 95% confidence interval of observed flow.