PARAMETER UNCERTAINTY AND SIMULATION OF DESIGN FLOODS IN SWEDEN

This paper examines the effects of parameter uncertainty on the simulation of recorded floods and design floods, using the HBV hydrological model. Two Swedish catchments with hydropower development were studied. A Monte Carlo procedure was used to generate parameter sets of different levels of uncertainty. The results showed that the most sensitive parameters in the calibration process were the snowfall correction factor and the recession parameters. Furthermore, when the model was extrapolated to simulate design flood and water stage hydrographs, the single most sensitive parameter was the highest recession coefficient. In addition, it was found that parameter uncertainty was associated with combinations of parameters rather than the absolute values of each.