Imagine you could use only three parameter sets of a hydrological model to represent the modelling uncertainty in the flood frequency analysis. Imagine how many more input scenarios could be then analysed at a lower computational cost. This study assists you in selecting such three parameter sets that could be assumed as representative for the full parameter ensemble.
Extreme-flood estimation based on simulation study often requires many model runs using different parameter sets of a hydrological model to account for modelling uncertainty. This could be prohibitive if a hydrological model is only a part of a complex modelling chain that includes also analysis of several meteorological or climate scenarios or hydraulic routing of simulated discharge time series at the chain end.
Such simulation frameworks are affected by high computational demands, particularly if floods with return periods > 1000 years are of interest or if modelling uncertainty due to different sources (meteorological input or hydrological model) is to be quantified.
In this study, three methods for reducing the computational requirements for the hydrological simulations for extreme-flood estimation have been proposed. These methods reduce the number of available parameter sets to a small number (three) that could be used within complex modelling chains. Thus, long streamflow time series or many input scenarios can be analysed at a much reduced computational cost.
Sikorska-Senoner, A.E., Schaefli, B., and Seibert, J. (2020) Downsizing parameter ensembles for simulation of rare floods, Natural Hazards and Earth System Sciences, 20, 3521–3549, https://doi.org/10.5194/nhess-20-3521-2020.
See related project: