The goal of this work is to comparatively evaluate the potential of both event-based automatic calibration and Dynamic Identifiability Analysis (DYNIA) as proposed by Wagener et al. This is combined with an investigation on a potential relation of a priori knowledge on event characteristics with optimal model parameters.
A joint application of DYNIA and automatic parameter estimation leads to implications considering the informational content of both methods. Optimal model parameters, identified on an event basis, are tested for statistical relations with physical characteristics of the rainstorm events (e.g. intensity). In this paper, we present results of a modelling study in the Rietholzbach catchment (Switzerland). We employed the hydrological model WaSiM-ETH, using a combined DYNIA (Dynamic Identifiability Analysis) and automatic parameter estimation (PEST) approach to investigate best parameter sets as well as parameter variability along the time series of the hydrograph. The results of the study indicate that the “drainage” parameter identifiability for the long-term simulation is linked to the event-based calibrated parameter for the flood events. However, the parameter sets obtained with single-event calibration could not be fully linked to the chosen characteristic features derived from the precipitation forecast.