Interdependence of Extreme Floods Peaks

Hydrological extremes such as strong precipitation and floods are phenomena which have distinct spatio-temporal features. In this subproject of the SPATE research unit, tools were developed that lead to a better understanding of the behaviour and the dependencies of precipitation and discharge in space and time for events that cause extreme floods. Large-scale atmospheric circulation patterns (CPs) which are associated with high rainfall amounts were identified for catchments in Germany and Austria. However, the CPs which consider the atmospheric conditions only partly explain the occurrences of large floods since the state of the catchment is not considered. Furthermore, two metrics which explain the simultaneous occurrences of extreme events were developed. One is based on the binary behaviour of combined flood event exceedance probabilities and the other focusses on the temporal variability of discharge time series expressed by the Fourier transform of precipitation and discharge time series. The results show that the simultaneous occurrence of extreme events can be clearly seen in the high frequencies of the Fourier cross-correlation spectra. This was exploited to formulate new precipitation and discharge time series generators that were better suited to simulate simultaneous occurrences of extreme events.