The role of flood wave superposition for the severity of large floods

Abstract. The severity of floods is shaped not only by event and catchment specific characteristics but also depends on river network configuration. At the confluence of relevant tributaries to the main river, flood event characteristics may change depending on magnitude and temporal matching of flood waves. This superposition of flood waves may potentially increase flood severity. However, this aspect is up to now not analysed for a large data set. To fill this gap, the role of flood wave superposition in determining flood severity is investigated. A novel methodological approach to analyse flood wave superposition is presented and applied to mean daily discharge data of 37 triple points from the four large river basins in Germany and Austria (Elbe, Danube, Rhine and Weser). A triple point consists of the three gauges at the tributary as well as upstream and downstream of the confluence to the main river. At the triple points, differences and similarities in flood characteristics are jointly analysed in terms of temporal matching and magnitudes of flood peaks. At many analysed confluences, the tributary peaks arrive consistently earlier than the main river peaks, but mostly high variability in the time lag is detected. No large differences in temporal matching are detected for floods of different magnitudes. In the majority of the cases, the largest floods at the downstream gauge occur not because of a perfect temporal matching of tributary and main river. In terms of spatial variability, the impact of flood wave superposition is site-specific. Characteristic patterns of flood wave superposition are detected for the flood peaks in the Danube, where peak discharge largely increases due to inflow from the alpine tributaries. Overall, we conclude that the superposition of flood waves is not the driving factor of flood peak severity in Germany, but a few confluences bear potential of strong flood magnifications in the case of temporal shift in flood waves.