Probabilistic (stylized) hydrological models to generalise flood changes and their attribution

The upper tail behavior of the probability distributions of flood peaks depends on a large variety of factors. Besides the derived flood frequency analysis for specific catchments, more general probabilistic models will be developed to describe these relationships. For this purpose, SP3 will extend the developed deterministic model chain by using a stylized model chain to obtain insights on the impact of the single components on the distribution tail. SP6 will develop a probabilistic model that explicitly focuses on the mechanisms of changes, such as increasing air temperatures leading to earlier spring snow melt and enhanced evaporation. Both models will consider flood types and characteristics of flood events and will be compared concerning their ability to model the probabilities of flood mechanisms and the linkages to the probability of flood events. The stylized models of change will draw on the experiences with the detection of flood rich periods in the first phase as well as on the findings of flood change attribution in the second phase. Specifically, the regionalisation results (i.e. the spatial patterns of flood change sensitives to their drivers) will be important for generalising the findings on flood change from individual gauged catchments to ungauged catchments and more general causal relationships that may also be applicable to catchments outside the study region. The findings of the uncertainty theme (U) from SP5 and 7 will also be very relevant for quantifying the uncertainties of generalised flood change attribution.

The head of the task force will be G. Blöschl (SP6), members are B. Merz (SP3) and A. Viglione (Mercator-fellow)