Derived flood frequency analyses: How to combine weather generators with hydrological models to simulate flood series

Derived flood frequency analyses (DFFA) are used to interlink the meteorological forcing with flood series by hydrological models. Since the aim of the research group is the development of a general framework to analyse and predict characteristics of extreme floods, methodological developments of components for the DFFA are planned outcomes in many subprojects. The aim of this task force is to combine these results of different subprojects and compare the developed DFFA models concerning their spatial applicability and transferability. A key tool for DFFA is the use of a stochastic weather generator. Such generators are developed by SP3 for large basins with a daily resolution and by SP7 together with SP5 for medium sized catchments with an hourly resolution. The latter model will be further improved by spectral simulation and the extension to regionalization and uncertainty considerations as well as by an alternating renewal rainfall model using radar data. Based on the results of regional event analyses, SP1 will estimated flood frequencies by a stochastic-deterministic event generator without explicitly referring to weather variables. Overall, three approaches to DFFA will be developed and compared:

- the model chain based approach of SP3 that will simulate daily series of meteorological (precipitation, temperature), hydrological (soil moisture, snowmelt, river discharge) and hydrodynamic (inundation) variables for large basins;
- the multivariate DFFA approach developed by SP7 together with SP5 for medium sized basins that operates at an hourly resolution; and
a stochastic-deterministic event generator developed by SP1 that differs between flood types and considers extreme events with a deterministic threshold approach,

The intercomparison of the three approaches will shed light on the relative strengths of the methods and the relevance of the individual process descriptions for estimated extreme flood probabilities. The intercomparison will also provide insights into the triggers (precipitation, temperature, soil moisture etc.) and the distributions of the flood events (peak, volume, duration etc.).

The head of the task force will be U. Haberlandt (SP7), members are A. Bardossy (SP5), B. Merz (SP3), A. Schumann (SP1) and A. Viglione (Mercator-fellow)