Abstract. Currently, a shift from classical flood protection as engineering task towards integrated flood risk management concepts can be observed. In this context, a more consequent consideration of extreme events which exceed the design event of flood protection structures and failure scenarios such as dike breaches have to be investigated. Therefore, this study aims to enhance existing methods for hazard and risk assessment for extreme events and is divided into three parts. In the first part, a regionalization approach for flood peak discharges was further developed and substantiated, especially regarding recurrence intervals of 200 to 10 000 years and a large number of small ungauged catchments. Model comparisons show that more confidence in such flood estimates for ungauged areas and very long recurrence intervals may be given as implied by statistical analysis alone. The hydraulic simulation in the second part is oriented towards hazard mapping and risk analyses covering the whole spectrum of relevant flood events. As the hydrodynamic simulation is directly coupled with a GIS, the results can be easily processed as local inundation depths for spatial risk analyses. For this, a new GIS-based software tool was developed, being presented in the third part, which enables estimations of the direct flood damage to single buildings or areas based on different established stage-damage functions. Furthermore, a new multifactorial approach for damage estimation is presented, aiming at the improvement of damage estimation on local scale by considering factors like building quality, contamination and precautionary measures. The methods and results from this study form the base for comprehensive risk analyses and flood management strategies.

No abstract

No abstract

Artificial neural networks (ANNs) provide a quick and flexible means of developing flood flow simulation models. An important criterion for the wider applicability of the ANNs is the ability to generalise the events outside the range of training data sets. With respect to flood flow simulation, the ability to extrapolate beyond the range of calibrated data sets is of crucial importance. This study explores methods for improving generalisation of the ANNs using three different flood events data sets from the Neckar River in Germany. An ANN-based model is formulated to simulate flows at certain locations in the river reach, based on the flows at upstream locations. Network training data sets consist of time series of flows from observation stations. Simulated flows from a one-dimensional hydrodynamic numerical model are integrated for network training and validation, at a river section where no measurements are available. Network structures with different activation functions are considered for improving generalisation. The training algorithm involved backpropagation with the Levenberg-Marquardt approximation. The ability of the trained networks to extrapolate is assessed using flow data beyond the range of the training data sets. The results of this study indicate that the ANN in a suitable configuration can extend forecasting capability to a certain extent beyond the range of calibrated data sets.

No abstract