The Severe 1374 Rhine River Flood Event in Present Times

Floods are one of the main natural hazards causing large economic damage and human casualties worldwide. To reduce risks and to protect the hinterland from inundations, many rivers are confined between levees, which height and strength is based on design discharges associated with a specific return period. These design discharges are commonly determined with the use of a flood frequency analysis (FFA). In an FFA, annual maximum measured discharges are used to identify the parameters of a probability distribution. From this fitted distribution, discharges corresponding to any return period can be computed.

However, the data sets of measured discharges are typically in the order of 100 years. Extrapolation to design discharges with a return period of >1,000 years results in large uncertainty bounds. Extending the data set of measured discharges by reconstructing historic flood events may reduce the uncertainty bounds.

We illustrate this in the Rhine delta. In 1374 the largest discharge of the last 1,000 years occurred. Water marks near the city of Cologne (Germany) give insight in the maximum discharge of this historic flood. However, before being used in FFA, this discharge should be homogenized for the current topography since the river system has altered significantly the last 1,000 years. Therefore, a 1D-2D coupled model is developed in which the main channel and its floodplains are discretized by 1D profiles and the hinterland by a 2D numerical grid. With this model, a Monte Carlo analysis is performed in which the upstream discharge flood wave and dike breach characteristics, such as final breach width and breach duration, are varied within realistic ranges. The Monte Carlo analysis results in an expected value of the mean maximum 1374 discharge with its uncertainty under current geometrical conditions. We will use this knowledge in the FFA to reduce the uncertainty bounds of the flood frequency curve.

This study proposes a method to enable homogenization of historic flood events for current topography such that the reconstructed maximum discharge can be included in an FFA. The results give insights in the effect of historic flood events in present times which improves the flood-frequency relation. This information is of high importance since it contributes to flood risk analysis and also provide information in the most vulnerable areas in Germany and the Netherlands.