Integration of extended information into the statistical analysis using the example of Flensburg: Assessment of the return period of the storm surge of October 20/21, 2024

On October 20-21, 2023, extraordinarily high water levels hit the Baltic Sea coast of Schleswig-Holstein. The question of the statistical return interval for this storm surge will be clarified in this article using the Flensburg gauge as an example.

The results of statistical analyses on the frequency or return intervals of extreme events such as storm surges or storm floods depend on the available database. The results change when additional sources of information such as analog gauge records, local historical records and spatial information from neighboring gauges are added to the database. Using the example of the Flensburg gauge, the possibilities associated with the inclusion of such water level information in the database are demonstrated. A central result is that the extension of the database results in a vertical shift of the distribution function and a reduction of the theoretical model uncertainties, especially for rare or extreme events, which in turn leads to changes in the estimation of recurrence intervals and flood risks. These results underline the need to include all available information as an integral part of the extreme value statistical analyses in order to achieve results that are as objective and robust as possible.

A recurrence interval of around 70 years is derived for the storm flood of 20/21/2023, which led to a water level of around 216 cm above mean water in Flensburg.

Particularly in view of climate change and the associated consequences for future extreme events, it is of the utmost importance that the design procedures and events for coastal protection facilities are based on high-quality water level data including historical information.