Assessment of storm surge risk in the German Bight: Modeling past events and projecting future changes

Abstract

Storm surges represent a persistent and growing challenge for coastal populations. These extreme events can cause severe flooding, substantial damage to property and critical infrastructure, and endanger human lives. The German Bight, located in the southeastern North Sea, is a region frequently affected by storm surges. With ongoing climate change, the frequency and intensity of extreme weather events are expected to increase, while rising sea levels further amplify the potential impact of storm surges. In this context, reliable information of future storm surge risk is essential for effective coastal planning and adaptation strategies.

From a scientific perspective, however, significant challenges remain, particularly in translating wind conditions into storm surge heights. This difficulty arises because the response of coastal waters to wind forcing is highly variable and influenced by multiple interacting physical factors. Current approaches rely on either computationally intensive regional or global climate models or highly complex statistical models, neither of which allow for comprehensive multi-model assessments based on climate projections. Consequently, projections of storm surge heights for the German Bight based on multi-model assessments do not yet exist. However, such assessments are essential for providing reliable and robust information on future storm surge risk. Moreover, the few existing projections of storm surge heights, which are based on a single climate model or scenario, do not incorporate sea-level rise, even though it plays a critical role in elevating the baseline water level. Rising sea levels make even moderate storm events increasingly dangerous. Despite its importance, the integration of sea-level rise into storm surge modeling remains an open issue.

As existing modeling approaches do not allow for projecting storm surge heights based on a multi-model ensemble of climate simulations, the first part of this dissertation focuses on developing a statistical storm surge model suitable for this purpose. The statistical model is based on a multiple linear regression approach that uses wind as the only predictor. By systematically tuning the model settings, the final model comprises only five terms – squared wind values at four lead times and an intercept as the fifth term. Despite its simplicity, it has a predictive skill comparable to more complex approaches. Moreover, the storm surge model delivers robust predictions for moderate and extreme storm surges. Its minimal input requirements and computational efficiency make it well suited for using large ensembles of climate model data, providing a valuable tool for projecting future storm surge heights.

Next, I explore the impacts of climate change on storm surge frequency and height in the German Bight. For this purpose, I apply the statistical model to a multi-model ensemble of climate simulations and present projections of storm surge heights for the German Bight across multiple emission scenarios. Based on changes in wind patterns alone I find an increase in the frequency of potential winter storm surge events by around 10% by 2100. Supported by recent studies, this increase is attributed to a higher frequency of westerly and northwesterly winds, linked to shifts in storm tracks under future climate conditions. When additionally accounting for sea-level rise, moderate and severe present-day storm surge thresholds in the German Bight are projected to be exceeded three to five times more often by the end of the century than historically.

Overall, in this thesis, I introduce a new statistical storm surge model for the German Bight. Its simplicity and efficiency make it a novel tool for investigating storm surge heights across a range of timescales. Moreover, the underlying modeling approach is transferable to coastal regions worldwide where wind is the dominant driver for storm surge development. I demonstrate the statistical model’s capabilities by applying it to a multi-model ensemble of climate projections to investigate future storm surge heights in the German Bight. The resulting projections across multiple emission scenarios provide robust insights into how storm surge risk in the German Bight may evolve under a changing climate.