Simulating major storm surge events in a complex coastal region

Elin Andrée,Jian Su,Morten Andreas Dahl Larsen,Kristine Skovgaard Madsen,Martin Drews

doi:10.1016/j.ocemod.2021.101802

Elin Andrée, Jian Su + Show 3 more

Open Access

https://doi.org/10.1016/j.ocemod.2021.101802

Copy DOI

Abstract

Climate change and increased coastal urbanization are causing low-lying coastlines to become increasingly susceptible to the threat of extreme water levels and coastal flooding. Robust decision-making on adaptation in the coastal zone, based on reliable ocean-modelling tools, is therefore crucially contingent on accurate assessments of current and future storm surge hazards. This accuracy relies considerably on the quality of the wind forcing used in the ocean models. In this paper, we use a high-resolution, regional 3D ocean model (HBM) covering the North Sea and Baltic Sea to simulate extreme water levels during three extreme storm surge events with different dynamics and patterns, in order to assess their impacts along Denmark’s coastlines, which are of varying levels of complexity. We demonstrate that the model is able to reproduce the observed extreme high-water levels accurately, indicating that the system is well suited for producing simulations of present and future projections of extreme storm surges with high resulting impacts and damage potentials. Additionally, we quantify the level at which acknowledged deficiencies in the otherwise most suitable atmospheric forcing data set influence the results of the storm surge simulations. We found that reducing the temporal resolution of the forcing data – that is, replacing two out of every six time stamps with linearly interpolated values – is preferable to using the original forcing data set when recurring noise is present in these time stamps. As a result, for given storm surge events, and depending on the stage reached in the storm’s evolution, mean absolute errors can be reduced by 4.5cm. This emphasizes the importance of considering such model fluctuations when coupling high-resolution atmosphere and ocean models.

Highlights

Storm surges and extreme water levels can cause devastating damage to low-lying coastlines
By comparing the correlation and root mean square error (RMSE) of our simulations with the ones reported by (Meier et al, 2004; Fernández-Montblanc et al, 2020), we find that the model system reproduces sea-level variability with a greater accuracy than what has previously been achieved for coarser-resolution, state-of-theart hindcast modelling in the specific region
We evaluate the simulated water levels, as well as the atmospheric regional reanalysis that forces the ocean model, against observations

Summary

Introduction

Storm surges and extreme water levels can cause devastating damage to low-lying coastlines. According to the Intergovernmental Panel on Climate Change (IPCC), the frequency of high water level events is expected to increase – for example, due to climate change-induced rises in sea level – as are areas and people exposed to coastal hazards worldwide (Oppenheimer et al, 2019; Vousdoukas et al, 2016). Without investments in coastal adaptation, the expected annual damage from extreme water levels is likely to increase by several orders of magnitude globally (Vousdoukas et al, 2020, 2018; Hallegatte et al, 2013) with both the current and future climate. According to Hinkel et al (2014), the globally projected annual damage costs may amount to as much as 10 % of global GDP by the end of the 21st century

Methods

Results

Discussion

Conclusion