Strategy Development Methods for Flood Risk
Modelling systems calculate storm situations that improve predictions of water levels, wave heights and currents in the North Sea.
Risks such as pollution from accidents in connection with oil tankers, oil rigs and wind farms exist in the German coastal regions. Due to the rising sea level, flooding is also increasingly a problem. In order to develop new strategies to protect us from these trends, a deeper understanding of the geophysical processes in the oceans is of utmost importance, and comprehending the coastal regions is particularly vital. Wave-current interaction describes processes such as the steepening of waves through opposing currents and the energy exchange between these two quantities. The energy exchange can produce strong currents in regions such as the Wadden Sea Islands.
Part of this interaction is schematically represented in Figure 1. When surface waves approach the coast at a particular angle, they will be diffracted toward the coast so that their wave fronts align parallel with it. The waves simultaneously begin to break in the surf zone whereby the wave height and thus the resulting wave energy is reduced. Some of this energy can pass over to a current that runs parallel to the coast and can reach relatively high velocity, particularly during extreme events such as storms. Other currents caused by waves that are particularly dangerous for swimmers are rip currents. Such currents can arise when the water that has been transported onto the beach by the waves can only recede into the sea through a narrow gap between obstacles such as sandbanks, thereby creating high current speeds. Swimmers caught in such rip currents often cannot swim against the current to make their way back to land.
Unfortunately, not much high-quality data exists for the German coastal areas to quantify the wave-current interactions. Nevertheless, in order to more precisely research these processes, the Institute of Coastal Research at the Helmholtz-Zentrum Geesthacht (HZG) utilises various numerical modelling systems. These modelling systems consist mainly of a hydrodynamic current model coupled to a surface wave model. They are operated on supercomputers and exchange information such as water levels, current velocities and wave heights while the calculations are running. It has been proven useful to employ unstructured models that allow the choice of a variable resolution throughout the modelling area and a particularly high resolution in the coastal regions. The coastlines can thereby be appropriately represented and the geophysical processes can be calculated and respectively modelled.
Using the modelling systems, storm situations in particular are re-computed to show that through the coupling of currents and waves, a better prediction of water levels, wave heights and currents can be made. For the 2006 Storm Britta, additional wave-driven currents of up to 1 m/s were simulated (see Figure 2), whereby the computed current velocities off the East Frisian Islands were in parts doubled. The water levels also increased in the model results when taking into consideration the influence of the waves. This can lead to an improved reproduction of observation data, such as during Storm Xaver in 2013 (see Figure 3).
These results show the importance of using coupled numerical models during storm events for optimised predictions of coastal processes. Sediment calculations can also be improved through coupling as they are strongly influenced by the prevailing current situation. The HZG is working toward a further step in this development by coupling atmospheric models and has already obtained promising results.
Grashorn, S., Lettmann, K.A., Wolff, J.-O., Badewien, T.H., Stanev, E.V. (2015) East Frisian Wadden Sea hydrodynamics and wave effects in an unstructured-grid model. Ocean Dyn. 65(3): 419–434 DOI 10.1007/s10236-014-0807-5
Staneva, J., Wahle, K., Günther, H., Stanev, E. (2015) Coupling of wave and circulation models in coastal-ocean predicting systems: a case study for the German Bight. Ocean Sci. Discuss. 12: 3169–3197