Hindcast of the flotsam levels of the 1 November 2006 storm in the Eastern Wadden Sea

This report describes the hindcast of the 1 November 2006 storm in the Eastern Wadden Sea. The storm was selected because of the relatively high water levels reaching record high values at the measuring station Delfzijl. The high water level in combination with high waves caused the wave run-up to reach the crest of the Emmapolderdijk. At the time of the storm no wave buoys were deployed in the Dutch Eastern Wadden Sea. In this study flotsam marks are used instead of wave buoys to assess the performance of SWAN. Flotsam levels are left behind on the outer slope of a dike after a storm surge when material from the foreshore has been transported up the dike surface by wave run-up. Five stationary SWAN simulations are performed at various moments during the 1 November 2006 storm. The SWAN runs are performed with the biphase breaker model of Westhuysen (2009) and a low bottom friction setting. The hindcast moment 1 November 2006 06:00 CET led to the highest wave conditions at the toe of the considered dike sections. These conditions correspond to the highest wind speed measured at Huibertgat and the highest water level measured at station Eemshaven. The observed flotsam levels show much more alongshore variation than the computations do. A comparison of the calculated and measured flotsam level showed that the computed flotsam level is overestimated by a much as 0.6 to 1.6 m at the four westernmost dike sections. Towards the Emmapolderdijk the difference between the calculated and measured flotsam level decreased. For dike section x = 234 km the calculated flotsam level showed a perfect agreement with the measured flotsam. At the Emmapolderdijk (x = 241 km) the measured flotsam level is underestimated by 1.4 m. The large difference between measurements and calculations is most likely caused by the translation of spectral wave results to wave run-up/flotsam levels, which is expected to be caused by the used wave height distribution in the wave run-up and flotsam formulations. In addition, a smaller part is caused by the combined errors in the water level and wave computations. It is therefore not possible to assess the performance of SWAN using flotsam measurements due to the combined uncertainty of the applied set of models/formulae. The flotsam levels are thus not a good indicator of the wave transformation in SWAN. We conclude that the overall performance of the set of models/formulae needs attention, specially the assumptions regarding the wave height distribution. The alongshore variation in measured flotsam level cannot be explained with SWAN because results showed almost no alongshore variation in wave conditions at the toe of the dike. Apparently, there are local processes that are not modelled with SWAN which cause the wave run-up/flotsam levels to vary spatially along the dike. It is expected that when accurate input data is available (wind, water level input and wave height distribution), flotsam levels can be valuable for assessing and improving the accuracy of the entire set of models/formulae used for deriving Hydraulic Boundary Conditions.