Comparison with Other Damage Functions. We want to compare our results with already published damage functions in literature. The city of London has been studied in detail by Xxxxxx et al. (2011) and a damage function for flood levels between 0 and 10 m has been provided (Xxxxxx 2014). Figure 9 shows a comparison between our macro-scale damage function and the one provided in Xxxxxx et al. (2011). It can be seen that the magnitudes of damages differ roughly by a factor of 2 to 4, which indicates a difference of estimated maximum damage values. One reason for that is certainly our underestimation entailed by the non-consideration of certain areas in our approach. However, the factor found appears to be too large to be explained by our methodology. Other reasons for a higher valuation of losses by Xxxxxx et al. (2011) remain unclear. Comparing the shapes of the two curves, a less steep increase of our curve for higher sea levels can be detected (this can be seen particularly in the linear-scale cross plot). A possible explanation could be the use of different small scale damage functions, particularly with respect to the inundation level where the maximum damage is reached. Another available macro-scale damage function has been elaborated by Xxxxxxxxxx et al. (2011) for the city of Copenhagen. Although this curve provides flood damages only for sea levels up to 4 m, we will use it as reference curve for our estimation. Figure 10 shows a comparison analogous to the London case beforehand. Here, we find very similar shapes which differ mainly by a certain factor (lower than in the London case). It can be concluded that the difference emerging from the other constituting factors, such as the orography and the relative damage functions, is comparably small. In general, the two comparisons show that, apart from a factor, our rather rough approach provides reasonable results comparable to existing more detailed curves. Since there are too few historical recorded storm surges and damages, it cannot be evaluated which of the functions performs better (see Part III).
