| 03 Nov 2025
Pan-European assessment of coastal flood hazards
Abstract. Coastal flooding is among the most damaging natural hazards in Europe, yet large-scale assessments have typically relied on simplified static "bathtub" models and coarse elevation data. Here, we present a novel pan-European methodology that applies a dynamic flood model (RFSM-EDA) at 25 m resolution, forced by location-specific total water level (TWL) hydrographs. These hydrographs integrate mean sea level, tides, storm surge, and wave setup with spatially varying foreshore slopes, allowing storm type, duration, and shape to be explicitly represented. More than 51,000 coastal target points were used to reconstruct events, and the methodology was validated against 12 historical floods across diverse coastlines. Sensitivity analyses quantified uncertainty from model selection, DEM resolution, hydrograph shape, and storm type. Results show that static flood models systematically overestimate inundation, with errors exceeding 25 % in low-lying coastal floodplains such as Belgium and the United Kingdom. At the continental scale, storm type variability explains 41 % of flood map uncertainty, while hydrograph shape has a smaller but measurable effect. Including coastal protection standards reduces the estimated exposed floodplain by more than half, underscoring the critical role of defenses. By bridging the gap between global static assessments and local dynamic models, this study establishes a methodological benchmark for continental-scale flood hazard mapping. The framework not only advances scientific understanding of large-scale coastal flooding but also provides actionable evidence to support the EU Floods Directive, adaptation planning, and climate risk management in the finance and insurance sectors.
The manuscript “Pan-European assessment of coastal flood hazard” introduces a new, high-resolution analysis of coastal flood hazard. The strength of the manuscript is an extensive sensitivity analysis and validation, which definitely increases the understanding of the limitations of coastal flood mapping. The work is comprehensive and solid, so my comments are only minor, in order of appearance:
Abstract: “The framework not only advances scientific understanding of large-scale coastal flooding but also provides actionable evidence to support the EU Floods Directive, adaptation planning, and climate risk management in the finance and insurance sectors.” I suggest to remove this, as the paper shows still serious limitations of the approach (out of 12 cases, 5 miss most flooding and in 5 most of the flood zone is overestimation). The data is not available publicly, so it can’t be easily reused. The actionability is reduced substantially by lack of future projections. The authors should mention the validation results, which are not mentioned here at all.
L72, L484: Unfortunately, an error crept into Groenemeijer et al. (2016) report, as in reality a resampled version of the original DEM in 100 meter resolution was used to calculate the coastal inundation. This was not made explicit as it should have been (I was the author of the data and that part of the report).
L108: “each located at a relative depth of 0.1.” – what does this mean?
L135: Corine Land Cover does not cover Russia either, so what data was there?
L153-158: this description is very vague, e.g. what historical storms were used, what was the calibration here, why only two events per year, what was the timespan of the historical data used or how the mean storm shape was derived.
L179: where did the Manning value come from and how were they linked in the specific land cover data used by the authors?
Table 2: some better naming scheme of the case studies needs to be applied, as it makes reading the text and analysing the graphs rather difficult.
L214: one additional important source of underestimation is the issue that higher resolution of DEM enables capturing some coastal defences, but still does not allow for defence failure mechanisms other than overtopping. In past coastal floods, dikes have failed without water levels reaching their crests, which is very difficult to represent in flood models (https://www.hkv.nl/wp-content/uploads/2020/07/Applications_of_VNK2_a_fully_probabilistic_risk_analyses_BM.pdf).
L437: datasets such as COASTPRO-EU rely on nominal or official protections levels, which are often much lower in practice due to e.g. inadequate maintenance or change in extreme water level probability (https://doi.org/10.1007/s11069-024-07039-5 ). This causes very large sensitivity of flood maps to protection level assumptions overshadowing all other (https://doi.org/10.1007/s11069-024-07039-5 ).
Table 6: it should be highlighted that the MFA results are not fully comparable across studies. The data in Paprotny et al. (2018) and Groenemeijer et al. (2016) are pretty much the same, but the figure from the former refers to a smaller number of countries than the latter.
L534-536: same remark as regarding the abstract.