Tidal wetlands are increasingly valued for their role in coastal defense. Nevertheless, in situ observations of storm surge attenuation within wetlands are still scarce. We present water level measurements along a 4 km intertidal channel and on the surrounding marsh platform for regular spring to neap tides and two major storm surge tides, showing the effects of flood wave height and marsh geomorphology on the amount of flood wave attenuation. Undermarsh tides with peak water levels below marsh platform elevation are mostly amplified (up to 4 cm/km) within the channels. Overmarsh tides with peak water levels above the marsh platform are generally attenuated along the channels, with maximum attenuation rates of 5 cm/km for tides that inundate the marsh platform by 0.5–1.0 m. For lower or higher flood waves, including storm surges, attenuation rates decrease. Furthermore, the observations show that the maximum attenuation occurs along narrow channel transects where the width of the platform is larger, whereas attenuation rates are lower along wider channels with smaller adjacent marsh platforms. These observations are confirmed by an analytical approximation of tidal wave propagation through convergent channels. The analytical model indicates that differences in attenuation rates are induced by variations in the cross‐channel averaged friction between channel sections and between tides with varying peak water levels. Finally, the highest attenuation rates of up to 70 cm/km are observed over short distances on the vegetated marsh platform. We conclude that this study provides an empirical basis for the wider implementation of nature‐based flood defense strategies.

As ecosystem-based adaptation to global change is gaining ground, strategies to protect coastal and estuarine areas from increasing flood hazards are starting to consist of natural tidal wetland conservation and restoration in addition to conventional coastal defense structures. In this study, the capacity of tidal wetlands to locally attenuate peak water levels during storm tides is analyzed using a two-dimensional hydrodynamic model (TELEMAC2D) for a 3000 ha intertidal marsh (SW Netherlands). Model results indicate that peak water level reduction largely varies between individual flooding events and between different locations in the marsh. Model scenarios with variable dike positions show that attenuation rates can be minimized by blockage and set up of water levels against dikes or other structures confining the marsh size. This blockage only affects peak water level attenuation across wetlands if the duration of the flood wave is long compared to the marsh size. A minimum marsh width of 6 to 10 km is required to completely avoid blockage effects for the storm tidal cases assessed in this study. If blockage does not affect flood wave propagation, variations in attenuation rates between different locations in the marsh and between tides with varying high water levels can be explained with a single relationship based on the ratio between the water volume on the marsh platform and the total water volume on the platform and in the channels. Attenuation starts to occur when this ratio exceeds 0.2-0.4 and increases from there on up to a maximum of 29 cm/km for a ratio of about 0.85. Furthermore, model scenarios with varying marsh channel depth show that marsh scale attenuation rates increase by up to 4 cm/km if the channel elevation is raised by 0.7 m on average. Conversely, marsh scale attenuation rates decrease by up to 2 cm/km for scenarios in which the channels are lowered by 0.9 m on average. The marsh platform elevation has little effect on the maximum attenuation, but it determines which tides are attenuated. In particular, only overmarsh tides that inundate the platform are attenuated, while undermarsh tides that only flood the marsh channels are not attenuated or even amplified. These findings may assist coastal communities and managers in the optimization of the coastal defense function of tidal wetlands in combination with dikes.

No abstract

No abstract

No abstract