Managing dyke retreat: Importance of century‐scale channel network evolution on storm surge modification over salt marshes under rising sea levels

Abstract

AbstractA hybrid coastal defense consisting of a salt‐marsh system at the seaward side and dykes on the landward side has been globally implemented to improve coastal resilience. In this approach, marshes are valuable ecosystems that dampen storm surges and have an intrinsic ability to keep up with sea‐level rise (SLR). Dykes instead, prevent the surge from penetrating inland. Therefore, dyke retreat is now considered a valuable option that allows the creation of new marshland in front of the relocated dyke, thus maximizing surge damping. The studies investigating the effect of dyke retreat on surge modification do not incorporate the morphodynamic expansion of tidal networks in the de‐reclaimed land, which is modeled as an unchannelized marsh. Here, we use the morphodynamic model Delft3D and a marsh evolution module to study the formation of tidal networks in a de‐reclaimed land after dyke retreat and evaluate their effect on surge modification (damping or amplification), and thus, the maximum water level reached at the dyke during a storm surge event. For this purpose, we consider different combinations of hydrodynamic (storm surge peak and duration, SLR), and morphodynamic (sediment input, dyke retreat distance, and de‐reclaimed land slope) parameters. Our results suggest that an increase in tidal prism, which depends on retreat distance, de‐reclaimed land slope, and SLR, drives the century‐scale morphodynamic evolution of the marsh. Results also show that surge damping is overestimated if the morphodynamic evolution of the tidal network after dyke retreat is neglected since the new creeks favor the landward propagation of the surge. Finally, a genetic algorithm is used to determine a relationship between surge modification, marsh morphology, and the morphodynamic and hydrodynamic parameters of the simulations. The relationship indicates that relative surge damping increases for shallower tidal networks, smaller surge peaks, higher marsh platforms, and smaller percentages of channelized area.