This article presents the current state-of-the-art understanding of underwater dilative slope failure (breaching). Experimental investigations are reviewed, providing critical insights into the underlying physics of breaching and pointing out knowledge gaps, which underscore the need for further research. Besides, field observations at several locations across the globe are outlined, highlighting the hazard of breaching and the need for effective coastal management strategies to mitigate the associated risks. Furthermore, existing methods for analyzing and predicting the slope failure evolution are discussed and reflected upon, including analytical approaches and numerical models, ranging from simplified 1D models to advanced 3D coupled flow-soil approaches. Lastly, open questions are posed and key future directions are identified to enhance our understanding of the breaching failure. Overall, this review paper provides a valuable resource for researchers and decision makers involved in slope stability and flow slide risk assessment.@en

In submerged sandy slopes, soil is frequently eroded as a combination of two main mechanisms: breaching, which refers to the retrogressive failure of a steep slope forming a turbidity current, and instantaneous sliding wedges, known as shear failure, that also contribute to shape the morphology of the soil deposit. Although there are several modes of failures, in this paper we investigate breaching and shear failures of granular columns using the two-fluid approach. The numerical model is first applied to simulate small-scale granular column collapses (Rondon et al., Phys. Fluids, vol. 23, 2011, 073301) with different initial volume fractions to study the role of the initial conditions in the main flow dynamics. For loosely packed granular columns, the porous medium initially contracts and the resulting positive pore pressure leads to a rapid collapse. Whereas in initially dense-packing columns, the porous medium dilates and negative pore pressure is generated stabilizing the granular column, which results in a slow collapse. The proposed numerical approach shows good agreement with the experimental data in terms of morphology and excess of pore pressure. Numerical results are extended to a large-scale application (Weij, doctoral dissertation, 2020, Delft University of Technology; Alhaddad et al., J. Mar. Sci. Eng., vol. 11, 2023, 560) known as the breaching process. This phenomenon may occur naturally at coasts or on dykes and levees in rivers but it can also be triggered by humans during dredging operations. The results indicate that the two-phase flow model correctly predicts the dilative behaviour and the subsequent turbidity currents associated with the breaching process.

@en

The ability to model bar migration in the nearshore zone is critically important for predicting coastal morphology. We present a cost-effective wave-averaged 3D model (CROCO) with parameterization of wave-related bedload transport (adapted SANTOSS), and evaluate/calibrate its performance in comparison with data measured in a two-dimensional wave channel. In this model, suspended-load transport (both in and above the wave bottom boundary layer) is resolved by the flow model. We assume that if the tuning parameters are not robust to varying forcing conditions, the parameterization can be considered a failure. The present work is therefore a first step towards a numerical model capable of predicting the onshore and offshore migration of a sandbar under storm and post-storm conditions using the same set of parameters. One way to achieve this goal is to treat the effects of waves and currents in separate formulations to avoid conflicts and redundancy. We show that the implementation and adaptation of SANTOSS in CROCO can achieve this goal, provided that the hydrodynamics are accurately reproduced and the separation between the wave-related bedload and current-related suspended load formulations is respected.@en

Reliable prediction of the erosion rate of sediment beds is important for many applications in coastal and river engineering. Theoretical understanding of empirically derived scaling relations is still lacking. This applies in particular for the scaling anomaly between low and high Shields number conditions. In this work, the erosion process is studied from the perspective of the phase-averaged turbulent kinetic energy (TKE) equations. The multi-phase TKE equations are written in a form that allows for a direct comparison with the TKE equation that appears for a stratified single-phase flow under the Boussinesq approximation. This reveals that next to buoyancy destruction, several other TKE modulation mechanisms become important at high Shields numbers and concentrations. Two scaling laws are derived for both moderate and high Shields numbers, and are tested against a wide range of experimental data.@en

The problem of sandbar migration on the storm timescale is revisited with a 3D wave-resolving hydro-sedimentary model. The latter presents an intermediate approach between expensive wave-resolving two-phase flow models and highly parametrized wave-averaged models. Innovative features include the use of weakly compressible assumptions in the hydrodynamics and morphological acceleration of bed changes to speed up numerical simulations. The model accurately simulates the successive offshore and onshore bar migration observed in a large-scale flume experiment in response to wave forcing representing storm and post-storm (recovery) conditions. The diagnosis of sand transport and the analysis of an ensemble-averaged asymmetric wave cycle reveal the migration mechanisms in each phase. In all cases, sediment resuspension is impacted by breaker-induced turbulence, while sediment transport and bed evolution are primarily the result of the undertow distribution – the breaker-induced seaward undercurrent – across the sandbar. There is also a significant contribution from asymmetric wave-related onshore fluxes, due to greater mobilization and currents during the wave crest period.@en