Search

Extreme hydrodynamic events, such as those driven by tsunamis, have a significant impact on coastal environments. The Smoothed Particle Hydrodynamics computational method gained popularity in modelling these phenomena. However, high resolution is needed in areas of interest, making coupling techniques popular to reduce computational costs. Herein, a new two-step offline coupling method was developed and validated in DualSPHysics. In step 1, the simulated velocity field and water depth are measured over a two-dimensional meshed surface of a generating domain. In step 2, the interpolated flow variables are used as boundary conditions in a receiving domain with equal or higher resolution. The method was validated by using two different laboratory experiments that are representative of tsunami propagation and inundation inland. The results show a reduction of computational time of up to 17.6 times, with decreasing savings for increasing resolution in the receiving domain. The validation tests showed that the developed method allows to simulate flows in the receiving domains at nearly the same accuracy of the generating domain while also decreasing computational time. When including debris transport, improvements in accuracy occur when doubling the resolution of the receiving domain with respect to the generating domain., This research was supported by grants from NVIDIA, United States and utilised NVIDIA RTX A6000 48GB. Dr. Corrado Altomare acknowledges funding from Spanish government and the European Social Found (ESF) under the programme ‘Ramón Cajal 2020’ (RYC2020-030197-I/AEI / 10.13039/501100011033). This work was partially supported by the project SURVIWEC PID2020-113245RB-I00 financed by MCIN/AEI/10.13039/501100011033 and by the project ED431C 2021/44 “Programa de Consolidación e Estructuración de Unidades de Investigación Competitivas” financed by Xunta de Galicia, Consellería de Cultura, Educación e Universidade, Spain. This study forms part of the Marine Science programme (ThinkInAzul) supported by Ministerio de Ciencia e Innovación and Xunta de Galicia with funding from European Union NextGenerationEU (PRTR-C17.I1) and European Maritime and Fisheries Fund ., Peer Reviewed, Postprint (author's final draft)

This work presents the results of an investigation on how wave overtopping at a near-vertical seawall at the back of a sandy foreshore is influenced by sequences of erosive storms. The experiments were carried out in the Large Wave Flume (GWK) at Leibniz University, Hannover (Germany). The tested layout consisted of a near-vertical seawall and a sandy foreshore with an initial slope. Three sequences of idealised erosive storms were simulated. Within each storm both the incident wave conditions and still water level were varied in time to represent high and low tide conditions. Each sequence started from a configuration and the beach was not restored in between storms. The measurements included waves, beach profile, wave overtopping volumes. The profile of the beach was measured after each sea state tested. Wave overtopping at each stage of the tested storms was significantly influenced by bed changes. This was linked to the measured evolution of the beach. Measurements showed that a barred profile developed quickly at the start of each sequence, and scour developed at the toe of the structure during high water level conditions, while accretion or partial backfilling developed during low water level conditions. Due to these processes, the position of a sea state in the tested sequence is shown to be an important factor in determining the wave overtopping volume. Remarkably, when a weaker idealised storm followed a more energetic one, nearly the same level of overtopping was recorded. This is explained by the foreshore erosion, leading to increased water depths and wave heights at the toe of the structure. This finding allows to quantify and to explain the variability of wave overtopping in storms following one another at intervals shorter than the recovery time of the foreshore., The work was supported by the European Union Horizon 2020 Research and Innovation Programme through the grant to HYDRALAB-PLUS, Contract no. 654110. We are grateful to all the staff in GWK for the hard work that made the experiments and analysis possible., Peer Reviewed, Postprint (published version)

Extreme hydrodynamic events, such as tsunamis, can transport debris of a very wide range of dimensions, from cohesive sediments to debris such as trees and cars. Large sized floating debris are particularly hazardous for two main reasons: (i) they can accumulate in narrow passages, such as the case of log jams at bridges, and obstruct the flow, creating potential for further flooding; (ii) they can impact directly on structures transferring significant energy to them, due to their mass and velocity, which can lead to damage and even failure. The transport and the interactions of this type of debris is studied experimentally, often in the context of tsunamis and flash floods. However, numerical studies on large floating debris impact on structures are rare. Therefore, the present study aims to address this gap by numerically modelling of the flow-debris-structure interactions. First, the experiments of Stolle et al. (2018) are simulated numerically. These experiments involve a single positively buoyant container impact a structure as a result of being transported by a dam-break flow. The numerical simulations are carried using the open source DualSPHysics model based on the Smoothed Particle Hydrodynamics method. First, the hydrodynamics results were validated with data from Stolle et al. (2018). Subsequently, DualSPHysics was coupled with the Multiphysics engine CHRONO to simulate the container and its impact on the structure. The dam break event described in Stolle et al. (2018), was generated by modelling the movement of a swing gate using the experimental time series, with the container and the structure positioned at 3.2 m and 7.03 m, respectively, downstream of the reservoir consistently with Stolle et al. (2018). The trajectory as well as the velocity of the centroid of the container were tracked throughout the simulation. The agreement between the model and the experiment results is quantitatively assessed and it is shown that the model is accurate in reprodu, Postprint (published version)