Numerical modelling of landslide-tsunami propagation in a wide range of idealised water body geometries

Abstract

Large landslide-tsunamis are caused by mass movements such as landslides or rock falls impacting into a water body. Research of these phenomena is essentially based on the two idealised water body geometries (i) wave flume (2D, laterally confined wave propagation) and (ii) wave basin (3D, unconfined wave propagation). The wave height in 2D and 3D differs by over one order of magnitude in the far field. Further, the wave characteristics in intermediate geometries are currently not well understood. This article focuses on numerical landslide-tsunami propagation in the far field to quantify the effect of the water body geometry. The hydrodynamic numerical model SWASH, based on the non-hydrostatic non-linear shallow water equations, was used to simulate approximate linear, Stokes, cnoidal and solitary waves in 6 different idealised water body geometries. This includes 2D, 3D as well as intermediate geometries consisting of "channels" with diverging side walls. The wavefront length was found to be an excellent parameter to correlate the wave decay along the slide axis in all these geometries in agreement with Green's law and with diffraction theory in 3D. Semi-theoretical equations to predict the wave magnitude of the idealised waves at any desired point of the water bodies are also presented. Further, simulations of experimental landslide-tsunami time series were performed in 2D to quantify the effect of frequency dispersion. This process may be negligible for solitary-and cnoidal-like waves for initial landslide-tsunami hazard assessment but becomes more important for Stokes-like waves in deeper water. The findings herein significantly improve the reliability of initial landslide-tsunami hazard assessment in water body geometries between 2D and 3D, as demonstrated with the 2014 landslide-tsunami event in Lake Askja.

Tsunamis Generated by Submerged Landslides: Numerical Analysis of the Near‐Field Wave Characteristics

“…3-D configurations (e.g., Abadie et al, 2010Abadie et al, , 2012Bellotti et al, 2008;Clous & Abadie, 2019;Grilli et al, 2017;Heidarzadeh et al, 2020;Heller et al, 2016;Kim et al, 2019;Liu et al, 2005;Løvholt et al, 2005;Lynett & Liu, 2005;Montagna et al, 2011;Ruffini et al, 2019;Shi et al, 2016Shi et al, , 2018Watts et al, 2003;Whittaker et al, 2017).…”

A Numerical Landslide-Tsunami Hazard Assessment Technique Applied on Hypothetical Scenarios at Es Vedrà, Offshore Ibiza

“…This is also observed in scenario 2, where the primary wave finally overtakes the landslide at t = 20.1 s, and continues to propagate outwards (Figure 6d). Both Figures 5 and 6 confirm that the landslide tsunamis are largest on the slide axis and significantly smaller at the peripheries [29,48]. Four wave probes in scenario 1 and five wave probes in scenario 2 were placed along the slide axes ( Figure 1) to record the landslide tsunamis, as shown in Figure 7.…”

“…This is the case at x = 750 m in scenario 1 and at x = 1000 m in scenario 2. Alternative coupling criteria, e.g., based on the location of the maximum wave amplitude defined in Heller and Hager [14], are discussed in Ruffini et al [48].…”

“…c f is the bottom friction coefficient defined by Manning's formula, and q(x, y, z, t) is the non-hydrostatic pressure. Equations (11)- (13) were expanded for the multi-layer case by Stelling and Zijlema [48] using a discretisation method based on the Keller-box scheme.…”

“…The use of SWASH v4.01 for the wave propagation allowed reducing the computational time significantly in both scenarios, and helped to avoid the wave decay problem of DualSPHysics (Section 4.2). Two different regular grids were created, one for each scenario, both with a grid resolution of ∆x = ∆y = 30 m, allowing at least 30 grid points per wavelength, which is sufficient to ensure the convergence of the solution [48,49]. The bathymetry was retrieved from the European Marine Observation and Data Network [36] with a resolution of approximatively 200 m. The topography was taken from the CNIG [37] with digital terrain model data with an accuracy of 25 m to match the grid resolution closely.…”