Landslide-generated tsunamis represent a significant hazard, specifically in coastal, fjord, and lake environments. The accurate modeling of these complicated phenomena is crucial for risk assessment and preparation of mitigation strategies. Landslide-generated tsunamis can be modeled using either three-dimensional (3D) computational fluid dynamics or depth-integrated two-dimensional (2D) numerical models. 3D models are valuable for the modeling in the near-field, however their computational cost is often prohibitive for simulating long-range wave propagation (far-field). 2D models are very efficient for simulating tsunami propagation over large areas, but cannot capture the full complexity of the generation process. In this paper, we propose a coupling strategy between 3D and 2D models for simulating tsunamis induced by granular landslides in complex geometries. Thus, it aims at maximizing near-field accuracy while minimizing computational costs, thereby enabling efficient far-field wave propagation. This strategy involves using the 3D model (OpenFOAM®) where nonlinear and confinement effects are dominant and feeding the 2D model (Mild Slope Equation) where the waves can efficiently be handled by such simplified tool. This model chain has been applied to the 2015 Taan Fiord, Alaska (USA), landslide-tsunami event, a complex and well-documented case, with extensive field data available. Beyond evaluating its performance for this specific event, the broader aim is to provide some general insights and practical guidance for the application of this modeling chain to landslide-tsunami events. To this end, a sensitivity analysis, by varying the position of the coupling boundary between the two models, is also presented and discussed.
Romano, A., Cecioni, C., Bellotti, G., Barajas, G., Lara, J.L. (2025). A Numerical 3D-2D Coupling Strategy for Simulating Landslide-Generated Tsunamis in Complex Geometries: The Taan Fiord, Alaska, Case Study. JOURNAL OF GEOPHYSICAL RESEARCH. OCEANS, 130(12) [10.1029/2025JC023042].
A Numerical 3D-2D Coupling Strategy for Simulating Landslide-Generated Tsunamis in Complex Geometries: The Taan Fiord, Alaska, Case Study
Romano A.
;Cecioni C.;Bellotti G.;
2025-01-01
Abstract
Landslide-generated tsunamis represent a significant hazard, specifically in coastal, fjord, and lake environments. The accurate modeling of these complicated phenomena is crucial for risk assessment and preparation of mitigation strategies. Landslide-generated tsunamis can be modeled using either three-dimensional (3D) computational fluid dynamics or depth-integrated two-dimensional (2D) numerical models. 3D models are valuable for the modeling in the near-field, however their computational cost is often prohibitive for simulating long-range wave propagation (far-field). 2D models are very efficient for simulating tsunami propagation over large areas, but cannot capture the full complexity of the generation process. In this paper, we propose a coupling strategy between 3D and 2D models for simulating tsunamis induced by granular landslides in complex geometries. Thus, it aims at maximizing near-field accuracy while minimizing computational costs, thereby enabling efficient far-field wave propagation. This strategy involves using the 3D model (OpenFOAM®) where nonlinear and confinement effects are dominant and feeding the 2D model (Mild Slope Equation) where the waves can efficiently be handled by such simplified tool. This model chain has been applied to the 2015 Taan Fiord, Alaska (USA), landslide-tsunami event, a complex and well-documented case, with extensive field data available. Beyond evaluating its performance for this specific event, the broader aim is to provide some general insights and practical guidance for the application of this modeling chain to landslide-tsunami events. To this end, a sensitivity analysis, by varying the position of the coupling boundary between the two models, is also presented and discussed.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


