DONATE

IAHR Document Library


« Back to Library Homepage « Proceedings of the 22nd IAHR APD Congress (Sapporo, 2020)

High Resolution Numerical Modelling of Tsunami Inundation Using Quadtree Method and Gpu Acceleration

Author(s): Shuang Gao; Greg Collecutt; William J. Syme; Philip Ryan

Linked Author(s):

Keywords: Tsunami; Numerical modelling; Quadtree; GPU acceleration; TUFLOW

Abstract: This study adapted the quadtree mesh in TUFLOW’s HPC 2D Shallow Water Equation (SWE) solver and investigated its feasibility of modelling tsunami inundation in coastal urban environments. The spatial mesh resolution was increased in the onshore urbanised area to reproduce the wave propagation at street scale, whereas a larger mesh size was used in offshore areas in order to reduce the total number of computational mesh elements. A ‘Sub-Grid-Sampling’ approach was also applied to mitigate the numerical head loss along wet/dry boundaries that are misaligned with the mesh. Model benchmark testing was conducted against a 1:50 scale flume experiment of the town Seaside, Oregon. The benchmarking results showed excellent agreement with the measured data and proved the numerical efficiency of the proposed quadtree mesh model. Furthermore, the combination of quadtree mesh and SGS approach offers great potential as an easy to construct, fast, high-resolution numerical tool to support tsunami hazard planning.

DOI:

Year: 2020

Copyright © 2024 International Association for Hydro-Environment Engineering and Research. All rights reserved. | Terms and Conditions