A comparison of a two-dimensional depth-averaged flow model and a three-dimensional RANS model for predicting tsunami inundation and fluid forces

Abstract

The numerical modeling of tsunami inundation that incorporates the built environment of coastal communities is challenging for both 2-D and 3-D depth-integrated models, not only in modeling the flow but also in predicting forces on coastal structures. For depth-integrated 2-D models, inundation and flooding in this region can be very complex with variation in the vertical direction caused by wave breaking on shore and interactions with the built environment, and the model may not be able to produce enough detail. For 3-D models, a very fine mesh is required to properly capture the physics, dramatically increasing the computational cost and rendering impractical the modeling of some problems. In this paper, comparisons are made between GeoClaw, a depth-integrated 2-D model based on the nonlinear shallow-water equations (NSWEs), and OpenFOAM, a 3-D model based on Reynolds-averaged Navier–Stokes (RANS) equation for tsunami inundation modeling. The two models were first validated against existing experimental data of a bore impinging onto a single square column. Then they were used to simulate tsunami inundation of a physical model of Seaside, Oregon. The resulting flow parameters from the models are compared and discussed, and these results are used to extrapolate tsunami-induced force predictions. It was found that the 2-D model did not accurately capture the important details of the flow near initial impact due to the transiency and large vertical variation of the flow. Tuning the drag coefficient of the 2-D model worked well to predict tsunami forces on structures in simple cases, but this approach was not always reliable in complicated cases. The 3-D model was able to capture transient characteristic of the flow, but at a much higher computational cost; it was found this cost can be alleviated by subdividing the region into reasonably sized subdomains without loss of accuracy in critical regions.