Application of a Three-Dimensional Unstructured-Mesh Finite-Element Flooding Model and Comparison with Two-Dimensional Approaches

Abstract

General rightsThis document is made available in accordance with publisher policies. Please cite only the published version using the reference above. AbstractUrban flood modelling plays a key role in assessment of flood risk in urban areas by providing detailed information of the flooding process (e.g. location, depth and velocity of flooding). Accurate modelling results are the basis of reliable flood risk evaluation. In this paper, a new 3D unstructured mesh urban flooding model developed in [1] has been applied to a flood event in a densely urbanized area within the city of Glasgow. Good agreement has been achieved when comparing the results with those published in other 2D shallow water models [2,3] in ponded areas. However, larger vertical velocity (> 0.2 m/s) and larger differences between the 3D and 2D model can be observed in areas with greater topographic gradients (> 3%). Through the modelling of a real flooding event this paper helps illustrate the case that 3D modelling techniques are promising to improve accuracy and obtain more detailed information related to urban flooding dynamics. This helps to obtain better assessments of flood damage and vulnerability of urban areas. To the * Corresponding author Email address: f.fang@imperial.ac.uk (F. Fang)Preprint submitted to Elsevier January 18, 2015 best of our knowledge, this is the first paper to apply a 3D unstructured mesh finite-element model to a real urban flooding event. It highlights some of the differences between the 3D and 2D urban flood modelling results.

Investigating the role of model structure and surface roughness in generating flood inundation extents using one‐ and two‐dimensional hydraulic models

“…Considering the increasing threat of frequently occurring high magnitude floods, there is a growing interest to create reach scale and regional scale flood inundation extents corresponding to near‐future storm forecasts (Bates, ; Bermúdez et al, ; Chang, Shen, Wang, Huang, & Lin, ; Nguyen, Thorstensen, Sorooshian, Hsu, & AghaKouchak, ; Pappenberger, Thielen, & Del Medico, ). Flooding in natural channels is a three‐dimensional hydrodynamic process that is typically simulated by using models that are simplified for idealised environmental systems based on certain assumptions (Marsooli, Orton, Georgas, & Blumberg, ; Zhang, Feng, Maksimović, & Bates, ). The structure of a hydraulic model can be described by its governing equations for river channel and floodplain as well as how these equations are solved in one (1D), two (2D), or three dimensions (3D).…”

Numerical Computation of Underground Inundation in Multiple Layers Using the Adaptive Transfer Method

“…Recently, various models and techniques were applied to predict flooding, inundation, and its vulnerability in advanced approaches [6][7][8][9]. However, there is a lack of hydrodynamic-based modeling that analyzes, in detail, the flow through underground access stairways or the flow characteristics through underground structures that link upper and lower layers in complex underground areas.…”

Stochastic modelling of a diffusive wave for flood propagation using the random walk particle tracking method in a hypothetical city

“…Many researchers used Eulerian formulation to obtain solutions of such equations (Fennema & Chaudhry, ; Kazezyılmaz‐Alhan & Medina Jr, ; Moussa & Bocquillon, ). Finite difference (Bates, Horritt, & Fewtrell, ; Haltas, Tayfur, & Elci, ; O'brien, Julien, & Fullerton, ), finite volume (Ali, Kimura, & Shimizu, ; Bradford & Sanders, ; Liang et al, ), and finite element (Cobby, Mason, Horritt, & Bates, ; Yu & Lane, ; Zhang, Feng, Maksimović, & Bates, ) schemes are well‐known numerical schemes, among many others. Solutions of Eulerian equations for flood modelling are not always reliable because they often display numerical instabilities or numerical oscillations and also produce artificial diffusions.…”