Modeling anisotropy in free-surface overland and shallow inundation flows

• Published in: Advances in water resources 2017-06, Vol.104, p.1-14
• Main Authors: Viero, Daniele Pietro, Valipour, Mohammad
• Format: Article
• Language: English
• Subjects: Agricultural environments; Agriculture; Anisotropic resistance; Anisotropy; Computation; Computer applications; Computer simulation; Ditches; Finite element method; Flooding; Floodplain inundation; Floodplains; Fluid dynamics; Furrows; High resolution; Hydrodynamics; Hydrologic models; Hydrology; Irrigation; Landscape; Large-scale models; Modelling; Overland flow; Physical simulation; Resolution; Shallow water; Shallow water equations; Simulation; Studies; Subgrid modeling; Urban areas; Water
• ISSN: 0309-1708; 1872-9657
• DOI: 10.1016/j.advwatres.2017.03.007

•Regular patterns are common in both natural and man-modified environments.•Oriented roughness forces anisotropic resistance that affects free-surface flows.•Anisotropic resistance is formalized and accounted for in a 2D hydrodynamic model.•A subgrid model, based on the concept of REA, ensures mesh-independentness.•Accuracy and efficiency make the model suitable to large-scale applications. Regular patterns, which are found in both natural and man-modified environments, are strongly interwoven with free-surface flows. Examples are ridge and slough landscapes, cultivated terrains with ditches and furrows, and urban areas, with many of them characterized by a marked anisotropy. Simulation of overland and shallow inundation flows in these contexts is complex and demanding, especially if very different spatial scales are involved. Anisotropic effects are introduced to cope with two-dimensional shallow water models and, particularly, with the subgrid modeling technique. Application to schematic test cases shows the key role played by anisotropy in shallow flows, and second, that anisotropy can be effectively captured by the subgrid model with low computational effort and preserving mesh-independentness. High-resolution model results are accurately reproduced on coarser meshes using one fiftieth of the original computational elements, with a speed-up of more than 20. The subgrid approach could serve in view of physically based, large-scale modeling of floodplain inundation processes, in irrigation science, and in high-resolution hydrodynamic-hydrological simulations at the basin scale.