High Resolution Numerical Simulation of Oscillatory Flow and Sediment Transport through Aquatic Vegetation: Using the Highly Scalable Higher- Order Incompressible Solver Nek5000
Paul Fischer and Som Dutta, University of Illinois at Urbana-Champaign
Aquatic vegetation provides a wide range of services to the ecosystem: improves water-quality through nutrient uptake and production of oxygen, promotes bio-diversity through creation of a heterogeneous velocity field by virtue of its own spatial heterogeneity, provides flood buffer and coastal protection services, and regulates erosion and deposition patterns on water bodies, playing a paramount role in habitat creation for other species. While interactions between flow and vegetation have been extensively studied for unidirectional flows, the impact on sediment is not yet fully understood, and relatively much less is known about such interactions for oscillatory flow. The proposed study is geared towards increasing our understanding of the interactions between vegetation, flow and sediment under oscillatory flows. We will conduct Direct Numerical Simulations and Large Eddy Simulations of oscillatory flow and sediment transport through various vegetation conditions: rigid and flexible, submerged and emergent, from a single element to dense arrays. The highly resolved and accurate velocity field from the simulations will allow us to: 1) quantifying vegetation impact on turbulence intensity and turbulent kinetic energy in the flow, 2) characterizing vegetated drag and drag coefficients, and 3) the effect of vegetation on the transport of sediment and the feedback of the sediment back to the flow. Results will be compared with experimental data from novel experiments at the Ven-Te-Chow Hydrosystems Laboratory. Simulations would be conducted using the highly scalable spectral element based incompressible Navier-Stokes solver Nek5000, which has shown strong scaling on Blue Waters for n/P ~ 7,000 (where n is number of computational points and P is number of processors). Sediment would be modeled under both the Lagrangian and Eulerian framework, in order to capture different aspects of the underlying physics.