Hydrodynamics of large-amplitude oscillatory flows over cylinder arrays

Abstract:

Coastal vegetation such as seagrasses and salt marsh vegetation can contribute to coastal defence by stabilizing the seabed and attenuating wave energy; as a result, the hydrodynamics of wave-vegetation interactions are the subject of much research. Several previous laboratory studies have investigated in-canopy hydrodynamics in small-scale wave flumes under moderate wave conditions. However, behaviour under large storm-scale wave conditions is still poorly understood.

Experiments were conducted in the Aberdeen Oscillatory Flow Tunnel (AOFT), a U-tube facility that can generate oscillatory flows with periods between 5-10 s and orbital amplitudes up to 1.5 m, similar to near-bed conditions under large storm waves. The vegetation canopy was represented by a submerged array of rigid cylinders (130 mm tall and 8 mm in diameter) which were fixed on a smooth PVC bed in three different ‘staggered’ arrangements with densities of 434, 578 and 1736 cylinders/m². Velocities were measured within and above the array using Laser Doppler Anemometry (LDA). In addition, one individual cylinder within the array was mounted on a 6-axis submersible load cell, measuring all forces and moments acting on the cylinder.

The included figure shows example results: vertical profiles of root mean square (RMS) horizontal velocity measured under different flow amplitudes for the ‘densest’ cylinder arrangement, each profile labelled with its RMS ‘free-stream’ velocity. The figure illustrates the impact of the array on the surrounding flow, with the formation of an oscillatory boundary layer on top of the canopy and a significant reduction in ‘in-canopy’ velocities at high flow amplitudes. Detailed hydrodynamics and turbulence within and above the array will be presented, focusing in particular on the interface between the array and the ‘free-stream’, where shear stresses are high. Drag coefficients (determined directly from the force data) will also be discussed, and experimental results will be compared to model predictions from a basic analytical model for in-canopy hydrodynamics.