Numerical Investigation of Solitary Internal Wave-Induced Global Instability in Shallow Water Benthic Boundary Layers

Abstract

Abstract The time-dependent boundary layer induced by a weakly nonlinear solitary internal wave in shallow water is examined through direct numerical simulation. Waves of depression and elevation are both considered. The mean density field corresponds to that typical of the coastal ocean and lakes where the lower fraction of the water column is subject to the stabilizing effect of a diffuse stratification. Sufficient resolution of the “inviscid” dynamics of the boundary layer is ensured through use of a Legendre spectral multidomain discretization scheme in the vertical direction. At higher Reynolds numbers, where the simulations become underresolved, because of restrictions in available computational resources, spectral accuracy and numerical stability at the scales of physical interest are preserved through use of a penalty scheme in the vertical and explicit spectral filtering. Thus, a highly accurate description of the qualitative dynamics of the wave-induced global instability is possible and finescale physical mechanisms critical to the appearance of this instability are not smeared out by the high artificial dissipation inherent in lower-order finite-difference schemes. Results indicate that, for a wave amplitude exceeding a critical value, the global instability occurs in regions near the bottom boundary where the wave induces an adverse pressure gradient. The structure of the associated separation bubble is modified through the establishment of coherent and synchronous dynamics, characterized by elevated levels of bottom shear stress and a periodic shedding of coherent vortex structures. Although details of the vortex shedding depend on the particular wave forcing involved, these vortical structures always ascend high into the water column. All findings suggest that this global instability is a potent mechanism for benthic turbulence, mixing, and possible sediment resuspension in shallow waters, presumably even more intense than the nominal turbulent boundary layer.