On nonhydrostatic coastal model simulations of shear instabilities in a stratified shear flow at high Reynolds number

Abstract

The nonhydrostatic surface and terrain‐following coastal model NHWAVE is utilized to simulate a continually forced stratified shear flow in a straight channel, which is a generic problem to test the existing nonhydrostatic coastal models' capability in resolving shear instabilities in the field scale. The resolved shear instabilities in the shear layer has a Reynolds number of about 1.4 × 106, which is comparable to field observed value. Using the standard Smagorinsky closure with a grid size close to the Ozmidov length scale, simulation results show that the resolved energy cascade exceeds 1 order of magnitude and the evolution and turbulent mixing characteristics are predicted well. Two different approaches are used to estimate the turbulent dissipation rate, namely using the resolved turbulent energy spectrum and the parameterized subgrid turbulent dissipation rate, and the predicted results provide the upper and lower bounds that encompass the measured values. Model results show significantly higher turbulence in braids of shear instabilities, which is similar to field observations while both the subgrid turbulent dissipation rate and resolved vorticity field can be used as surrogates for measured high acoustic backscatter signals. Simulation results also reveal that the surface velocity divergence/convergence is an effective identifier for the front of the density current and the shear instabilities. To guide future numerical studies in more realistic domains, an evaluation on the effects of different grid resolutions and subgrid viscosity on the resolved flow field and subgrid dissipation rate are discussed.

Observations and Modeling of a Buoyant Plume Exiting Into a Tidal Cross‐Flow and Exhibiting Along‐Front Instabilities

“…F. Shi et al (2017) used this same model to simulate an internal hydraulic jump at the mouth of the Columbia River, which also exhibited along-jump propagating instabilities (Honegger et al, 2017). Also, Zhou et al (2017) previously demonstrated that this non-hydrostatic model can resolve Kelvin-Helmholtz shear instabilities in the vertical shear layer underneath a fresh water plume simulated at field scale.…”

“…Shi et al, 2015) and accuracy (Derakhti et al, 2016) and to produce increasingly realistic simulations of internal waves (J. Shi et al, 2019), and non-hydrostatic frontal processes in river plumes (F. Shi et al, 2017;Zhou et al, 2017). F. Shi et al (2017) used this same model to simulate an internal hydraulic jump at the mouth of the Columbia River, which also exhibited along-jump propagating instabilities (Honegger et al, 2017).…”

“…For comparison to the field data, we utilize the NHWAVE non‐hydrostatic model (Ma et al., 2013; J. Shi et al., 2015). Since these fine‐scales, along‐front instabilities are turbulent coherent structures, a non‐hydrostatic, 3D eddy‐resolving numerical model such as NHWAVE is needed to simulate the frontal structures (F. Shi et al., 2017; Zhou et al., 2017).…”

Application of Airborne Infrared Remote Sensing to the Study of Ocean Submesoscale Eddies

“…Our view tends toward the latter; but how the banding arises, and why it has the spacing and spatial extent that it does, are questions for future work. At this point, a reasonable conjecture is that such features arise from vertical shear instability (e.g., Zhou et al, 2017).…”

Tidally Induced Instability Processes Suppressing River Plume Spread in a Nonrotating and Nonhydrostatic Regime

“…First, the grid spacing might remain insufficient for resolving K-H billows. Recently, Zhou et al (2017) reproduced the shear instabilities developed at the bottom of a freshwater plume using an idealized nonhydrostatic numerical model with horizontal and vertical resolutions less than one third those used in the present study. A folded-wave pattern was observed clearly at the bottom of the plume, as in laboratory experiments (e.g., Yuan & Horner-Devine, 2013).…”