Langmuir turbulence in shallow water. Part 1. Observations

2007 ◽  
Vol 576 ◽  
pp. 27-61 ◽  
Author(s):  
ANN E. GARGETT ◽  
JUDITH R. WELLS

During extended deployment at an ocean observatory off the coast of New Jersey, a bottom-mounted five-beam acoustic Doppler current profiler measured large-scale velocity structures that we interpret as Langmuir circulations filling the entire water column. These circulations are the large-eddy structures of wind-wave-driven turbulent flows that occur episodically when a shallow water column experiences prolonged strong wind forcing. Many observational characteristics agree with former descriptions of Langmuir circulations in deep water. The three-dimensional velocity field reveals quasi-organized structures consisting of pairs of surface-intensified counter-rotating vortices, aligned approximately downwind. Maximum downward velocities are stronger than upward velocities, and the downwelling region of each cell, defined as a pair of vortices, is narrower than the upwelling region. Maximum downward vertical velocity occurs at or above mid-depth, and scales approximately with wind speed. The estimated crosswind scale of cells is roughly 3–6 times their vertical scale, set under these conditions by water depth. The long axis of the cells appears to lie at an angle ∼10°–20° to the right of the wind. A major difference from deep-water observations is strong near-bottom intensification of the downwind ‘jets’ found typically centred over downwelling regions. Accessible observational features such as cell morphology and profiles of mean velocities, turbulent velocity variances, and shear stress components are compared with the results of associated large-eddy simulations (reported in Part 2) of shallow water flows driven by surface stress and the Craik–Leibovich vortex forcing generally used to represent generation of Langmuir cells. A particularly sensitive diagnostic for identification of Langmuir circulations as the energy-containing eddies of the turbulent flow is the depth trajectory of invariants of the turbulent stress tensor, plotted in the Lumley ‘triangle’ corresponding to realizable turbulent flows. When Langmuir structures are present in the observations, the Lumley map is distinctly different from that of surface-stress-driven Couette flow, again in agreement with the large-eddy simulations (LES). Unlike the LES, observed velocity fields contain two distinct and significant scales of variability, documented by wavelet analysis of observational records of vertical velocity. Variability with periods of many minutes is that expected from Langmuir cells drifting past the instrument at the slowly time-varying crosswind velocity. Shorter period variability, of the order of 1–2 min, has roughly the observed periodicity of surface wave groups, suggesting a connection with the wave groups themselves and/or the wave breaking associated with them in high wind conditions.

Author(s):  
J. Boudet ◽  
A. Cahuzac ◽  
P. Borgnat ◽  
E. Lévêque ◽  
F. Toschi

2016 ◽  
Author(s):  
Brian M. Griffin ◽  
Vincent E. Larson

Abstract. The subgrid-scale representation of hydrometeor fields is important for calculating microphysical process rates. In order to represent subgrid-scale variability, the Cloud Layers Unified By Binormals (CLUBB) parameterization uses a multivariate Probability Density Function (PDF). In addition to vertical velocity, temperature, and moisture fields, the PDF includes hydrometeor fields. Previously, each hydrometeor field was assumed to follow a multivariate single lognormal distribution. Now, in order to better represent the distribution of hydrometeors, two new multivariate PDFs are formulated and introduced. The new PDFs represent hydrometeors using either a delta-lognormal or a delta-double-lognormal shape. The two new PDF distributions, plus the previous single lognormal shape, are compared to histograms of data taken from Large-Eddy Simulations (LES) of a precipitating cumulus case, a drizzling stratocumulus case, and a deep convective case. Finally, the warm microphysical process rates produced by the different hydrometeor PDFs are compared to the same process rates produced by the LES.


2015 ◽  
Vol 137 (5) ◽  
Author(s):  
Dario Amirante ◽  
Nicholas J. Hills

Large-eddy simulations (LES) of wall bounded, low Mach number turbulent flows are conducted using an unstructured finite-volume solver of the compressible flow equations. The numerical method employs linear reconstructions of the primitive variables based on the least-squares approach of Barth. The standard Smagorinsky model is adopted as the subgrid term. The artificial viscosity inherent to the spatial discretization is maintained as low as possible reducing the dissipative contribution embedded in the approximate Riemann solver to the minimum necessary. Comparisons are also discussed with the results obtained using the implicit LES (ILES) procedure. Two canonical test-cases are described: a fully developed pipe flow at a bulk Reynolds number Reb = 44 × 103 based on the pipe diameter, and a confined rotor–stator flow at the rotational Reynolds number ReΩ = 4 × 105 based on the outer radius. In both cases, the mean flow and the turbulent statistics agree well with existing direct numerical simulations (DNS) or experimental data.


Recent advances in the mathematical theory of the Navier-Stokes equations have produced new insight in the mathematical theory of turbulence. In particular, the study of the attractor for the Navier-Stokes equations produced the first connection between two approaches to turbulence that seemed far apart, namely the conventional approach of Kolmogorov and the dynamical systems theory approach. Similarly the study of the approximation of the attractor in connection with the newly introduced concept of approximate inertial manifolds has produced a new approach to large eddy simulations and the study of the interaction of small and large eddies in turbulent flows. Our aim in this article is to survey and describe some of the new results concerning the functional properties of the Navier-Stokes equations and to discuss their relevance to turbulence.


Author(s):  
Spencer R. Alexander ◽  
Peter E. Hamlington

As ocean current turbines move from the design stage into production and installation, a better understanding of oceanic turbulent flows and localized loading is required by researchers and members of industry. Consideration of realistic ocean turbulence environments, in particular, is essential for obtaining accurate and reliable predictions of ocean turbine lifetime and performance. In this study, large eddy simulations (LES) are used to model the turbulent boundary layer in which an ocean current turbine operates. The LES model captures current driving due to winds, waves, and tides, thereby providing a high degree of physical realism. Inflow and boundary conditions are designed to represent conditions during an observational campaign at Admiralty Head in Puget Sound, and comparisons are made between the LES results and available observational measurements. Further statistical measures of the LES flow fields are outlined, including vertical profiles of Reynolds stresses, turbine loading, and two point correlations. The ability of the synthetic turbulence generator TurbSim to reproduce realistic ocean turbulence is qualitatively assessed through comparisons with LES results. Finally, preliminary simulation results are presented for an ocean current turbine represented by an actuator disk.


2011 ◽  
Vol 23 (1) ◽  
pp. 015101 ◽  
Author(s):  
C. Duprat ◽  
G. Balarac ◽  
O. Métais ◽  
P. M. Congedo ◽  
O. Brugière

Sign in / Sign up

Export Citation Format

Share Document