Elsevier

Ocean Modelling

Volume 149, May 2020, 101601
Ocean Modelling

The effect of Langmuir turbulence under complex real oceanic and meteorological forcing

https://doi.org/10.1016/j.ocemod.2020.101601Get rights and content

Highlights

  • Swell dominate over young wind sea reduce more /create less Shear /Stokes production.

  • Langmuir circulation counteract on deep convection and reduce the total turbulence.

  • Large wind-wave misalignment reduces resolved Reynold stress and eddy viscosity.

  • LES solution is more sensitive to the SGS model used for large wind-wave misalignment.

Abstract

In this study, we expand previous large eddy simulation (LES) modeling investigations of Langmuir turbulence (LT) to real ocean conditions using field observations collected under the multi-platform field campaign “Coupled Air–Sea Processes and Electromagnetic (EM) ducting Research (CASPER-East)”. The measurement site has strong local variabilities of temperature and salinity and experienced large variations in wind forcing and several cooling events.

Although LT enhances the turbulence in the water column and deepens the mixed layer during most of the simulation period, being consistent with previous studies, strong reduction of turbulent kinetic energy (TKE) in the mixed layer is observed in the simulation with Stokes drift compared to that without Stokes drift during a short period. Analysis of the meteorological forcing and the TKE budget have revealed that in the circumstance of swell dominated wave fields with young wind seas, the presence of Stokes drift reduces shear production more than the Stokes production it generates, and a reduction of total TKE in the mixed layer may be expected whether or not the Stokes drift is aligned with the wind.

Weak reduction of TKE due to the inclusion of Stokes drift is also observed beneath the mixed layer during a cooling event possibly due to the fact that the upwelling associated with Langmuir circulation at the base of the mixed layer counteracts on the downwelling associated with the deep convection and reduces the total turbulence level in the water column.

While both resolved Reynold stresses and the bulk eddy viscosity decrease with the increase of wind-wave misalignment angle θww and become smaller than that in the case without Stokes drift when θww exceed 60°, the subgrid scale (SGS) part of the momentum flux increases with the increase of θww, suggesting that the LES solutions in cases with large wind-wave misalignment become more sensitive to the SGS models used and need to be dealt with caution.

Introduction

Langmuir turbulence (LT) is believed to be one of the leading order causes of turbulent mixing in the upper ocean, which is important for momentum and heat exchange across the air–sea interface and between the mixed layer and the thermocline. Both observational studies (D’Asaro, 2001, D’Asaro, 2014) and large-eddy simulation (LES) investigations (Li et al., 1995, Kukulka et al., 2009, Kukulka et al., 2010, Skylingstad and Denbo, 1995, McWilliams et al., 1997, Hamlington et al., 2014) have shown enhanced vertical mixing within the ocean surface boundary layer in the presence of LT through the enhanced vertical turbulent velocity variance.

The dynamical origin of Langmuir circulation is understood as wind-driven shear instability in combination with surface wave influences related to their mean Lagrangian motion, called Stokes drift. The prevailing theoretical interpretation of Langmuir circulation is derived by Craik and Leibovich (1976), where they introduced the effect of waves on Eulerian mean flow into the Navier–Stokes equations. Even though the theory was developed four decades ago, scientists were not able to adequately measure or simulate LT until thirty years ago. While LES models have been used to simulate Langmuir circulation in the upper ocean, yielding new insights that could not be obtained from field observations or turbulent closure models, most of these studies were conducted under idealized conditions with simplified oceanic and wind conditions (Skylingstad and Denbo, 1995, McWilliams et al., 1997, McWilliams et al., 2012, Sullivan et al., 2012, Hamlington et al., 2014, Reichl et al., 2016). Idealizing and isolating individual processes makes it easier to study their effects, but can also unrealistically magnify or underestimate their impact, due to the lack of complex and nonlinear interactions of multiple dynamical processes taking place in the real ocean. Thus, parameterizations that have developed from these idealized studies can have limited practical application in ocean modeling. For example, evaluation of three of the K profile parameterizations (KPP, developed by Large et al., 1994) with LT modifications (McWilliams and Sullivan, 2000, Smyth et al., 2002, Qiao et al., 2004) in the GFDL climate model have shown that none of the schemes give consistent improvement to ocean circulation models globally most likely due to their lack of interaction with ocean physics (Fan and Griffies, 2014). While in Li et al. (2016) and Li and Fox-Kemper (2017), substantial improvements are observed when more physical processes are considered in the scaling, such as Harcourt and D’Asaro (2008) and Van Roekel et al. (2012).

In this study, we expand the previous LES modeling investigations of LT to real ocean conditions. Model forcing and initial conditions are obtained from a multi-platform field campaign, “Coupled Air–Sea Processes and Electromagnetic (EM) ducting Research (CASPER-East)” that took place off the coast of North Carolina in late October to early November of 2015. The study location, approximately 63 km east of Duct, N.C., is frequently influenced by fresher and cooler water inflow from nearby rivers and bays and warmer and saltier water intrusion from the Gulf stream, and experienced several cooling events and dramatic turning of wind directions due to storm passage during the observation period. Temperature (T) and salinity (S) profiles, surface gravity wave spectra, and meteorological forcing data were collected during the CASPER-East campaign, providing a rich data set to study the effect of LT on the dynamics and structure of the oceanic mixed layer under complex oceanic and meteorological conditions. The outline of this paper is as follows. A brief description of the observations during the CASPER-East experiment, the LES model used for this study, and the experiment set up are given in Section 2. Results are analyzed in Section 3, and discussion and concluding remarks are presented in Section 4.

Section snippets

Observations

The field data used in this research was collected under the CASPER project aimed to improve the characterization of the propagation of radio frequency signals through the marine atmosphere (Wang et al., 2018a). CASPER-East, the first of two major field campaigns during the CASPER project, was conducted from October 10 to November 6 of 2015 off the coast of North Carolina, eastward from the US Army Corps of Engineers Field Research Facility pier at Duck.

Atmospheric and oceanic measurements used

The Turbulent Kinetic Energy (TKE) budget

The turbulent kinetic energy (TKE) budget is usually analyzed in previous studies (Grant and Belcher, 2009, McWilliams et al., 2012, Van Roekel et al., 2012) to examine the effect of LT on the mixing. The horizontal domain averaged TKE equation can be written as: et=TurT+ShearP+BuoyP+PrsT+StokesPε+sgswhere, TurT=12u2wz+v2wz+w2wz is the turbulenttransport term,ShearP=uwuzvwvz is the shear production term,BuoyP=αgθwβgsw is the buoyancy production term,PrsT=1ρ0pw

Summary and discussions

Langmuir turbulence (LT) is believed to be one of the leading causes of turbulent mixing in the upper ocean (Li et al., 1995, Skylingstad and Denbo, 1995, Kukulka et al., 2009, Kukulka et al., 2010, McWilliams et al., 1997, Hamlington et al., 2014). Large eddy simulation (LES) models that solve the Craik–Leibovich equations are used to study LT, yielding new insights that could not be obtained from field observations or turbulent closure models alone. However, these studies have been mostly

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

The authors thank Dr. Adam Christman for providing the R/V air–seaflux data for this study. This work was funded by the Office of Naval Research, United States of America under program element 0601153N. This paper is a contribution of​ NRL/JA/7320-19-4508 and has been approved for public release. We would like to express our appreciation to the anonymous reviewers for their constructive comments. Observations presented in the manuscript are open-access data. Please contact Dr. Ivan Savelyev

References (48)

  • FanY. et al.

    Simulated global swell and wind-sea climate and their responses to anthropogenic climate change at the end of the twenty-first century

    J. Clim.

    (2014)
  • GargettA.E. et al.

    Langmuir turbulence in shallow water. Part 1. Observations

    J. Fluid Mech.

    (2007)
  • GargettA.E. et al.

    Langmuir supercells: A mechanism for sediment resuspension and trans- port in shallow seas

    Science

    (2004)
  • GrantA.L.M. et al.

    Characteristics of Langmuir turbulence in the ocean mixed layer

    J. Phys. Oceanogr.

    (2009)
  • HamlingtonP.E. et al.

    Langmuir–submesoscale interactions: Descriptive analysis of multiscale frontal spindown simulations

    J. Phys. Oceanogr.

    (2014)
  • HarcourtR.R. et al.

    Large-eddy simulation of Langmuir turbulence in pure wind seas

    J. Phys. Oceanogr.

    (2008)
  • KukulkaT. et al.

    Influence of Stokes drift decay scale on Langmuir turbulence

    J. Physic. Oceanogr.

    (2017)
  • KukulkaT. et al.

    Significance of Langmuir circulation in upper ocean mixing: Comparison of observations and simulations

    Geophys. Res. Lett.

    (2009)
  • KukulkaT. et al.

    Rapid mixed layer deepening by the combination of Langmuir and shear instabilities: A case study

    J. Phys. Oceanogr.

    (2010)
  • KukulkaT. et al.

    The influence of crosswind tidal currents on Langmuir circulation in a shallow ocean

    J. Geophys. Res.

    (2011)
  • LargeW.G. et al.

    Oceanic vertical mixing: a review and a model with a nonlocal boundary layer parameterization

    Rev. Geophys.

    (1994)
  • LiQ. et al.

    Assessing the effects of Langmuir turbulence on the entrainment buoyancy flux in the ocean surface boundary layer

    J. Phys. Oceanogr.

    (2017)
  • LiM. et al.

    Role of Langmuir circulation in the deepening of the ocean surface mixed layer

    Science

    (1995)
  • LiangJ.-H. et al.

    Parameterizing bubble-mediated air-sea gas exchange and its effect on ocean ventilation

    Glob. Biogeochem. Cycles

    (2013)
  • Cited by (10)

    • Exploring the use of machine learning to parameterize vertical mixing in the ocean surface boundary layer

      2022, Ocean Modelling
      Citation Excerpt :

      With continued efforts using advanced observational techniques and high-fidelity computer simulations, it is now understood that OSBL turbulence is primarily driven by three processes at the sea surface, viz. wind (e.g., Skyllingstad et al., 1999), heating/cooling (e.g., Li et al., 2005; Pearson et al., 2015), and ocean surface gravity waves (e.g., D’Asaro et al., 2014; Qiao et al., 2004). It is also altered by other factors including density stratification (Price and Sundermeyer, 1999), the earth’s rotation (Liu et al., 2018), the depth of the water column (e.g., Tejada-Martinez and Grosch, 2007; Yan et al., 2022), larger-scale horizontal density gradient (Fan et al., 2018, 2020) and lateral currents such as tidal and submesoscale currents (e.g., Hamlington et al., 2014; Sullivan and McWilliams, 2019; Yuan and Liang, 2021). Given the aforementioned challenges in traditional physics-based OSBL mixing parameterizations, this study explores the use of machine learning, specifically, a deep neural network (DNN) model, to parameterize the vertical mixing effects of OSBL turbulence.

    • A new K-profile parameterization for the ocean surface boundary layer under realistic forcing conditions

      2022, Ocean Modelling
      Citation Excerpt :

      We find that the effect of misaligned Stokes drift and wind direction has significant impact on the diagnosed eddy viscosity profiles in idealized LES under weakly convective conditions. This is obscured in realistic LES configurations where strong convection is often dominant over LT, but important differences in turbulent intensities are still observed between aligned and misaligned wind–waves in realistic LES experiments (Fan et al., 2020). To assess the overall performance of the models, results are validated with observed sea surface temperature (SST) and mixed layer depth (MLD), which provide a bulk measure of the total mixing strength.

    • Vertical coordinate and resolution dependence of the second moment turbulent closure models and their limitations

      2021, Ocean Modelling
      Citation Excerpt :

      LES models differ from SMTC models in that the large turbulent eddies are explicitly resolved in the LES down to the grid resolution and only the smaller eddies are modeled or parameterized, so it can better resolve the turbulence in the water column. The NCAR LES model has been shown to accurately reproduce observed upper ocean responses to different meteorological forcing and oceanic conditions (Kukulka et al., 2009; Liang et al., 2013, 2017; Fan et al., 2018, 2020). In the simulation at OSP, the observed temperature and salinity variations in the water column are also well represented by the LES model during the 15-day simulation period (Fig. 1d–f).

    View all citing articles on Scopus
    View full text