On the topic of oceanic variability near the Coriolis frequency; generation mechanisms, observations, and implications for interior mixing
MetadataShow full item record
Internal waves with frequency near the Coriolis frequency, the frequency of oscillations due to the Coriolis acceleration (commonly denoted by the symbol f), are ubiquitous throughout the world’s oceans. However, observational constraints on their global energetics and impact on subsurface mixing remain sparse. This study investigates near-inertial (inertial and Coriolis frequency are used interchangeably) waves in the Southern Ocean using measurements of water velocity made by Electromagnetic Autonomous Profiling Explorers (EM-APEX). Initial observations from the eastern Pacific showed that coherent near-inertial waves were episodic and enhanced at mid-depth between 500 and 1000 m. The observed waves showed depth-integrated horizontal kinetic energy between 1 and 7 kJ m^−2, with an average of 1.6 kJ m^−2, and a typical group velocity of 40 m d^−1. These observations imply an average energy flux of 3 mW m^−2 at the mixed layer base decreasing to approximately 25% of that value at 1500 m. Simulations of near-inertial surface currents forced with reanalysis winds along each float track agree with observed surface currents from EM-APEX, provided that mixed layer depth is restricted to the layer of weakest observable stratification, interpreted as the maximum depth which remains mixed over an inertial period. Simulations using the Price-Weller-Pinkel model, which permits time varying stratification, provides a better match to the observations; emphasizing the importance of near-surface stratification in amplifying wind power input. These simulations indicate an average wind power input of 3mW m^−2 in the eastern Pacific sector of the Southern Ocean. The thickness of the active mixing-layer, the turbulent layer in contact with wind stress, is needed to accurately estimate wind power input. Vertical shear, Langmuir cells, and buoyant convection were investigated as possible mechanisms for maintaining turbulent mixing within the mixing-layer. Over 90% of the observed variance of the mixing-layer thickness is explained by either shear-driven entrainment, which is simulated using the Price-Weller- Pinkel model, or by a parameterization of downwelling plumes due to Langmuir cell convergence. In general, surface buoyancy fluxes are too weak to drive mixed-layer turbulence. The density profiles shown here indicate that a fine-density threshold between the finest possible criterion of 0.002 kg m^−3 and the previous finest criterion of 0.01 kg m^−3 is needed to better diagnose the surface mixed-layer thickness. The results of wind driven simulations suggest that this criterion is 0.005 kg m^−3. Comparison of National Oceanographic Data Center (NODC) climatological mixed-layer thickness to those determined using the 0.005 kg m^−3 density threshold suggests a multiplicative seasonally varying correction of 1.5 to 3.5 should be applied to estimates of wind work made using the NODC climatological mixed-layer thickness in the Southern Ocean. The vertical structure of the inertial-band of the internal wave field was observed by EM-APEX in the eastern Pacific, Scotia Sea, and western Atlantic sectors of the Southern Ocean. Downward propagating internal wave variance is shown to vary with the seasonal cycle in both wind stress and surface mixed-layer depth. Mixed-layer depth was found to inhibit the formation of near-inertial waves in the Pacific during the austral winter. Inertial-band internal-wave energy in the Scotia Sea is dominated by upward propagating waves, likely generated by interaction of the strong currents found in this region with rough sea-floor topography. Inertial-band internal wave energy in the Atlantic was observed to vary with the seasonal cycle in near-inertial wind-stress. Diapycnal diffusivity in the 300 to 1500 m depth range, estimated using observed vertical shear in the Gregg et al. (2003) parametrization, is 1.96±0.36×10^−5 m^2 s^−1, 1.08±0.08×10^−4 m^2 s^−1, and 7.96±1.17×10−^5 m^2 s^−1 in the eastern Pacific, Scotia Sea, and western Atlantic sectors of the Southern Ocean. In the eastern Pacific and Scotia Sea these values similar to direct observations of diapycnal diffusivity inferred from the vertical spread of tracer in the Diapycnal and Isopycnal Mixing Experiment in the Southern Ocean.
- Oceanography