Schematic of internal wave mixing processes in the open ocean that are considered as part of this CPT. Tides interact with topographic features to generate high-mode internal waves (e.g., at midocean ridges) and low-mode internal waves (e.g., at tall steep ridges such as the Hawaiian Ridge). Deep currents flowing over topography can generate lee waves (e.g., in the Southern Ocean). Storms cause inertial oscillations in the mixed layer, which can generate both low- and high-mode internal waves (e.g., beneath storm tracks). In the open ocean, these internal waves can scatter off of rough topography and potentially interact with mesoscale fronts and eddies until they ultimately dissipate through wave–wave interactions. Internal waves that reach the shelf and slope can scatter or amplify as they propagate toward shallower water.
Depth-averaged diffusivity κ from (a) the upper ocean (from MLD to 1000-m depth) and (b) the full water column, updated from Waterhouse et al. (2014). The background diffusivity map in (a) comes from the strain-based inferences of diffusivity from Argo floats, updated from Whalen et al. (2015) with observations included from 2006 to 2015. (c) Compiled observations of mixing measurements with blue and green squares and diamonds denoting microstructure measurements. Green represents full-depth profiles, while blue denotes microstructure profiles. Purple circles represent inferred diffusivity from a finescale parameterization using lowered acoustic Doppler current profiler (LADCP)/conductivity–temperature–depth (CTD) profiles [dark purple, Kunze et al. (2006); medium purple, Huussen et al. (2012)] and High Density Sounding System (HDSS) shipboard shear (light orange). Dark orange circles are diffusivities from density overturns in moored profiles.
(a) A snapshot of baroclinic velocity (m s−1) from a two-dimensional numerical simulation of internal tides forced by M2 (semidiurnal) tidal velocities over rough topography for parameters corresponding to the Brazil Basin (Nikurashin and Legg 2011). (b) Observational time series of internal wave breaking over tall steep topography; here, we see (top) northward velocity and (bottom) turbulent dissipation rate oscillate twice a day as the tide flows over Kaena Ridge, Hawaii (Klymak et al. 2008). (c) Global energy flux from the M2 tide into internal tides (log10 W m−2) estimated using (top) the topography resolved in the Shuttle Radar Topography Mission (SRTM) global bathymetry and elevation data at 30 arc s resolution with data voids filled (SRTM30_PLUS) bathymetry database and (bottom) a statistical representation of unresolved abyssal hill topography estimates (Melet et al. 2013b). (d) The vertical structure of dissipation from Brazil Basin observations (thick solid curve) and the Polzin (2009) [Eq. (4)] parameterization of near-field internal tide dissipation (thin solid curve), as well as associated observed values of stratification (N2) and diapycnal diffusivity (Kρ). (e) The impact of the Polzin parameterization in the GFDL CM2G coupled climate model: (top) the Indo-Pacific meridional overturning streamfunction (Sv; 1 Sv = 106 m3 s−1; averaged over the final 100 years of a 1000-yr simulation) using the Polzin (2009) parameterization and (bottom) the differences in Indo-Pacific meridional overturning streamfunction (Sv) between the simulations with the Polzin (2009) parameterization and the St. Laurent et al. (2002) parameterization as implemented by Simmons et al. (2004b) (from Melet et al. 2013a).
Far-field internal tide: (a) SSH amplitude (mm) of global mode-1 M2 internal tides from multisatellite altimetry (Zhao et al. 2016). The light blue color indicates regions of high mesoscale activity, which make extraction of internal tides from altimetry difficult. Modeled semidiurnal tidal fluxes and comparison to observations: (b) HYCOM-modeled semidiurnal internal tide barotropic-to-baroclinic conversion rates (background color) and vertically integrated energy flux vectors (black arrows, plotted every 768th grid point for clarity) and (c) depth-integrated semidiurnal mode-1 energy fluxes in HYCOM (red arrows) and high-resolution mooring observations to the north of Hawaii (blue arrows) (Ansong et al. 2017). Impact on thermosteric sea level of using different spatial distributions of remote internal tide energy dissipation in GFDL ESM2G climate model: (d) thermosteric sea level (m) in a reference simulation using a constant background diapycnal diffusivity for remote internal tide dissipation. Anomalies (m) of thermosteric sea level from the reference case in (d) for simulations where (e) all internal tide energy is dissipated locally, over the generation site and (f) 20% of the internal tide energy is dissipated locally and 80% is dissipated uniformly over the ocean basins with a vertical profile proportional to buoyancy squared N2 (Melet et al. 2016).
Internal lee waves: (a) observations from DIMES showing (left) turbulent dissipation rates (logarithmic scales from 10−10 to 10−7 W kg−1) for the Phoenix Ridge (circles in right inset) and (right) average height above bottom profiles of turbulent kinetic energy dissipation [see details in St. Laurent et al. (2012)]. (b) Power conversion into lee waves [Nikurashin and Ferrari (2011) used in Melet et al. (2014)],(c) consequences of parameterized lee-wave mixing on the global ocean meridional overturning circulation [Sv; averaged over the final 100 years of 1000-yr simulations, from Melet et al. (2014)], and (d) global map of depth-integrated dissipation due to parameterized topographic wave drag inserted inline into global 1/25° HYCOM simulation, from Trossman et al. (2016).
Near-inertial internal waves: (a) observational example from Alford et al. (2012) showing a (top) 2-yr record of wind work and (bottom) near-inertial kinetic energy in the northeastern Pacific. (b) One estimate of global power input (shading) and low-mode NIW energy fluxes (arrows; Simmons and Alford 2012). (c) Impact of near-inertial waves on annual-mean precipitation in ocean climate models: (top) the mean precipitation (mm day−1) from an experiment where the NI flux is set to 0.34 TW and (bottom) the same experiment, but with a doubling of the NI flux to 0.68 TW. The total tropical precipitation in the two experiments differs by less than 1%. An increase in near-inertial energy flux within observational uncertainties ameliorates the double ITCZs in the Atlantic and Pacific Oceans and creates the South Pacific convergence zone, three significant improvements for climate simulations of tropical precipitation.
