Submesoscale Eddies and Internal Waves: from the base of the mixed layer to the sea surface microlayer Jose da Silva
Submesoscale eddies (SMEs) are at an awkward size (< 10 – 30 km at mid-latitudes) which presents an observational barrier that delayed an appreciation of their abundance in the ocean. They are large for shipboard instrument detection, small and rapidly evolving for typical ship surveys, small for many satellite remote sensing footprints, and their currents are often difficult to distinguish from inertia-gravity waves in single-point time series or individual vertical profiles. Submesoscale processes/features are those that operate on horizontal scales less than the first baroclinic Rossby radius of deformation, around and above which the currents are more geostrophic. Their evolution and time scales are typically hours-days, at scales that can be realistically sampled by modern robotic systems (AUVs). What makes SMEs dynamically distinct from geostrophic mesoscale dynamics is a Rossby number and a Froude number that are not asymptotically small, nor large (i.e., approximately unity). SMEs can be viewed as a manifestation of turbulence in the regime of marginal control by planetary rotation and stable density stratification. Submesoscale features have been intensively studied lately (McWilliams, 2017), in part due to their important contribution to vertical fluxes of mass, tracers, and buoyancy. While SMEs may be cyclonic or anti-cyclonic, “Spirals” were accounted for as broadly distributed over the world's oceans, 10-25 km in size, with streaks of film slicks hundreds of meters wide, and overwhelmingly cyclonic (Munk et al., 2000). These vortices have been since associated with horizontal shear instability, modified by rotation, in regions where the shear is comparable with the Coriolis frequency. One question that remains outstanding in the oceanographic community is, ‘‘What are the relative roles of horizontal and vertical shear in generating SMEs in the global ocean?’’ While we know from analytical and numerical modeling solutions that SMEs can be generated by both (1) horizontal shear instability (barotropic instability) (Munk et al., 2000; Munk, 2001) and (2) vertical shear (baroclinic instability) (Eldevik and Dysthe, 1999, 2002; Gula et al., 2015), their relative importance may differ with geographic location and time of year. In this paper we review some of the current knowledge of SMEs generation mechanisms, and point out our lack of knowledge about the SME’s dynamics near the equator where f→0. A case study is presented in the tropical Atlantic off the Amazon region. Cyclonic “Spirals” are commonly associated with film slicks of linear structure in remote sensing images. As shown in Munk et al. (2000), this has to do with the visibility of the spiral arms, presumably the result of the alignment and concentration of surfactants. Because cyclonic shear on the cyclonic side of the ocean front that generates the eddie effectively reduces an elementary surface area of the sea, and the films “stick-up” to the water surface (the sea surface microlayer), the surfactant concentration therefore increases there, and film slicks can form. As it is well known, the density of molecules in monolayers changes with the surface area, and slicks spontaneously form when the concentration of insoluble surfactants exceeds some unknown threshold value. The slick becomes visible due to capillary wave-damping, which importantly acts to reduce the air–sea exchange of material, gas, heat and momentum (see Frew et al., 2007). On the contrary, for anti-cyclonic eddies, the same theory predicts that the elementary area expands significantly, and film slicks are usually absent. In this paper we show several examples of cyclonic “Spirals” in remote sensing satellite images and discuss sampling problems. Internal Solitary Waves (ISWs) have attracted a great deal of attention in recent years because they have been identified as playing a significant role in mixing shelf waters (e.g. Gregg and Klymak, 2014). This mixing is particularly effective for mode-2 ISWs because the location of these waves in the middle of the pycnocline plays an important role in eroding the barrier between the base of the surface mixed layer and the stratified deep layer below. An urgent problem in physical oceanography is therefore to account for the magnitude and distribution of ISW-driven mixing, including mode-2 ISWs. Several generation mechanisms of mode-2 ISWs have been identified and are illustrated with Synthetic Aperture Radar (SAR) case studies that portray evidence of those generation mechanisms. Some of the SAR images correspond to numerical simulations with the MITgcm in fully nonlinear and nonhydrostatic mode (in a 2D configuration) with realistic stratification, bathymetry and other environmental conditions. A particular reference case study is that of the Mascarene Ridge of the Indian Ocean (da Silva et al., 2015). We also document interaction of ISW trains with mesoscale eddies in the South China Sea, illustrated by remote sensing data.