Subduction by Submesoscales

V. M. Canuto,Y. Cheng,A. M. Howard

DOI: https://doi.org/10.1029/2018jc014142

2018-01-01

Journal of Geophysical Research Oceans

Abstract:Several studies have demonstrated the relevance of submesoscales (SM) to the production of phytoplankton and the ocean absorption of anthropogenic CO2. One variable that thus far has not been fully quantified is the SM-induced rates of subduction and obduction at the bottom of the mixed layer. In this work, we use an SM parameterization that had been previously assessed to estimate the rates of subduction-obduction using a coarse resolution stand-alone ocean code to compute the large-scale fields. To separate the contributions of baroclinic instabilities from those due to wind stress, we derive an analytic expression for the SM subduction-obduction rates. Since there are no numerical simulations and/or measurements to assess the results, we compare them to subduction rates due to mesoscales for which an eddy-permitting southern ocean state estimate is available. An interesting result is that in the Antarctic Circumpolar Current the maximum SM subduction rates are not significantly smaller than those due to mesoscales, but the SM subduction rates exhibit a topology opposite to that of mesoscales: obduction is equatorward and subduction is poleward. Sensitivity studies are presented to explore the dependence of the results on the depth of the SM vertical extent and the SM kinetic energy. On the basis of these results, it is suggested that the SM-induced subduction should not be neglected because topology and magnitude indicate that mesoscales and submesoscales may add to or subtract from one another. Plain Language Summary The ocean contains much more heat and carbon dioxide than the atmosphere, but in most places the transfer, called subduction, is slow between the large volume of the deep ocean, where almost all is stored and the much smaller volume near the surface that exchanges with the atmosphere. North of Antarctica is a region where ocean current circles the globe, known as the Antarctic Circumpolar Current (ACC). There subduction is faster and much is done by very large ocean eddies similar to 10-100km in size called mesoscales and somewhat smaller ones similar to 1-10km called submesoscales. Because it links the ocean reservoir with the atmosphere, subduction by ACC eddies plays an important role in climate. Estimates had previously been made of mesoscale, but not submesoscale, subduction. Using a physical model of submesoscales, taking into account their energy and how wind stress drives them, we have derived expressions for the submesoscale subduction in terms of the density and velocity obtained from a large-scale ocean simulation suitable for climate studies. Our result is that the rate of ACC submesoscale subduction is comparable to that of mesoscales but the pattern is roughly opposite, so it is important to include both.