The ocean mixed layer regulates all exchanges of heat, freshwater, and trace gases between the ocean interior and the atmosphere, thus playing a key role in the Earth's climate. The mixed layer variability is governed jointly by air-sea fluxes and ocean submesoscale processes. Air-sea fluxes are vertically one-dimensional by driving turbulent motions of the mixed layer locally. By contrast, ocean submesoscales are inherently two- to three-dimensional by gaining energy from lateral buoyancy gradient fields in the mixed layer to restratify the upper ocean. Consequently, mixed layer variability is routinely quantified based on air-sea fluxes provided by atmospheric reanalyses, which possess fine temporal resolution but coarse horizontal grid spacing, supplemented by fluxes estimated from parametrization of submesoscale restratification. However, ocean submesoscales are also featured by sharp ocean fronts, which necessarily imprint on vertical air-sea fluxes that are unresolved by any atmospheric reanalysis product. How such additional one-dimensional processes induced by ocean submesoscales influence the mixed layer variability remains elusive. 

In this work, we combine in-situ observations with a high-resolution model to isolate one-dimensional air-sea fluxes at the ocean submesoscale. High-frequency measurements of upper-ocean and lower-atmosphere properties collected by a Saildrone in the Southern Ocean are analyzed, which reveal intense air-sea flux variability. We then employ a coupled numerical model to disentangle air-sea fluxes due to ocean submesoscales from those contributed by high-frequency atmospheric forcing. Virtual Lagrangian and Eulerian Saildrone tracks resembling those in the real ocean are set up, with along-track oceanic variables filtered for scale separation before quantification of the resulting vertical fluxes. Air-sea fluxes across simulated submesoscale fronts show variations comparable with Saildrone-derived results. We further quantify the potential impacts of ocean submesoscales on one-dimensional air-sea fluxes and hence on mixed layer variability through a suite of process-based mixed layer models in idealized configurations. Our findings suggest that the uncertainties resulting from air-sea fluxes at ocean submesoscales are comparable to those due to diverging performance of different mixed layer parametrization schemes.
 

All Are Welcome!