Ocean overturning circulation requires a continuous thermodynamic transformation of the buoyancy of seawater. The steeply sloping isopycnals of the Southern Ocean provide a pathway for Circumpolar Deep Water to upwell from mid depth without strong diapycnal mixing 1-3 , where it is transformed directly by surface fluxes of heat and freshwater and splits into an upper and lower branch 4-6 . While brine rejection from sea ice is thought to contribute to the lower branch 7 , the role of sea ice in the upper branch is less well understood, partly due to a paucity of observations of sea-ice thickness and transport 8,9 . Here we quantify the sea-ice freshwater flux using the Southern Ocean State Estimate, a state-of-the-art data assimilation that incorporates millions of ocean and ice observations. We then use the water-mass transformation framework 10 to compare the relative roles of atmospheric, sea-ice, and glacial freshwater fluxes, heat fluxes, and upper-ocean mixing in transforming buoyancy within the upper branch. We find that sea ice is a dominant term, with di erential brine rejection and ice melt transforming upwelled Circumpolar Deep Water at a rate of ∼22 × 10 6 m 3 s −1 . These results imply a prominent role for Antarctic sea ice in the upper branch and suggest that residual overturning and wind-driven sea-ice transport are tightly coupled.The Southern Ocean State Estimate (SOSE) is an ice/ocean data assimilation produced for the time period January 2005 through December 2010. (See Methods and Supplementary Information for SOSE details and validation.) The bulk freshwater fluxes at the ocean surface south of 50 • S, as estimated by SOSE, are summarized in Fig. 1a. When sea ice forms, nearly all of the salt remains behind in the underlying seawater (a process called brine rejection); when the ice melts, liquid freshwater is returned to the ocean. Direct open-ocean precipitation minus evaporation (0.28 fwSv) and glacial ice melt (0.05 fwSv) are both smaller than net sea-ice melt (0.50 fwSv) and brine rejection (0.36 fwSv). (Bulk freshwater volume fluxes are given in units of freshwater sverdrups 11 , 1 fwSv = 10 6 m 3 freshwater s −1 3.15 × 10 4 Gt freshwater per year.) Melt exceeds brine rejection because sea ice incorporates snowfall at a rate of 0.14 fwSv. Moreover, wind-driven sea-ice transport creates a freshwater conveyor belt from the Antarctic coast to the open ocean 12 , leading to sharp gradients in freshwater flux. The spatial structure of the sea-ice redistribution is assessed in Fig. 1b by comparing the annual mean freshwater flux leaving the atmosphere, land and glaciers (left panel) with that entering the ocean (right panel); the difference is due to sea-ice freshwater redistribution (middle panel, vectors show the ice thickness transport). From the atmosphere, widespread precipitation over the Southern Ocean leads to a broadly distributed downward freshwater flux with a characteristic magnitude of 0.5 m yr −1 , and glacial ice melt provides a stronger freshwater source near the Antarctic coas...

Meridional cross sections of effective diffusivity in the Southern Ocean are presented and discussed. The effective diffusivity, K eff , characterizes the rate at which mesoscale eddies stir properties on interior isopycnal surfaces and laterally at the sea surface. The distributions are obtained by monitoring the rate at which eddies stir an idealized tracer whose initial distribution varies monotonically across the Antarctic Circumpolar Current (ACC). In the absence of observed maps of eddying currents in the interior ocean, the advecting velocity field is taken from an eddy-permitting state estimate of the Southern Ocean (SOSE). A threedimensional advection-diffusion equation is solved and the diffusivity diagnosed by applying the Nakamura technique on both horizontal and isopycnal surfaces. The resulting meridional sections of K eff reveal intensified isopycnal eddy stirring (reaching values of ;2000 m 2 s 21 ) in a layer deep beneath the ACC but rising toward the surface on the equatorward flank. Lower effective diffusivity values (;500 m 2 s 21 ) are found near the surface where the mean flow of the ACC is strongest. It is argued that K eff is enhanced in the vicinity of the steering level of baroclinic waves, which is deep along the axis of the ACC but shallows on the equatorial flank. Values of K eff are also found to be spatially correlated with gradients of potential vorticity on isopycnal surfaces and are large where those gradients are weak and vice versa, as expected from simple dynamical arguments. Finally, implications of the spatial distributions of K eff for the dynamics of the ACC and its overturning circulation are discussed.

The Southern Ocean, with its large surface area and vigorous overturning circulation, is potentially a substantial sink of anthropogenic CO(2) (refs 1-4). Despite its importance, the mechanism and pathways of anthropogenic CO(2) uptake and transport are poorly understood. Regulation of the Southern Ocean carbon sink by the wind-driven Ekman flow, mesoscale eddies and their interaction is under debate. Here we use a high-resolution ocean circulation and carbon cycle model to address the mechanisms controlling the Southern Ocean sink of anthropogenic CO(2). The focus of our study is on the intra-annual variability in anthropogenic CO(2) over a two-year time period. We show that the pattern of carbon uptake is correlated with the oceanic vertical exchange. Zonally integrated carbon uptake peaks at the Antarctic polar front. The carbon is then advected away from the uptake regions by the circulation of the Southern Ocean, which is controlled by the interplay among Ekman flow, ocean eddies and subduction of water masses. Although lateral carbon fluxes are locally dominated by the imprint of mesoscale eddies, the Ekman transport is the primary mechanism for the zonally integrated, cross-frontal transport of anthropogenic CO(2). Intra-annual variability of the cross-frontal transport is dominated by the Ekman flow with little compensation from eddies. A budget analysis in the density coordinate highlights the importance of wind-driven transport across the polar front and subduction at the subtropical front. Our results suggest intimate connections between oceanic carbon uptake and climate variability through the temporal variability of Ekman transport.

No abstract