Bottom Water Formation and Overturning in the Southern Ocean










Movie loading...














Video of an eddy-resolving simulation conducted using the MITgcm, showing the potential temperature, which for simplicity is linearly related to the density. We impose an idealised patterns of surface wind forcing and heat fluxes: strong westerlies and surface heating in the north, and weaker easterlies and surface cooling in the south. We parametrise bottom water formation by providing a source of cold/dense water on the shelf. 



The formation of Antarctic Bottom Water (AABW) occurs through a combination of ice-shelf melt, brine rejection during sea-ice formation, and mixing with Circumpolar Deep Water. Export of AABW constitutes a key component of the ocean's meridional overturning circulation (MOC). The AABW sinks down over the Antarctic continental slope, then spreads throughout the deep ocean. This process ventilates carbon from and transports oxygen to the abyssal ocean. Stagnation of this circulation is associated with anoxic events in past climates, and exchanges of CO2 between the deep MOC and the atmosphere may control climate change on glacial timescales. Shifts in the mid-latitude westerly winds have been hypothesised to change the strength and extent of the deep overturning circulation, and thus modify atmospheric pCO2 levels, between glacial and inter-glacial periods (e.g. Toggweiler, 2009).

Modern theories for the meridional transport in the Southern Ocean invoke a near-balance between a wind-driven Ekman overturning circulation and a counter-acting "eddy" circulation, sustained by the bolus velocities of eddies generated by baroclinic instability. The "residual" of these circulations describes the net latitudinal and radial mass transport. This circulation is almost entirely adiabatic in the interior, but it is closed by strong diapycnal mixing in the surface and bottom boundary layers. We are investigating the sensitivity of the MOC in the Antarctic Circumpolar Current (ACC) and over the Antarctic continental shelf to changes in surface forcing. Our approach combines develop idealised, eddy-resolving simulations using the MIT general circulation model with an analytical residual-mean theory.



Overturning streamfunction calculated for the simulation shown above, calculated using the mean meridional transports within isopycnal layers (e.g. Doos and Webb, 1994).  A pair of overturning cells is visible in the meridional/vertical plane: in these coordinates the upper cell overturns clockwise,  and the lower cell overturns anti-clockwise. This circulation corresponds qualitatively to the upwelling of North Atlantic Deep Water (NADW) and downwelling of Antarctic Intermediate Water (AAIW) and AABW observed in the Southern Ocean.




Sensitivity of the upper and lower overturning cells to changes in the mid-latitude westerlies (top) and polar easterlies (bottom). For comparison we also plot ΨACC and ΨASF, the theoretical mean overturning circulation due to wind-driven Ekman pumping in the ACC and ASF respectively (e.g. Marshall and Radko, 2003). The upper overturning cell exhibits a modest sensitivity to changes in the mid-latitude westerlies, in approximate agreement with previous work (e.g. Abernathey et al., 2011). The deep overturning cell is around four times as sensitive to changes in the polar easterlies. Thus changes in the polar easterlies may modify the deep overturning circulation substantially over glacial time-scales, and much more so than changes in the mid-latitude westerlies.



References:

  1. •Sensitivity of the ocean’s deep overturning circulation to easterly Antarctic winds, A.L. Stewart and A.F. Thompson, Geophysical Research Letters (2012), 39, L18604.

  2. •Shifting Westerlies, J.R. Toggweiler, Science (2009), 323, 1434-1435.

  3. •The Deacon cell and the other meridional cells of the Southern Ocean, K. Doos and D.J. Webb, Journal of Physical Oceanography (1994), 24, 429-442.

  4. •Residual-mean solutions for the Antarctic Circumpolar Current and its associated overturning circulation, J. Marshall and T. Radko, Journal of Physical Oceanography (2003), 33, 2341-2354.

  5. •The dependence of Southern Ocean meridional overturning on wind stress, R. Abernathey, J. Marshall and D. Ferreira, Journal of Physical Oceanography (2011), 41, 2261-2278.