Modification of the deep salinity-maximum in the Southern Ocean by circulation in the Antarctic Circumpolar Current and the Weddell Gyre

Abstract

The evolution of the deep salinity-maximum associated with the Lower Circumpolar Deep Water (LCDW) is assessed using a set of 37 hydrographic sections collected over a 20-year period in the Southern Ocean as part of the WOCE/CLIVAR programme. A circumpolar decrease in the value of the salinity-maximum is observed eastwards from the North Atlantic Deep Water (NADW) in the Atlantic sector of the Southern Ocean through the Indian and Pacific sectors to Drake Passage. Isopycnal mixing processes are limited by circumpolar fronts, and in the Atlantic sector, this acts to limit the direct poleward propagation of the salinity signal. Limited entrainment occurs into the Weddell Gyre, with LCDW entering primarily through the eddy-dominated eastern limb. A vertical mixing coefficient, κ_V of (2.86 ± 1.06) × 10⁻⁴ m² s⁻¹ and an isopycnal mixing coefficient, κ_I of (8.97 ± 1.67) × 10² m² s⁻¹ are calculated for the eastern Indian and Pacific sectors of the Antarctic Circumpolar Current (ACC). A κ_V of (2.39 ± 2.83) × 10⁻⁵ m² s⁻¹, an order of magnitude smaller, and a κ_I of (2.47 ± 0.63) × 10² m² s⁻¹, three times smaller, are calculated for the southern and eastern Weddell Gyre reflecting a more turbulent regime in the ACC and a less turbulent regime in the Weddell Gyre. In agreement with other studies, we conclude that the ACC acts as a barrier to direct meridional transport and mixing in the Atlantic sector evidenced by the eastward propagation of the deep salinity-maximum signal, insulating the Weddell Gyre from short-term changes in NADW characteristics.

Appendix

The LCDW in the ACC exists at a fairly constant neutral density layer of 28.04–28.08 kg m⁻³ throughout its circumpolar journey as earlier detailed. However, when LCDW is entrained into the Weddell Gyre as WDW the salinity maximum associated with the LCDW and WDW shallows dramatically and moves to deeper density levels. A simple one-dimensional model is used to illustrate how the transport of the LCDW/WDW and its interaction with overlying and underlying water masses act to modify the density profile of the water column and help explain why, outside of the ACC regime, the tracking of the salinity maximum remains an acceptable technique for assessing water mass changes in the LCDW/WDW throughout its Southern Ocean journey.

Imagine that a column of water is being carried along by the ACC and as it does so it is slowly being drawn southwards towards Antarctica, whilst mixing with the overlying and underlying water masses. As a result of the poleward movement, the characteristics of these boundary water masses will change. The northward Ekman transport—out of the Weddell Gyre—results in near-surface waters occupying a shallower depth range south of the ACC. This includes the absence of AAIW which only subducts north of the Polar Front within the ACC (Sallée et al. 2010).

Dense water which forms over the Antarctic continental shelf mixes with deeper waters as it descends the continental slope to the deep ocean. This leads to the formation of topographically constrained WSDW (−0.7 °C < θ < 0 °C) and WSBW (θ < −0.7 °C) (Fahrbach et al. 1995; Gordon et al. 2001), the former of which is present at depths as shallow as 1100 m compared to the salinity maximum of the LCDW in the ACC which can lie at depths of up to 2500 m. The presence of WSDW and WSBW acts to force the LCDW/WDW upwards from beneath, resulting in the shoaling of isopycnals and the rise of LCDW/WDW in the water column.

The standard framework for considering the ACC includes transport being dominated by streamlines of eastward flow aligned with the position of circumpolar fronts and a strong dependence on the sloping isopycnal surfaces associated with these fronts for tracking conservative tracers. This is against the background of a northward Ekman transport near the surface and an opposing southward flux below by the mesoscale eddy field. However, whilst this approach is suitable for the ACC, the Weddell Gyre represents a different dynamical regime characterised by the recirculation of gyre water masses, water mass transformation, and interaction with the ACC along the northern and eastern boundaries of the Gyre. This appendix explores a simple example to explain the observed changes in the properties of the LCDW as it circulates around Antarctica and is entrained into the Weddell Gyre.

To provide realistic initial conditions, the hydrographic profile from station 25, cast 1 of the 2002 A12 cruise track undertaken during FS Polarstern’s expedition ANT-XX/2, was used as the starting hydrographic profile. The station was located at 49° S, 2.8° E placing it south of the Sub-Antarctic Front but north of the Polar Front.

At the surface boundary, the sea surface temperature decreases by 0.005 K per time step to represent the decreasing surface temperature moving towards Antarctica, whilst the sea surface salinity is constant to represent the general stability and relative freshness of the Antarctic Surface Water. The potential temperature at the bottom boundary of 4000 m decreases by 0.0018 K per time step and the salinity decreases by 0.00005 per time step, mimicking the transition into the WSBW/WSDW regime at depth.

These linear changes were derived from the total meridional temperature and salinity gradient over the A12 section as a representative trend for surface and bottom waters. As both surface and bottom waters are formed locally in the Southern Ocean, these boundary conditions provide a reasonable first-order approximation to the meridional gradient in oceanographic conditions, although clearly this is an idealised approach.

In the vertical, the grid was set at 101 depth levels, spaced every 40 m from 0 to 4000 m. The initial A12 profile was used for the model down to a depth of 4000 m (of a total cast depth of 4090 m), whilst the shallowest data (at 9 dbar) was used as the surface value. Except the surface values, there were no missing data.

The one-dimensional model is based around simple equations for the modification of potential temperature and salinity:

$$ {\theta}_t={\theta}_{t-1}+\frac{d\theta}{d t}\Delta t,\mathrm{where}\frac{d\theta}{d t}={\kappa}_z\frac{\left({\theta}_{i-1}-2{\theta}_i+{\theta}_{i+1}\right)}{{\left(\Delta z\right)}^2}+ w\frac{\theta_i-{\theta}_{i-1}}{\Delta z} $$

(6)

$$ {S}_t={S}_{t-1}+\frac{dS}{dt}\Delta t,\mathrm{where}\frac{dS}{dt}={\kappa}_z\frac{\left({S}_{i-1}-2{S}_i+{S}_{i+1}\right)}{{\left(\Delta z\right)}^2}+ w\frac{S_i-{S}_{i-1}}{\Delta z} $$

(7)

where θ is potential temperature, S is practical salinity, t is time, i is the vertical counter, positive downwards, z is the vertical co-ordinate, positive downwards, κ _z is the diapycnal diffusivity coefficient, w is the uplift rate, Δt is the time step and Δz the vertical grid spacing.

A 120-year transport duration was derived from an estimated 40,000 km track from the station location in the ACC eastward around Antarctica, subsequently returning to a more southerly position on the Greenwich Meridian, and finally entrainment into the Weddell Gyre at an assumed 1 cm s⁻¹, rounded down to the nearest whole year. The number of time steps was set at 365 and the time step size set to 120 days for a simple scaling of the duration. This is consistent with the distances and times discussed in the main body of this text, whereby LCDW is shown to recirculate in the ACC before entering the polar gyres.

The vertical thermal diffusivity coefficient, κ, was set to 5 × 10⁻⁵ m² s⁻¹ which falls within the range of many studies of vertical mixing in the ocean (e.g. Munk 1966; Naveira Garabato et al. 2004b; Zika et al. 2009). In addition, to represent the rise of LCDW/WDW in the water column as a result of Ekman suction, a rate of water column uplift of 20 m year⁻¹ was applied.

Both potential density and neutral density were calculated based upon the new potential temperature, practical salinity and pressure at each time step.

The initial and final profiles of the 1D model shows an overall decrease in temperature at all depths, a marked increase in salinity in the upper water column but a slight decrease in deeper waters resulting in the upward shift of the salinity maximum and an overall increase in the density of the profile.

The salinity and neutral density time-series section shown in Fig. 11 illustrates the rapid upward shift of the salinity maximum as the profile moves into the modelled regime of the Weddell Gyre. Salinity stabilises at depth reflecting the shallow salinity gradient in the WSDW and WSBW whilst a fresh surface layer is retained. The potential temperature (not shown) decreases steadily at all depths, with the temperature maximum remaining at about 100 m depth.

Fig. 11

Time-series contour plot for the top 2500 m of the water column showing the upward migration of the LCDW salinity core (filled-colour contours) and the shoaling of neutral density (γ_n) surfaces towards the surface (black contour lines)

Neutral density increases at depth resulting in a decreasingly stratified deep water column, whilst the surface boundary constraint results in a strongly stratified upper water column. The key result is that density increases at depth such that the salinity maximum initially uplifts at a similar rate to the neutral surfaces until about year 70 when the density at the salinity maximum increases whilst the depth of the salinity maximum stabilises at about 400 m depth after 80 years. The intervening period represents the entrainment of the water mass into the Weddell Gyre proper.

The framework for examining the ACC described earlier is well established; however, this framework does not extend to the transition to the hydrographic regime of the Weddell Gyre (Schröder and Fahrbach 1999). This 1D model shows an initial decrease in the depth of LCDW from ~2500 to ~400 m, paralleling the north-south distribution of temperature, salinity and density evident on sections crossing the ACC into the Weddell Gyre.

Whilst simple, this model—when viewed as a representation of the mean-state of more complex Southern Ocean dynamics over long time scales—serves to illustrate the broad-scale effect of crossing from one regime to another upon the core of the LCDW. The core of a traditional tracer can be mixed and advected over multi-year time scales to produce a result which is consistent with observations. Whilst within the core of the ACC the salinity maximum remains within a constant density layer, this is not the case as the regime shifts from a circumpolar flow to the Weddell Gyre circulation. The normal framework for examining the ACC on streamlines and isopycnal surfaces does not apply beyond the Southern Boundary of the ACC, where complicated processes of advection, mesoscale eddies and frontal meandering together modify the water column. Recognition must be given to the effects of the contrasting hydrographic regimes of the ACC and Weddell Gyre, which demonstrate a situation in which the signal of the deep salinity maximum crosses isopycnal surfaces.