WATER-MASS VARIABILITY IN THE SOUTHERN OCEAN:
Collaborators: Agus Santoso, Alex Sen Gupta, Chris Aiken, Steve Rintoul and Tony Hirst
SEPARATING NATURAL FLUCTUATIONS FROM LONG-TERM CHANGE
The goal of this long-term project was to examine the magnitude and dynamics of natural variability in the Southern Ocean
and the overlying atmosphere. We set out to analyse how each of the major Southern Ocean
water-masses varied in both space and time. We also wanted to discover what physical mechanisms
were at play: such as the relative importance of air-sea heat and freshwater fluxes,
ocean circulation, and sea-ice exchange.
Of particular interest was the nature of variability of sea surface temperature, interior
oceanic water masses, and variability associated with the Southern Annular Mode (SAM), a climatic fluctuation in the
wind field over the Southern Ocean. Extended integrations of global climate models were analysed to
determine the nature of seasonal, interannual, decadal and centennial variability in upper-ocean circulation,
water-mass properties, Antarctic sea-ice, and atmospheric circulation. Wherever possible, observational data
were analysed and compared to the model-simulated variability.
The detection and attribution of climate change depends critically on knowledge of natural modes of variability in the
ocean-atmosphere system. This is because any signal of apparent change needs to be compared, in both magnitude
and time-scale, to that due to internally generated climate oscillations. This project aimed to improve our ability
to detect long-term Southern Hemisphere climate change by obtaining a better characterisation of
natural ocean-atmosphere variability in the region. Our knowledge of natural climate oscillations in the tropics
and in the Northern Hemisphere is more advanced than for the Southern Oceans, due to better data coverage
and a greater effort in these geographic areas. One of our goals was to quantify modes of climate variability in the
Southern Hemisphere and to estimate how these modes impact on water properties in the Southern Ocean.
The model experiments assessed comprised multi-millennia natural variability simulations run
with constant atmospheric CO2. The duration of the model runs provides a longer time-scale
estimate of natural climate variability than that available from observations alone. Systematic dynamical
processes and feedback loops were explored using statistical analyses, including empirical orthogonal functions
and wavelets, to extract relationships between the different climate parameters. In addition, momentum and
property budgets were employed to elucidate the dominant processes controlling the frequency and
magnitude of ocean-atmosphere variations. Variables analysed included atmospheric conditions
(air temperature, cloud cover, precipitation, winds, sea-level pressure), sea-ice extent, air-sea fluxes,
ocean circulation and hydrography.
The main results from this work relate to a comprehensive assessment of the natural variability of
water masses in the Southern Ocean [Rintoul and England, 2002; Santoso and England, 2004,
Santoso, Hirst and England, 2006; Santoso and England, 2006]. The water masses analysed
include Antarctic Bottom Water (see Fig. 1 below), which forms under sea ice off the Antarctic coast,
sinks to the ocean’s abyss, and then flows northward into each of the major ocean basins.
Another focus was Antarctic Intermediate Water and Subantarctic Mode Water (see also Fig. 1),
both of which contribute to a substantial component of the ocean’s uptake of carbon dioxide.
Quantifying the natural variability of Southern Ocean water masses, including their properties and
overturning rates, is vital for detecting anthropogenic climate change.
Figure 1. Schematic depth-latitude diagram showing the major circulation and water masses
of the Southern Ocean. The following water masses are highlighted: (1) Antarctic Bottom Water flowing along the
abyssal ocean, (2) Circumpolar Deep Water upwelling into the Antarctic Divergence Zone, (3) Antarctic Intermediate
Water in the temperature range 4-6°C, and (4) Subantarctic Mode Water in the upper ocean north of the
Subantarctic Front (SAF).
Antarctic Bottom Water variations were found to be controlled by Antarctic sea-ice fluctuations,
with a dominantly decadal to multi-decadal time-scale (see Figure 2). The variability mechanism for
Subantarctic Mode Water, in contrast, was discovered to be linked to wind-driven northward flow of
cool Subantarctic Water, with the rate of mode water formation controlled by interannual variability
in the circumpolar westerly winds (Rintoul and England, 2002; see Figure 3).
Previously, SAMW variability was thought to be controlled mainly by air-sea heat fluxes. Instead, we found that
wind-driven northward Ekman flow of colder Subantarctic Water competes with southward flowing warmer saltier
water from western boundary current extensions, with variability resulting in upper ocean water-mass characteristics
and mixed layer depth. Under this mechanism, interannual variability in the circumpolar westerly winds drives variability in Ekman fluxes,
and therefore variability in the rate of mode water formation near 40°S.
Figure 2. Schematic diagram showing the formation mechanism
for Antarctic Bottom Water. Cold dry air blows off the Antarctic continent,
cooling the ocean to sub-freezing temperatures. When seawater freezes to form
sea-ice, salt is rejected so that the surface waters become highly saline. This
dense salty water sinks over the Antarctic continental shelf and eventually
becomes Antarctic Bottom Water, which flows into the abyssal Atlantic, Indian, and
Figure 3. Schematic diagram showing cross-front Ekman transport as a result of
westerly winds over the subpolar Southern Ocean. Cold fresh Antarctic Surface Water (ASW) is advected in
the surface Ekman layer across the Subantarctic Front (SAF). Ocean-atmosphere heat loss during winter
also acts to cool Subantarctic Mode Water (SAMW). However, the dominant driver
of SAMW T-S variations was found to be due to variability in the circumpolar westerly winds
and the associated northward Ekman transport [Rintoul and England, 2002].
These analyses were extended to include an assessment of Antarctic
Intermediate Water and Circumpolar Deep Water variability. Antarctic
Intermediate Water variations were found to be mainly linked to fluctuations in air-sea and ice-sea fluxes, whereas
Circumpolar Deep Water variability was found to be sourced primarily in North Atlantic Deep Water (NADW)
variations. For full details of these analyses and results, refer to the series of papers appearing in the
Journal of Physical Oceanography during 2002-2006 (see reference list at bottom of page).
The impact of the Southern Annular Mode:
Another focal point of this project has been documenting the impact of the Southern Annular Mode
on natural oceanic variability [Sen Gupta and England, 2006].
The Southern Annular Mode is the leading mode of climate variability over
the Southern Ocean, manifesting as a circumpolar pressure oscillation between
Antarctica and southern midlatitudes. The Southern Annular Mode thus
controls the latitude and strength of the Southern Hemisphere subpolar westerly winds.
These winds, often termed the ‘Roaring Forties’, play a pivotal role in controlling ocean
circulation and water-mass formation [Oke and England, 2004].
In this project we analysed a natural variability model experiment run over 200 years
as well as measurements dating back to the 1950s to ellucidate the way the Southern Annular Mode impacts
upon the Southern Ocean and regional climate.
Vacillations in the position and strength of the
circumpolar winds via the Southern Annular Mode result in a dynamic and thermodynamic
forcing of the ocean, summarized below in Fig. 4 [taken from Sen Gupta and England, 2006].
Both meridional and zonal components of ocean circulation are modified through
Ekman transport which in turn leads to anomalous surface convergences and
divergences that strongly affect the meridional overturning circulation, and
potentially the pathways of intermediate water ventilation.
Figure 4. Schematic representation of the climate system response to a positive phase
of the Southern Annular Mode. Warm and cold anomalies are denoted by red and blue regions respectively.
Arrow heads / tails denote flow out of / into the page respectively. The corresponding diagram of circulation,
properties and fluxes for the negative phase of the Southern Annular Mode exhibits the same patterns as displayed
here, only with reversed directions of circulation and the opposite sign for property anomalies and fluxes
(from Sen Gupta and England, 2006).
Summary of main findings to date:
Southern Ocean water mass variability conclusions can be summarised as follows:
Subantarctic Mode Water (SAMW) variability was found to be controlled by interannual variations
in wind-driven northward Ekman transport [Rintoul and England, 2002]
Antarctic Intermediate Water (AAIW) variations were found to be driven by air-sea heat
fluxes and ice-ocean salt fluxes on a 3-5 year time-scale [Santoso and England, 2004]
Circumpolar Deep Water (CDW) variations in T-S were discovered to be sourced from changes in
transport/properties of NADW, dominated by time-scales centennial and beyond [Santoso, Hirst and England, 2006]
Antarctic Bottom Water (AABW) variability of the order of 3-5 Sv was found to be dominantly
multi-decadal and linked to Antarctic sea-ice formation rates [Santoso and England, 2006]
Findings from the Southern Annular Mode aspects of the project to date include: The Southern Annular Mode forces an organized circumpolar response in ocean circulation, SST, mixed layer depth and sea-ice,
The sea surface temperature response is significant due to a conspiring of ocean circulation effects and air-sea heat fluxes,
There is a link between the Southern Annular Mode and rainfall in Tasmania, parts of Victoria and South Australia, and
The atmosphere responds to Southern Ocean SST and sea-ice variations by prolonging phases of the Southern Annular Mode.
Aiken, C.M., and M.H. England, 2005:
On the stochastic forcing of modes of interannual Southern Ocean sea surface
temperature variability, J. Climate, 18, 3074-3083.
Aiken, C.M., M.H. England, and C.J.C. Reason, 2006:
Optimal growth of Antarctic Circumpolar Waves, J. Phys. Oceanogr.,
Oke, P.R., and M.H. England, 2004:
Oceanic response to changes in the latitude of the
Southern Hemisphere subpolar westerly winds, J. Climate,
Rintoul, S.R., and M.H. England, 2002:
Ekman transport dominates air-sea fluxes in driving variability of Subantarctic Mode Water, J. Phys. Oceanogr.,
Sen Gupta, A., and M.H. England, 2006:
Coupled ocean-atmosphere-ice response to variations in the Southern
Annular Mode. J. Climate, in press.
Santoso, A., and M.H. England, 2004:
Antarctic Intermediate Water circulation and variability
in a coupled climate model. J. Phys. Oceanogr., 34, 2160-2179.
Santoso, A., and M.H. England, 2006:
Antarctic Bottom Water Circulation and Variability in a
Coupled Climate Model. J. Phys. Oceanogr.,
manuscript in revision.
Santoso, A., M.H. England, and A.C Hirst, 2006:
Circumpolar Deep Water Circulation and Variability in a
Coupled Climate Model. J. Phys. Oceanogr., in press.
Return to Matthew England's home page