and analysis techniques have allowed the circulation, formation and modification of Southern
Ocean water masses to be quantified for the first
time, as discussed in Section 4.6.4. Observations,
theory and modelling over the last decade have led
to new insights into the meridional circulation of
the Southern Ocean and its link to the global
overturning circulation (Section 4.6.5). Finally, we
identify a number of important questions that
remain open, despite the substantial progress made
during the WOCE era (Section 4.6.6). These open
questions will provide the challenges for the next
generation of field programmes and modelling
efforts to follow WOCE.
4.6.2 Observations of the Antarctic
Circumpolar Current
A decade ago our knowledge of the structure of the
ACC was largely built on detailed measurements in
Drake Passage and coarse-resolution hydrographic
sections at other longitudes. The increase in the
number of high-quality, high-resolution sections
and advances in remote sensing have revealed a
number of new features of the ACC.
4.6.2.1 Fronts of the ACC
An example of the high-quality sections spanning
the ACC collected during WOCE is shown in
Fig. 4.6.3. This winter section between Australia
and Antarctica near 140°E (WOCE repeat section
SR3) illustrates a number of features that are
characteristic of the ACC system. As noted first by
Deacon (1937), isopleths of all properties generally slope upward to the south across the ACC in a
series of steps, or fronts. The locations of the
major fronts are indicated above the plots in Fig.
4.6.3. (For the criteria used to define the fronts at
SR3, see Rintoul and Bullister (1999).) A number
of prominent property extrema in Fig. 4.6.3 define
the well-known water masses of the Southern
Ocean, as discussed in Section 4.6.4.
Orsi et al. (1995) and Belkin and Gordon
(1996) have described the circumpolar path and
characteristics of the major fronts of the ACC
based on careful analysis of a large number of
hydrographic sections across the ACC (e.g. Fig.
4.6.4, see Plate 4.6.4, p. 300). In addition to the
two main fronts of the ACC, the Subantarctic
Front (SAF) and Polar Front (PF), Orsi et al. (1995)
identified two other fronts that were circumpolar
in extent, the ‘southern front’ and ‘southern
boundary’ of the ACC (Fig. 4.6.4, see Plate 4.6.4,
p. 300). They also showed that while in Drake
Passage the fronts are almost always distinct features separating zones of quieter flow and uniform
water properties, at other longitudes the fronts of
the ACC merge or split. The merging of fronts
is particularly dramatic in the southwest Indian
Ocean (Fig. 4.6.1), where the Subantarctic and
Polar Fronts of the ACC and the Subtropical Front
and Agulhas Return Current are all in close proximity and together produce some of the largest temperature and salinity gradients in the world ocean
(Olbers et al., 1992; Park et al., 1993; Read and
Pollard, 1993; Belkin and Gordon, 1996; Sparrow
et al., 1996). The SR3 section shown in Fig. 4.6.3
provides another example of a frontal structure
more complex than the classical description based
on Drake Passage experience: both the SAF and PF
are split into two branches, and the southern SAF
and northern PF have merged near 53°S.
While studies like those above have shown that
at most longitudes it is possible to identify particular fronts in hydrographic sections, and so verify
their circumpolar extent, advances in remote sensing and modelling of the ACC have provided a new
view of the rich structure of the current. By representing the SAF and PF as meandering Gaussianshaped jets, Gille (1994) was able to map the full
circumpolar path of the fronts from GEOSAT
altimeter data and illustrate the extent to which
the fronts were steered by topography. Streamfunction maps from eddy-resolving numerical
models reveal a more complex, filamented structure to the ACC than that generally inferred from
hydrographic climatologies (e.g. Maltrud et al.,
1998). Mean gradients of sea-surface temperature
(Fig. 4.6.4, see Plate 4.6.4, p. 300) also show a
complex, filamented structure similar to that seen
in the models (Hughes and Ash, 2001).
4.6.2.2 ACC transport
The need to measure the transport of the ACC was
a key motivation for the International Southern
Ocean Studies (ISOS) experiment in the 1970s.
Early attempts to use a small number of current
meters to provide a reference for geostrophic calculations were frustrated by the banded nature of
the flow: the resulting transport estimates varied
dramatically depending on whether a particular
instrument was in a front or not. During ISOS, the
SECTION 4 THE GLOBAL FLOW FIELD
274
Ocean water masses to be quantified for the first
time, as discussed in Section 4.6.4. Observations,
theory and modelling over the last decade have led
to new insights into the meridional circulation of
the Southern Ocean and its link to the global
overturning circulation (Section 4.6.5). Finally, we
identify a number of important questions that
remain open, despite the substantial progress made
during the WOCE era (Section 4.6.6). These open
questions will provide the challenges for the next
generation of field programmes and modelling
efforts to follow WOCE.
4.6.2 Observations of the Antarctic
Circumpolar Current
A decade ago our knowledge of the structure of the
ACC was largely built on detailed measurements in
Drake Passage and coarse-resolution hydrographic
sections at other longitudes. The increase in the
number of high-quality, high-resolution sections
and advances in remote sensing have revealed a
number of new features of the ACC.
4.6.2.1 Fronts of the ACC
An example of the high-quality sections spanning
the ACC collected during WOCE is shown in
Fig. 4.6.3. This winter section between Australia
and Antarctica near 140°E (WOCE repeat section
SR3) illustrates a number of features that are
characteristic of the ACC system. As noted first by
Deacon (1937), isopleths of all properties generally slope upward to the south across the ACC in a
series of steps, or fronts. The locations of the
major fronts are indicated above the plots in Fig.
4.6.3. (For the criteria used to define the fronts at
SR3, see Rintoul and Bullister (1999).) A number
of prominent property extrema in Fig. 4.6.3 define
the well-known water masses of the Southern
Ocean, as discussed in Section 4.6.4.
Orsi et al. (1995) and Belkin and Gordon
(1996) have described the circumpolar path and
characteristics of the major fronts of the ACC
based on careful analysis of a large number of
hydrographic sections across the ACC (e.g. Fig.
4.6.4, see Plate 4.6.4, p. 300). In addition to the
two main fronts of the ACC, the Subantarctic
Front (SAF) and Polar Front (PF), Orsi et al. (1995)
identified two other fronts that were circumpolar
in extent, the ‘southern front’ and ‘southern
boundary’ of the ACC (Fig. 4.6.4, see Plate 4.6.4,
p. 300). They also showed that while in Drake
Passage the fronts are almost always distinct features separating zones of quieter flow and uniform
water properties, at other longitudes the fronts of
the ACC merge or split. The merging of fronts
is particularly dramatic in the southwest Indian
Ocean (Fig. 4.6.1), where the Subantarctic and
Polar Fronts of the ACC and the Subtropical Front
and Agulhas Return Current are all in close proximity and together produce some of the largest temperature and salinity gradients in the world ocean
(Olbers et al., 1992; Park et al., 1993; Read and
Pollard, 1993; Belkin and Gordon, 1996; Sparrow
et al., 1996). The SR3 section shown in Fig. 4.6.3
provides another example of a frontal structure
more complex than the classical description based
on Drake Passage experience: both the SAF and PF
are split into two branches, and the southern SAF
and northern PF have merged near 53°S.
While studies like those above have shown that
at most longitudes it is possible to identify particular fronts in hydrographic sections, and so verify
their circumpolar extent, advances in remote sensing and modelling of the ACC have provided a new
view of the rich structure of the current. By representing the SAF and PF as meandering Gaussianshaped jets, Gille (1994) was able to map the full
circumpolar path of the fronts from GEOSAT
altimeter data and illustrate the extent to which
the fronts were steered by topography. Streamfunction maps from eddy-resolving numerical
models reveal a more complex, filamented structure to the ACC than that generally inferred from
hydrographic climatologies (e.g. Maltrud et al.,
1998). Mean gradients of sea-surface temperature
(Fig. 4.6.4, see Plate 4.6.4, p. 300) also show a
complex, filamented structure similar to that seen
in the models (Hughes and Ash, 2001).
4.6.2.2 ACC transport
The need to measure the transport of the ACC was
a key motivation for the International Southern
Ocean Studies (ISOS) experiment in the 1970s.
Early attempts to use a small number of current
meters to provide a reference for geostrophic calculations were frustrated by the banded nature of
the flow: the resulting transport estimates varied
dramatically depending on whether a particular
instrument was in a front or not. During ISOS, the
SECTION 4 THE GLOBAL FLOW FIELD
274