basis of Worthington’s (1981) volumetric census,
and T–S properties at the sills bounding the
Australian–Antarctic Basin, that the Adélie Land
Bottom Water accounts for up to 25% of the global
volume of AABW. The concentration of CFC-11 in
plumes of Adélie Land Bottom Water is high
(92.9 pmol kg
91 in 1991; Rintoul and Bullister,
1999), and the CFC saturation (35%) is similar to
that in plumes observed in the southwest Weddell
Sea. High CFC concentrations observed over the
continental slope at 30°E in the eastern Weddell Sea
(Mantisi et al., 1991) indicate that a source exists
along the Enderby coast or in Prydz Bay as well, as
suggested earlier by Jacobs and Georgi (1977).
Recent studies have also identified regions of the
Antarctic coastline where AABW is not formed.
For example, Fahrbach et al. (1994b) show that
AABW is not formed in the eastern Weddell Sea,
primarily because of the inability to confine shelf
water (and increase its salinity) over the narrow
shelf there. In the Amundsen and Bellingshausen
Seas, nearly undiluted warm Circumpolar Deep
Water intrudes onto the shelf and melts large
amounts of continental ice, preventing formation
of dense saline water (Hellmer et al., 1998).
Foster (1995) confirmed the observation of
Carmack and Killworth (1978) that the Wilkes
Land coast east of 150°E forms water that is not
quite dense enough to sink to the seafloor. Sinking
of Antarctic shelf waters to deep, rather than
abyssal, layers is potentially of significant importance to ventilation of the deep ocean at lower latitudes, since these lighter layers are less confined by
bathymetry and are free to spread northward. The
formation of ‘not quite bottom’ water around
Antarctica has so far received little attention, and
the transport is unknown. Weppernig et al. (1996)
conclude from the distribution of stable isotopes
that a large fraction (980%) of the dense Ice Shelf
Water leaving the continental shelf in the Weddell
Sea supplies the WSDW at intermediate depth,
rather than sinking to the bottom to supply WSBW.
A powerful demonstration of the utility of transient tracer measurements is the use of CFC inventories to estimate the formation rate of AABW
(Orsi et al., 1999). They find that about 8 Sv
(4.9 Sv in the Atlantic and 3.2 Sv in the
Indian–Pacific) of new bottom water (a 50 : 50 mix
of dense shelf water and entrained deep water)
must sink across the 2500 m isobath to explain the
observed CFC-11 inventory in the AABW layer
(water with neutral density greater than
28.27 kg m
93 ). Much of the 3.2 Sv formed in the
Indian–Pacific sectors is likely produced by sources
in the Australian Antarctic Basin (e.g. Adélie
Land), since the average CFC-11 concentration
there is 2.5 times as large as that of the Atlantic
and Pacific basins (Fig. 4.6.13).
The CFC-11 inventory of Orsi et al. (1999) provides a strong integral constraint on the formation
of AABW. Broecker et al. (1997) have noted that
PO 4 * (PO 4 *:PO 4 ;O 2 /17591.95 mol kg
91
, where
175 is the average molar Redfield ratio of O 2 consumption to remineralization in the deep sea and
1.95 is an arbitrary constant.) also provides a constraint on the relative production of dense ventilated water in the Southern Ocean and the North
Atlantic: roughly equal contributions from the two
source regions are required to explain the PO 4
*
value observed in the deep Indian and Pacific
Oceans. The PO 4
* budget thus requires sinking
of about 15 Sv of ventilated shelf water around
Antarctica, a factor of three to four higher than
implied by the CFC-11 inventory or by transport measurements near the sources. Part of this
discrepancy may be explained by exchange of
ventilated shelf water with deep water lighter than
neutral density of 28.27 kg m
93
. However, given
that the CFC content of the lighter deep water is so
much lower than that of the dense abyssal layer it
is difficult to see how this process can account
for the discrepancy. Another possibility is that
AABW production in recent decades is substantially smaller than the average during the last
millennium (Broecker et al., 1997). The conflict
between estimates of ventilated deep water production based on CFC and PO 4 * remains
unresolved.
WOCE sections, moored arrays, and other
recent measurements have helped to map out the
system of deep western boundary currents and
through-passage flows that carry AABW and
LCDW to lower latitudes (e.g. Mantyla and Reid,
1995; Rhein et al., 1998a; Whitworth et al., 1999;
Zenk et al., 1999a; see also Hogg, Chapter 4.5).
4.6.5 The Southern Ocean and the global
overturning circulations
The global overturning circulation is often taken
to be synonymous with the circulation loop
formed by sinking and export of NADW from the
SECTION 4 THE GLOBAL FLOW FIELD
296
and T–S properties at the sills bounding the
Australian–Antarctic Basin, that the Adélie Land
Bottom Water accounts for up to 25% of the global
volume of AABW. The concentration of CFC-11 in
plumes of Adélie Land Bottom Water is high
(92.9 pmol kg
91 in 1991; Rintoul and Bullister,
1999), and the CFC saturation (35%) is similar to
that in plumes observed in the southwest Weddell
Sea. High CFC concentrations observed over the
continental slope at 30°E in the eastern Weddell Sea
(Mantisi et al., 1991) indicate that a source exists
along the Enderby coast or in Prydz Bay as well, as
suggested earlier by Jacobs and Georgi (1977).
Recent studies have also identified regions of the
Antarctic coastline where AABW is not formed.
For example, Fahrbach et al. (1994b) show that
AABW is not formed in the eastern Weddell Sea,
primarily because of the inability to confine shelf
water (and increase its salinity) over the narrow
shelf there. In the Amundsen and Bellingshausen
Seas, nearly undiluted warm Circumpolar Deep
Water intrudes onto the shelf and melts large
amounts of continental ice, preventing formation
of dense saline water (Hellmer et al., 1998).
Foster (1995) confirmed the observation of
Carmack and Killworth (1978) that the Wilkes
Land coast east of 150°E forms water that is not
quite dense enough to sink to the seafloor. Sinking
of Antarctic shelf waters to deep, rather than
abyssal, layers is potentially of significant importance to ventilation of the deep ocean at lower latitudes, since these lighter layers are less confined by
bathymetry and are free to spread northward. The
formation of ‘not quite bottom’ water around
Antarctica has so far received little attention, and
the transport is unknown. Weppernig et al. (1996)
conclude from the distribution of stable isotopes
that a large fraction (980%) of the dense Ice Shelf
Water leaving the continental shelf in the Weddell
Sea supplies the WSDW at intermediate depth,
rather than sinking to the bottom to supply WSBW.
A powerful demonstration of the utility of transient tracer measurements is the use of CFC inventories to estimate the formation rate of AABW
(Orsi et al., 1999). They find that about 8 Sv
(4.9 Sv in the Atlantic and 3.2 Sv in the
Indian–Pacific) of new bottom water (a 50 : 50 mix
of dense shelf water and entrained deep water)
must sink across the 2500 m isobath to explain the
observed CFC-11 inventory in the AABW layer
(water with neutral density greater than
28.27 kg m
93 ). Much of the 3.2 Sv formed in the
Indian–Pacific sectors is likely produced by sources
in the Australian Antarctic Basin (e.g. Adélie
Land), since the average CFC-11 concentration
there is 2.5 times as large as that of the Atlantic
and Pacific basins (Fig. 4.6.13).
The CFC-11 inventory of Orsi et al. (1999) provides a strong integral constraint on the formation
of AABW. Broecker et al. (1997) have noted that
PO 4 * (PO 4 *:PO 4 ;O 2 /17591.95 mol kg
91
, where
175 is the average molar Redfield ratio of O 2 consumption to remineralization in the deep sea and
1.95 is an arbitrary constant.) also provides a constraint on the relative production of dense ventilated water in the Southern Ocean and the North
Atlantic: roughly equal contributions from the two
source regions are required to explain the PO 4
*
value observed in the deep Indian and Pacific
Oceans. The PO 4
* budget thus requires sinking
of about 15 Sv of ventilated shelf water around
Antarctica, a factor of three to four higher than
implied by the CFC-11 inventory or by transport measurements near the sources. Part of this
discrepancy may be explained by exchange of
ventilated shelf water with deep water lighter than
neutral density of 28.27 kg m
93
. However, given
that the CFC content of the lighter deep water is so
much lower than that of the dense abyssal layer it
is difficult to see how this process can account
for the discrepancy. Another possibility is that
AABW production in recent decades is substantially smaller than the average during the last
millennium (Broecker et al., 1997). The conflict
between estimates of ventilated deep water production based on CFC and PO 4 * remains
unresolved.
WOCE sections, moored arrays, and other
recent measurements have helped to map out the
system of deep western boundary currents and
through-passage flows that carry AABW and
LCDW to lower latitudes (e.g. Mantyla and Reid,
1995; Rhein et al., 1998a; Whitworth et al., 1999;
Zenk et al., 1999a; see also Hogg, Chapter 4.5).
4.6.5 The Southern Ocean and the global
overturning circulations
The global overturning circulation is often taken
to be synonymous with the circulation loop
formed by sinking and export of NADW from the
SECTION 4 THE GLOBAL FLOW FIELD
296