weeks. Hughes et al. (1999) used results from two
eddy-permitting numerical models to show that
transport correlated better with pressure measured
on the south side of Drake Passage than with pressure difference across the passage. Pressure to the
south was also highly coherent around the coast of
Antarctica. The model transport variations were
well correlated with zonally averaged wind stress
(with a lag of less than 3 days) near the south
of Drake Passage, and occurred in currents that
are strongly steered by f/H contours, rather than
following the path of the ACC. The circumpolar
coherence of pressure at the Antarctic continental
margin is also observed in the WOCE pressure
records, as is the relationship (also noted from the
ISOS measurements) between bottom pressure and
wind stress, for semiannual and shorter periods.
The pressure record at the northern side of the
passage is dominated by local effects, resulting in
the relatively weak correlation between pressure
difference and transport.
While much has been learnt about the circumpolar structure of the ACC in the last decade, we
have not yet made much progress in refining our
estimate of the mean absolute transport of the
ACC. Improved estimates of absolute transport
are likely to come from inverse models capable of
synthesizing the complete suite of WOCE observations (hydrography, Eulerian and Lagrangian
velocity measurements, and remote sensing) with
dynamical constraints. Development of such models is an active research area. Several recent models
give absolute transport estimates that are similar to geostrophic estimates relative to the bottom (e.g. Macdonald, 1998; Sloyan and Rintoul,
2001a; Yaremchuk et al., 2000). However, given
that each of these models start with a first guess
of zero barotropic flow, and no such calculations have yet included the full WOCE data
set including direct velocity measurements, it is
inappropriate to conclude that the barotropic
contribution to the mean absolute transport of
the ACC is small.
4.6.2.3 Antarctic Circumpolar Wave
The ACC is of interest in part because it allows
communication between the ocean basins. One
phenomenon that depends on the oceanic teleconnection provided by the ACC is the Antarctic
Circumpolar Wave (ACW) identified by White and
Peterson (1996). The ACW consists of anomalies
in sea-surface temperature, sea-level pressure, and
sea-ice extent that propagate eastward around
the Southern Ocean. The patterns have zonal
wavenumber 2 and circle the globe in about 8–9
years, so the apparent period at any location is
about 4 years.
The discovery of the ACW has sparked considerable interest. Part of this interest lies in the
potential predictability offered by the ACW. Two
recent studies suggest that the ACW has a substantial impact on rainfall in Australia and New
Zealand, and may provide some predictive skill
(White and Cherry, 1998; White, 2000). The
physics of the ACW, in particular the extent to
which it represents a coupled mode of the ocean–
atmosphere system, has also been a topic of active
debate. The initiation of the ACW may be the
result of atmospheric teleconnections related to the
El Niño-Southern Oscillation (ENSO) (Peterson
and White, 1998; Baines and Cai, 2000). Other
studies suggest the ACW arises from, or is at least
maintained by, atmosphere–ocean coupling within
the Southern Ocean (Qiu and Jin, 1997; White
et al., 1998a; Goodman and Marshall, 1999;
Talley, 1999c; Baines and Cai, 2000). Several
recent model experiments suggest, on the other
hand, that the ACW is a passive ocean response to
atmospheric forcing, and not a true coupled mode.
These studies themselves differ as to the nature of
the atmospheric forcing that drives the ACW, with
ACW-like oscillations resulting from stochastic
forcing (Weisse et al., 1999), standing patterns
in the atmosphere (Christoph et al., 1998; Cai
et al., 1999), or ECMWF (European Centre for
Medium Range Weather Forecasts) re-analysis
fluxes (Bonekamp et al., 1999). In summary, a
variety of dynamical hypotheses have been proposed to explain the ACW, each of which succeeds
in explaining at least some of the characteristics
of the ACW. Longer time series of observations
(including subsurface ocean measurements) and
further modelling studies will likely be required to
improve our understanding of the mechanism of
the ACW.
4.6.2.4 Eddy fluxes of heat and momentum
The large-scale heat budget of the area south of
the ACC implies a significant poleward eddy heat
flux across the current (de Szoeke and Levine,
1981), and observed fluxes in Drake Passage
(Bryden, 1979; Nowlin et al., 1985) and southeast
4.6 The Antarctic Circumpolar Current System
279
Rintoul, Hughes and Olbers
eddy-permitting numerical models to show that
transport correlated better with pressure measured
on the south side of Drake Passage than with pressure difference across the passage. Pressure to the
south was also highly coherent around the coast of
Antarctica. The model transport variations were
well correlated with zonally averaged wind stress
(with a lag of less than 3 days) near the south
of Drake Passage, and occurred in currents that
are strongly steered by f/H contours, rather than
following the path of the ACC. The circumpolar
coherence of pressure at the Antarctic continental
margin is also observed in the WOCE pressure
records, as is the relationship (also noted from the
ISOS measurements) between bottom pressure and
wind stress, for semiannual and shorter periods.
The pressure record at the northern side of the
passage is dominated by local effects, resulting in
the relatively weak correlation between pressure
difference and transport.
While much has been learnt about the circumpolar structure of the ACC in the last decade, we
have not yet made much progress in refining our
estimate of the mean absolute transport of the
ACC. Improved estimates of absolute transport
are likely to come from inverse models capable of
synthesizing the complete suite of WOCE observations (hydrography, Eulerian and Lagrangian
velocity measurements, and remote sensing) with
dynamical constraints. Development of such models is an active research area. Several recent models
give absolute transport estimates that are similar to geostrophic estimates relative to the bottom (e.g. Macdonald, 1998; Sloyan and Rintoul,
2001a; Yaremchuk et al., 2000). However, given
that each of these models start with a first guess
of zero barotropic flow, and no such calculations have yet included the full WOCE data
set including direct velocity measurements, it is
inappropriate to conclude that the barotropic
contribution to the mean absolute transport of
the ACC is small.
4.6.2.3 Antarctic Circumpolar Wave
The ACC is of interest in part because it allows
communication between the ocean basins. One
phenomenon that depends on the oceanic teleconnection provided by the ACC is the Antarctic
Circumpolar Wave (ACW) identified by White and
Peterson (1996). The ACW consists of anomalies
in sea-surface temperature, sea-level pressure, and
sea-ice extent that propagate eastward around
the Southern Ocean. The patterns have zonal
wavenumber 2 and circle the globe in about 8–9
years, so the apparent period at any location is
about 4 years.
The discovery of the ACW has sparked considerable interest. Part of this interest lies in the
potential predictability offered by the ACW. Two
recent studies suggest that the ACW has a substantial impact on rainfall in Australia and New
Zealand, and may provide some predictive skill
(White and Cherry, 1998; White, 2000). The
physics of the ACW, in particular the extent to
which it represents a coupled mode of the ocean–
atmosphere system, has also been a topic of active
debate. The initiation of the ACW may be the
result of atmospheric teleconnections related to the
El Niño-Southern Oscillation (ENSO) (Peterson
and White, 1998; Baines and Cai, 2000). Other
studies suggest the ACW arises from, or is at least
maintained by, atmosphere–ocean coupling within
the Southern Ocean (Qiu and Jin, 1997; White
et al., 1998a; Goodman and Marshall, 1999;
Talley, 1999c; Baines and Cai, 2000). Several
recent model experiments suggest, on the other
hand, that the ACW is a passive ocean response to
atmospheric forcing, and not a true coupled mode.
These studies themselves differ as to the nature of
the atmospheric forcing that drives the ACW, with
ACW-like oscillations resulting from stochastic
forcing (Weisse et al., 1999), standing patterns
in the atmosphere (Christoph et al., 1998; Cai
et al., 1999), or ECMWF (European Centre for
Medium Range Weather Forecasts) re-analysis
fluxes (Bonekamp et al., 1999). In summary, a
variety of dynamical hypotheses have been proposed to explain the ACW, each of which succeeds
in explaining at least some of the characteristics
of the ACW. Longer time series of observations
(including subsurface ocean measurements) and
further modelling studies will likely be required to
improve our understanding of the mechanism of
the ACW.
4.6.2.4 Eddy fluxes of heat and momentum
The large-scale heat budget of the area south of
the ACC implies a significant poleward eddy heat
flux across the current (de Szoeke and Levine,
1981), and observed fluxes in Drake Passage
(Bryden, 1979; Nowlin et al., 1985) and southeast
4.6 The Antarctic Circumpolar Current System
279
Rintoul, Hughes and Olbers