by Keeling and Heimann (1986) and Keeling et al.
(1989) inferred that during the preindustrial
period, atmospheric CO 2 was approximately 1 atm
lower in the northern hemisphere than in the
southern hemisphere. This natural gradient is now
reversed due to fossil fuel emission, with the
northern hemisphere pCO 2 now being approximately 3 atm higher. Assuming that net terrestrial
biosphere carbon storage was in steady state for
the preindustrial era, the natural gradient must
have been supported by either (1) an interhemispheric southwards oceanic transport of order
1 PgC yr
91 (Keeling et al., 1989); (2) effects of seasonal co-variation of atmospheric CO 2 concentration and atmospheric circulation patterns that are
not resolved by atmospheric transport models (the
seasonal rectifier of Keeling et al., 1989; see also
Denning et al., 1995), or (3) a combination of
both. Resolution of this issue is central to the
interpretation of present-day meridional atmospheric pCO 2 gradients (Fig. 6.3.2). Modelling of
such gradients has led to the inference that there is
a large, northern hemisphere, terrestrial carbon
sink (e.g. Tans et al., 1990). However such inference is clearly sensitive to the assumed or modelled magnitude of the oceanic carbon transport
(Broecker and Peng, 1992) and/or the seasonal rectifier effect (Law et al., 1996).
The geographical distribution of contemporary
air–sea CO 2 flux estimates has also sometimes
been used as a constraint in order to resolve better
terrestrial sink distributions (e.g. Tans et al.,
1990). Indeed, the regional terrestrial sink distribution inferred by atmospheric inverse modelling
can be extremely sensitive to the distribution and
magnitude of the air–sea carbon flux (Fan et al.,
1998). Problems of data scarcity, variability and
accuracy involved with estimating air–sea CO 2
fluxes (Takahashi et al., 1997) currently limits the
power of this approach. Improved surface observation strategies and gas-exchange parameterizations can significantly strengthen this constraint in
the future. However, alternative approaches to
estimating regional air–sea exchanges should also
be considered, in a manner analogous to what
has been done previously using the heat and freshwater transport divergence in the ocean
(cf MacDonald and Wunsch, 1996; Bryden and
Imawaki, Chapter 6.1; Wijffels, Chapter 6.2).
Notably, the possibility of defining meridional
oceanic transports of carbon, and assessing carbon
transport divergence and convergence on regional
scales should be explored. However, it must be
remembered that (in contrast to heat and fresh
water) it cannot be assumed that regional carbon
storage is negligible over decadal time scales.
Rather, storage of Excess CO 2 is highly significant
for the contemporary ocean.
In Section 6.3.4.3, we discuss methods whereby
the anthropogenic or Excess component of the
contemporary carbon transport can be separated
from the total transport. This permits estimation
of the preindustrial or natural oceanic carbon
transport, which should prove of value for assessing the magnitude of the seasonal rectifier effect
within atmospheric transport models, and for better resolving the contemporary distribution of
terrestrial sinks.
6.3.3 The JGOFS/WOCE Global CO 2
Survey
6.3.3.1 The JGOFS–WOCE collaboration
During the late 1980s planning was underway for
two major international oceanographic research
programmes related to global change: WOCE
(part of the World Climate Research Programme,
see Thompson et al., Chapter 1.3) and JGOFS
(part of the International Geosphere–Biosphere
Programme). A key component of WOCE
(e.g. Woods, 1985a) was the WOCE Hydrographic Programme (WHP), which provides an
unprecedented ‘snapshot’ of the distribution of
water masses and chemical tracers throughout the
World Ocean during the 1990s. Included with this
snapshot was the measurement of boundary currents and hydrographic properties at high density
along zonal heat-flux lines where transports of
heat and fresh water could be estimated. In addition to temperature and salinity, traditional hydrographic properties such as dissolved oxygen and
inorganic nutrients were measured in order to
define water mass characteristics. Transient tracers, including the CFCs and
3 H–
3 He, give a picture
of the formation and spreading of water masses
away from the sea surface, and their distribution
act as constraints on ocean circulation and climate
modelling. As part of WOCE, a concerted effort
was made to define accuracy standards and quality
assurance procedures to ensure that the WHP data
set would be of the highest quality possible (King
et al., Chapter 3.1).
6.3 Storage and Transport of Excess CO 2 in the Oceans
495
Wallace
(1989) inferred that during the preindustrial
period, atmospheric CO 2 was approximately 1 atm
lower in the northern hemisphere than in the
southern hemisphere. This natural gradient is now
reversed due to fossil fuel emission, with the
northern hemisphere pCO 2 now being approximately 3 atm higher. Assuming that net terrestrial
biosphere carbon storage was in steady state for
the preindustrial era, the natural gradient must
have been supported by either (1) an interhemispheric southwards oceanic transport of order
1 PgC yr
91 (Keeling et al., 1989); (2) effects of seasonal co-variation of atmospheric CO 2 concentration and atmospheric circulation patterns that are
not resolved by atmospheric transport models (the
seasonal rectifier of Keeling et al., 1989; see also
Denning et al., 1995), or (3) a combination of
both. Resolution of this issue is central to the
interpretation of present-day meridional atmospheric pCO 2 gradients (Fig. 6.3.2). Modelling of
such gradients has led to the inference that there is
a large, northern hemisphere, terrestrial carbon
sink (e.g. Tans et al., 1990). However such inference is clearly sensitive to the assumed or modelled magnitude of the oceanic carbon transport
(Broecker and Peng, 1992) and/or the seasonal rectifier effect (Law et al., 1996).
The geographical distribution of contemporary
air–sea CO 2 flux estimates has also sometimes
been used as a constraint in order to resolve better
terrestrial sink distributions (e.g. Tans et al.,
1990). Indeed, the regional terrestrial sink distribution inferred by atmospheric inverse modelling
can be extremely sensitive to the distribution and
magnitude of the air–sea carbon flux (Fan et al.,
1998). Problems of data scarcity, variability and
accuracy involved with estimating air–sea CO 2
fluxes (Takahashi et al., 1997) currently limits the
power of this approach. Improved surface observation strategies and gas-exchange parameterizations can significantly strengthen this constraint in
the future. However, alternative approaches to
estimating regional air–sea exchanges should also
be considered, in a manner analogous to what
has been done previously using the heat and freshwater transport divergence in the ocean
(cf MacDonald and Wunsch, 1996; Bryden and
Imawaki, Chapter 6.1; Wijffels, Chapter 6.2).
Notably, the possibility of defining meridional
oceanic transports of carbon, and assessing carbon
transport divergence and convergence on regional
scales should be explored. However, it must be
remembered that (in contrast to heat and fresh
water) it cannot be assumed that regional carbon
storage is negligible over decadal time scales.
Rather, storage of Excess CO 2 is highly significant
for the contemporary ocean.
In Section 6.3.4.3, we discuss methods whereby
the anthropogenic or Excess component of the
contemporary carbon transport can be separated
from the total transport. This permits estimation
of the preindustrial or natural oceanic carbon
transport, which should prove of value for assessing the magnitude of the seasonal rectifier effect
within atmospheric transport models, and for better resolving the contemporary distribution of
terrestrial sinks.
6.3.3 The JGOFS/WOCE Global CO 2
Survey
6.3.3.1 The JGOFS–WOCE collaboration
During the late 1980s planning was underway for
two major international oceanographic research
programmes related to global change: WOCE
(part of the World Climate Research Programme,
see Thompson et al., Chapter 1.3) and JGOFS
(part of the International Geosphere–Biosphere
Programme). A key component of WOCE
(e.g. Woods, 1985a) was the WOCE Hydrographic Programme (WHP), which provides an
unprecedented ‘snapshot’ of the distribution of
water masses and chemical tracers throughout the
World Ocean during the 1990s. Included with this
snapshot was the measurement of boundary currents and hydrographic properties at high density
along zonal heat-flux lines where transports of
heat and fresh water could be estimated. In addition to temperature and salinity, traditional hydrographic properties such as dissolved oxygen and
inorganic nutrients were measured in order to
define water mass characteristics. Transient tracers, including the CFCs and
3 H–
3 He, give a picture
of the formation and spreading of water masses
away from the sea surface, and their distribution
act as constraints on ocean circulation and climate
modelling. As part of WOCE, a concerted effort
was made to define accuracy standards and quality
assurance procedures to ensure that the WHP data
set would be of the highest quality possible (King
et al., Chapter 3.1).
6.3 Storage and Transport of Excess CO 2 in the Oceans
495
Wallace