geographically uniform as that in the atmosphere, it
is a great deal less heterogeneous than the distribution of carbon, and carbon inventory changes, in
the terrestrial biosphere. Furthermore, most of the
carbon in the oceans is present as inorganic carbon,
comprising three main chemical species (dissolved
CO 2 , HCO 3
9 and CO 3
29
). This can be contrasted
with the single predominant carbon species found
in the atmosphere (CO 2 ) and the vast multitude of
carbon species, both inorganic and particularly
organic, which are significant for carbon storage on
land. Consequently estimates of carbon storage on
land are extremely difficult or impossible to constrain by direct measurement on a global scale.
Excess CO 2
The budget summarized above is a budget for
‘Excess CO 2 ’. This refers to carbon inventory
changes within an environmental reservoir relative
to the inventory that existed during the ‘preindustrial’ era. Analysis of high-resolution ice cores
(Indermuehle et al., 1999; Smith et al., 1999a)
reveals that atmospheric levels of CO 2 have varied
by no more than about 20 atm (1 atm:
0.101325Pa) through most of the Holocene.
Around the year 1750 atmospheric levels started
to rise from a late-Holocene level of about
280 atm: initially due to Excess CO 2 releases
associated with land-use changes and later due to
fossil-fuel combustion. It is generally assumed that
prior to 1750 the global carbon cycle was in a
steady state that has now been significantly perturbed as a direct result of human activity. Hence
the preindustrial era against which Excess CO 2
levels are assessed ended around 1750.
We refer to the rate of carbon inventory (or
concentration) change within an environmental
reservoir since 1750 as ‘storage of Excess CO 2 ’. In
order to be significant for storage, an environmental reservoir must have a large capacity to absorb
or release carbon and must be able to exchange
carbon on time scales relevant to the time scale of
the anthropogenic perturbation (i.e. years to centuries). For example, the enormous oceanic carbon
reservoir exchanges with the atmosphere over time
scales ranging from 1 year up to the ventilation
time of the deep ocean (several hundred years).
Hence the oceans are a potentially very significant
reservoir for storing Excess CO 2 , as are the atmosphere and the terrestrial biosphere (including carbon contained in soils).
In estimating oceanic levels of Excess CO 2 , it is
almost invariably assumed that the major factor driving changes in the ocean’s carbon inventory since
1750 has been air–sea exchange and the increased
partial pressure of CO 2 in the atmosphere (pCO 2 ).
Possible interannual variability in CO 2 uptake
driven by other, presumably natural, causes is
currently debated (amplitude:2 PgC yr
91
: Francey
et al., 1995; Keeling et al., 1995; Lee et al., 1998;
Battle et al., 2000). However, it is generally
assumed that on decadal and longer time scales
there is much less natural variability.
Uptake versus storage
An important distinction should be made between
the rate of uptake and the rate of storage of Excess
CO 2 . Uptake refers to the flux or net transfer of
carbon from one major environmental reservoir to
another. Storage refers to changes of inventory
within the reservoir. On the global scale these
quantities must be equivalent, of course, but this is
not necessarily the case on local or regional scales.
For most of the land surface, the rate of uptake
and rate of storage of Excess CO 2 will be near
identical at (almost) all spatial geographical scales.
This is because lateral carbon transport within the
terrestrial biosphere is severely restricted (trees
don’t move much).
This is certainly not the case for carbon in the
atmosphere and the ocean. For the atmospheric
reservoir, the uptake of Excess CO 2 (emissions) is
spatially non-uniform, being concentrated in industrial regions, yet the inventory change (storage) is
almost globally uniform as a result of rapid atmospheric transport and mixing. For the ocean, uptake
of Excess CO 2 refers to a change in the air to sea
(and, possibly, riverine) flux of carbon relative to
the preindustrial flux. Once again, on the global
scale these must balance carbon inventory changes
(i.e. storage), but on the local and regional scale
uptake of Excess CO 2 across the air–sea interface
may be significantly decoupled, geographically,
from storage due to transport effects (Sarmiento
et al., 1992; Caldeira and Duffy, 2000).
Net versus Excess fluxes
The discussion of uptake raises a related issue that
frequently confuses. Both the uptake (flux) and
storage of Excess CO 2 refer to changes evaluated
with respect to preindustrial conditions. In the
case of the terrestrial biosphere this distinction
SECTION 6 LARGE-SCALE OCEAN TRANSPORTS
490
is a great deal less heterogeneous than the distribution of carbon, and carbon inventory changes, in
the terrestrial biosphere. Furthermore, most of the
carbon in the oceans is present as inorganic carbon,
comprising three main chemical species (dissolved
CO 2 , HCO 3
9 and CO 3
29
). This can be contrasted
with the single predominant carbon species found
in the atmosphere (CO 2 ) and the vast multitude of
carbon species, both inorganic and particularly
organic, which are significant for carbon storage on
land. Consequently estimates of carbon storage on
land are extremely difficult or impossible to constrain by direct measurement on a global scale.
Excess CO 2
The budget summarized above is a budget for
‘Excess CO 2 ’. This refers to carbon inventory
changes within an environmental reservoir relative
to the inventory that existed during the ‘preindustrial’ era. Analysis of high-resolution ice cores
(Indermuehle et al., 1999; Smith et al., 1999a)
reveals that atmospheric levels of CO 2 have varied
by no more than about 20 atm (1 atm:
0.101325Pa) through most of the Holocene.
Around the year 1750 atmospheric levels started
to rise from a late-Holocene level of about
280 atm: initially due to Excess CO 2 releases
associated with land-use changes and later due to
fossil-fuel combustion. It is generally assumed that
prior to 1750 the global carbon cycle was in a
steady state that has now been significantly perturbed as a direct result of human activity. Hence
the preindustrial era against which Excess CO 2
levels are assessed ended around 1750.
We refer to the rate of carbon inventory (or
concentration) change within an environmental
reservoir since 1750 as ‘storage of Excess CO 2 ’. In
order to be significant for storage, an environmental reservoir must have a large capacity to absorb
or release carbon and must be able to exchange
carbon on time scales relevant to the time scale of
the anthropogenic perturbation (i.e. years to centuries). For example, the enormous oceanic carbon
reservoir exchanges with the atmosphere over time
scales ranging from 1 year up to the ventilation
time of the deep ocean (several hundred years).
Hence the oceans are a potentially very significant
reservoir for storing Excess CO 2 , as are the atmosphere and the terrestrial biosphere (including carbon contained in soils).
In estimating oceanic levels of Excess CO 2 , it is
almost invariably assumed that the major factor driving changes in the ocean’s carbon inventory since
1750 has been air–sea exchange and the increased
partial pressure of CO 2 in the atmosphere (pCO 2 ).
Possible interannual variability in CO 2 uptake
driven by other, presumably natural, causes is
currently debated (amplitude:2 PgC yr
91
: Francey
et al., 1995; Keeling et al., 1995; Lee et al., 1998;
Battle et al., 2000). However, it is generally
assumed that on decadal and longer time scales
there is much less natural variability.
Uptake versus storage
An important distinction should be made between
the rate of uptake and the rate of storage of Excess
CO 2 . Uptake refers to the flux or net transfer of
carbon from one major environmental reservoir to
another. Storage refers to changes of inventory
within the reservoir. On the global scale these
quantities must be equivalent, of course, but this is
not necessarily the case on local or regional scales.
For most of the land surface, the rate of uptake
and rate of storage of Excess CO 2 will be near
identical at (almost) all spatial geographical scales.
This is because lateral carbon transport within the
terrestrial biosphere is severely restricted (trees
don’t move much).
This is certainly not the case for carbon in the
atmosphere and the ocean. For the atmospheric
reservoir, the uptake of Excess CO 2 (emissions) is
spatially non-uniform, being concentrated in industrial regions, yet the inventory change (storage) is
almost globally uniform as a result of rapid atmospheric transport and mixing. For the ocean, uptake
of Excess CO 2 refers to a change in the air to sea
(and, possibly, riverine) flux of carbon relative to
the preindustrial flux. Once again, on the global
scale these must balance carbon inventory changes
(i.e. storage), but on the local and regional scale
uptake of Excess CO 2 across the air–sea interface
may be significantly decoupled, geographically,
from storage due to transport effects (Sarmiento
et al., 1992; Caldeira and Duffy, 2000).
Net versus Excess fluxes
The discussion of uptake raises a related issue that
frequently confuses. Both the uptake (flux) and
storage of Excess CO 2 refer to changes evaluated
with respect to preindustrial conditions. In the
case of the terrestrial biosphere this distinction
SECTION 6 LARGE-SCALE OCEAN TRANSPORTS
490