Geoscience Reference
In-Depth Information
to invade the ocean causing the CO
3
2-
concentration
to decrease and exposing increasing areas of marine
carbonate sediments to undersaturated waters.
Second, the dissolution rates are very small. As a
result, it will take thousands of years for this com-
pensation to occur (e.g. Archer 2005). Sundquist
(1990) estimated that about 60% of the total buffer-
ing of an atmospheric CO
2
perturbation by ocean
processes can be attributed to circulation with char-
acteristic timescales of several centuries. The inter-
action with carbonate sediments will neutralize the
remaining 40%. The majority of ocean biogeochem-
ical models that are concerned about projecting the
evolution of ocean acidii cation over the next cen-
turies do not include the sediment pool. In strong
contrast, if one wants to consider the long-term
consequences of ocean acidii cation or model past
geological events, it is absolutely critical to include
this compartment as well (e.g. Ridgwell and
Hargreaves 2007 ).
The mean global saturation state of surface-ocean
waters with respect to calcite (Ω
c
) decreases from
Ω
c
> 5 at year 0 to Ω
c
= 2 at the end of the acidii ca-
tion scenario (Fig. 12.2B). The model projects a
decrease in CaCO
3
production of 27% (Fig. 12.2C).
In experiment CAL01, CaCO
3
dissolution decreases
by 16%, rel ecting a reduction in CaCO
3
production
and thus the availability of particles for dissolution
(Fig. 12.2.D). When normalized to production, car-
bonate dissolution increases from 61% at 1×
p
CO
2
to
72% at 4×
p
CO
2
. In experiment CAL02, water-col-
umn dissolution increases by 19% relative to the
pre-industrial state. The reduction in CaCO
3
pro-
duction (CAL01) drives an additional uptake of 5.9
Gt C relative to CAL03 over the course of the simu-
lation (Fig. 12.2.E). This corresponds to a very mod-
est decrease in atmospheric CO
2
of about 2.8 ppmv,
i.e. equivalent to 2 years of current levels of growth
of atmospheric CO
2
. The increase in dissolution l ux
alone (CAL02) gives rise to an excess uptake of 1.2
Gt C. In this particular model study, the substantial
decrease in CaCO
3
production combined with the
increase in relative dissolution translates to an over-
all modest negative feedback to atmospheric CO
2
.
How robust are these future projections of CaCO
3
production and dissolution across models? At the
same atmospheric CO
2
level of about 1100 ppmv,
Heinze (2004) predicts a global decrease in CaCO
3
production of approximately 38%, slightly larger
than the 27% simulated by Gehlen
et al.
( 2007 ). This
difference may be the result of the former study
reaching this
p
CO
2
level after 420 years compared
with 140 years in the case of the latter one. The longer
duration of the experiment allows for an amplii ca-
tion of ocean chemistry changes and contributes to
the stronger decrease in calcii cation (see Chapter
14 ). Ridgwell
et al.
( 2007 ) adjusted their parameteri-
zation to reproduce a dependency of CaCO
3
produc-
tion on Ω
c
similar to Gehlen
et al.
( 2007 ). The
corresponding calcii cation feedback ranged from 6.5
to 7.7 Gt C at 3×
p
CO
2
, compared with 5.6 Gt C at 4×
p
CO
2
reported by Gehlen
et al.
( 2007 ).
At i rst sight, one might thus conclude that the
projected decrease in pelagic calcii cation and asso-
ciated increase in atmospheric CO
2
uptake by the
ocean converge to modest levels of less than 10 Gt C
over the next century. However, these models rely
largely on the same small experimental dataset for
12.2.2.4 Future projections: impacts and feedbacks
Given that calcii cation increases surface-ocean
p
CO
2
(Eq. 12.1), a decrease in calcii cation in res-
ponse to a decrease in the CaCO
3
saturation state
would translate into an additional uptake of CO
2
.
The opposite would be the case if calcii cation were
to increase. Thus, a decrease in calcii cation tends to
act as a negative indirect group 1 feedback.
To illustrate the evolution of carbonate chemistry
and its impact on the marine carbonate cycle in
more detail, we use output from an ocean biogeo-
chemistry model that was run following the
standard scenario of the Coupled Model Inter-
comparison Project (CMIP;
http://www-pcmdi.llnl.
gov/projects/cmip/index.php)
. In this scenario,
atmospheric
p
CO
2
increases at a rate of 1% yr
-1
from
286 (referred to as 1× CO
2
) to 1144 (4× CO
2
) ppmv
over a 140 year time period (Fig. 12.2A). Three sen-
sitivity experiments were undertaken during this
study (Gehlen
et al.
2007 ). In experiment CAL01,
both CaCO
3
production and dissolution responded
to changes in carbonate chemistry, while in experi-
ment CAL02, CaCO
3
production was kept constant
at pre-industrial levels, but dissolution responded
to ocean acidii cation. Finally, in experiment CAL03
production and dissolution of CaCO
3
were kept at
pre-industrial levels.
