Geoscience Reference
In-Depth Information
model (Fig. 14.2C) and the CSM1.4-carbon (Fig.
14.4B), projected 21st century warming is lower in
CSM1.4-carbon than in the Bern2.5CC model. This
difference is primarily related to the difference in
climate sensitivity; 2°C for a nominal doubling of
CO
2
in the CSM1.4 versus 3.2°C for the Bern2.5CC
best estimate.
Atmospheric CO
2
concentration increases by
300% and by 190% over this century in the high
'A2_c' and low 'B1_c' case, respectively (Fig. 14.4B).
Thereafter, atmospheric CO
2
decreases only very
slowly, although carbon emissions are (unrealisti-
cally) reduced to zero in 2100. Atmospheric CO
2
concentration is still twice as high by 2500 than in
pre-industrial times in the 'A2_c' case. On the other
hand, CO
2
falls below 350 ppmv within a few dec-
ades in the 'Hist' case. The global-mean surface
temperature anomaly peaks at 3°C in the 'A2_c'
case and at 1.7°C in the 'B1_c' case and remains
elevated for centuries (Fig. 14.4C). In the 'Hist' case,
global-mean surface temperature remains only
slightly perturbed (0.2°C warming) by 2500. The
global average saturation state of aragonite in the
surface ocean closely follows the evolution of
atmospheric CO
2
. Mean surface Ω
a
is reduced by
about half in 2100 for the 'A2_c' case and remains
reduced over the next centuries.
The long perturbation lifetime of CO
2
is a conse-
quence of the centennial to millennial timescales of
overturning of various carbon reservoirs. Most of
the excess carbon is taken up by the ocean and
slowly (on a multicentury to millennial timescale)
mixed down to the abyss. Ultimately, interaction
with ocean sediments and the weathering cycle
will remove the anthropogenic carbon perturba-
tion from the atmosphere on timescales of millen-
nia to hundreds of millennia (Archer
et al.
1999; see
Chapter 2 ).
In conclusion, the results from the commitment
scenarios show that the magnitude of 21st century
CO
2
emissions pre-determines the range of atmos-
pheric CO
2
concentrations, temperature, and ocean
acidii cation for the coming centuries, at least in the
absence of the large-scale deployment of a technology
to remove excess CO
2
from the atmosphere. In other
words, the CO
2
emitted in the next decades will per-
turb the physical climate system, biogeochemical
cycle, and ecosystems for centuries.
14.5 Regional changes in surface ocean
acidii cation: undersaturation in the
Arctic is imminent
The impacts of climate change and ocean acidii ca-
tion on natural and socio-economic systems depend
on local and regional changes in climate and acidii -
cation rather than on global average metrics. It is
important to recognize that the global-mean metrics
discussed in the previous sections lead to different
changes regionally (Fig. 14.5). Fortunately, the spa-
tial patterns of change in pH
T
and surface Ω
a
scale
closely with atmospheric CO
2
( Figs 14.6 and 14.7 ).
This eases the discussion of local changes for the
scenario range and enables us to make inferences
for local and regional changes from projected atmos-
pheric CO
2
. This section presents regional changes
in the saturation state of aragonite, Ω
a
, and pH
T
for
the three commitment scenarios introduced in the
previous section and as evaluated in the NCAR
CSM1.4-carbon model.
There are large regional differences in the surface
saturation state for pre-industrial conditions and in
its change over time (Figs 14.5 and 14.6). The sur-
face ocean was saturated with respect to aragonite
in all regions under pre-industrial conditions
( Kleypas
et al.
1999 ; Key
et al.
2004 ; Steinacher
et al.
2009); the lowest saturation levels are simulated in
the Arctic and in the Southern Ocean, whereas sur-
face water with saturation values above 4 can be
found in the tropics.
Surface-water saturation is projected to decrease
rapidly in all regions until 2100 and remains
reduced for centuries for all three zero-emission
commitment scenarios (Fig. 14.6). The largest
ocean-surface changes are found in the tropics
and subtropics for Ω
a
. In the high 'A2_c' case, Ω
a
in
the tropics and subtropics decreases from a satu-
ration state of more than 4 in pre-industrial times
to saturation below 2.5 at the end of the 21st cen-
tury. The saturation of tropical and subtropical
surface waters remains below 3 until 2500.
Although experimental evidence remains scarce,
these projected low saturation states in combina-
tion with other stress factors such as increased tem-
perature pose the risk of the irreversible destruction
of warm-water coral reefs (Kleypas
et al.
1999 ;
Hoegh-Guldberg
et al.
2007).
