Emissions, Concentrations, and Related Factors - Climate Stabilization Targets - page 80

20th and 21st centuries forced with specified trajectories for fossil fuel and
land-use CO 2 emissions; carbon-climate sensitivity is assessed by comparing
simulations with and without coupling between model atmospheric CO 2
and atmospheric radiation (and thus climate). Some simulations exhibit a
large amplifying effect whereby climate effects lead to increased release of
carbon dioxide, mostly from terrestrial processes. By the year 2100, simula-
tions with strong feedbacks have higher atmospheric CO 2 levels as much as
200 ppm above the companion simulation without climate feedbacks. The
airborne CO 2 fraction tends to increases in models where uptake processes
are more sensitive to climate, where uptake processes are less sensitive to
atmospheric CO 2 , and where physical climate sensitivity to CO 2 is large.
The land modules in the C 4 MIP simulations did not include interactions
between carbon and other limiting nutrients such as nitrogen. When soil
organic matter is respired due to warming, nitrogen is released, stimulating
enhanced plant growth that can offset the CO 2 released from the soil matter.
More recent coupled simulations that include this effect indicate a small
negative climate-carbon feedback on atmospheric CO 2 (Thornton et al.,
2009a). Interestingly, the carbon-nitrogen model had only a very weak CO 2
fertilization effect on land photosynthesis because of nitrogen limitation.
As a result, the model land carbon uptake responded only weakly to rising
atmospheric CO 2 and had as a result one of the largest atmospheric CO 2
levels at the end of the 21st century. The degree to which CO 2 fertiliza-
tion is modulated by nitrogen is an important unresolved science question,
and for the carbon cycle, sensitivity of the sinks to CO 2 is as important as
sensitivity to climate. The C 4 MIP simulations did not address the full suite
of interactions between the carbon cycle and other changes in the Earth
System; for example, ocean carbon storage can be influenced by ozone-
driven changes in Southern Ocean winds (Le Quéré et al., 2008; Lovenduski
et al., 2008; Lenton et al., 2009). Finally, the current generation of coupled
climate-carbon cycle models neglect a number of carbon reservoirs, such
as high-latitude peats, permafrost and methane clathrates that could be
important on longer time scales (see Section 6.1).