Environmental Engineering Reference
In-Depth Information
With 1,000 GtC cumulative emissions the median estimate of the warm-
ing falls to 1.6°C at 1,000 years. The high end warming is still above 2°C
at this time, though it falls below 2°C by 5,000 years. With 2,500 GtC cu-
mulative emissions the median warming is still 2.7°C at 5,000 years, while
the high-end warming exceeds 4°C. If cumulative emissions reach 5,000
GtC, even the lowest estimated warming remains above 2°C after 10,000
years, while the high end is above 7.5°C. In all cases, the warming decays
very little between 5,000 years and 10,000 years, and in fact the warming
remaining after 10,000 years would take 100,000 years or more to recover
under the slow action of silicate weathering processes, even in the absence
of destabilizing long-term Earth System feedbacks.
The large range of possible very long-term warming exhibited in the
preceding discussion is due in large measure to the uncertainty in climate
sensitivity. To what extent do past climate variations help to constrain this
spread? The study of the instrumental record of climate provides, at best, a
window into transient climate sensitivity. To obtain observational constraints
on equilibrium climate sensitivity or Earth System Sensitivity, one must look
into the more distant past. There are many ways to make use of the past
climate record as a guide to the future, and in evaluating the published re-
sults, one must take care to distinguish the kind of climate sensitivity being
estimated, which categories of climate forcings are regarded as feedbacks,
and which categories of climate forcings are regarded as known or diag-
nosed forcings to be used in determining the sensitivity of the rest of the
climate system.
Part of the cooling during the Last Glacial Maximum was due to a reduc-
tion in atmospheric CO
2
, and this can be used to estimate climate sensitivity.
To accomplish this, one must estimate and subtract out the portion of climate
change forced by changes in Earth's orbit, by growth of the Northern Hemi-
sphere ice sheets, and by dust radiative effects. In the end, this provides an
estimate of climate sensitivity rather than Earth System Sensitivity, because
the non-CO
2
forcings are treated diagnostically instead of as feedbacks. On
this basis, Hargreaves and Annan estimate a most likely climate sensitivity
corresponding to
Δ
T
2x
= 2.5C, with low likelihood that
Δ
T
2x
exceeds 6°C.
When additional observational constraints are incorporated, the maximum
likely value is reduced to 4°C, in line with the range seen in the IPCC en-
semble of models. Crucifix (2006) cautions, however, that the sensitivity of
Earth's climate to reductions in CO
2
may not be a good indication of the
sensitivity to increases.
The warm climates of Earth's more distant past provide our most impor-
tant guide as to Earth System Sensitivity. There are three times of particular
