Variability of the Earth’s climate - Ice Ages and Interglacials: Measurements, Interpretation, and Models - page 34

temperature changes during past climate episodes, palaeoclimatology currently
has a unique opportunity to fundamentally contribute to understanding climate
sensitivity. At present, one of the standard tools for estimating climate sensitivity
is the use of numerical climate models. Unfortunately, model-derived climate
sensitivities are subject to large uncertainties

. Studying past climates to
estimate climate sensitivity inarguably has one great advantage over theoretical
computer models: it is based on actual data. Unfortunately, palaeo data-derived
climate sensitivities have large uncertainties as well. Errors can arise from issues
such as dating, alteration of the climate signal after deposition, insucient spatial
and/or temporal coverage, and various uncertainties associated with the proxies
for environmental variables such as temperature and past atmospheric CO 2
concentrations.''

...

Climatologists use two different definitions of climate sensitivity. The political
definition is the temperature rise ( D T G ) in the 21st century resulting from doubling
CO 2 from the pre-industrial level of 280 ppm. The scientific definition is the
temperature rise ( D T G ) resulting from a forcing of 1W/m 2 at the top of the
atmosphere. The scientific definition is applicable for both short- and long-term
periods.
It is common for paleoclimatologists to distinguish between two different
types of political climate sensitivity: fast feedback sensitivity and Earth system
sensitivity (Royer et al., 2011; Zeebe, 2011). The former includes ''water vapor,
clouds, snow, and sea ice'' operating on timescales of less than 100 years, whereas
the latter includes these fast feedbacks as well as longer term ''changes in non-CO 2
greenhouse gases, vegetation, dust/aerosols, ice sheets, ocean circulation, marine
productivity, weathering and more'' (Zeebe, 2011). Climate models are aimed at
estimating the temperature rise due to doubling CO 2 from the pre-industrial level
of 280 ppm during the 21st century, and thus deal with the political fast feed-
back sensitivity. However, consider this hypothetical scenario. Suppose climate
models indicate that doubling CO 2 in the 21st century will produce an increase in
T G of X degrees (we need not specify X, except that it is a relatively large
number). Suppose further (as some climatologists believe) that this temperature
increase will gradually erode the ice sheets on Greenland and to some extent
Antarctica, and produce other long-term effects that will be manifested well after
the 21st century. These changes will add further warming leading to a higher long-
term political Earth system sensitivity than the short-term political fast feedback
sensitivity. However, scientific climate sensitivity is the same in both eras. The only
thing that changes from the short term to the long term is the magnitude of the
forcing.
Royer et al. (2011) provide a list of previous attempts to estimate the
long-term political Earth system sensitivity from paleoclimatic data on climate
and CO 2 levels. In the present chapter, I have attempted to provide insights and
assessment of various models for paleoclimatic estimates of scientific climate
sensitivity.