Global warming, models and language - Climate Change: The Facts - page 33

Multi-decadal Oscillation—all of which de pend critically on this coupling and all of
whicharepoorlyrepresentedincurrentmodels.Seeingwhetherreducingsensitivitywould
improve this situation seems like an obvious test, but to the best of my knowledge, this has
not been attempted. One is left with the suspicion that the modeling community prefers to
defend high sensitivity even when it means failure to simulate known phenomena.
Nir J. Shaviv used correlations between the solar cycle and ocean temperature to
estimate the flux needed to produce the change in ocean temperature. 6 This allowed
him to estimate the amplification of solar flux (potentially related to cosmic rays and
their secondary effects on clouds and albedo), as well as climate sensitivity. Once again
sensitivity was around 1°C.
The climate forcing associated with added greenhouse gases manifests itself as
radiative forcing at the top of the atmosphere. This forcing also acts on the surface, but
no longer in the form of simple radiative forcing. Rather the primary forcing is associated
with what is called latent heat flux. This is simply the flux of energy associated with the
heat of vaporisation of water vapour that is evaporated from the surface. Here, it has been
noted byFrank Wentz et al. that whereas inmodels evaporation changes about 1-3percent
per degree of warming, it is observed to change 5.7 per cent per degree of warming. 7 The
former is consistent with model sensitivities of about 1.5-4.5°C , but the latter is consistent
with much lower sensitivity on the order of 0.8°C.
Intheabsenceofwhatarereferredtoasfeedbacks,theexpectedresponsetoadoubling
of CO 2 is about 1°C. In current general circulation models, this warming is accompanied
by feedbacks to water vapour and clouds (and to a much smaller extent to the
reflectivity—known as albedo—of snow) which serve (in the models but not necessarily
in nature) to amplify the response to CO 2 alone. Feedbacks that amplify the response are
known as positive feedbacks while those that diminish the response are known as negative
feedbacks. The situation is crudely described by the following equation (see Roe's 2009
paper for a detailed discussion): 8
Where ∆T is the response to a doubling of CO 2 , ∆T 0 is the response in the absence
of feedbacks, and ∑f i is the sum of the various feedback factors. A negative f i clearly
diminishes the response while a positive f i amplifies the response. Should ∑f i reach +1,
the response becomes infinite, but it would take literally forever to reach this value. As we
have already noted, high sensitivity is associated with long response times. It is common