Global warming, models and language - Climate Change: The Facts - page 35

Time scales on the order of one to three months are, thus, appropriate for sensitivity
studies. Longer time scales also involve 'pollution' from seasonal effects, etc. Restricting
consideration to such short time scales is the approach taken in papers by Lindzen and
Choi, Spencer and Braswell, and Trenberth and Fasullo. 10
The second problem is more difficult. Outgoing radiation varies (especially in the
visible) for reasons other than changing surface temperature (volcanoes, non-feedback
cloud fluctuations). Such changes are not responses to surface temperature fluctuations but
they do cause surface temperature fluctuations. As the 2014 paper by Yong-Sang Choi et
al. shows, the high noise to signal ratio makes determination of the shortwave feedback
unreliable. 11 However, the situation is better for the longwave feedback and the results of
papers by Lindzen and Choi, Spencer and Braswell and Trenberth and Fasullo all point to
either the absence of longwave feedback or negative longwave feedback. 12
Apart from basic physical issues, there are other practical problems such as the
presenceofsignificantgapsintheoutgoingradiationdata.Also,theradiationdatainvolves
two satellite systems (ERBE and CERES) with different properties. The 2011 paper by
Lindzen and Choi describes how we deal with these issues. 13
The fact that the data shows the absence of the longwave feedback is extremely
important. In all current GCMs, the contribution of the longwave feedback (generally
incorrectly referred to as the water vapour feedback) in the equation above is f longwave ~0.5.
This alone doubles the sensitivity from the no-feedback value. In these models, the
contributions of shortwave feedbacks can be as great as 0.3. However, the addition of this
to0.5leads toanamplification byafactor offive.Thevariation oftheshortwave feedback
factor between zero and 0.3 (in models) is what basically leads to the IPCC claim that
sensitivity is 2-5°C. However, if the longwave feedback factor is zero or even negative,
then the sensitivity with a shortwave feedback factor of 0.3 is no more than 1.4°C and
with a shortwave feedback of zero, the sensitivity can well be below 1°C. The absence of
a longwave feedback strongly suggests that something like the iris effect discovered in a
2001 paper by Lindzen, Ming-Dah Chou and Arthur Hou is causing the variations in upper
level cirrus to cancel any positive water vapour feedback. 14 Consistent with all this is the
finding by Lindzen and Choi that no model correctly depicts the observed variations in
outgoing radiation. 15
As we have seen, the basis for the possibility of high sensitivity is either models or
clearly incorrect approaches to data. In reality, the data all points to low sensitivity. It is
sometimes asserted that paleoclimate data points to high sensitivity. This stems from a
profound misunderstanding of the climate system. We will return to this last point later in
this essay.