Partial Results - Climate Change: Identification and Projections - page 81

Camp for this analysis, but from data covering more than a millennium, that
dynamic parameters are common to all causality chains, and that their
identification results from a processing of the collection of all data, including
CO 2 and volcanism.

However, there is a difference in the nature of these two “sensitivities”.
Ours corresponds to a given point of the frequential response of the climate
process (frequency 1/11 years ), and not only manifests itself by a gain, but
also with a phase change, which here is of 47°. The correlation established
by Camp regards values without phase changes, which would only make
sense for a static model (reduced to sG ), or for signals which vary very

slowly. The result is that the response to the Schwabe cycles does not
prejudge any level of static sensitivity. This is much higher than the
dynamic sensitivity, given the ratio

( ) = 1

. Camp's estimation
technique ( composite mean difference projection ) is not in question
here: it is the object of his estimation itself which is inconsistent. Likewise,
the temperatures in Figure 3.1 (the hockey stick as well as boomerang
curves) do not vary sufficiently slowly in relation to the time constants
which are of the order of a century. In fact, the only significant
“correlations” of a cause and effect relationship are those which are
expressed via the identification of a dynamic causal model. This is the entire
purpose of this work.

ρ = 0 . 9 / 17 . 5 ~ 1 / 20

Finally, apart from the fact that sensitivity to irradiance and its
contribution to warming are incompatible with the IPCC's assumptions,
there is nothing wrong to be added about the simulation as a whole. For
instance, it can be noted that the estimated equilibrium temperature is T E =
0.027 ° C, close to the reference of zero from 1850-1900, the assumed
pre-industrial equilibrium temperature.

7.3. Forced identifications

Contrary to “free” identification, here the model is now constrained
to comply with arbitrary assumptions on given parameters. The
parameters which remain free are adjusted to optimize temperature
reproduction.