Introduction - Climate Change: Identification and Projections - page 9

Chapter 8 extends this analysis to the sixteen possible combinations
among the four paleotemperatures and the four reconstructions of solar
irradiance. The overwhelming majority of these confirm the previous
analysis. The exceptions all arise from the same temperature reconstruction,
that of Phil Jones and Michael Mann [JON 04], who are active protagonists
in the graph war mentioned above. Even though processing their
reconstruction does not allow us to reject the hypothesis of weak sensitivity
to solar irradiance, it does not confirm it either. On the other hand, high
sensitivity to solar activity, statistically validated, cannot be contested with
the argument that the mechanisms of action are not accurately known. In
terms of sensitivity to CO 2 , the IPCC window is very wide, with the extreme
values differing by a factor of six. Unfortunately, the nature of historic and
paleoclimatic data is such that identification is unable to narrow this
window. Instead, it moves the whole range downwards, reducing the highest,
and seemingly most exaggerated, values. It cannot even be excluded that
human activity has a negative impact on global temperatures.
Chapter 9 compares results of the IPCC simulation, over the historical
period mentioned above (1850 to today), with simulations from the identified
models. Temperature reconstructions observed are similar, but contributions of
natural and anthropogenic factors are turned upside-down. It would appear that
the human influence conclusion for global warming is predetermined by the
mode of generation of the input data as defined by the IPCC.
Chapter 10 offers long-term climate projections. To do so, scenarios
created by the IPCC itself are used. Representative Concentration Pathways,
or RCPs, put forward a series of profiles for future CO 2 concentrations.
Unsurprisingly, simulations with constraints are alarming, especially in the
worst case scenario, the so-called “business as usual” (RCP 8.5). Projections
which result from free identification are much less worrying. Only a
minority of cases exceed the allegedly critical threshold of a temperature
increase of two degrees above pre-industrial levels by 2100.
Chapter 11 uses the short-term predictive capabilities (from several years
to several decades) of the energy balance models. The climate models
identified can be converted into state space models (in the systems theory
sense). Observations available presently can be used to estimate this state
through traditional techniques (Kalman filtering), from which it is then
possible to carry out a prediction on the future climate using the current state
estimated. The state in question includes both the climatic process itself and