NZCLIMATE TRUTH NO 187
26TH OCTOBER 2008
PROBABILITY AND UNCERTAINTY
In Newsletter No 183 on "Chaos" I pointed out that The Intergovernmental Panel on Climate Change (IPCC) admits that the forecasting of weather is Impossible (though actually only difficult) because of "chaos", but, by changing the word "weather" to "climate". suddenly "prediction" as long ahead as 100 years becomes "much more manageable" and "chaos" disappears.
Actually, since their models are never tested, they admit that they cannot "predict" anything; they can only "project", but they can rely on friends such as the likes of Stern, Garnaut or Gore to change the word "projection" back to "prediction". The IPCC can also rely on the well-paid providers of the computer climate models to think of a number, double it, and when it reaches 90% then that can be called the "probability" that the "projections" are right.
The IPCC admit that a correlation, however convincing, does not prove cause and effect, but they carry out another semantic triumph when they rename "correlation " "attribution", and suddenly it does prove cause and effect.
Even the supposedly convincing correlations can only be "attributed" without massive fudging. Most of the parameters and equations that make up the models have huge uncertainties, and some are nothing but guesswork. It is thus possible to try to fit some of the climate sequences by simple fudging. They only partially succeeded with the "mean global surface temperature anomaly" by leaving out the two main influences on this record, the ocean oscillations, and its bias from urbanization. They also downplay the possible influence of the sun.
The recent CCSP Report "Temperature Trends in the Lower Atmolsphere" at
tried to claim that discrepancies between surface, satellite and radiosonde temperature records have been "removed" by manipulation of "uncertainties".
I have published a paper at
Gray, V R. 2007. Temperature Trends in the Lower Atmosphere Energy and Environment 17 707-714
which refutes this claim, but I am forced to admit that if you really get down to the uncertainties associated with the "mean global surface temperature anomaly record" you would end up proving that the supposed global warming did not happen.
The above CSSP Report provided proof that the models do not work from two graphs (attached) which compare the actual temperatures in the lower troposphere with the calculations from models. Climate sceptics have published this comparison widely..
But the IPCC climate scientists had to get out of this somehow. Now
"Consistency of modelled and observed temperature trends in the tropical troposphere by
|B. D. Santer 1 *, P. W. Thorne 2, L. Haimberger 3, K. E. Taylor 1, T. M. L. Wigley 4, J. R. Lanzante 5, S. Solomon 6, M. Free 7, P. J. Gleckler 1, P. D. Jones 8, T. R. Karl 9, S. A. Klein 1, C. Mears 10, D. Nychka 4, G. A. Schmidt 11, S. C. Sherwood 12, F. J. Wentz 10"|
no less than seventeen of them have fossetted around to find enough uncertainties they had previously neglected, plus a dose of "noise" , to show that the difference between these two graphs can be explained by these uncertainties.
They "revisit such comparisons" with "new observational estimates", There is "no longer a serious discrepancy" which means there still is a npn-serious discrepancy. There is an "emerging reconciliation". There is an "improved procedure for adjusting for inter-satellite biases"
So after all this. thay have only "attributed" temperatures in the lower atmosphere to climate model results with some unstated level of "significance":
We look forward to a revisit of the notoriously biased "mean global surface temperature anomaly". Maybe "global warming" will disappear.
![CCSPN1.1 - Chapter5, Figure 7E]( "Snapshot 2008-10-26 17-01-45.tiff")