Monday, September 13, 2010

How they do the warming trick

I think I'm getting there. The official figures say double CO2 from around 300ppm to 600ppm and you'll get a 1'C rise, then subject to positive feedback which could be as high as 6.5' at the top estimate. On paper the first figure would work as is the same as filling a greenhouse with CO2 and doubling it, and then extending that to the atmosphere. They even added that every other factor is irrelevant as the CO2 automatically raises the temperature by an exact base figure per amount increase.

This is a pretty clear equation even I have trouble disputing. But as I could see from such a simple (and purely hypothetical mental wank by statisticians and climatologists) formula so much more had been assumed I had to look into it further. Not being qualified in any area beyond O level, although I took (and passed) one maths/stats unit at degree level, I was no better at deciphering the following studies than anyone else, but had to try.

What I found may well explain how they can start (as the devil does) with 90% truth and then hide a lie in it. They have to have a base of fact or there's nothing to get confidence, and then when everyone is convinced they can extend the good stuff to cover basically whatever they like. This is done in two very clever ways, adjusting data and reducing uncertainty. I can't see that a graph is created from smoothing, selecting or rebasing but statisticians can. Some have even gone to the effort of pointing it out to help people like me who can't do it. Three so far have explained how the hockey stick, the holy grail of Al Gore and the IPCC, is created by all such methods, and universally ignored by anyone in charge for obvious reasons.

The hockey stick is created from three sets of data. Proxy data until 1860, adding thermometers till 1979 and then mainly satellite. The IPCC even admit the earliest data was partial and needed to be filled in (how do they manage that then?) meaning the error margin was far wider, but narrows up by the sharp rise in 1980. So they explain the coexistence of such methods, but however you look at it there's still this almost vertical line at the end. So uncertainty can no longer be used as a way to rebut it as according to the red satellite part it simply isn't uncertain and that is all higher than the last millenium.

But get a degree in maths and it all looks different, as they rebased it. I'll admit here I don't know what that means, except that they start the mean at a point away from the beginning, in this case around where it begins to rise. Further they have adjusted or smoothed the data in this and nearly every other graph of ice and sea levels as well as temperature to make it look exactly the same way, with ice falling (although the raw data which is available shows otherwise) and sea level rising (although many measurements show it is now virtually static). It seems that they can (and do) adjust every graph they create without question as besides their own no one can tell. Therefore the emails at CRU admitting they refuse (and continue to do so) to release the raw data behind the hockey stick means without raw data you can't do the before and after and see how they got it to change.
But someone else has been caught out, New Zealand's national climate people have got a hockey stick of their own, and when raw data was put in the angle dropped by about 75%. This is currently under trial and the results should follow soon. But it's a wonderful example and if proven to have cheated then ought to represent how the tricks are always done. Like the magicians who are exposed on TV, once you know it's no longer clever or an illusion, it's just someone fiddling the data.

So I think now people are finding ways to figure out how so many graphs rise in 1980 and most peak right now (highly in dispute) then if you remove the adjustments they ought to fall back by the same amount. So they start with good physics (the truth) and hide a lie in it (the adjustment).

Part two is known and a lot easier for anyone to follow. The levels of uncertainty are so great that it may be beyond a measurable significance whatever the IPCC say. Michael Shermer of the Skeptic magazine shows how they underestimated their uncertainty by about a factor of 10, making any predictions meaningless (as they all are in an open system). Star movements are fairly fixed so predictable for millions of years, but climate movements are not predictable beyond a rough cyclic framework. But if doubling CO2 raises base levels by 1'C then you can factor in feedback mechanisms to see how that will increase. What? Can you really? As man has never pumped extra CO2 in the atmosphere there's no history there at all. Nothing to go on. So we have the present (0.7'C rise in 150 years, quite typical coming out of a cold spell) and the future ( between 0.5-6.5'C by 2100).
This has become like the Lord's Prayer to warmists, as if thy kingdom come, thy warm be done. But it assumes the IPCC can actually predict climate boundaries within a reliable measure. Where's their track record for this? They haven't got one- it's never been done. But who thinks about that when thinking about the first psalm 'We've got to protect our unborn grandchildren'. That's fear and primal survival. Forget maths, stats and any other science, this is raw fear data. Pure emotion with no space left for any thought or science at all.

There are more and more detectives unravelling the tricks built into graphs worldwide, and besides Michael Shermer the logic behind the uncertainty is for anyone who looks to see. You can see the New Zealand trick here
The hockey stick summary is here, linked to a book-length PDF
Thanks to James Delingpole for those two.
Here's the analysis of the uncertainty.

Crime relies on being guilty beyond reasonable doubt. Fraud always involves an element of truth, if well made enough will hide the lies until at least after the money is long gone, sometimes forever. How anyone can read these articles alone and not have some doubt is truly a test of their mental power.

No comments: