Making it up as they go along

Remember how the AGW alarmists had to find a way to get rid of the Medieval Warming Period? This inconvenient period of previous high temperatures didn't fit with their narrative so they had to get rid of it. They came up with the infamous Hockey Stick graph which they claimed "proved" global warming was unprecedented and that man-made global warming was genuine.

It was hokum - on a grand scale. So dodgy that even the IPCC won't use it any more - but it did it's job for a while.

More recently, the AGW crowd have had a problem with something harder to get rid of. Satellite data. All their models tell them that the upper atmosphere is where we should be seeing the tell tale signs of global warming, but, unfortunately for the alarmists, the satellite data says it isn't.

The alarmists say that this can't be right so the satellites must be wrong. So guess what? Yep - they've come up with a way to "prove" they are wrong.

Over the last two decades, temperature readings from the upper troposphere -- 12 to 16 kilometres (7.5 and 10 miles) above Earth's surface -- based on data gathered by satellites and high-flying weather balloons showed little or no increase.

Damn - we can't have that shrieked the alarmists.

Oft cited by climate change sceptics, these findings were known to be flawed but still challenged the validity of computer models predicting warming trends at these altitudes, especially over the tropics.

So accurate readings taken by a satellite are "known to be flawed"? But their computer models are perfect, eh? They can't be wrong so the satellite data must be flawed. I kid you not - this is how these people think.

So they had to get rid of the satellite data - and now they have.

[C]limate scientists Robert Allen and Steven Sherwood of Yale University use a more accurate method to show that temperature changes in the upper troposphere since 1970 -- about 0.65 degrees Centigrade per decade -- are in fact clearly in sync with most climate change models.

Uh huh - more accurate than a satellite reading? More accurate than directly observed readings?

Rather than measuring temperature directly, which had yielded inconsistent results, they used wind variations as a proxy. (My emphasis)

"Inconsistent" with their models, by the way - the actual results are very consistent in saying the models are rubbish.

"We take an alternative approach by using trends in winds to infer those of temperature," say the authors.

What a load of codswallop. How can they genuinely claim that using a trend to "infer" a temperature is more accurate than satellite data? They are making it up as they go along. The whole point of the study was to prove that the computer models aren't crap after all - which they are.

Well, guess what. Two can play at that game.

I've found that rather than measuring global surface temperature directly I've developed a system of using trends in fridge freezers to infer those of outside temperature.

Apparently we're in an ice age.