Friday, December 4, 2009

CRU and alternative data sets

So the major defence of the global warming theory now comes down to this: granted there were problems with the CRU data base, even before the dog ate it, but there are at least three other data bases that are truly independent of the CRU mess. They are all reliable, and they all agree, so global warming is real.

(Stipulating that by "the global warming theory" I mean the belief that: temperature is going up dangerously; this is linked to/caused by man-made CO2; there is something realistic that governments can do about this; and billions of dollars should be spent accordingly. Different skeptics question different parts of this theory).

Slate has jumped on this bandwagon, with credit for information being given to Gavin Schmidt, one of the people implicated in the CRU e-mails.

Veronique de Rugy of the Corner offers a strange take on this: She cites people saying don't worry, science is de-centralized, climate science couldn't be controlled by a small group, anyone with a computer can generate a model, all the models pretty much agree. Her answer is that there are more Democrats than Republicans involved in producing the models. But this is not really the point. Models ultimately need to be tested against a complex mass of data. The big question is whether there are any other sets of data that are truly independent of the lost CRU data.

As to surface temps, Ron Bailey says No.

What about satellite temps, available since the 1970s? Lord Monckton says they are tied to the suspect surface temp data.

I have my doubts about Monckton, especially his fears that the warmists are trying to bring about a world government. I actually think the warmists have done little harm so far, since actual governments are so hesitant to actually pay an economic price over this issue. Many of the countries that have flamboyantly promised to achieve CO2 levels that go back decades are ex-Communist countries that have lost a lot factories, and will have little difficulty in keeping their promises. Canada used to make promises based on the idea that forests are carbon sinks, so we can get away with the oil sands.

Having said that, another point from Monckton: On "hiding the decline". Actual temperature readings do not go very far into the past. Going further back than 1900, or at best 1850, requires the use of proxies such as tree growth. The warmists discovered that for the period during which they had both tree ring data and temperature readings, the tree ring data was completely out of whack. In the most recent years, the tree rings showed decline whereas the temperature readings did not, and in earlier periods there were other problems indicating that tree rings were totally unreliable. So they wanted to stick to temperature data, which itself has probably been fudged to exaggerate warming in the 20th century, while concealing the fact that the tree ring data is of no use, and there is really no good data from before 1850 or 1900. "Hiding the decline" shown by tree rings in recent years was just the tip of the iceberg: more of the iceberg was exaggerating warming in the 20th century, and pretending to have good data from before the 20th century.

The IPCC report says tree rings are a questionable proxy for temperature, since many factors other than temperature can affect tree growth. Despite this warning, the warmists have relied very heavily on a very naïve reading of tree ring data.

Today on Anthony Watts' site: someone who works with stats, who has previously not been involved in the climate debate, shows that the CRU program literally told the computer to align all temperature data with a hockey stick.

No comments:

Post a Comment