Thursday, September 17, 2015

An Even Easier Way to Get a Hockey Stick: Mere Exponential CO2 Growth

When I was gone last week, someone pointed out to me an excellent (and obvious) point: Superexponential growth in CO2 isn't needed to give a hockey stick -- mere exponential growth suffices. 

That is, by the exact same reasoning I gave before, exponential CO2 growth gives a linear increase in temperature -- i.e. a hockey stick.

Since, all else being equal, temperature change is proportional to the logarithm of CO2 concentration, CO2 increasing exponentially leads to a linear increase in temperature.

So a constant atmospheric CO2 concentration from, say, 1000 AD to the year 1850, results in no temperature change.

After that, an exponential change in CO2 results in a linearly increasing temperature.

That is, a hockey stick.

So, for Steyn's book to be at all true, basic physics must be violated. Whose side would you take in that fight -- a music reviewer, or the fundamental laws of the universe?


David in Cal said...

David -- the problem with the hockey stick is the handle, not the blade. Temperatures have risen since the mid-1970's. I think it's reasonable to believe that the current hiatus will end and temperatures will again begin to rise, although the rate of future rise is unclear.

The problem with the hockey stick is the implication that temperatures were fairly level in the past. This is where the faulty statistical formulas and faulty proxies gave a faulty result, which also contradicted much existing evidence of past temperature variation.

David in Cal

David Appell said...


Why are you afraid to comment using your real name? Who do you think will hurt you for our opinion?

If you think the blade is a "faulty result," then prove it. Just prove it.

Because you once saw such a diagram from Lamb in grade school, or whatever the Brits call those grades, is not such a reason. Can you point me to Lamb's published data, by the way? Where did his graph come from?

Prove it.

David in Cal said...

Who is Lamb?

Mann's Hockey Stick used an incorrect statistical formula. Apparently Mann monkeyed with some complex formula and bollixed it up. Wiki says, "Barton and U.S. Rep. Ed Whitfield requested Edward Wegman to set up a team of statisticians to investigate, and they supported McIntyre and McKitrick's view that there were statistical failings, although they did not quantify whether there was any significant effect."

My understanding is that the hockey stick purports to tell us the actual shape of the past global warming anomaly. If so, then a different projection of future temperatures would be important, but wouldn't affect the hockey's stick's claim that past temperatures had been relatively stable.

David in Cal

David Appell said...

Lamb = Hubert Lamb, of this graph:

David Appell said...

David in Cal wrote:
"...which also contradicted much existing evidence of past temperature variation."

Which evidence (reconstructions of Northern Hemisphere temperatures)? Something besides's Lamb's curve?

JoeT said...

It's interesting that David in Cal left out the following two sentences from the Wikipedia entry on the hockey stick and the Wegman report:

"They also produced an extensive network analysis which has been discredited by expert opinion and found to have issues of plagiarism. Arguments against the MBH studies were reintroduced as part of the Climatic Research Unit email controversy, but dismissed by eight independent investigations."

Mann did not "monkey with some complex formula" as you claim. He did a principal component analysis. The question in dispute is with you center the data based on an average over the past 50 years or so or whether you center it over a much longer time period. The difference between the centerings means that the warming trend moves from a leading principal component to one further down in the hierarchy. If you do the analysis correctly, it doesn't really make a difference.