CIP Explains It: Yamal

Steve McIntyre ventures once again into hockey sticks and thinks he finds something amiss. The usual suspects jump up and down in in their cages and throw some shit around. Real Climate (aka Gavin Schmitt?) responds but the Pig remains puzzled by significant obscurities in each's utterances. (Steve's and Gavin's that is).

So here is the the Pig's attempt to decipher all.

Tree ring temperature reconstructions always seemed like a black art to me, but so far as I can tell, it goes a bit like this. To start with, you have a collection of old living trees and trunks from fossil trees. It most places, the living trees are at most a few hundred years old, so to get information about earlier times you need to add fossil trees, and they need to be matched up in a continuous sequence up to the present. To present a simplified example, you might have a fossil tree that lived from 1200 to 1500, another that lived from 1400 to 1750, and a living tree that sprouted in 1675. Because different years have different effect on the growth rings, it's possible to match up the overlaps and create a continuous sequence. After one has collected a significant number of such sequential growth rings, it is necessary to try to get temperature information from them. Tree rings depend on temperature, but they also depend on a lot of other things - water, insect infestation, etc).

The blackest of the black arts in this is the elimination of cores that don't appear to be good indicators of temperature (or alternatively, selection of those that do appear to be good indicators.) Possible reasons for elimination might be fire or insect damage or just growth ring behavior significantly different from ones peers. I call this a black art because it is precisely here that potential selection bias raises its ugly head.

Finally, the cores are statistically analyzed to produce a temperature sequence.

McIntyre's argument, if I understand it correctly (and I don't find his prose that pellucid) is that (a)the sequence used at Yamal has a very small number of tree ring cores, and (b)that there are a whole bunch of nearby cores from living trees, and (c) when he throws them into the mix, he gets much different results for the last 100 years or so. Because his cores are all from short lived living trees, older results aren't affected except in that the divergence of recent results suggests more noise than the earlier samples admit.

The hysterical have concluded that the failure to include McIntyre's cores is some kind of scientific fraud, and that meme has become a cause celebre in the dimmer regions of the denial-o-sphere. McIntyre makes the more modest claim that ignoring the other cores makes the conclusion untrustworthy.

Real Climate's response devotes a lot more space to sarcasm than detailed critique, but here is about the clearest part:

So along comes Steve McIntyre, self-styled slayer of hockey sticks, who declares without any evidence whatsoever that Briffa didn’t just reprocess the data from the Russians, but instead supposedly picked through it to give him the signal he wanted. These allegations have been made without any evidence whatsoever.

McIntyre has based his ‘critique’ on a test conducted by randomly adding in one set of data from another location in Yamal that he found on the internet. People have written theses about how to construct tree ring chronologies in order to avoid end-member effects and preserve as much of the climate signal as possible. Curiously no-one has ever suggested simply grabbing one set of data, deleting the trees you have a political objection to and replacing them with another set that you found lying around on the web.

My take away from that: Of course McI's signal is noisy, he didn't do any quality control. Which is where the black art part comes in.

Finally, Keith Briffa, the author critiqued, has responded in a measured fashion, refuting some of McI's claims, but allowing that his data set deserves further consideration - which brings us back once again to selection criteria.

The "hockey stick" is one small piece of the case for AGW, but the criteria and statistical details are intricate enough for me to be glad that it's hardly an essential one. Am I missing anything important?