The Marcott-Shakun Dating Service
By Paul Homewood
Figure 1. Reconstructions from alkenone proxies in Marcott style. Red- using published dates; black- using Marcott-Shakun dates.
For those who have been following the saga of the Marcott Hockey Stick MkII, Steve McIntyre has followed up his series of posts, which have already raised serious questions about the paper’s credibility.
This time he has discovered that some of the proxies used by Marcott have been redated. The result can be seen in the above graph. Based on the original published dating of the proxies, there should be a sharp dip downwards in the temperature reconstruction. After Marcott has altered the dates, the uptick appears.
I always steer clear of these sort of statistical discussions, as I am not a statistician and people like Steve Mac are.
Nevertheless, I wondered how a bit of redating could make this much difference. So, apparently did one commenter, Clay Marley, on WUWT. Thanks to another commenter, K Scott Denison, there is a very simple explanation.
k scott denison says:
Clay Marley says:
March 16, 2013 at 5:22 pm
OK so I read the post but I’m not seeing where the hockey stick comes from.
In case no one else has answered, think of it this way.
I have two proxies.
Proxy one’s anomalies are: (-2, -2, -2, -2, n/a), the last one because I redacted the proxy and it doesn’t extend to the fifth period.
Proxy two’s are: (2, 2, 2, 2, 2).
The average is (0, 0, 0, 0, +2)
There’s your up tick.
Worth bearing in mind.