View Single Post
Old 04-16-2013, 06:29 AM   #740 (permalink)
Arragonis
The PRC.
 
Join Date: Oct 2009
Location: Elsewhere.
Posts: 5,304
Thanks: 285
Thanked 536 Times in 384 Posts
Quote:
Originally Posted by NeilBlanchard View Post
I believe that the "hide the decline" phrase is in reference to a known anomaly in tree ring data that is a proxy for temperatures, that we have concurrent thermometer data for, as well? I may not be correct, but that is my recollection.
Yep. And the fact that the reconstruction doesn't match reliable instrument data leads automatically to questions about whether it has happened before during the reconstructed period, and if so why ? Also are those divergencies enough to devalue the reconstruction ?

Its worth noting for balance the reconstruction which did diverge has since moved back closer to the instrumental temps suggesting an external factor not yet isolated.

Quote:
Originally Posted by jamesqf View Post
And so? Point here is that the reconstructions, taken all together, are the best information we have. That information - from many different and unrelated sources - is pretty much all giving us the same picture, and that picture matches what is predicted from physics and observed CO2 increases.
So why were the tree rings highlighted diverging ? Are you stating that this could never have happened before during the period reconstructed ? How do you know ?

Quote:
Originally Posted by jamesqf View Post
What you seem to be saying is that we should just ignore all the reconstructions (and the physics) because you don't like the conclusions people draw from them.
Absolutely not, but they can be compared to historical records as a verification - the MWP, LIA, RWP etc. Some show the MWP, some do not, some show it as a local thing, some as a worldwide thing and some as a worldwide thing but not at the same time - so use them but acknowledge the uncertainties.

Also when selecting which ones to use clearly identify why some archives are included and others are specifically excluded.

Use standard maths methods like PCA and don't make up some other kind of PCA which is not recognised by statisticians.

And finally include your verification stats.

Quote:
Originally Posted by jamesqf View Post
This is apparently because you (along with a lot of other like-minded folks lumped under that term you so dislike) have the (utterly mistaken, IMHO) belief that the contemporary western lifestyle represents the best of all possible worlds, so anything that threatens to upset the status quo must be silenced.
__________________
[I]So long and thanks for all the fish.[/I]