MattE

With the amount of individual proxy response variation over time I would have to think that either the CIs for the final reconstruction would have to reach from floor to ceiling or that a large number of the proxies are not responding or responding weakly to temperature over time and should not be used in a reconstruction. Obviously for the second consideration to be addressed would require selecting the “correct” proxies to use. That the paper above and that from Loehle and McCulloch can show differing past temperatures relative to current times than those in many other reconstructions (with those other reconstructions also varying one from another ) only shows me that a selection process of proxies can significantly change the picture of past temperatures but that that says little about the validity of using the proxies as thermometers.

The lack of statistical or scientific rigour in paleoclimate science… or climate science for that matter… shocks many of us, when we start to “peek under the bonnet”. But, if you’re shocked by that particular discussion, maybe it would be better for your sanity if you don’t look any further!!!

I don’t recall offhand which figure he’s referring to. But, it most likely was for Chapter 6 of the IPCC 2007 Working Group 1 report (the “4AR” Steve refers to). If you’re interested in finding out more about it, have a read of Osborn & Briffa, 2006 first (you can find a .pdf in the link I gave you above). He used a similar logic there. You might not agree with it. But, you’ll understand his thinking a bit better. Then, you can have a look through the section of Chapter 6 in the IPCC report for the section on millennial temperature change. The figure you’re inquiring about is probably there.

But, to be honest, there are far more troubling difficulties with that chapter. You’re rightly concerned about the error bars associated with estimates. But, measurement error bars are only relevant if your measurements actually mean something!!!

You’ll be more interested when you find out what their measurements are of. Maybe start with the “Yamal” controversy. There’s a brief summary at Bishop Hill here: http://bishophill.squarespace.com/blog/2009/9/29/the-yamal-implosion.html .
He makes a few mistakes & misrepresentations, but a lot of what he says is valid.

P.S. Look up the difference between “accuracy” and “precision”. You seem to be using them interchangeably, but they’re quite different.

P.P.S. Be careful not to get to drawn into conspiracy theories. Most of these guys, e.g., Briffa, etc. genuinely seem to believe a lot of what they’re saying.

You could have highlighted this odd infilling on the Monday following their release on a long weekend.

Very decent of you to give them a week to bask in their ‘glory’ .

Certainly a sporting approach to give a chance to let one of their ‘peers’ flag this.
I guess none of them noticed anything fishy!
Yikes!
RR

LinkText Here

We deal in different industries. In the space business this kind of clouding of the results would be a shock.

Then again, when we look at errors and probabilities we tend to be concerned with things like how much time do we have to self-destruct a launch before it can kill people living nearby. The physics dictates the cone of safety as a rocket ascends. The farther it goes the more time to respond. But we have to get that probability cone right.

In those cases, screwing up the precision is truly criminal.

I have concluded my standards are probably to high for such mundane things as comparing tree rings between the last 50 years and 1500 years ago…

#3468

Maybe look at what publications Briffa was invovled with at that time and see it you can see evidence of what he describes. He gives quite an accurate description so it may be possible to recognise it.