[OT] So, where is your climate blog? I clicked on your name and got lost in Make Money Fast!, Supermom, family vacations… ???

TIA, Best for 2008, Pete Tillman

The smooths in the two figures are different – one is 25 years and one is 40 years.

Thanks for responding Steve, but that is not a linear transformation, look at the two curves between 1640 and 1700 for example.
Your inset for Fig 3 shows the truncation of the light blue curve at a minimum, there’s no minimum in the Briffa curve until the 1970s and as you say it goes way down.
If those curves started life from the same data there has been rather more than a linear transformation performed.
Depending on the smoothing algorithm the ends of a smoothed series can be distorted which could justify truncation (should be indicated of course).

The Briffa graphic you posted is for summer mean temp. for land north of 20N, while the TAR graphic is also for extratropical N hemisphere it is clearly not the from the same data. Unless we are shown the same data we’re unable to judge what has been truncated.

Make you sure that you understand what these folks do. First they create a time series going from 1402-1990 and then they fit it to a “target” series. All that happens in this step is that the series is re-scaled and re-centered. It’s a linear transformation, but the underlying series is the same. I know the data intimately and we’re talking apples and apples here.

Even so the truncation you refer to, if it had been of the Briffa data you presented, was at the very point that their recon had reached a minimum and was starting to rise, an unusual choice if they were attempting to ‘conceal a mismatch’.

Nope. You’re way wrong. The data goes down a lot after the truncation point. That’s why they truncated it. Read the linked posts as well.

In any case, I would submit that there is a distinction between truncation of measured data and truncation of data smoothed over long time periods (24 yrs in Briffa and 30 in Loehle) which have problems due to the treatment of the end condition. (See discussion of Loehle recon. for example).

What is the distinction that you have in mind and how does it justify IPCC concealing the mismatch?

Liselle #82, that was what I was trying to point out in #25 that, the IPCC Fig. 2. presentation looks like a different or smoothed
dataset when compared to the TAR.

Likewise I pointed out in #22 that the Briffa data in the original post obviously wasn’t the data used in the TAR but got no response.
I also pointed out that the truncation was at the point where the data was turning up with the same result.

There are two Courtillot data sets in play and I think that there is some confounding in the present discussion. I examined the temperature data set, where Courtillot used the Jones version used in Briffa et al 2001. He did not actively truncate it; he just used an obsolete version. Is this worse than Mann’s use of obsolete data – an issue raised as long ago as MM03? Or with Hughes not using the Ababneh update? I’m OK with a condemnation of the use of obsolete data versions as long as this criterion is applied consistently.

The other issue is the solar data version. I’ve been able to locate digital versions of the SOLAR2000 data for comparison.

Regarding Courtillot’s use of the Briffa data it’s not just that he used an obselete version it’s that he used data that wasn’t what he says it is, i.e. it’s not global mean temperature it’s NH extra-tropical summer T. When there are available data sets for Tglobe (as shown in B&D’s recasting of his graph one wonders why he didn’t use them instead, as I recall in his response he still doesn’t acknowledge that it’s not a global temp.
I’d be interested in seeing the SOLAR2000 data although there do we know what data was used?
According to the source for SOLAR2000 the following acknowledgement should have been used:
‘Solar Irradiance Platform historical irradiances are provided courtesy of W. Kent Tobiska and Space Environment Technologies. These historical irradiances have been developed with partial funding from the NASA UARS, TIMED, and SOHO missions.’
Why they used that ~50 yr data instead of the 100 yr data that they referred to in their paper is still not known.

Regarding the Ababneh data I didn’t follow that but as I recall it poses a rather tricky situation in academia, using data from a colleague’s grad student’s thesis before it has been formally published. Someone did that to me once and I wasn’t very happy about it, to say the least!

Steve: Courtillot’s data handling is unsatisfactory as I’ve observed. How is it worse than Mann’s? There’s more going on with the Abaneh data than just another grad student’s thesis. Hughes didn’t mention Ababneh at AGU or her results (which had been published by that point.) Given that the Sheep Mountain data is the most important item in the MBH reconstruction, Hughes’ failure to confront the Ababneh results is highly questionable. As is Briffa’s failure to report on the updated Polar Urals data. And what’s happened to the publication of the Grudd update of Tornetrask. These are important data sets that are used in every reconstruction. If people are going to be self-righteous about Courtillot, then be consistent. I’m consistent about this: Courtillot should have used properly documented data but so should everyone else. You’re being inconsistent in your application of the rules.

Steve is concerned with setting and applying standards. Consistently and scrupulously. I think that is something worth fighting for.

” … The author proposes that the Biffra data was truncated deliberately to avoid revealing some ‘truth’ that the current climate is no warmer than the past or that we can’t tell what the current climate is compared to the past. And then carries on to claim that this is deliberate deception and that no clear reason is given as to why they did it.”

An important thing about Briffa’s work is that it fails to confirm the recent upswing evident in other series. That alone should send it straight to the front of the queue. It is far more interesting than just another series which shows the same pattern as others.

Despite what you say, the authors appear to have attempted to cover this up by truncating the most interesting part of the Briffa series. A reaonable mind would wish to understand why.

If it was a simple mistake, why not publish an erratum showing the graphic with the Briffa series in full? Easy enough to do on the internet these days.

“IPCC clearly states that … they’ll have to keep looking at it, but they can’t at the moment because of some problems with data collection.”

Are you suggesting that the IPPC has slipped from passively reviewing peer reviewed material into an active role as a reviewer?

If there was any particular doubts about Briffa, surely it ought to be emerging from the normal peer review procedures. If so, the correct approach would include the full Briffa series in the graphic and add a footnote with respect to Briffa. That would have fairly represented “the state of the science”.

“Why do people persist with the conspiracy theory?”

Why do you try to play-down important questions?