Muscheler versus Solanki

realclimate has posted up a discussion of a recent Brief Communications Arising in Nature by Muscheler et al., commenting on Solanki et al. [2004]. They haven’t posted up the Solanki et al. reply, which argues that Muschelerr et al. have screwed up their normalization. However here it is . There are two points of interest to this: one substantive and one bitchy.

I’ve not looked at C14 and Be10 before, but it looks interesting and I’ll try to make some time to look at it. The C14 data comes from tree rings; there are some interesting aspects to C13/C12 fractionation in tree rings that I’ve read about in connection with CO2 fertilization and water use efficiency. I presume that the C14/C12 fractionation would be similar and even more so. Whether it affects anything in these calculaitons would take some time to find out.

Solanki et al.[2004] argued that solar activity is anomalously high in the late 20th century, although they disclaim the idea that high solar activity could have caused climate change. Muscheler et al. [2005], based on different C14 calculations, argue that solar irradiance was as high ~1150 and ~1600. I am unfamiliar with these calculations and have no views on who is right. A connection that intrigues me: remember that IPCC [1990] was unable to apportion the current warming between anthropogenic and natural variation, because it believed that there had been equal changes in the past as in the last century without AGW impact. IPCC [1995] overturned this position, relying heavily on Wigley and Kelly [1990], who stated that there had not been equivalent changes in the past 10,000 years. Wigley and Kelly [1990] relied on C14 levels, which were relatively new then. If there are substantive differences between Muscheler et al. on the one hand and Solanki et al on the other, I wonder if they can be both reconciled with the Wigley and Kelly [1990] position relied upon by IPCC [1995] and whether that argument, already weakly argued in Wigley and Kelly [1990], has any remaining legs.

Now for a purely personal bitch. Muscheler et al. is classified by Nature as a "Brief Communications Arising" and is about 1525 words with two figures. Our submission last year was also treated as a "Brief Communications Arising". After we got positive reviews in our first submission (undoubtedly to Nature’s consternation) and a revise-and-resubmit, we were told to shorten to 800 words as follows:

26th Mar 2004
Dear Mr McIntyre
Thank you for your revised manuscript entitled "Global-scale temperature patterns and climate forcings over the past six centuries: A comment" that is under consideration for our Brief Communications Arising section.

Before we can proceed further, I am afraid that it will be necessary for you to shorten your manuscript substantially, in accordance with our author guidelines on http://www.nature.com/nature/submit/gta/index.html#8. You will have seen that submissions to this section of the journal have a strict length limit (up to 800 words, with one multipanelled figure and no more than 15 references), although you are welcome to include supplementary material for reviewing purposes only.

We shall of course impose the same length limits on Mann et al. when we ask them to revise their reply in response to your revised manuscript and the referees’ comments….

We hope to see your revised manuscript within two weeks; please let us know if the delay is likely to be longer.

Yours sincerely
Rosalind Cotter
Editor, Brief Communications

It was hard to condense our position to 800 words and we had to take out a lot. Then, in August 2004, we were told that our article could not be condensed into 500 words/1 figure. The referee comments also mentioned word length, so it sure looks like Nature specifically put the referees on notice that word length was an issue to be taken into consideration in reviewing the article. Our GRL article was not simply a "re-submission" of our Nature article as has been misrepresented by Bradley and others The simulations showing the generation of the hockey stick from random data were re-done using more persistent red noise and much more dramatic similarities; we quantified the tendency to yield hockey sticks (through one-sigma and 1.5-sigma hockeysticks) and had an entirely new argument on spurious RE statistics. The GRL article is better than our Nature submissions, in part because it deals with responses and arguments raised by Mann et al. in our correspondence (e.g. the RE statistic.) Here’s the Nature rejection notice, mentionoing 500 words:

Sent: Tuesday, August 03, 2004 11:11 AM
Subject: Decision on 2004-01-14277B
Dear Mr McIntyre

Thank you for your revised comment on the contribution by Mann et al., which
I am afraid we must decline to publish. As is our policy on these occasions,
we showed your revised comment to the earlier authors, and their response is
enclosed. We also sent the exchange to 3 referees, whose comments are
attached.

In the light of this detailed advice, we have regretfully decided that
publication of this debate in our Brief Communications Arising section is
not justified. This is principally because the discussion cannot be
condensed into our 500-word/1 figure format (as you probably realise,
supplementary information is only for review purposes because Brief
Communications Arising are published online) and relies on technicalities
that do not bring a clear resolution of the underlying issues.
….
Yours sincerely

Rosalind Cotter
Editor, Brief Communications

I suppose that the word limit could be construed as them just being polite, but they reject hundreds of articles a year and, unless this is a standard formula, you’d have to take them at face value and assume that word length actually was a consideration. So it’s irritating to see Muscheler et al. being given 1500+ words and 2 figures.

25 Comments

  1. John A
    Posted Aug 5, 2005 at 2:14 PM | Permalink

    Yes, but Muscheler et al were defending the AGW theory that was being pushed over by Solanki et al – so they must be given priority. You on the other hand were being critical of another key piece of that theory, the one that is constantly used when linked to another suspiciously unaudited hockey stick called the CO2 curve for the last 1000 years.

    I think the behavior of Nature is a disgrace to science, and in that view I have very distinguished company.

  2. Dave Dardinger
    Posted Aug 5, 2005 at 3:57 PM | Permalink

    And why in the world is it important to hold to strict word lengths if something’s being published online in the first place? It’s not like they need to be worried about protecting trees and their precious rings, is it?

  3. Steve McIntyre
    Posted Aug 5, 2005 at 4:04 PM | Permalink

    They do refer to 500 words/1 figure or 700 words/0 figures here, but they waived this for Muscheler, I guess.

  4. Posted Aug 7, 2005 at 3:22 AM | Permalink

    Thanks for the link to the comment of Solanki (which was subscriber protected and the Nature issue was not yet in the library). I have the impression that Solanki has adequately responded to Muscheler.
    Yesterday late evening, I sent some comment to RealClimate, but this morning, it seems that the whole discussion disappeared from their web site!

    Here follows my comment to RealClimate:

    There is some discrepancy between the tree ring 14C data used by Usoskin ea. and the calculated 14C production rate by Muscheler in the peak values of around 1780, in the same period as the discrepancy between calculated 14C production and group sunspot number is largest. Neither is there any important change visible in the 10Be Antarctic Taylor Dome data from Steig ea. for that period.

    I was looking for the comments on the different articles, but most of them are subscriber protected (and the library here has not yet the latest Nature issues).

    Further, I don’t see a direct relation between DNA and climate (as now deleted in comment #1), but the link to the article of Tinsley is interesting, as he proposes three different ways that solar activity may influence cloud cover and hence climate.

    At last, I disagree with the last paragraph. While there is no increase in different solar activity indices for the last decades, its activity now is continuous higher than up to 1935, when there was a fast warming. As far as I remember, in climate models, an immediate doubling of CO2 leads to an immediate increase in temperature plus a long tail of additional increase over longer term. I don’t see why this shouldn’t be applicable for the longer term (oceanic) effect of continuous high solar activity.

  5. Chas
    Posted Aug 7, 2005 at 5:39 AM | Permalink

    Ferdinand; Is the relative size of Muscheler’s 1780 spike more down to a reduction of the present C14/ solar activity [through modern dilution and a reducing geomagnetic field] rather than an increase in the 18C activity? -He is seeking a proxy for irradience from ‘up there’
    I probably have this horribly wrong, though!

  6. Steve McIntyre
    Posted Aug 7, 2005 at 5:46 AM | Permalink

    As Ferdinand pointed out, the entire segment has disappeared from realclimate. I wonder why.

  7. Greg F
    Posted Aug 7, 2005 at 5:56 AM | Permalink

    Do a Google search using:

    Federal Reserve Bank Working Paper on Replication

    It should show 1 hit, you can then use the Google cache.

    Show Google’s cache of http://www.realclimate.org/index.php?p=178

  8. John A
    Posted Aug 7, 2005 at 6:51 AM | Permalink

    I blame global warming.

  9. Posted Aug 7, 2005 at 7:02 AM | Permalink

    Chas,

    As far as I know, correction for geomagnetic variations are already included in the tree-ring 14C data series.
    If I interprete the reaction of Solanki well, Muscheler indeed uses the dilution of 13C/12C ratio to correct 14C for the increase of CO2 by fossil fuel use. But Solanki points out that this is not completely correct, as land use change also affects 13C/12C ratio and 14C/12C ratio in a different way. Thus the (linear) correction factor that Muscheler used over the whole period is too small. Further, the correction factor is scaled against cosmic ray flux, measured in ionisation chambers, which had a large drift in the first period used for scaling and don’t reflect the real cosmic rays intensity. Both errors lead to an overestimate of solar activity in the pre-1950 period…

  10. Chas
    Posted Aug 7, 2005 at 9:42 AM | Permalink

    Ferdinand; as I imagine it, if geomagnetic variations are used in producing the C14 series they would be used to tune it, whilst Muscheler would be trying to strip out variation caused by the earths field so as to get to the activity on the sun [as a luminosity proxy].
    The bit I cant fathom is why Gavin’s plot of Muscheler’s C14 production series shows no 1780 spike ,whilst Muschelers plot of his C14 production series in the Nature Matters Arising [Fig 2], does.
    [ http://www.cgd.ucar.edu:80/ccr/raimund/publications/Muscheler_et_al_Nature2005.pdf ]

    Maybe M gave Gavin the wrong dataset?

  11. Chas
    Posted Aug 7, 2005 at 9:48 AM | Permalink

    [Gavin’s plot of M’s C14-production series is in ‘The lure of solar forcing’ ]

  12. Posted Aug 7, 2005 at 12:29 PM | Permalink

    The Commissar (er, blog, data) vanishes re #4, 6, 7 above.
    I find the trimming of MBH related data and blogs a habit more than annoying.

    Google cache has RC page 178 plus comments 1 – 4. MSN Search cache has the RC page plus comments 2 – 11. I infer a missing 20-30 hours of comments after RC, p178, #11 on 8/5 to some time after the submission that FE mentions in #4 above, on the evening 8/6. Does anyone have any missing comments after RC p178, comment #11. I would appreciate a fair use posting of any of the comments after RC p178, #11.

    RC p 178, Tinsley link in #4 above, and Solanki are drifting into the geological and astrophysical discussion vs CO2 discussion that TAR skipped. This is an area of interest to me from 15 years ago, dormant these last 10 years. Just when RC p178 comments’ give and take are warming up, poof. What seems to have aroused such sensitivity from RC or Raimund Muscheler?

    It’s nice to see some of the more recent papers in the geological+astrophysical area available on the internet now. One thing I don’t notice much on is the ionspheric modulation of light/energy transmission associated with the earth’s ebbing magnetic field/pole reversal and sunspots/solar storms.

  13. Chas
    Posted Aug 7, 2005 at 1:31 PM | Permalink

    I can only imagine that RC checked Muscheler’s calcs in the light of Solanki’s reply and found Solanki correct
    I hope that it doesnt put RC off
    BTW Muscheler has a rather interesting article:

    Click to access Muscheler_et_al_ISCS2003.pdf

  14. Posted Aug 7, 2005 at 2:15 PM | Permalink

    Bruce, I did send my comment this morning 7 August around 2 AM (CET), at that moment the page still was there, up to comment #11. At 9 AM the page was gone…

    Just read that the question “why” is posted on RealClimate as comment #51 in the “record breaking” topic, no answer yet, but the reason indeed may be the reply of Solanki, as Chas said.

    Fascinating…

  15. ClimateAudit
    Posted Aug 7, 2005 at 2:39 PM | Permalink

    The Solanki reply said that Muscheler erred in their normalization. Maybe that struck too close to the bone for Mann and his PC methodology and he wanted the whole topic of disputes over normalization off the air.

  16. Posted Aug 7, 2005 at 2:58 PM | Permalink

    Chas,

    I didn’t see any 14C spike in Solanki’s graph at PRL (page 3), but if you look at the similar graph in Nature (page 2), you will notice that the uncertainty range around 1870 is very wide (thus there a spike may be hidden). But no spike in the 14C Solanki data around 1780, neither in the 10Be data of different ice cores…
    If you read the Solanki paper in Nature, they explain that they corrected the raw 14C tree ring data for the (large) change in geomagnetism, before applying the data in a model that translate that into group sunspot numbers. The variance of the 14C data attributable to the solar magentic field is small compared to the change in geomagnetic field, but this changes rather smoothly and slowly…

    Thus I have no idea where the Muschler 1780 14C spike is based on. But the article he wrote indeed is interesting.

  17. Chas
    Posted Aug 8, 2005 at 12:58 AM | Permalink

    Re #13
    Its back!!! ( a tech glitch )
    -so I was wrong

  18. Bruce Alan
    Posted Aug 8, 2005 at 9:42 AM | Permalink

    re:The Commissar (er, blog, data) vanishes
    RealClimate p178 has returned as p180 with comments 5 – 11, EF’s submission (#4, above) missing, and a 98 hour correspondence gap appears, where previously the last 6 comments posted came in 8 hr. The MSN cache on p178 is now emptied. Comments #6, 8, 11 were not favorable to RC. #6 complained about data selection on the Medival Warming Period, mike and gavin condescends, threatens deletion.

    However I do have the direct MSN link for RC p178:
    http://cc.msnscache.com/cache.aspx?q=2128203089103&lang=en-US&FORM=CVRE
    While not especially enlightening, I was interested to see the dialogue where the 98 hr hole was headed.

    The strange parts of comment #1, previously edited out, have returned, and a new #6 – to make rabid looking strawmen?

    I really don’t like how RC does business. Selection, deletion and control – great Soviet medicine. As I wrote earlier, I really wanted to see some discussion of geological-astrophysical effects (with linked references). This is debate with your finger on the scale. The real problem is resource based, these guys have government monies, cheap student labor and scads of unpublished/archived results so they can snow or blindside any critic. They’re paid to argue – just like lawyers except they can keep the pot, too.

    Perhaps RC will criticize Climate Audit’s comical struggle with the spam filter but that looks entirely different.
    sign me – B Annoyed

  19. TCO
    Posted Sep 21, 2005 at 8:12 PM | Permalink

    Steve were you Charlie T? Or anyone else on this threa different on this thread or the RC thread? (Not trying to be all Lambert, just wondered…)

  20. John A
    Posted Sep 21, 2005 at 9:03 PM | Permalink

    Bruce Alan,

    Despite its claims, RC is not a scientific blog, it’s a political advocacy blog run by climate modelers. I stopped bothering a long time ago when RC would selectively delete perfectly reasonable questions and comments without any explanation.

  21. TCO
    Posted Sep 21, 2005 at 9:22 PM | Permalink

    I think they would post stuff by Steve if he went there (under his own name of course). Steve has a bit more credibility as a critic than any of us. He would have to avoid the sarcasm born of frustration (from this site), but I think they would publish his stuff. If not, he could post here and it would look pretty embarrasing when anyone could see the snip…

    Ps. where did cytochrome and per go? Was cytochrome, Steve?

  22. Steve McIntyre
    Posted Sep 21, 2005 at 9:41 PM | Permalink

    Not me. I should have asked them about R2’s and things like that. They would never have posted them. Maybe I will the next time, thier discussion verges that way.

  23. TCO
    Posted Sep 21, 2005 at 9:47 PM | Permalink

    I think you should post there. While they have an editorial heavy hand, they would likely allow you a little more slack as a principle in the kerfuffle.

    Actually they’ve never edited me.

  24. Dave Dardinger
    Posted Sep 21, 2005 at 10:24 PM | Permalink

    I think all my posts there have been left alone too. But I try not to be insulting when I post there. I just wish they had some interesting threads. The threads they do have seem designed to stay as far away from hockeysticks as they can figure out how to do.

  25. TCO
    Posted Sep 21, 2005 at 10:35 PM | Permalink

    They could always come here. Mano y mano stats competition. no fair sneaking an ARMA function into the cage match…