“Delayed Oscillator” on divergence

Anyone who has blogged, or even just hangs around blogs more than they should, is familiar with the “pingback”. You know, it’s that automatically generated comment that signals that another blog has referred to a particular post. Yesterday, I got this “pingback” from a blogger called Delayed Oscillator (or “delayed.oscillator” as it is formatted there) and decided to follow it up.

And am I ever glad I did. DO (pronounced Dee-Oh), as I will call this blogger (I hope that’s OK!) has brought a welcome expert perspective to the discussion of the Steve McIntyre-Keith Briffa controversy, said by certain economists and business section editors to expose the global warming sham once and for all. In a two-part series of posts, DO shows why it “ain’t necessarily so” (to say the least).

In the first post, DO sets the stage:

Steve McIntyre has once again stirred the hornet’s nest of online climate change denial with a hasty modification of the Yamal tree ring data published by Keith Briffa and colleagues in 2008 as part of a paper in Philosphical Transactions of the Royal Society (Phil. Trans. R. Soc. B (2008) 363, 2271–2284). Normally, I ignore McIntyre’s blog because of the juvenile name calling, repetitive nonsense, and the general misunderstanding of huge swaths of proxy paleoclimatology. However, I knew when Roger Pielke Jr. jumped in with support for his collaborator, it merited some attention [insert smiley face emoticon here].

[Update, Oct. 6: I’ve added McIntyre’s plot and DO’s zoomed plot and clarified sources of differences between DO and McIntyre versions.]

DO emulates McIntyre’s combination of Briffa’s original Yamal site data with the modern, relatively short series from Khadyta, and obtains the following chart, using a standard implementation of RCS (Regional Curve Standardization).

[Correction, Oct. 6: Here is McIntyre’s version (the green line is the combined version). McIntyre’s merged version apparently omits Yamal period from 1991-1996.]:

DO comments:

… the actual impact on the chronology is still far less than being implied by non-scientist partisans on one side. Why is that?

Part of the difference appears to be McIntyre’s use of a 21 year Gaussian low pass filter … one influence of the filter is such that it helps create the appearance of a massive rise, when annual values in mid-century are actually similar to those in the late 20th century.

In other words, the very end spike of Yamal’s “enormous HS blade”, said by McIntyre to be like “crack cocaine” for paleoclimatoligists, is quite dependent on choice of smoothing. As noted by DO in comments below, the apparent 21-point Gauussian smoothing with end-point “reflection” exaggerates the upturn at the end.

DO’s “zoom in” shows how the filter exaggerates the final decade relative to the mid-20th century.

[Update, Oct. 6: Here is McIntyre’s version excluding recent so-called “picked” Yamal trees in the “Schweingruber variation”, as displayed at Tim Lambert’s Deltoid post on the subject:

The “Schweingruber Variation” series excludes what McIntyre calls the “12 picked trees” from Yamal in the combined series, and substitutes the 34 cores from the Khadyta site – a choice that really begs the question of whether these two series should be considered part of the “same site” in the first place. (Most of Khadyta only goes back to the 1700s, so the rest is all Yamal in all variations). This was also the only version shown at James Delingpole’s “MASSIVE lie” piece, with a comment that didn’t even mention that samples had been removed from the combination, or that the series was really a grafting of series from two different sites.

When finally McIntyre plotted in a much larger and more representative range of samples than used those used by Briffa – though from exactly the same area – the results he got were startlingly different.

The scary red line shooting upwards is the one Al Gore, Michael Mann, Keith Briffa and their climate-fear-promotion chums would like you to believe in. The black one, heading downwards, represents scientific reality.

I don’t think so.]

In the second post, DO explores the “divergence” problem and compares the two series to actual temperature as recorded in the appropriate CRUTEM temperature gridcell. But first, there are a few caveats to get out of the way (gee, just like a real scientist). Example:

The problem with treating tree ring chronologies as nothing more than received time series downloaded from the internet to be manipulated in various ways is that the context of the original investigators can be lost. Moreover, there are several subdisciplines within dendrochronology that collect tree ring data for different reasons, and in their fieldwork emphasize different site or individual tree characteristics during sampling.

Now here’s the key chart from DO, comparing the two series to the CRUTEM temperature gridcell at [65-70N, 65-70E]:

As can be clearly seen, Yamal tracks temperature quite well, while Khadyta suffers from a noticeable divergence problem from about 1970 on. And keep in mind that the two series were quite similar up to that point. DO’s conclusion:

… my quick review of these data here shows that including Khadyta River raw data in the Yamal chronology does not result in a more accurate nor precise understanding of past temperatures in the region. This isn’t to say that some time in the past that Yamal didn’t experience divergence (this after all is a large part of the concern about divergence), but we can clearly see that Khadyta River does exhibit modern divergence. [Emphasis added]

[Update, October 6: I would be remiss if I did not point out DO’s parting gift – a new paper in Global Change Biology from Esper et al, entitled “Trends and uncertainties in Siberian indicators of 20th century warming” (available online). From the abstract:

Despite these large uncertainties, instrumental and tree growth estimates for the entire 20th century warming interval match each other, to a degree previously not recognized, when care is taken to preserve long-term trends in the tree-ring data. We further show that careful examination of early temperature data and calibration of proxy timeseries over the full period of overlap with instrumental data are both necessary to properly estimate 20th century long-term changes and to avoid erroneous detection of post-1960 divergence. [Emphasis added]

Obviously a paper of great interest for all working in this field. I wonder if McIntyre has read it yet. h/t to Hank Roberts for pointing out the paper.]

DO’s posts are very readable and clear, and are recommended to anyone who wants to understand better the serious scientific issues behind the latest attack on the “hockey stick”. It’s all there – logical progressive exposition, clear charts and appropriate scientific references. It’s a pleasure after spending too much time wading through the rhetorical swamps at ClimateAudit. Well done!

Oh, one last thing. In case you’re wondering about the name of the blog, it turns out that a standard theory of the essential physical mechanism behind El- Niño Southern Oscillation (ENSO) is known as “delayed oscillator”. I don’t think that’s a coincidence, but I’m open to other theories. And maybe DO will come by and set us straight on that too.

[Update  Oct. 7: DO has a third update and summary post and has promised a future post:

However, reading blog posts about this topic, I’ve become aware that there persist misunderstandings in how field and lab dendrochronology is actually done. So, hopefully there will a post about this from me in the near future.

We hope that Steve McIntyre will be reading, and learning from, Delayed Oscillator.]


14 responses to ““Delayed Oscillator” on divergence

  1. Great Post DC. Can’t speak to the details/specifics, but if what DO did is correct, then McIntyre’s hypothesis is in trouble.

  2. delayedoscillator

    Hi Deep Climate,

    Thanks for the links, and for the very readable summary of my posts.

    One clarification — McIntyre’s RCS program appears to perform similarly to the standard software (although I haven’t had time to spin up R and take his code for a drive yet), but I think any differences will probably be minor.

    However, the rather dramatic late 20th century rise shown in the graph (reproduced at Deltoid), is at least in part due to the choice of the 21 Gaussian filter with reflected endpoints — now, objectively smoothing series is a non-trivial task, but it does bear keeping in mind that what McIntyre has graphed represents only on particular way of doing so, and one that tends to emphasize the end of the series. If one use all the Yamal data, the two chronologies reconverge at the end, but this is because one is left with only the original Yamal trees between 1991 and 1996.

    Oh, and yes, the name of the blog isn’t a coincidence, you’ve got its origin right

  3. Interesting post DC. Have added you to my must read list.


  4. Andrew Dodds

    It’s almost as if what McIntyre really needed was some mechanism by which experts could critique and help improve his results before they were publicly presented – like, perhaps, a review by his peers?

    Anyone know of such a process?

  5. Not being a scientist or mathematician, this was a bit of a tough blog war to slog through. It seems Andy Revkin was confused too – he aired some of McIntyre’s grievances over at DotEarth, and then defended the decision by saying he had to give McIntyre a fair chance to defend himself. So getting some clarifications from DeepClimate on the last post, I’ve tried to distill the Briffa controversy to a simple narrative with some key takeaways for laypersons like myself. In particular, McIntyre’s deceptive insinuations and rhetorical questioning are strikingly similar to tactics used by the Right in America’s health care debate.


    I don’t have the training to comment on the actual science (fortunately, McIntyre not being a scientist either, I have no problem questioning him) – so I’d love a little “peer review” to make sure I got my story straight.

  6. DeepClimate, you are embarassing yourself with this post. Take another look at what Steve posted, just the Yamal divergence thread, then DO’s thread, and your own commentary. Major major mistake. Maybe ask Tamino for advice, since he’s good at finding flaws.

    [DC: Tamino’s got better things to do, I’m sure. But I think I fixed the mistake – I’ve shown both of McIntyre’s charts now.]

  7. Pingback: Yamal III: Summary and Update « delayed.oscillator

  8. Pingback: Let the backpedalling begin « Deep Climate

  9. OK, so there is some hope for you. I went to his post thru your site, and your bad graph made me misidentify the problem with his work. I assumed it was the change in RCS.

    The zoomed chart would be a bit better. That’s what made me realize, he hasn’t said anything of significance. It’s the same chart.

  10. Sir Rupert Studwell

    “McIntyre’s merged version apparently omits Yamal period from 1991-1996.”

    No s**t.

  11. Here’s what Steve knew in 2006 from his website:

    which is that Briffa’s RCS produced a hockey stick whereas the RCS did not. He had to wait until now to get the raw data. Here is the 2006 thread from which the graphic above came :

  12. Sorry meant to say the RCS produced a hockey stick, but the mean ring width did not…

  13. Sorry meant to say the RCS produced a hockey stick, but the mean ring width did not…

    Just mean ring width without adjusting for age-related changes is growth?

    That would be meaningless but wouldn’t surprise me.

  14. Pingback: Enron and the Case of the Zombie Fungus « The Policy Lass