Wegman (and Rapp) on tree rings: A divergence problem (part 1)
Posted on December 22, 2009 by Deep Climate| 68 Comments
Edward Wegman
There has been renewed interest in the Wegman Report, which purported to critique the work of paleoclimatolgists Michael Mann, Raymond Bradley and Malcolm Hughes and their controversial “hockey stick” millennial temperature reconstruction.
Today we’ll take a closer look at Wegman et al’s key passage on tree-ring proxies and do a detailed side-by-side comparison with its apparent main antecedent, chapter section 10.2 in Raymond Bradley’s classic Paleoclimatology: Reconstructing Climates of the Quaternary.
That comparison leaves no doubt that Wegman et al’s explication was substantially derived from that of Bradley, although the relevant attribution appears to be missing. There are, however, several divergences of note, also in the main unattributed, and some of Wegman’s paraphrasing introduces errors of analysis.
But the real shocker comes in two key passages in Wegman et al, which state unsubstantiated findings in flagrant contradiction with those of Bradley, apparently in order to denigrate the value of tree-ring derived temperature reconstructions.
My first post on the Wegman Report showed that all of Wegman’s proxy explication had been incorporated almost verbatim into Donald Rapp’s text book Assessing Climate Change.
But I also noted that the opening passage of Wegman’s tree ring explication, starting at p. 13, was mainly derived from Bradley. So I decided a detailed analysis of the entire section might be revealing. (For purposes of this discussion, Wegman and Rapp are essentially identical; interesting “divergences” between Wegman and Rapp will be discussed in part 2.)
To perform the analysis, I downloaded the two documents (linked at the end of this article), found all the relevant passages in Bradley, and placed the two passages side-by-side in a single PDF document. The text has been reformatted into sub-paragraphs for ease of direct comparison; each page of the comparison corresponds to one paragraph from Wegman et al (three in all). The text has been formatted as follows:
Regular font for sentences with substantial wording in common between Wegman and Bradley
Italic for Bradley material paraphrased in Wegman
Bold for significant departures in Wegman with respect to Bradley
Bold italic to indicate passages where Wegman and Bradley contradict each other
With that material in hand, let’s begin.
As noted in my previous post, Wegman’s first paragraph is very close to Bradley, with near-identical wording dominating, along with some close paraphrasing. One of the paraphrased sections introduces an error of perhaps some consequence.
Bradley states:
However, optimum climatic reconstructions may be achieved by using both ring widths and densitometric data to maximize the climatic signal in each sample (Briffa et al., 1995).
Wegman’s version implies the practice of combining indices is the norm rather than specific to particular situations:
Both tree ring width and density data are used in combination to extract the maximal climatic temperature signal. [Emphasis added]
Also as noted in my previous post, one departure is the addition of carbon dioxide at the end of the list of “climatic factors”, which seems misplaced. Another, immediately following, is the insistence (and repetition) of the role of “confounding factors”:
Obviously there are many confounding factors so the problem is to extract the temperature signal and to distinguish the temperature signal from the noise caused by the many confounding factors. [Emphasis added]
Bradley’s original uses a more conventional signal-noise formulation:
The problem facing dendroclimatologists is to extract whatever climatic signal is available in the tree ring data and to distinguish this signal from the background noise.
Wegman’s second paragraph, covering Bradley’s discussion of site and sample selection, contains more paraphrasing and less of the same direct wording. There is, however, one pair of sentences that are very close, but with a crucial difference. Bradley has:
Trees growing near to the latitudinal or altitudinal treeline are mainly under growth limitations imposed by temperature and hence ring-width variations in such trees contain a strong temperature signal.
Now here’s Wegman, with a small, but significant change:
Trees growing near to their ecological limits either in terms of latitude or altitude show growth limitations imposed by temperature and thus ring width variations in such trees contain a relatively strong temperature signal. [Emphasis added]
Is any comment necessary?
Wegman concludes the paragraph with a passage about the effects of carbon dioxide and nitrous oxide on tree rings. The first is attributed to Graybill and Idso (1993), while the second is unattributed. The relevance is unclear, to say the least, since the discussion concerns varying climatic conditions at that affect tree ring growth at different sampling sites. It may be that this topic is introduced in order to cast doubt on the value of tree ring reconstruction.
The final paragraph discusses standardization, as well as touching on calibration. Wegman starts out:
Wider rings are frequently produced during the early life of a tree. Thus the tree rings frequently contain a low frequency signal that is unrelated to climate or, at least, confounded with climatic effects such as temperature. In order to use tree rings as a temperature signal successfully, this low frequency component must be removed.
This is not completely wrong, although the odd usage of a “signal” that is “confounded” with climatic effects seems backward, and the description does not clearly state that we are speaking here of an individual tree core sample. Bradley is much clearer, referring to the removal of “the growth function peculiar” to a “particular tree”.
Perhaps this confusion explains the very next sentence in Wegman (emphasis added in this and following quotes):
Because the early history of tree rings confounds climatic signal with low frequency specimen specific signal, tree rings are not usually effective for accurately determining low frequency, longer-term effects.
Nowhere does Wegman refer to Bradley’s explanation of “regional curve” standardization, whereby a “regional curve” is derived in order to preserve low frequency climatic information, in contradiction of the assertion above.
… The resulting “regional curve” provided a target for deriving a mean growth function, which could be applied to all of the individual core segments regardless of length (Fig. 10.13). Averaging together the core segments, standardized in this way by the regional curve, produced the record shown in Fig. 10.12b. This has far more low frequency information than the record produced from individually standardized cores (Fig. 10.12~) and retains many of the characteristics seen in the original data (Fig. 10.12a).
Wegman’s final passage discusses calibration of the proxy to the instrumental record, before pointing to Bradley (finally) for further information. Just before that, though, we have this extraordinary, unattributed (and unsubstantiated) statement:
As pointed out earlier, many different sets of climatic conditions can and do yield similar tree ring profiles. Thus tree ring proxy data alone is not sufficient to determine past climate variables.
It’s hard to know what to make of that statement. The most charitable explanation is that it is the trivial observation that a tree ring proxy must be calibrated to the temperature record over some portion of the chronology to provide a useful reconstruction. But this calibration would still be required even if tree rings only responded to temperature and not to other “climatic conditions” (which apparently even include CO2 and NO2).
So, in context, Wegman appears to imply that tree ring proxy data must be supplemented by other proxies.
Bradley, as one might expect, draws the completely opposite conclusion.
If an equation can be developed that accurately describes instrumentally observed climatic variability in terms of tree growth over the same interval, then paleoclimatic reconstructions can be made using only the tree-ring data.
What’s next - a finding that black is, after all, white? One can only be thankful that a discussion of the albedo effect was beyond Wegman’s scope.
[Update, Dec. 24: Lest readers imagine that problems in the background section on proxies have no impact on Wegman's findings, here is one key finding (number 7 on p. 49):
7. Our committee believes that the assessments that the decade of the 1990s was the hottest decade in a millennium and that 1998 was the hottest year in a millennium cannot be supported by the MBH98/99 analysis. As mentioned earlier in our background section, tree ring proxies are typically calibrated to remove low frequency variations. The cycle of Medieval Warm Period and Little Ice Age that was widely recognized in 1990 has disappeared from the MBH98/99 analyses, thus making possible the hottest decade/hottest year claim. However, the methodology of MBH98/99 suppresses this low frequency information. The paucity of data in the more remote past makes the hottest-in-a-millennium claims essentially unverifiable. [Emphasis added]
This finding apparently rests at least in part on the fundamental misunderstanding, or at best an oversimplification, of tree-ring proxy standardization methodology (as well as an apparent confusion concerning previous knowledge about the MWP).]
Clearly, the problems I have exposed here go well beyond lack of proper attribution. Wegman et al have followed closely Bradley’s exposition, but have still managed to introduce mistakes and even gross distortions.
That such a shoddy misrepresentation of another author’s work has been used as part of a baseless, politically motivated attack on that author is beyond shameful. Perhaps it is time to interrupt the incessant braying about so-called Climategate, and examine a real outrage for once. “Sound science” indeed!
===========================
[Correction and reversion, Dec. 24: The accidentally missing word "signal" has been restored in the first quote from Wegman. Commentary has reverted to the original post, as Arthur Smith's criticism was based on a misunderstanding engendered by the missing word. Thanks to Michael Smith for pointing out the error.]
===========================
References:
Ad Hoc Committee Report on the ‘Hockey Stick’ Global Climate Reconstruction, Edward J. Wegman, David W. Scott and Yasmin H. Said. [PDF]
Paleoclimatology: Reconstructing Climates of the Quaternary (Second Edition), Raymond S. Bradley (Academic Press, 1999) [Link] [Chapter 10: Google cache]
Wegman-Bradley Tree Ring Comparison [PDF]
deepclimate.org |
