Another large batch of tree ring chronologies was archived on June 21, 2006, this time with records up to 2003. Are the results for the 1990s and 2000s off the chart?
On June 21, Connie Woodhouse, Stephen Gray and David Meko archived 44 measurement data sets and site chronologies from Colorado, discussed in Woodhouse et al 2006, at WDCP. The data was published in Woodhouse et al 2006; it is gratifying to see such a thorough and prompt archiving. Woodhouse is a NOAA employee. (On a previous occasion, I’d expressed frustration about non-archiving in connection with communications her, but that pertained to an article in which she was a co-author with a Hockey Team author, who had control over non-archiving. Here she seems to have had jurisdiction, rather than a Hockey Team author, and commendably archived promptly.)
The chronologies covered a range from 1126 to 2003 (not all chronologies covering the full range.) The detrending method was described as follows:
Detrend method: Negative exponential or linear regression of any slope, or 2/3-length spline. Detrending was done interactively, and if a negative exponential or linear regression fit poorly, a 2/3-length spline was used.
If you are familiar with non-linear methods, you might feel that this is hardly an adequate description of methodology, but this is an expansive methodological description by dendro standards.
A simple average of the 44 chronologies is shown below. Personally I don’t see a loud signal in the 1990s for this data set.
Unweighted average of co589.crn to co632.crn
At Jean S’s request, I did a Mannian PCA on the data. Now you don’t always get a HS from Mannian PCA (as we’ve mentioned before) . For example, if you take out the bristlecones from Mann’s network, you don’t get a HS even using the strongly data mining mannomatic – we menitoned this in one or more of our articles. In Mannian PCA, mining for a HS competes with an inherent signal. Thus, if there actually is a signal in the data, either Mannian PC or regular PC will both get similar results. In the von Storch and Zorita case, there is a very strong signal (in paleoclimate terms) and either method looks about the same (as does the mean). Also for data mining, the larger the network, the more series to mine for (so the instability increases with the number of series, rather than being reduced.) Anyway the PC1 doesn’t have an HS, the PC4 shown below has an HS. If this went into the regression stage (mannomatic part 2), the PC4 of this network would probably be enough to impart a HS to the reconstruction. With all the attention on the PCs, the other part of the mannomatic – spurious regression – sometimes gets lost.
PC4 from co589.crn to co632.crn
Climate Audit 
22 Comments
There also does seem to be much of the LIA in these data. Given the evidence for that event in many records and proxies (as just asserted once again by the NAS panel) one has to wonder about the de-trending methods used here, or again question whether tree rings are reflecting tempertaure variation alone or whether any relationship they have to temperature is linear. Nevertheless, one interesting feature of the twentieth century record here is that the peak in the early part of the century exceeds the peak in the latter part. My understanding of the surface records is that this is also true of remote and particularly polar locations. The surface measurements give a higher peak at the end of the century only if UHI-contaminated urban sites are included. It would be interesting to see what rural locations in Colorado near where these samples were taken show for twentieth century temperature (and precipitation) trends.
There is also a peak around 1900 which is indicated by polar explorers descriptions of sea ice conditions around the turn of the last century.
I went to that link, at WDCP, but how do I find the specific deposit by woodhouse? Also, citation (or pdf) for the woodhouse paper?
On the science was this lower or upper treeline? How did the paper compare to the results?
I would guess no new methodology is needed, Mannian PCA will do just fine (please, plot also Mannian PC1!).
So there should be good fodder in publishing papers that “reconstruct” w/out bristlecones. Wait, we already have one. CENSORED!
Mark
Actually, what I would rather see, sans bristlecones of course, would be a reconstruction with tracking. I.e. an attempt to track the non-stationarity of the data. Is the degree of non-stationarity > 1 and therefore not trackable? Or do the other proxies have enough stationarity that they could be tracked? I haven’t run any numbers but I bet a windowed sample variance would be rather erratic over time.
Mark
#2
Consider this quote from the Royal Society, London for 1817, “The ice which has this year surrounded the northern coast of Ireland in unusual quantity and remained there unthawed till the middle of August..” Notice this is more than usual, thus extreme but not an unusual event. This comment is important because the eruption of Tambora in 1815 certainly was a factor in exceptionall cold conditons in the subsequent few years. However global temperatures were lower and had been declining since at least 1809 in connection with the Dalton Minimum. There are well documented rports of Inuit (Eskimos) in Kayaks showing up off the coast of Scotland in the period from 1700 to 1740 apparently because they could travel along the extensive and southerly location of the ice pack.
This comment is not about any of your ideas or articles.
Your site is very interesting but i have a problem reading it. your index (on the right side of the page)often expands to the left and covers the bulk of the article. any ideas on how I can read the articles?
I find that happens when using Internet Explorer. Try Firefox web browser instead.
Thanks Steve!
Now the remaining recipe for the HS is:
1) make up a PC selection rule that includes the four first PCs
2) regress with the CRU temperature series in such a manner (try detrending/short normalization etc if
needed) that the 4th PC gets a high weight.
That’s it, folks! You may try this at home, although only qualified “climate scientists” are allowed to publish such results. Please, remember to delete all unsuccesful attempts. Under no circumstances, do not store any failed results under “CENCORED” directory, and do not publish your code.
RE: #7 – I have long suspected that a significant impetus for the settlement of the US / southern American colonies, especially the SE US / colonies, during the late 1700s and early 1800s were the horrible climate conditions in the British Isles at the time. The Carolinas would have been a compelling destination, from the point of view of, say, a struggling North York farmer (maybe even one named Mr. Heathcliff, LOL … ).
re: 10. There’s a better way. Truncate at 1960 and splice the surface temperature data.
Re#7
George Washington and Thomas Jefferson wrote diaries on weather. They are interesting to read. Some of them are hosted online.
From online search :
Washington’s temperature records begin Jan. 1785
Some of his extremely cold readings may indicate that the thermometer was outdoors. (they are not sure of others..my comment)
He wrote on 5 Feb. 1788 of weather so cold that the mercury did not rise out of the bulb of the thermometer all day. But he was writing about one of the coldest days of the century, when near Philadelphia the temperature registered–17° F.
Holy freejoles! Just that one example of applied Mannian PCA should tell any half-witted scientist they need to seriously question the (flawed) methodolgy. Amazing how that big dip circa-2000 suddenly becomes the peak of the warming spike.
In the Mannian case, there actually is an increasde in bristlecone pines. PC series don’t have any orientation. Indeed, one of the theoretical weaknesses of the method is that it disregards information on whether a series is trending up or down – not exactly ideal informatino to “throw out”.
Steve:
1. Could you please answer the citation and tree-line questions from #3?
2. When adding content like the Jean S remark would be useful to clarify with update or better yet do new post.
3. On that material:
a. why does the displayed PC4 have a different start time then the overall average?
b. Is the simple average that you show, that derived by Woodhouse or you? If by you, what does Woodhouse do in their paper (do they have some weighting scheme)?
c. How much do the HS indices differ from your displayed PC4 and simple average. (It does not look visually stunning). What are the HS indices of PC1, 2, 3.
d. What is Preisendorfer’s “n” for this example.
4. Criticism:
a. I’ve had some influence…based on the comment on result in reconstruction versus just the PC.
b. Usually though, you (or Ross) cite purely the PC1 as your example. This does not seem such a dramatic example of data mining. In fact, in this example we are right at the PC4, which usually you scoff at!
c. Why would being in the PC4 drive a reconstruction to be hockey-stickish. We are entering deep into the metaphysical. But, I don’t see how this would do so if this data combined with larger data or even if this were the only data. If you are going to make that hypothetical point, you need to show how the average differs from the average of the PC1/PC2/PC3/PC4/…PCn.
d. You are subtley conflating different issues with the example here of what “Mannian PCA did”. Your readers are used to seeing that term used to indict the thing that was bizzare about it (off-centering). However, in this example you’ve just done, we don’t know to what extent the “Mannian aspect” (off-centering) drives any data mining and to what extent it is variance normalization driving things or even just PCA itself (and PCA itself must “mine” in that it selects some of the series more then others).
Editorial:
I realize that my comments are not new logical insights to you. My point is that to look at this example fairly, those are the questions that must be answered. This is both for my own sake and for people like #15, who are jumping to conclusions based on an example where you picked the PC4!
1. The paper itself is cited in one of the *.txt files. See say co589.txt.
2. OK, I usually do, but I forgot this time.
3. aFor the average, I took the average of available series. So it stretched over the max of all the sites. For a Mannian PC,I wanted to have enough sites to permit data mining and a long enough period to permit leverage – this wasn’t entirely possible. As kind of a compromise, I chose the same length of time as the 14th century step (581 years) and ended up with 23 or sites.
b. The PC1 looks like the mean. We pointed out that the Mannian PC doesn’t always produce a HS. Indeed, the NOAMER network without the bristlecones doesn’t have a HS.
c. The HS for this is about 0.9 (I should have posted this up, but this was a pretty incidental post) and low for the higher PCs.
4a. I get really tired of you “taking credit” for attending to the reconstruction. Our E&E article has a lengthy discussion of the impact of various permutations and combinations on MBH reconstruction. We referred to this in our GRL article. We subdivided the discussion into two papers since, in keeping with much advice to focus academic articles on individual points, we wanted to keep each article to relatively bite-size points.
c. In the regression phase after the PC phase, the regression phase doesn’t care whether the series comes as a PC1 or PC4; both get the same weighting. If you use covariance PCs and 5 PCs (AS WE OBSERVED IN OUR EE ARTICLE), you get a HS shape in the MBH reconstruction because the bristlecones impart a HS shape to the PC4 which now gets in and is overweighted in the reconstruction.
b. Both PCs and the Mannian twist have some odd properties. In this case, I haven’t explored the matter to see whether the Mannian data mining specifically contributes to whatever HS-ness we see in the PC4. That could be thrown up by the PC method seeking series orthogonal to the 3 higher PCs. I’d suspect that the Mannian weights would, in this case, have something to do with it. If I pursue this post past an incidental post, I’ll look at it, and might look at it anyway some time in the future, but I’m chock-ablock with higher priority items right now.
re #16/3(d): TCO, don’t worry about Preisendorfer’s rule-N. As Steve said many times before it’s NOT THE way of selecting PCs, there are many other rules available (that’s what I’m fererring to in #10). No matter what Mann says Preisendorfer’s method should never be applied to his “PCA”, and not IMO even to the normal PCA in the situation at hand. Moreover, the Preisendorfer’s rule is IMHO a more or less ad-hoc procedure not much in use outside of the “climate field”. Also, I think there are much better criteria seeminly unknown in the climate circles. Those interested should search for “model order selection”, see, e.g., this paper for the basic references.
A quick search on IEEE and Alltheweb (Google equivalent) for “Preisendorfer’s rule” turns up nothing but mentions from the debate between here and RC. Nothing at IEEE, btw, and PCA is primarily an image processing mechanism, i.e. if it were used in image processing, we’d have a link or two. Talk about obscurity. Heck, there are more mentions of MY last name on Alltheweb and there have only been 15 people in the history of humankind with my last name. 🙂
Mark
Re: #19
I didn’t think it was much in use *inside* the climate field either. Didn’t it get mentioned only as a (presumably) post-hoc justification for including the bristlecone PC in Mann’s analyses?
Thanks for the dedicated response. Seriously.
A. FYI: my request for a citation (the specific file is great!) was because “Woodhouse et al 2006” is not sufficient to tell what paper it is (even if I go to a uni library and look for a hardcopy).
B. Yes, I’m aware that all of the PCs (which are retained) get same weighting. This was my complaint when Ross would show egregious PC1s as examples of data mining and not show the other ones.
C. You may be tired, but tough, Steve. I KNOW that you mentioned it in your EE article. I’m not gigging your knowledge. I’m gigging your fairness in debate. My complaint was on the blog where individual posts and discussions were misleading because of not clarifying the difference. Becuase the impact on a suspect PC is an OVERSTATEMENT of the impact on the overall result! I think one of the John’s (Cross or Hunter) gigged you for actually referring to a PC as a reconstruction and I had to persevere to get the clarification–I did not get an answer to this question when I first asked it a long time ago. So I’m sensative to this and will watch it. And I guess…other things as well.
D. I understand that the post is incidental. My questions are reasonable ones from someone trying to understand what “take-aways”, what inferences to draw. As opposed to number 15 jumping up and down and cackling over a PC4.
E. Jean, I understand that Preisendorfer’s n may be arbitrary. I’d still like to know it in this example. I think what readers need to realize is that “Mannian” PC1 (what Steve and Ross refer to often when asked about mining by readers on the blog) looked like the mean. Steve didn’t show it. Then he looked at PC2…no, still no hockey stick. Then he looked at PC3…still no hockey stick. Then at PC4, he saw a mild hockey stick. And posted that. Just going down to the 4 is interesting (given that there are other times when Ross makes the argument that PC1, while not mathematically special in the reconstruction, is important as the “dominant mode of variation. Similarly, it’s interesting to think just how far Steve went to get his “bad to Mann looking” example. Did he go past Preisendorfer’s n? And then, was it really the “Mannian” nature of the PCA (the off-centering which in the free world is unique to Michael Mann) which led to the restult on PC4 or was it PCA in general. Finally, I think if you look at orbitals (which are decompositions in to parts of the density functional pattern of electrons) you will see some interesting things–symmetric, anti-symmetric modes, etc. So it does not surprise me that individual PCs have some funny patterns. (Note that this also indicts Mann for touting PC1). But I’m about truth. Not about only gigging Mann and never gigging Steve.