The Gift That Keeps on Giving Continued

Here is a little present for Jean S and UC. Something new from the gift that keeps on giving. Something that even I never noticed before.

What we’re going to see is that in Mann-world, U.S. tree ring series are capable not merely of reconstructing world temperature, but 9 different “climate fields”. In effect, using Mann’s algorithm, one doesn’t need to leave the U.S. to reconstruct world climate in as much detail as one could ever want. It’s funny that actual U.S. temperature readings in the 20th century are held to be unrepresentative of world climate, but U.S. tree rings have such remarkable properties.

Mann, followed by Wahl and Ammann, say that they can “get” a HS without using PCs, by using all proxies. I’ve previously commented on the overfitting problems inherent in this approach, but there’s more when this web begins to get tangled.


Today’s post is about a puzzling corner of MBH98 that even Wahl and Ammann 2007 made no attempt to replicate – the determination of how many “climate fields” could be reconstructed. Wahl and Ammann despaired of even beginning the ascent of this particular mountain and simply adopted the schedule published in the MBH98 SI. Neither Jean S, UC nor myself have made any headway either.

In all our discussions of Mannian principal components until a couple of days ago, we’d only discussed two forms of PC calculation: temperature PCs (which Mann reifies as “climate fields”) and tree ring PCs. When I re-parsed the source code again a couple of days ago, I noticed a third PC calculation (also Mannian short-centered) which we hadn’t discussed before.

In order to determine how many climate fields could be reconstructed, Mann did a Mannian short-centered PC calculation on the proxy network for each step, calculating the number of normalized eigenvalues that exceeded the square root of 2/M, where M is the number of proxies. As noted in my post, although Mann described this as an implementation of Preisendorfer’s Rule N, the rule is not set out in this form in Preisendorfer. Nonetheless, I’ve done some experiments with networks with low-order AR1 coefficients, of the type that Mann uses. It is a more restrictive rule than say the Guttman Rule which would effectively use the square root of 1/M, so it’s not something that I take particular issue with (other than the general point that “significance” under Rule N is nothing more than notice that the PC should be investigated and is not itself a test of scientific significance, contrary to anything that Mann or Ti-mann might say.

The graphic below show an emulation of this test using the MBH AD1400 set of proxies (short-centered tree ring PCs) in a short-centered PC test. As you can see, this calculation shows a calculation which in Mann-world is used as a rationale for only reconstructing one “climate field” in the AD1400 step. According to this test, the network contains only one significant pattern and therefore only one “climate field” is reconstructed. So the calculation here yields the reported result – something that should not pass without a small celebration for Mann crossword puzzle solvers.

thegif12.gif
Figure 1. All proxy AD1400 network normalized eigenvalues (squared). Dashed line shows the 2/M test.

Mann, followed by Wahl and Ammann and Tamino, says that they get the same results without using PCs, by using the 19 non-PC proxies in the AD1400 network together with the 76 tree ring proxies in the NOAMER and Stahle SWM networks – added to a few North American tree ring sites in the AD1400 network used directly.

The above calculation was done with 22 series- 19 proxies and 3 PC series. So let’s do the same calculation using the no PC network – using 95 proxies individually, first with Mannian short-centering and then in a centered calculation. In the short-centered calculation, the test states that there are seven “significant” climate fields that the proxy network can reconstruct. Using a centered test, it’s even better: now there are nine “significant” climate fields that can be reconstructed.

 thegif11.gif  thegif13.gif

Mann said that his erroneous PC method didn’t “matter” because he got the “same” answer without using PCs. Well, did he get the same results? Taking his absurd methodology at face value, had he followed the “no PC” alternative and without short-centering, he would have had to continue through the calculation with nine climate fields. I haven’t followed through this leg of the calculation to its end. There is a process for selecting which “fields” to reconstruct which is still a mountain to climb. But with this new foothold, we may be able to advance on it.

But Mann’s claim that the error had no impact is clearly falsified here. Because of this error, Mann failed to reconstruct nine climate fields in the no PC case that were “permitted” under his methodology. Think of all the lost “information”.

The problem is far more than bristlecones merely being antennae for world temperature. U.S. trees, analysed according to Mannian methods, are supposedly capable of reconstructing ENSO, the Chinese monsoon, the East Asian monsoon, the PDO, the North Atlantic Oscillation, the Indian Ocean Dipole – did I leave anything out?

Think of the time and effort that can be saved for paleoclimate research. Using Mann’s methodology, there’s no need to collect data from remote corners of the world. You can just keep adding to the U.S. tree ring network. More Starbucks for everyone.

A few readers may argue: maybe so, but this is a “10 year old” paper and doesn’t apply to Mann et al 2007. Well, the same issues apply to it, perhaps even more so. Mann et al 2007 uses exactly the same network as MBH98 – right through to the erroneous PCs. Mann has cheekily not changed a comma in his data base in 10 years. The approach in Mann et al 2007 is to calculate an enormous correlation (covariance) matrix between all the proxies in the entire world and use that for reconstruction. In the early stages where the network is dominated by U.S. tree rings, you end up with exactly the same sort of problem as the MBH “all proxy” network – the absurd conceit that patterns in a matrix of U.S. tree rings can be used to reconstruct climate details all over the world.


53 Comments

  1. Anthony Watts
    Posted Mar 12, 2008 at 7:44 PM | Permalink

    This is what should be called “climate divining”.

  2. Andrew
    Posted Mar 12, 2008 at 8:12 PM | Permalink

    It’s funny that actual U.S. temperature readings in the 20th century are held to be unrepresentative of world climate, but U.S. tree rings have such remarkable properties

    Hasn’t one of your’s and Anthony’s explorations been the pronounced upward trends apparent in certain stations in the America Southwest? Perhaps this is the reason that region is considered “teleconnected” to global temperatures? ;)

  3. steven mosher
    Posted Mar 12, 2008 at 8:44 PM | Permalink

    UC. I think shen98 needs to be ammanded. you dont need 60 optimal sites to characterize the
    global temp feild. you need a few good trees.

  4. Steve H
    Posted Mar 12, 2008 at 9:12 PM | Permalink

    Steve McIntyre:

    Perhaps on or off topic, but where are the results of your tree ring samples?

    You showed some preliminary data, but it has been several months now.

    I trust your data, and wanted to use it as a reference.

    Steve: Thanks for reminding me. I need to visit the lab and then put a push on this.

  5. Dave Dardinger
    Posted Mar 12, 2008 at 10:03 PM | Permalink

    re#3

    you need a few good trees.

    As opposed to determining marine temperatures where all you need is a few good bouys.

  6. Posted Mar 12, 2008 at 10:51 PM | Permalink

    #2: One doesn’t need to look to conspiracy jokes to find something funny, though… remember that the BCPs’ growth is not correlated to the local temperatures around them, which was the main reason for the “teleconnection” invention.

  7. Armin
    Posted Mar 13, 2008 at 3:26 AM | Permalink

    Mann said that his erroneous PC method didn’t “matter” because he got the “same” answer without using PCs

    During math classes at our high-school, getting the ‘right’ answer using the wrong method, was still considered as being wrong …

    Note, I don’t mention this as a joke, but serious. Let us be honest. Much of the attacks on M&M and other people who dared mentioning flaws in MBH98 and successors are attacked because people want to protect the answer, not so much the method. The answer is not at stake there. (Yes, the answer is most likely plain wrong also, but that is a different story.)

    The second step is, that using different mathematiscs on the same data, indeed also produces a HS. But, I recall that one of the main conclusions from Wahl and Ammann – regardless of how you word it in a press release – is that when removing certain datasets (the famous Bristlecone) the result is statisticly seen “non-meaningful” as it lacks statistical significance. Howqever, we know that Bristlecone “should be avoided” as they are also non-meaningful as a temperature proxy.

    So regardless of the issue of (presumed) faulty statistics, the MBH98 and fraternal twin brother, are “non-meaningful” in every case. So Mann, is right indeed, his PC’s don’t “matter”: ragrdless if the answer (the HS) is true or not, MBH98 is no proof for it.

  8. Scott-in-WA
    Posted Mar 13, 2008 at 4:28 AM | Permalink

    The problem is far more than bristlecones merely being antennae for world temperature. U.S. trees, analysed according to Mannian methods, are supposedly capable of reconstructing ENSO, the Chinese monsoon, the East Asian monsoon, the PDO, the North Atlantic Oscillation, the Indian Ocean Dipole – did I leave anything out?

    The more assumptions one needs to make to support an argument, the weaker is the case for that argument. But this seemingly simple principle doesn’t seem to have any relevance in the world of Mannian climate science. (Or, in the world of climate science generally, for that matter.)

    Concerning the most important assumptions Mann relies upon, there is the issue of the tree ring proxy divergence problem, and also the related issue of whether or not trees are linearly sensitive to temperature in ways that can be separated from their sensitivity to other environmental factors – precipitation, availability of water, C02 fertilization, physical location, etc.

    As far as trees being antennae for world temperatures, let’s add another “thinking-outside-the-box” type of assumption, since there seems to be little penalty in the world of climate science for doing so: Perhaps it is possible that the bristlecone pines have some sort of natural on-board data processing capability, which, with the passage of time, allows them to sort through the wealth of environmental signals they receive.

    Being trees, of course, it takes them a long time to process the data they accumulate, perhaps hundreds of years. This means that after the passage of several hundreds of years, it may be possible to use these trees to sort out the rich variety of climate signals that exist today, but which presently seem to be of some difficulty in clearly interpreting.

    A further assumption here is that our descendants of several hundred years hence (say in the year 2408) will be able to apply the appropriate calibration methods so as to properly interpret the temperature data which was processed by the tree antennae over those previous several hundreds of years.

  9. Jean S
    Posted Mar 13, 2008 at 4:29 AM | Permalink

    The problem is far more than bristlecones merely being antennae for world temperature. U.S. trees, analysed according to Mannian methods, are supposedly capable of reconstructing ENSO, the Chinese monsoon, the East Asian monsoon, the PDO, the North Atlantic Oscillation, the Indian Ocean Dipole – did I leave anything out?

    Well, yes. They are capable of reconstructing not only different regional (European, SH, latitude bands, ENSO etc.) annual temperatures, but also warm/cold season splits. See here and here. Which leads to another puzzle: why in the cold season (and warm season) one has different number of eigenvectors (climate fields) for these seasons than for the annual reconstruction although the proxy network appears to be the same? In particular, why there are two “climate fields” in AD1400 step for the cold season?

  10. Lobert
    Posted Mar 13, 2008 at 6:37 AM | Permalink

    #3. You say:

    As opposed to determining marine temperatures where all you need is a few good bouys.

    An interesting observation about sexual orientation if true. I still think that a few girls would be more effective even in today’s armed forces.

  11. George M
    Posted Mar 13, 2008 at 7:02 AM | Permalink

    General question about tree rings for the dendro gang. The rings, we are told, are from the seasonal changes, growing in the summer, dormant in the winter, etc., etc. Do trees growing in the vicinity of the equator develop rings? How do they know what season it is, when it is warm and sunny year around? Teleconnection?? Or do they show 2 rings per year? I do not believe I have ever seen this discussed anywhere.

    Steve
    : there are technical studies of tropical trees. They do not have rings which is why they haven’t been used much. Its possible that there may be some dry season wet season differentiation that will aid dating; dendros are working on such problems.

  12. Larry T
    Posted Mar 13, 2008 at 7:50 AM | Permalink

    Since there is a strong correlation between bristlecones and solar activity we should use them to perdict future solar activity trends

  13. BarryW
    Posted Mar 13, 2008 at 7:51 AM | Permalink

    I thought it was the Marines that wanted a few good ‘Manns’.

  14. pat
    Posted Mar 13, 2008 at 7:55 AM | Permalink

    All these statistics are mind-numbing! I think it was Lord Kelvin who said “If you have to do statistics in your experiment, you should have done a better experiment” I like the view that the “function of science is to predict the outcome of experiments”, whether the experiment is done in a lab, or by observing the universe.
    The reason we have all this confusion is that, at its present state, Climate Science is no more a science than Political Science* or Social Science – its predictive power is nil! No one tilts lances (ref:Hansen’s “jousting with jesters”) at electrodynamics or relativity. Until they make some serious breakthrough, the so-called climate scientists should be much more modest in their claims.
    *BTW is everyone aware that the the National Science Foundation awards grants to people in Political Science? Incredible, what’s next astrology and alchemy?

  15. Jerry
    Posted Mar 13, 2008 at 8:14 AM | Permalink

    Pat – Right on, all this statistical mumbo-jumbo to try and use tree data/temperature data to reconstruct past temperatures of the entire world is nice and great for discussions, but on the whole not very useful for directing econmoic policy. We’ve already seen how good our temerature taking is right now (thanks to Anthony Watts and his group for his review of as many temperature stations as best they can).

    I’m sorry Jean S., but the papers you refer to aren’t very convincing arguments (in fact they even convince me further that we don’t have a clue what is going on). I know I’m not a climatologist, just a lowly Civil Engineer, but neither of these papers convince me that we can predict, with any clarity (or at least not very accurately), past temperature reconstructions or future temperature reconstructions, based soley on tree rings in the Northern Hemishpere and their correlations to measured current temperatures. At best everything is an educated guess (which are nice but to base economic policy on is a farce, IMHO). And climate alarmists, do themeselves no favors by trying to imply that they have more confidence in these results than they do.

    Good lord, I get so tired of the extreme confidence that Mann, Hansen and the rest of his ilk have about any percieved warming trend and what might occur 100 years in the future or 10 years in the future. Just because you think you have the right answer, doesn’t necesarily make it so. But I guess the old adgae, Close enough for Government work, falls right in line with this thinking.

  16. Jeff A
    Posted Mar 13, 2008 at 9:11 AM | Permalink

    My wife is a Horticulturist and tells me that Pine trees in general rely greatly on heavy fall rains to be able to survive winter freezes. Seems that rings would should which years had heavier rainfall more than they would temperature, unless the temp were particularly cold.

  17. Posted Mar 13, 2008 at 10:26 AM | Permalink

    @Pat

    All these statistics are mind-numbing! I think it was Lord Kelvin who said “If you have to do statistics in your experiment, you should have done a better experiment” I like the view that the “function of science is to predict the outcome of experiments”, whether the experiment is done in a lab, or by observing the universe.

    This is still the prevailing rule for lab work. That said, the rule is: if you have to do fancy statistics, you should have done a better experiment.

    People do use statistics when calibrating instruments, finding trends, stating uncertainties etc. But if generally speaking, in laboratory work, if at all possible, you design a better experiment when at all possilble. It quite often is.

    Unfortunately, with climate, it often is not possible to redo the experiment. With respect to the problem discussed in this thread, the ideal experiment would have required humans to set out thermometers in a number of well distributed spots on the planet, relatively far from urban areas and record temperatures at regular intervals.

    We didn’t invent thermometer long enough ago to start measuring as far back as required to do a good enough experiment! More tree samples in more places would help too. More recent cores would help.

  18. Posted Mar 13, 2008 at 10:37 AM | Permalink

    Some time ago, I tried to reconstruct all 16 ‘fields’ with MBH data. The results were disappointing -Denzil Dexter. Negative verification REs for AD1000 step, negative sparse verification RE for AD1400, 1509 was the warmest year of the millennium..

  19. jae
    Posted Mar 13, 2008 at 10:51 AM | Permalink

    Tropical trees generally have rings, and there are some tropical tree proxies. Here’s a particularly interesting study.

  20. Retired geologist
    Posted Mar 13, 2008 at 11:34 AM | Permalink

    LIke many other readers of this post, I have some trouble following all the details of the statistical arguments whether by Mann, McIntyre, or any others. Maybe this is way too simplistic a view, but if global warming is really happening, shouldn’t I be able to go to the data collected at any weather station (without regard to whether it is sited on asphalt, near an A/C unit, in an urban or rural area, etc.) and simply look at the data over the recent few tens of years and notice at least a slight up tick in the temperatures recorded? If they changed the type of thermometers used shouldn’t the early style show at least slight temp rises and then when the newer type came into use it too should show slight temp increases. I appreciate the statistical work but I’m still not sure why its needed at all. I understand the need for proxies for estimations of the distant past but again, if global warming is happening now, what difference does it make if there was a little ice age or medieval warm period, the damn temperature should show heating now.

  21. Gary
    Posted Mar 13, 2008 at 11:43 AM | Permalink

    #11 & 19 – lake sediments provide proxies in the topical lowlands where tree rings aren’t available. An example is here.

  22. MarkW
    Posted Mar 13, 2008 at 11:48 AM | Permalink

    The problem is that the signal that we are looking for is more than an order of magnitude smaller than the variation you will see in any given day. It is almost 2 orders of magnitude less than what you will see over a calendar year.

    And this signal is spread out over a century or more.

    Even if it does exist, you aren’t going to be able to see it just eyeballing the raw temperature record.

  23. MarkW
    Posted Mar 13, 2008 at 11:50 AM | Permalink

    Additionally, even if you do see an increase at a particular station, how do you know it’s a real climate signal, and not the result of gradual encroachment by housing subdivisions? Or any of the dozens of other things that can slowly change the nature of the region that a sensor may be located in?

  24. Jon
    Posted Mar 13, 2008 at 12:01 PM | Permalink

    Retired geologist:

    The day-to-day, month-to-month, year-to-year, decade-to-decade variation is substantially larger than the AGW signal.

    Thus, when we do as you say we are just as likely to see substantial cooling as substantial warming…

    Following the IPCC, AGW amounts to a ~0.05C/decade trend. Short time frames have stronger noise. Longer time periods amplify the AGW relative to the noise but then suffer worse instrumentation problems.

  25. SteveSadlov
    Posted Mar 13, 2008 at 12:01 PM | Permalink

    The magical orcwood trees, on the upper slopes of Mount Doom, amazing teleconnected tools of discernment and sorcery – but be advised, only an elite group of wizards has the “skill” to use these tools.

  26. SteveSadlov
    Posted Mar 13, 2008 at 12:12 PM | Permalink

    95% of the available moisture for Great Basin BCP subspecies in California comes from whatever moisture is delivered by the snow pack. There is very little rain during the growing season, some years there is almost none.

  27. G Alston
    Posted Mar 13, 2008 at 12:14 PM | Permalink

    #20 Retired Geologist

    For all we know the surface temp recordings don’t show CO2 warming anyway
    so much as they seem to be just as much a proxy for land use as anything else.

    Not to add any confusion or anything…

  28. LadyGray
    Posted Mar 13, 2008 at 12:46 PM | Permalink

    The magical orcwood trees, on the upper slopes of Mount Doom, amazing teleconnected tools of discernment and sorcery – but be advised, only an elite group of wizards has the “skill” to use these tools.

    And any who dare to disagree with that elite group of wizards will have the Nine Riders to deal with . . .

    But we have the White Wizard Steve McIntyre doing battle against them all, reaching out to fell the orcwood trees and renounce their black magic which threatens to destroy the world.

    Now, if only I could find that Palantir I had lying around . . .

  29. Sam Urbinto
    Posted Mar 13, 2008 at 2:07 PM | Permalink

    Okay, here’s the tree milllyun dollar question. If I tell you today the mean temperature was 14 C in my backyard, what was the day like in the 100 kilometer radius around my backyard?

    Armin: “…people want to protect the answer, not so much the method. ”

    It appears that the method is the first line of defence for the answer. One would believe it would be the case that if you can keep the method unknown, confusing or not replicable, then how can the answer be attacked? Seems plausible. (I’m trying to be diplomatic here.)

  30. JP
    Posted Mar 13, 2008 at 2:19 PM | Permalink

    “Okay, here’s the tree milllyun dollar question. If I tell you today the mean temperature was 14 C in my backyard, what was the day like in the 100 kilometer radius around my backyard?”

    According to Mann, all you would need is a Bristlecone pine at any point along the radius, and that the Bristlecone would teleconnect to your backyard, regardless of differences in geography. So, when do I pick up my millyun?

  31. Sam Urbinto
    Posted Mar 13, 2008 at 2:41 PM | Permalink

    Ah, sorry, the millyun is only available to those with climate scientist degrees.

  32. Gene
    Posted Mar 13, 2008 at 3:11 PM | Permalink

    In math the following rule applies
    multiplication and division:
    The LEAST number of significant figures in any number of the problem determines the number of significant figures in the answer.
    This means you MUST know how to recognize significant figures in order to use this rule.
    Example #1: 2.5 x 3.42.
    The answer to this problem would be 8.6 (which was rounded from the calculator reading of 8.55).
    2.5 has two significant figures while 3.42 has three. Two significant figures is less precise than three, so the answer has two significant figures.
    This wouuld seem to me to make .01 c adjustments bogus or any temp reconstruction only
    accurate to 1 degree(?) or am I mixing trees with tempature

    Steve: I wouldn’t worry too much about rounding. There are many other issues.

  33. Posted Mar 13, 2008 at 4:46 PM | Permalink

    I cannot seem to find the trackback link for this article. When I click on “trackback” I get sent to the original article

  34. John Norris
    Posted Mar 13, 2008 at 6:03 PM | Permalink

    re: JP March 13th, 2008 at 2:19 pm

    Here: millyun

    It’s every bit as good as tree ring proxy data.

  35. Another Larry
    Posted Mar 13, 2008 at 7:28 PM | Permalink

    I am not a statistician, scientist or engineer (and I have the grade-slips to prove it). I have however lasted near 3-score and ten mostly trying to do useful work with computers of one kind or another and among the things I have learned are things like “if the corrections and adjustments are bigger than the thing you are trying to measure, one of the three is a huge waste of my time”.

    Another is that somebody telling me that I don’t need to look under the hood, I can just take there word for how good it is, is a clear signal to run like hell, and if I can’t run away, to take the thing apart and mike every last piece. And then run away if I find a chance to.

    I’ve said it before, and I’ll say it again.

    The thing that worries me most about the current panic is the likelihood that our long history of doing the worst possible thing will again be repeated.

  36. kim
    Posted Mar 13, 2008 at 7:42 PM | Permalink

    The four D’s of the tort of negligence are 1. Duty, 2. Dereliction of Duty 3. Damages, and 4. Direct causation from dereliction to damages. With respect to the whole business of Kyoto limits, and carbon caps there are clearly those with a duty to care. There has been dereliction. There are ongoing and increasing damages, and there is direct cause between the dereliction of duty and the damages.

    Now I’m not sure that Gore presently has a duty to anything. But there are people who do have a duty to care about this matter and get it right.

    This is probably a policy matter, and subject to editing. I don’t mind if this post disappears; the facts of the case won’t vanish with it.
    ===================================================================

  37. kim
    Posted Mar 13, 2008 at 8:08 PM | Permalink

    Raven, granted such talk probably won’t go anywhere, but I look upon it as the 2X4 for the mule. It might concentrate a few minds upon their duty to science and the public.
    =======================================

  38. Pat Frank
    Posted Mar 13, 2008 at 8:46 PM | Permalink

    Paul wrote, “It’s my belief that many of the AGW scientists have come to their original conclusions honestly.

    Let’s say they came to their original conclusions sincerely.

    Speaking before Congress as a scientist in 1988, Jim Hansen could have had no valid scientific basis whatever to claim 99% certainty that the warming climate trend was caused in large part by human-produced GHG’s.

    No scientist speaking before Congress today, 13 March 2008, could, with scientific validity, claim that certainty either. Or 50% certainty. Or 10%.

  39. Bob Mc
    Posted Mar 13, 2008 at 9:10 PM | Permalink

    Mann…..looking for a few good trees.

  40. Posted Mar 13, 2008 at 10:30 PM | Permalink

    I read and partly comprehend what you guys are talking about.I never have enough time to really study this stuff well.

    But I have to say that I am surprised that most you guys have NOT read up on the Bristlecone Pine tree themselves.

    It is a SUBALPINE zone growing tree in very high elevations.In just a few areas of the American west.Has one of the slowest growth trend of any tree in the world.

    Here is a good link that shows that the pines live in a zone that has a very short growing season.Meaning that for most of the year it barely grows at all.A tree that will grow in such areas where NO OTHER TREE SPECIES WILL!

    A quote from the link below:

    “The bristlecone has adjusted to places on earth that no other tree wants to inhabit, and in these harsh environments, has flourished, free of competition.”

    A exclusive habitat all to its own.

    http://www.sonic.net/bristlecone/growth.html

    A map shows just how minimal this tree’s habitat is.

    The point is that the species grows in very rare climatic zone.Hardly a reasonable represenative of past climate information repository for the Northern Hemisphere.

    Personally I find it hard to believe that this tree can produce tree rings comparable to the other tree species tree ring growth.When it’s growth pattern is so far off the beaten path of common growth patterns of most tree species.

    So what made it valid to use the tree rings of an exclusive tree that does NOT grow the way trees usually grow.In a small and rare climate zone.To build a 1,000 year + long temperature proxy from?

    A tree that grows in a regional climatic area that is far less than minimal represenative SAMPLE for North America.

    I think it needs further examination.

    Cheers

    Steve: In our 2005 EE article, we provide an extensive list of references to bristlecone pine botany. It’s not that we didn’t cite bristlecone literature: the problem is that Mann and other reconstructions using bristlecones haven’t and it’s to them that your criticism lies.

  41. Raven
    Posted Mar 13, 2008 at 11:10 PM | Permalink

    Sunsettommy says:

    So what made it valid to use the tree rings of an exclusive tree that does NOT grow the way trees usually grow.

    My understanding is trees growing near the limits of their range are the most likely to be a proxy for something (not necessarily temperature). I found this guide to dendro: http://sonic.net/bristlecone/principles.html

  42. MrPete
    Posted Mar 14, 2008 at 8:10 AM | Permalink

    Raven, that’s correct within limits. There are many assumptions made, and little reason (at least with high altitude BCP’s) to have confidence in the results. A few challenges relating to BCP’s and the principles you linked to:
    Uniformitarian Principle: problem is, climate changes. If limiting factors change over time in a given place, there’s no reason to believe a tree’s growth will consistently reflect any particular attribute of climate. What if growth is precip-limited for 100 years, then temp-limited, then nutrient-limited? Very reasonable proposition, and no way to know. See Aggregate Tree Growth below.
    Limiting Factors: a lot of assumptions are made here. What applies in one place for one species and microclimate may not at all apply elsewhere. But such assumptions are made all the time. (Search for data provenance showing the connection between sampled trees and specific analysis of limiting microclimate factors in each of these proxy studies. Rarely will you find any such information. Instead you’ll see statements along the lines of “high tree line will be temp limited.” That’s an assumption, not a fact nor even a well-tested hypothesis.
    Aggregate Tree Growth: even the linked page notes that this challenge, of disaggregating various growth factors, is (not) simple.
    Site Selection: (I’m less confident about asserting this, but my strong sense is…) BCP’s don’t grow in many places. So sites are limited to where BCP’s are found, not where climate is ideal from the Limiting Factors perspective. Thus, good Site Selection is difficult to demonstrate.
    Replication: study the multiple BCP samples collected from single trees in our 2007 Almagre update. With such huge intra-tree sample variance, replication CI’s are quite wide… to be nice about it. Update – to say this using nickel instead of dollar words :-) — We found growth variations of up to 400 percent within a single tree, for certain time periods but not others. Who’s to say what the “correct” answer is, when one sample says 1-2-1-1-1-2-1 and another says 1-2-1-1-3-8-5 in the same tree?. That’s the “story” told by strip bark BCP’s.

  43. SteveSadlov
    Posted Mar 14, 2008 at 11:49 AM | Permalink

    RE: #40 – I would take it even further. BCPs are essentially freak organisms, that just happen to share certain genetic characteristics with other members of Pinus. At some point they may have been evolutionarily close to other Pinus species. But, as hopeful monsters, they came to occupy a very rare niche. This is where most botanists get wrapped around the axle. They presume that what may hold true for Jack Pine, Cedar, Scotch Pine, what have you, holds true here. When in fact, there is probably more in common, from the standpoing of growth dynamics, between BCPs and the button cacti of Estes Park, than there is between BCPs and other conifers in more common settings.

  44. GeorgeStanley
    Posted Mar 14, 2008 at 2:17 PM | Permalink

    BCP’s? perhaps it’s the PCB’s found in BCP’s, but I digress…
    It is currently 64F outside my office, but when I get home tonight(12 miles away) if the temperature there is below 60F will there be evidence of glaciation by morning? How does this relate to the rate of ice melt in the average Mint Julep?
    It’s all so oblivious, even to the casual hors doerve-er.

  45. SteveSadlov
    Posted Mar 14, 2008 at 7:41 PM | Permalink

    This panorama is at the southern end of the Inyo Mountains (a southerly spur of the Whites). This is just to give everyone a sense of what we are talking about here. Yes, those are BCPs.

    Steve Mc: Also http://www.virtualguidebooks.com/SouthCalif/EastOfTheSierra.html has many background photos. I’ve collected some very interesting site info on old roads in the bristlecone areas, by searching on mining ghost towns from the 19th century. There are some great panoramas at this website. This panorama shows dead bristlecones above the present treeline http://www.virtualguidebooks.com/SouthCalif/EastOfTheSierra/Patriarch/DeadBristlecones.html -great web presentations.

  46. Willis Eschenbach
    Posted Mar 14, 2008 at 8:32 PM | Permalink

    MrPete, thanks for your clear statement:

    (Search for data provenance showing the connection between sampled trees and specific analysis of limiting microclimate factors in each of these proxy studies. Rarely will you find any such information. Instead you’ll see statements along the lines of “high tree line will be temp limited.” That’s an assumption, not a fact nor even a well-tested hypothesis.)

    Actually, it’s even worse. There is a complex interaction between trees, temperature, and water. Given enough water, trees can endure and even thrive when it’s hot.

    But even at lower temperatures, the same tree may wilt and suspend growth if there is not enough water. I have observed this at the high altitude tree line. Up there, for part of the year, the water matches the temperature and the tree grows. But at some point in the summer, there’s not enough water to match the temperature, and the growth slows, sometimes to a crawl. The trees show the characteristic signs of “heat stress” (e.g. loss of turgor, browning starting from the tips of branches), even at the cool high altitude … because it’s not the heat that’s causing the “heat stress”, it’s the lack of water.

    And of course, actually, it’s much worse. Tree ring growth can be expressed as some kind of function like

    RW = Temp + H + Th – C – Tc + M + Tm + N + Tn + S + Ts + E + Te + interactions + …

    Where

    RW = Ring Width

    Temp = Temperature (annual average)

    H = Heat (degree-days)

    Th = Timing/amount/form of heat (e.g. does it come in the appropriate season and quantity, soil heat vs. air heat, hot wind or hot sun)

    C = Cold (degree-days below freezing)

    Tc = Timing/amount/form of cold (does it come in the appropriate season and infrequency, etc)

    M = Moisture (total available water/yr)

    Tm = Timing/amount/form of Moisure (as above)

    N = Nutrients

    Tn = Timing/amount/form of Nutrients

    S = Sunlight (hours/yr)

    Ts = Timing/amount of Sunlight

    E = One-Time Events (e.g. the tree that shaded your tree for two centuries dies, and sunlight streams in)

    Te = Timing/amount/form of change

    interactions = interactions of the various timings (e.g. is the moisture available when there’s lots of sunlight, or little sunlight)

    … = a bunch of other requirements (e.g. soil stability, presence of mist)

    First thing we can say about this function? Non-linear, and occasionally step-wise.

    Next, regardless of the method used (PCA, OLS, RegEM, whatever), the underlying assumption of all purely statistical methods (i.e. methods without explicit treatment of confounding variables) is that in the ring width equation, everything else will average out except “Temp”, the annual temperature. I question that assumption very, very strongly.

    But let’s assume that it is true, that in some mysterious fashion everything averages out but the temperature, we are still left with a very thorny problem. A concise statement of the problem would be:

    All else being equal, a year that is too hot and a year that is too cold will both result in narrow rings.

    That is a tough nut to unravel.

    To me, the only true statements you can say about BCP ring widths are things like “those were good years for that tree (or stand)” or “those decades, the tree (or stand) didn’t thrive”, or “wow, what happened in 1877?”.

    To summarize:

    1. There is a temperature signal in ring widths.

    2. Unless there is always ample moisture, there is a water signal in ring widths.

    3. There is a combined water-temperature limit on growth.

    4. These signals are sometimes related (e.g. hot years tend to be drier).

    5. The temperature signal is an upside-down quadratic, which is by nature not invertible.

    Best to everyone,

    w.

  47. MrPete
    Posted Mar 15, 2008 at 2:49 AM | Permalink

    (I’ve added a simpler way of describing the last point in my little comment about dendro principles vs BCP’s… The rest was intended to be easily read by lurkers; apologies for the inconsistency.)

  48. Stu Miller
    Posted Mar 15, 2008 at 9:17 AM | Permalink

    Willis,

    You will be encouraged to hear that the University of Arizona has taken over Biosphere and is using that very expensive facility to, hopefully among other things, investigate the theory that drought combined with high temperatures kills more trees than drought combined with lower temperatures.

    My wife and I recently took a tour of Biosphere where we heard a very earnest professor explain that during a recent drought, quite a number of pines had died in Arizona. After thinking about this for a while they came up with the hypothesis that heat plus drought might be bad and set about testing this with dozens of pine trees in the different climate zones in Biosphere. We can only hope the botanists talk to the dendros. We can only wish that the botanists knew anything about gardening.

  49. Steve McIntyre
    Posted Mar 15, 2008 at 9:34 AM | Permalink

    #48. This is OT, but it’s my blog. I remember reading years ago about the failure of the first Biosphere. As I recall, the first Biosphere failed because they ran short of oxygen but this occurred without a concurrent buildup of CO2 which they were carefully monitoring. This really puzzled the operators who were baffled as to where the CO2 went. They eventually determined that CO2 was being absorbed into the concrete pad – something that struck me as an interesting result even if it wasn’t what they intended. This is only a vague recollection from many years ago and I don’t vouch for it.

  50. Dave Dardinger
    Posted Mar 15, 2008 at 9:52 AM | Permalink

    Steve, you’re right that the CO2 was being absorbed into the concrete. It’s interesting in that when calculations are made of anthropic releases of CO2 the production of concrete is a semi-major factor as limestone has to be converted into lime with the release of CO2. Eventually, if every bit of the concrete were broken down, all the CO2 would be re-absorbed, but insofar as it stays in lumps, not too much of the CO2 is absorbed, just at the surface.

  51. Pat Frank
    Posted Mar 15, 2008 at 1:12 PM | Permalink

    #49 Steve, et al., Wikipedia has a pretty good description of what happened at Biosphere 2. You’re right about the inner concrete surfaces absorbing the CO2 that the soil bacteria were producing.

    I recall reading that many of the people involved in building Biosphere 2 were Gaia-type environmentalists who had pushed aside the biologists with regard to constructing the ecosystem. They insisted on putting in rich loamy soil instead of the sandy soil recommended by the botanists and soil scientists. Sandy soil would have drained well. But the rich black soil favored by the mother-earthers retained the water and provided the rich carbon/nitrogen sources that supported heavy bacterial growth. It was the extremely high rates of bacterial metabolism that removed the oxygen and that produced the CO2 that ended up in the concrete. Apparently oxygen levels were reduced to about 14%, equivalent to a 4000 meter altitude. The occupants of the Biosphere ended up not only almost starving, but also almost brain-dead.

    The Wiki article also gives a little flavor of what must have been a most incredible psycho-pathologically obsessive inter-tribal conflict, among people trapped in one another’s company. There must be one or two fascinating insider books about that, somewhere.

  52. jae
    Posted Mar 15, 2008 at 5:55 PM | Permalink

    46, Willis: that’s a post worth tacking to my wall!

  53. John F. Pittman
    Posted Mar 16, 2008 at 6:26 AM | Permalink

    #52 jae. Keep in mind that many of the symbols in #46 Willis report are actually complex equations. For example O2 uptake of a living aquatic system is del(O2) = alpha(T2-T1)^1.x where T1 is the refernce temperature for the system or species, T2 a tmeperature not to either heat or cold inhibition, and x is the measured enhancement for O2 uptake. Also note as expounded several times on this blog, and in Mr. Pete’s quotes, it is actally a quadratic. An interesting example of this can be done in a lab. As COD of a biologically usable substance as measured by BOD goes up from 0 mg/L to 100%, it makes a nice 0 to max to 0 curve. It is called inhibition. So, not only do you have limiting factors, too much of a good thing (inhibition) can cause a negative response wrt the increase of a otherwise positive corrlation. From numerous biological studies, species have optimum temperature ranges. Some temperature ranges for some unique species are about 3C between good life and death at either extreme. One has to wonder with BCP’s being in such an extreme environment what the species has had to evolve in order to live where it does.

    But these proxy data sets are the data we have. That is why I think some express such concern about the correct use of statistics or lack there of. And of course, any assumptions as to there actaully being a physical relationship to the observed correlation(s).

Follow

Get every new post delivered to your Inbox.

Join 3,254 other followers

%d bloggers like this: