Principal Components and Tree Ring Networks

I’m finding some benefit to having spent some time on station histories prior to my present re-visit to Mannian proxies. Digging into the handling of station histories gives some interesting perspectives on network handling that are worth considering for tree ring networks.

For example, assume for a moment that North American tree ring chronologies used in MBH98 actually were little thermometers. Yeah, yeah, I know all the problems with them. But let’s suppose that they actually were little thermometers, noisy little station histories going back hundreds of years. What would Phil Jones or Hansen or USHCN would do if they had dozens of North American station histories going back 600 years?

You know right away what they’d do: they’d lay out a grid over North America; they’d allocate each station to its respective gridcell; they’d form anomaly series by subtracting the mean over a common interval, take an average over each gridcell of available records and then take an average over all the gridcells.

Would they do principal components on the raw geographically inhomogeneous station data? Of course not. What meaning could one possibly attach to geographically inhomogeneous station data? Would they do stepwise principal components when there was missing data? The mind boggles at CRU doing something like that.

As an exercise, I thought that it would be interesting to do a CRU-type “gridcell history” for Mann’s NOAMER network.

An interesting issue arose in starting this – one that we reflected on a bit in MM2003, but haven’t discussed much. The MBH “NOAMER” network is actually a subset of about 75% of the North American tree sites used in MBH98. The first figure shows a location map of all the tree ring sites mentioned in the original MBH SI, color coded to illustrate the following.

The pink dots show the Graybill bristlecone sites in the AD1400 network – the sites that so much controversy attaches to. Geographically they obviously in a restricted range in the US southwest; the blue dots show the rest of the 70 sites in the AD1400 NOAMER network; the green dots show the other 141 sites used in later steps of the NOAMER network; the cyan dots show the Jacoby network (each site being used individually in MBH), with the Gaspe site used in the AD1400 regressions shown a little larger. The brown dots show the Stahle SWM and Stahle OK networks, many sites being between “NOAMER” sites and a couple of sites in the Four Corners area being used in both networks. The red dots show the sites listed in the original SI but not used in the NOAMER network – for reasons that have been never been properly explained. Their non-use was admitted in the MBH Corrigendum , which provided a false excuse for their non-use:

These series, all of which come from the International Tree Ring Data Bank (ITRDB), met all the tests used for screening of the ITRDB data used in ref. 1 (see ref. 5), except one—namely, that in 1997, either it could not be ascertained by the authors how these series had been standardized by the original contributors, or it was known that the series had been aggressively standardized, removing multidecadal to century-scale fluctuations.

This is untrue. Some of the excluded series (red) were Schweingruber series; all sorts of Schweingruber series were used in MBH98 and the excluded series came from identical publications as included series. This was pointed out to Nature, but they didn’t care. When I plotted this up, I noticed two red dots in Alberta – these are both locations where Rob Wilson has worked; one of the series was used in Esper et al 2002.

noamer3.gif
Figure 1. North American tree ring sites used in MBH98.

Anyway, continuing with my development of a “station history” type procedure. As a test, I did one on the MBH NOAMER network without worrying about exclusions and Stahle and that sort of stuff. I standardized the series on 1613-1900 as a long period over which the MBH NOAMER network has values to the beginning. My guess is that standardization with 1613-1970 values wouldn’t make much difference and I’ll probably do this calculation as well if I pursue this any more.

I allocated all the series to 5×5 Jones-style gridcells and averaged all available standardized chronologies within each gridcell, thereby forming a gridded network of 31 gridcells. I then calculated an annual average over all available values using a truncated mean (not using the two extreme values on either end). This yielded the following North American Tree Ring Index, shown to 1980 the final year of MBH tree ring calculations.

I’ve marked a few years with extreme values. 1934, known to be an exceptionally hot year in the U.S., had the lowest “Tree Ring Index” in the period 1880-1980. 1946 had the highest. The decade in the 1840s had exceptionally low growth – something that we also noticed in our Almagre samples. There’s certainly no hockey-stick in this Tree Ring Index.

noamer4.gif
Figure 2. North American Tree Ring Index

Intrigued by this result, I download a CRU gridded temperature history (using HadCRU2 for this since it’s a little more contemporary to MBH98) and calculated the correlation of the gridded growth index to CRU temperature history for each gridcell. I then made a contour map using the akima package that I had previously used for station history plots. For this calculation, I re-did the grids using all the available stations – Stahle, Jacoby, excluded sites.

noamer55.gif
Figure 3. Correlation to Temperature

Obviously one feature that sticks out like a sore thumb: the gridded growth histories in the U.S. Great Plains are negatively correlated to temperature. It looks to me like the core of this negative correlation is pretty near Crawford, Texas.

There’s a strip along western Canada that shows positive correlation. If you look back at the location map, you’ll see that this has mostly been filled in by interpolation as there are no MBH network stations. Also something curious – this positive correlation area is bounded at either end by stations that Mann deleted from his network, stations that weren’t used.

So we can see one reason why a station history approach doesn’t work very well – we don’t know whether our little thermometers are reading up or down. For example, suppose that a new treasure trove of instrumental measurement data decoded from Aztec or Maya glyphs were delivered to Phil Jones – the only problem was that you didn’t know whether the numbers ran up or down. Even CRU wouldn’t just dump all this data into their data base and hope that some algorithm could sort it out. Before the Aztex instrumental data was incorporated into station history data bases, one would hope that some sort of technical study would be done showing how to convert the Aztec instrumental data into modern terminology, demonstrating which direction was up in their nomenclature and what their scale was? Tree rings should be no different.

And what if Phil Jones found that some of the Aztec data was actually instrumental precipitation measurements. Would he just dump that into his data base and hope that it improved things? That maybe there was a teleconnection between Aztec instrumental precipitation and temperatures in some other part of the world. As soon as you even write this down, you realize that someone would first have to demonstrate that there was a solid relationship between modern measured precipitation in the Yucatan and modern instrumental temperatures in (say) Timbuktu or wherever, and demonstrate that this information actually aided in the estimate in a way that rose above data mining before assuming that Aztec instrumental precipitation measurements contained useful information for temperature reconstructions.


Comparison to the MBH PC1

The above graphic showing the relationship of gridded tree ring growth and gridcell temperature may provide a helpful perspective on bristlecones and the Mannian PC1 (or the PC4 or whatever).

In the graphs below, I show the geographical properties of the MBH weighting so that readers can appreciate how the locations of the famous MBH PC1 fit in the above map. On the left I’ve done a contour map in which each value of the MBH98 PC1 eigenvector is located spatially at its site. I’m experimenting a little with this still. On the right is a plot in which the weight of each site is shown by the area of the dot. Graybill bristlecone/foxtail chronologies are shown in red; all others are shown in green. The key Sheep Mt site is near the CA-NV border in California.

The Graybill bristlecone chronologies (especially the key Sheep Mountain chronology) are, on this coarse scale, in areas where there is neither a strong positive nor strong negative correlation of growth to temperature – in a shoulder zone. This agrees with site specific analyses, which show little positive correlation of bristlecone growth to temperature (nor negative correlation.)

For someone that’s looked at a lot of geophysical maps in my life, the supposed occurrence of bristlecone chronologies measuring world climate in such a shoulder zone raises red flags. Why should a site chronology in a U.S. shoulder zone have a loud response to world climate, when the chronology (and the gridded “station history”) have negligible correlation. One would surely investigate the possibility of some artifact in the Graybill chronology. Given Linah Ababneh’s failure to replicate the Graybill chronology, the alarm bells should be ringing even in Mann-world.

noamer31.gif pcsht41.gif

UPDATE: Woodhouse and Overpeck (BAMS 1998) show the following comparison between a tree ring reconstruction of drought and observed Palmer PDSI in 1934 – a year of low overall growth. It’s not that relevant specialists are unaware of the connection between U.S. tree rings and drought. Exactly why none of them ever commented on these issues in connection with MBH is something you’d have to ask them.

noamer16.jpg
From Woodhouse and Overpeck BAMS 1998

Here’s the MBH98 PC1 (bristlecones) again marking 1934. Given that bristlecone ring width are allegedly responding positively to temperature, it is notable that the notoriously hot 1934 is a downspike.
noamer47.gif


62 Comments

  1. Bill F
    Posted Mar 17, 2008 at 12:58 PM | Permalink | Reply

    The fact that the best positive correlation with temperature on your map occurs in areas where there is no tree-ring data is a very eloquent statement on the value of Mann’s chronology work.

  2. MarkW
    Posted Mar 17, 2008 at 1:07 PM | Permalink | Reply

    That’s the beauty of teleconnections. The further away from what you are measuring, the better your results.

  3. SteveSadlov
    Posted Mar 17, 2008 at 1:13 PM | Permalink | Reply

    A whole new level of vapor ware. No trees? No problem. We got correlation. And where there are trees? But of course, correlation does not matter, as THOSE trees are not local treemometers, they are GLOBALLY TELECONNECTED treemometers. Makes sense to me, what’s the problem? /s

  4. Pat Keating
    Posted Mar 17, 2008 at 1:26 PM | Permalink | Reply

    Steve
    Looking at Fig 3, it seems to me that the positive correlations are obtained where the climate is damp and cooler, while the negative correlations are obtained in regions where the climate is hot and dry.

  5. Bill F
    Posted Mar 17, 2008 at 1:39 PM | Permalink | Reply

    Pat, I suspect that what you are seeing is sites where precipitation is not the limiting factor in growth. Where there is always plenty of rain, things like temperature then become the limiting factor. In more arid climates, precipitation is often the limiting factor.

  6. Willis Eschenbach
    Posted Mar 17, 2008 at 1:44 PM | Permalink | Reply

    I got to thinking about PCs last night, and about the use of PC1 as a temperature proxy. What I realized was this:

    Suppose we plot tree ring width in some kind of n-space, where each dimension is one of the factors affecting ring width (e.g., temperature, moisture, nutrients, CO2, wind, shade, whatever).

    Now, the thing about PC analysis is that it comes up with new vectors which most parsimoniously express the original relationships. These vectors are combinations of the original variables.

    So what I was thinking was, it seems almost inevitable that in this case PC1 represents a combination of the temperature and the moisture variables. I say this because as any farmer knows, the most important things for a plant are sun and water.

    Thus, the PC1 is not a temperature proxy, it is a proxy for some unknown combination of temperature and moisture plus some minor players. Its use in later studies as a temperature proxy seems totally unwarranted to me.

    Anyhow, that was my thoughts … off to work now.

    w.

  7. Mark T.
    Posted Mar 17, 2008 at 1:45 PM | Permalink | Reply

    E.g., if the latter were BCPs.

    There’s actually a distinct sawtooth in Figure 2 above. A Daubechies mother wavelet would tease that structure out.

    Mark

  8. Mark T.
    Posted Mar 17, 2008 at 1:48 PM | Permalink | Reply

    Thus, the PC1 is not a temperature proxy, it is a proxy for some unknown combination of temperature and moisture plus some minor players. Its use in later studies as a temperature proxy seems totally unwarranted to me.

    Preaching to the choir, Willis. That those initial vectors also contain a certain amount of correlation between them only complicates things.

    Mark

  9. Pat Keating
    Posted Mar 17, 2008 at 1:52 PM | Permalink | Reply

    5 Bill
    Yes, that thought occurred to me. However, it could also be that the tree has an optimum temperature for growth: if the temperature goes above that, growth is diminished, if the temperature is below it, then a rise in temperature is good.

    Kind of like a lot of plants, trees, and animals, including humans.

  10. Gerald Machnee
    Posted Mar 17, 2008 at 1:54 PM | Permalink | Reply

    Yes, eye-balling the chart, I would go along with Pat. Start with moisture and see what else may fit in.

  11. Anthony Watts
    Posted Mar 17, 2008 at 1:55 PM | Permalink | Reply

    “It looks to me like the core of this negative correlation is pretty near Crawford, Texas.”

    Are you saying this stunning Mannomatic feature of MBH98 to correlate tree ring growth negatively to temperature is all Bush’s fault? ;-)

  12. Mark T.
    Posted Mar 17, 2008 at 2:02 PM | Permalink | Reply

    However, it could also be that the tree has an optimum temperature for growth: if the temperature goes above that, growth is diminished, if the temperature is below it, then a rise in temperature is good.

    There was what I would call an unfinished study a while back that showed some upside-down parabola response to temperature, though I do not recall any of the details other than it was from the mid/late 90s or so. The guy that posted it tried to use it as “evidence” that tree-rings accurately measure temperature and I quietly (yeah, right) responded that he had only shown the relationship was non-linear, which proved my point in the first place. He commented no further. :)

    Mark

  13. deadwood
    Posted Mar 17, 2008 at 2:13 PM | Permalink | Reply

    Anthony@11
    Come on, everybody knows its all Bush’s fault. Like everything else that ails the world, it starts in Crawford Texas.

  14. Chris Harrison
    Posted Mar 17, 2008 at 2:22 PM | Permalink | Reply

    Willis,

    Of course, your example factors wouldn’t be unique dimensions in your n-space because they are not linearly independent (moisture being partially dependant on temperature, for example). Your n variables would actually be in some m-space where m is less than n.

    One of the advantages of eigen vectors is that they do form a basis (coordinate system) for an n-space, being mutually perpendicular.

    Like you, I don’t see how an individual PC can be used as a temperature proxy

  15. Steve McIntyre
    Posted Mar 17, 2008 at 2:26 PM | Permalink | Reply

    Folks, you’re missing an important nuance on this post which is raising a point different than the usual one about trees and temperature and/or PCs. It’s the problem of trying to do PCs on geographically inhomogeneous data sets and pretending that it’s continentally valid. PC will pick out patterns but if you’ve concentrated local patterns you’ll get local results.

  16. MarkW
    Posted Mar 17, 2008 at 2:31 PM | Permalink | Reply

    Others have noted that in many trees, the width of individual rings can vary tremendously as you go around the circumference of the tree.

    To examine enough of the tree to compensate for this variance, you either have to drill so many holes that you are likely to kill the tree, or you have to cut the tree down. Which isn’t to good for it’s health either.

    I believe I’ve come up with a solution. And the equipment to do it is also light weight enough that it can be carried on a quick Starbucks based excursion.

    Sound waves.

    Steve should already be familiar with this method, it’s used extensively in geological exploration.

    They set off an explosion and measure the sound waves as they bounce of layers of differing density. The dark and light bands in a tree trunk have differing density, so they should reflect sound waves as well.

    The solution is simple.

    Get a sensitive microphone and tape it securely to one side of a bristlecone pine.
    Get a stick of dynomite, and tape it to the other side.

    Set off the stick of dynomite, record the sound waves.

    Viola, problem solved.

  17. Bill F
    Posted Mar 17, 2008 at 2:48 PM | Permalink | Reply

    Steve,

    Once again, I think what you have done is translated what was already known in many other sciences into climate science. As a geologist, if I start drilling 6 inch diameter boreholes thousands of feet apart in a heterogeneous environment and examining 2″ diameter samples, I would be a fool to assume that I could apply a statistical calculation to tell me what the stratigraphy of the area looked like as a whole. Before I can make any guesses about the stratigraphy between my sample points, I have to form a hypothesis about the geologic environment and then prove that what I see in my borehole samples conforms with that hypothesis. Then I can go ahead with the assumption that I can predict what goes on between the boreholes with some confidence.

    What Mann has done is made the assumption that he doesn’t need to have a physical hypothesis that explains the reaction of his trees to local temperature and he skipped the part about verifying that his data matches that physical explanation…he just went straight to the statistics and played with it until it made a shape that matched his needs, without ever proving that the results of the statistics actually matched some physical explanation…unless you count “teleconnections” as a valid physical explanation.

  18. Mark T.
    Posted Mar 17, 2008 at 3:02 PM | Permalink | Reply

    Good point Steve. PCA finds the vectors that represent the space spanned by the inputs, but that space may not be representative of the actual 3D “space” (i.e. the actual physical space) that is used to represent the input space.

    Mark

  19. jae
    Posted Mar 17, 2008 at 3:12 PM | Permalink | Reply

    5, Bill F.

    Pat, I suspect that what you are seeing is sites where precipitation is not the limiting factor in growth. Where there is always plenty of rain, things like temperature then become the limiting factor. In more arid climates, precipitation is often the limiting factor.

    Bingo. Anyone who has spent much time in the Inland West knows that growth is controlled primarily by precipitation. And it is no different at treeline, IMHO.

  20. Pat Keating
    Posted Mar 17, 2008 at 3:40 PM | Permalink | Reply

    15 Steve

    It’s the problem of trying to do PCs on geographically inhomogeneous data sets and pretending that it’s continentally valid

    I think most of us got that point, but had nothing useful to add to it. You are way ahead of most of us, there. However, the geographical distribution is interesting and meaningful in itself.

  21. DocMartyn
    Posted Mar 17, 2008 at 3:54 PM | Permalink | Reply

    I wonder if you could run a Fourier Transform on your figure 2 data set. It looks like a 180 standing wave is in there.

  22. SteveSadlov
    Posted Mar 17, 2008 at 4:06 PM | Permalink | Reply

    There have got to be some old trees on a certain greenie’s ranch in Crawford. So, in addition to “update the proxies” I would add “fill in the grid cells.”

  23. LadyGray
    Posted Mar 17, 2008 at 4:07 PM | Permalink | Reply

    This technique should have an official name. I think it should be called The Homeopathic Approach to Climate Tree-Ring Correlation.

  24. Posted Mar 17, 2008 at 5:18 PM | Permalink | Reply

    An old article on 19′th Century climate in Arkansas is here . In scanning the reports it indeed seems that Figure 2 is negatively correlated with temperature. I did not check precipitation.

  25. Charlie Young
    Posted Mar 17, 2008 at 6:20 PM | Permalink | Reply

    I am curious if one could measure the amount of carbon 14 in each ring. If the sun does determine the amount of cosmic radiation that strikes the earth, then I would assume there would be variations in the amount of carbon 14. Would the ratio be the same at different locations, etc? The chronology of sunspot activity is known, and finding a correlation between sunspots and carbon 14 would be interesting

    Steve:
    C14 in bristlecones was measured a long time ago and is the basis for one of the solar proxies.

  26. Dave Dardinger
    Posted Mar 17, 2008 at 6:31 PM | Permalink | Reply

    re: #25 Charlie,

    As far as C14 it’s quite well mixed, at least over periods of a year or two. And many studies have been done correlating C14 with tree-ring sequences. Indeed it’s standard practice to adjust C14 ages based on treering age. And I think the differences have been used to calculate pre-historic sunspot levels.

    Of course the combination of extra C14 produced from nuclear blasts and C14less CO2 from fossil-fuel burning has made recent aging rather problematic. I think there are some years where there are ambiguous results.

  27. Pofarmer
    Posted Mar 17, 2008 at 7:27 PM | Permalink | Reply

    So what I was thinking was, it seems almost inevitable that in this case PC1 represents a combination of the temperature and the moisture variables. I say this because as any farmer knows, the most important things for a plant are sun and water.

    Thus, the PC1 is not a temperature proxy, it is a proxy for some unknown combination of temperature and moisture plus some minor players. Its use in later studies as a temperature proxy seems totally unwarranted to me.

    Anyhow, that was my thoughts … off to work now.

    Finally!!!! Don’t leave out growing season length, either. A mild, early spring, and a mild, late fall, could make an awful lot of difference. Most crops that I’m familiar with respond fairly favorably to heat as long as they have adequate moisture. If moisture starts becoming a problem, that’s when you see symptoms. Most plants can handle a lot of heat as long as they have moisture. Last year we had a faily cool, fairly dry summer. In fact, the lack of extreme heat through the first week of August is the only thing that saved us a little bit. You can’t say there’s a perfect correlation between temperature and precipitation, either.

  28. John Lang
    Posted Mar 17, 2008 at 8:02 PM | Permalink | Reply

    A negative correlation just implies that Mann placed a negative coefficient on those particular tree samples. For Texas, it might be rationale to assume that reduced temperature leads to increased tree ring growth given increased temperatures are also correlated with drought conditions in Texas.

    In the last year when Texas had cooler temperatures and lots of rain, I assume that tree ring growth would have been extremely high.

    But again, you have to prove that these particular local conditions/temperature are positively/negatively correlated with tree ring growth.

    Steve: You say: “A negative correlation just implies that Mann placed a negative coefficient on those particular tree samples. ” No, it doesn’t. There are many possibilities. For several instrumental temperature series, Mann places a negative coefficient on actual temperature records – so that higher past temperatures in some stations in the past entail lower reconstructed temperatures. All possibilities exist.

  29. Lloyd Graves
    Posted Mar 17, 2008 at 9:19 PM | Permalink | Reply

    Tree ring sites are no more related through “teleconnection” than any other specific real estate. It’s still location,location,location.

  30. David Holland
    Posted Mar 18, 2008 at 4:23 AM | Permalink | Reply

    Re #25 Charlie,

    This old paper is quite interesting, perhaps it “was the SUN what did it”.

    A Statistical Study of the Relationship between the Solar Cycle Length and Tree-Ring Index Values
    Z. Keqian and C.J. Butler
    Journal of Atmospheric and Solar-Terrestrial Physics 1998

    It is available at the Armagh site in .ps form, and converted here (at least for a while) to .pdf

  31. David Holland
    Posted Mar 18, 2008 at 4:33 AM | Permalink | Reply

    Unfortunately that link I gave in #30 has timed out but you can convert the .ps here

  32. danbo
    Posted Mar 18, 2008 at 4:58 AM | Permalink | Reply

    I just want to make sure I’m reading fig 3 correctly.

    The greatest coorelation between tree rings and tempertures is a neg .3. So .7 doesn’t coorelate to temperture or coorelates to other unknown or undefined variables? I’m assuming that’s based on 1 as a total coorelation or exact match.

  33. John A
    Posted Mar 18, 2008 at 5:06 AM | Permalink | Reply

    The solution is simple.

    Get a sensitive microphone and tape it securely to one side of a bristlecone pine.
    Get a stick of dynomite, and tape it to the other side.

    Set off the stick of dynomite, record the sound waves.

    Viola, problem solved.

    The problem being “How to demolish a tree using a stick of dynamite”.

    One would assume, just for the sake of argument, that if you hit a nail into a tree rather than use dynamite, you’d get an interesting result from propagation which could be calibrated from a single tree core taken immediately afterward.

    One question remains: now that you know the full 360 degrees story of ring widths, how do you combine this into something that measures some climate proxy variable that makes any sense? In other words, is there any useful information in there?

  34. MarkW
    Posted Mar 18, 2008 at 5:13 AM | Permalink | Reply

    I was actually thinking of using some kind of automated thumper. Sort of like a small hammer hitting a metal plate that is in contact with the bark, for the sound source.

    I agree that it is highly unlikely that trees can perform the treemometer function, I was just thinking that if we are going to gather data, it is best to gather the most accurate data possible, and doing so in a way that has almost no chance of damaging the tree in question.

  35. Willis Eschenbach
    Posted Mar 18, 2008 at 5:16 AM | Permalink | Reply

    danbo, the correlation between temp and tree ring in fig 3 varies between about -0.3 and 0.3. There is no place where the correlation is 0.7.

    w.

  36. danbo
    Posted Mar 18, 2008 at 5:29 AM | Permalink | Reply

    I agree. There is no .7. What I was asking. Assuming “1″ is a total and exact corelation. In the best of the coorelations it’s a .3. To therefore to find a total coorelation. Even in the .3 we would have to look to variables other than temperture for .7 not found.

    Seems to me even a .3 isn’t a high degree of coorelation.

  37. Geoff Sherrington
    Posted Mar 18, 2008 at 6:02 AM | Permalink | Reply

    Yes, there are about 7 sharp declines in the graph of Fig 2, with slow rises between them. I have no idea whatsoever as to what could cause that. A new 100 year cycle.

    To me, these are just noise. One cannot do much with tree data on a yearly basis, when a single unseasonal event like a hot or cold snap with defoliation or disbudding can affect the tree for several years. Trees produce treelings and there are reasons why these are sparse stands and not dense thickets, so one has to assume that damaging events are not uncommon. Unless you have the fine structure of the data, like daily temperatures and notable events, you cannot hope to reconstruct with confidence.

    Since PCs are all the go, I’d love to see the Mother of All PCAs, where as much temporal data as is possible to find in the USA is lumped into one huge analysis. Not just climate factors, but stock exchange results, metal prices, crop yields, human and animal births and deaths, average weight at birth (I was 12 lb 7 ozs), sunspots, tide levels, immigrant numbers, exchange rates, petroleum consumption, funding to Universities, etc etc. It would be valuable to show how the methods can produce unexplained results and place significance tests into a context of “You have to do better than this”.

    Steve, inhomogeneity equals heterogeneity. Neater word.

  38. Charlie Young
    Posted Mar 18, 2008 at 6:38 AM | Permalink | Reply

    Reply to David Holland:#30-31
    David,
    Thanks for the information. I downloaded the paper and will give it a read.
    Charlie

  39. Steve McIntyre
    Posted Mar 18, 2008 at 6:39 AM | Permalink | Reply

    Woodhouse and Overpeck (BAMS 1998) show the following comparison between a tree ring reconstruction of drought and observed Palmer PDSI in 1934 – a year of low overall growth. It’s not that relevant specialists are unaware of the connection between U.S. tree rings and drought. Exactly why none of them ever commented on these issues in connection with MBH is something you’d have to ask them.


    From Woodhouse and Overpeck BAMS 1998

  40. welikerocks
    Posted Mar 18, 2008 at 6:42 AM | Permalink | Reply

    I would suggest contacting the Osmond Brothers Family. They just had their 50th Reunion Special on PBS and Oprah and the family now numbers one hundred people or so. Certainly all that ample tooth enamel could be used as a proxy to tell us something about the environment! :D

  41. Paul Foote
    Posted Mar 18, 2008 at 8:01 AM | Permalink | Reply

    Willis: “it seems almost inevitable that in this case PC1 represents a combination of the temperature and the moisture variables. I say this because as any farmer knows, the most important things for a plant are sun and water.

    Thus, the PC1 is not a temperature proxy, it is a proxy for some unknown combination of temperature and moisture plus some minor players.”

    May I suggest that the fertility of the soil has a major effect on tree growth along with precipitation? I am fairly certain that the trees’ roots are responsible for taking in more nutrition than the leaves or needles. And I don’t believe trees respond to temperature at all except when it gets so high that they burst into flames as in a forest fire. So the only thing tree rings are a proxy for is soil condition and precipitation. Anything else that effects a trees growth (wind) is minor and cannot be resolved from looking at tree rings.

    You guys have bought into this nonsense about temperature and are now trying to herd cats.

  42. Posted Mar 18, 2008 at 8:29 AM | Permalink | Reply

    re 39:
    Looks like the Almagre BCP flourish with drought, or something is wrong with the data. :-D

  43. Chris Knight
    Posted Mar 18, 2008 at 10:12 AM | Permalink | Reply

    Apologies if this has been covered before, but are tree ring growth rates sensitive to increasing CO2 levels?

    Steve: Some say yes. Based on the sampling at Almagre, I’m inclined to think that this issue has been overrated and that the effect is slight. However, there are other BIG issues with the Graybill chronologies which purport to show increased growth rate – largely related to sampling problems with strip bark trees. See posts on Almagre,

  44. gdn
    Posted Mar 18, 2008 at 5:33 PM | Permalink | Reply

    Folks, you’re missing an important nuance on this post which is raising a point different than the usual one about trees and temperature and/or PCs. It’s the problem of trying to do PCs on geographically inhomogeneous data sets and pretending that it’s continentally valid. PC will pick out patterns but if you’ve concentrated local patterns you’ll get local results.

    The nuance I picked out of this was that here we have yet another example of uncorrelated series (CO2 and tree-rings) appearing correlated when you over-smooth the series with generally similar trend-lines, and then adjust the scales to match. In this case, smoothing simply hides the absurd details.

  45. MrPete
    Posted Mar 18, 2008 at 9:47 PM | Permalink | Reply

    Paul Foote,
    Trees do respond to temp in a sense — too high or too low and their growth is stunted.

    OTOH, “too low” can in many cases mean “no H2O available because it is frozen” — which is actually a water issue not a temp issue.

  46. ferris
    Posted Mar 19, 2008 at 7:43 AM | Permalink | Reply

    Re: #41 Paul

    In order for soil fertility to have a varying affect on tree growth, nutrient availability would have to vary over time. Given the environment that trees such as the BCP grows, this is not possible. Mineralization rates will remain fairly constant on year to year basis so unless you have an influx of nutrients from an offsite source, you will not see any appreciable affect on growth rates.

    Re: #45 MrPete

    Keep in mind that “too high” is variable and dependent upon plant available moisture. As an example, you could have 85 degree temps with limited soil moisture and have restricted growth. On the other hand, you could have 90 degree temps with abundant soil moisture and have unrestricted growth. The upper growth limits are wholly dependent upon soil moisture which is why ring widths don’t correlate well with temperature. If you had abundant soil moisture throughout the growing season, the correlation would be good. But that’s not the case in the environment that most of these tree are found.

  47. Paul Foote
    Posted Mar 19, 2008 at 8:55 AM | Permalink | Reply

    RE: #45 Mr. Pete

    Hi Mr. Pete, Please explain to me how you think high temperatures stunt tree growth. Higher temperatures mean they are going to draw more moisture from their roots to their needles, but the small surface area of their needles are the trees’ defense against that causing them to dry up and die. Wind has more of an effect causing the trees to grow at odd angles and hugging the ground as you get higher in altitude. I just don’t believe you can pull a temperature signal out of the noise of louder signals such as wind.

    RE: #46 ferris:

    Mineralization rates will remain fairly constant on year to year basis so unless you have an influx of nutrients from an offsite source, you will not see any appreciable affect on growth rates.

    Hi Ferris, Are you considering root growth in your theory? Surely root growth, even slowly, leads to new sources of nutrients for the trees that will provide a varying source of nutrition. Plus you have other sources such as animal droppings, decaying carcasses and vegetation where nutrients from those sources get washed into the mix.

  48. MarkW
    Posted Mar 19, 2008 at 9:00 AM | Permalink | Reply

    In order for soil fertility to have a varying affect on tree growth, nutrient availability would have to vary over time. Given the environment that trees such as the BCP grows, this is not possible. Mineralization rates will remain fairly constant on year to year basis so unless you have an influx of nutrients from an offsite source, you will not see any appreciable affect on growth rates.

    Given the marginal nature of the BCP environment, a single Big Horn Sheep, taking a p*** or c*** could have a big impact on the amount of growth for that season.

  49. MrPete
    Posted Mar 19, 2008 at 2:29 PM | Permalink | Reply

    Let’s see… some quick responses in the middle of other Real World things :)

    * I’m not saying trees are good temp proxies at all. Please don’t misunderstand!
    * Upper temp challenges: even with plenty of water, when it gets hot enough, various plants/trees can get rather crispy. Not healthy. Yes, some kinds have protection (else perhaps there’d be no trees in Arizona?) I was speaking in general, not about specific species.
    * Please don’t mix precip-limited with temp-limited. Remember, a dendro principle is to (try hard to) find conditions where only one factor is limiting. So a discussion of temp-limited growth presumes plenty of moisture, while a discussion of moisture-limited growth presumes good temps. And of course, those are pretty tough conditions to find, particularly for BCP’s at 10000+ feet in Colorado ;)
    * Nutrients: there’s no reason to assume constant nutrient availability over the years. Sure, some trees have that. But not all. In the harder places, as the root system grows it is possible that it will find pockets of nutrients here and there and sometimes be nutrient-limited. I’m not saying this is a certainty, just a reasonable possibility. Another very good possibility: storm damage causing limb breakage and bark-strip, leading to “recovery” growth. Look at the photos in our Almagre gallery. These trees are not growing in anything close to green house conditions :)

  50. Mike Davis
    Posted Mar 19, 2008 at 2:53 PM | Permalink | Reply

    Paul Foote:
    Heat causes stress in trees! Try growing a tree in the desert even if you give them enough water they will generally slow down and start growing again in the cooler temps.

  51. ferris
    Posted Mar 19, 2008 at 2:54 PM | Permalink | Reply

    Re: #47, #48 Paul & MarkW

    I appreciate what you are saying but in order for root growth to have a sudden impact on growth rates, one would have to assume that the roots would encounter varying concentrations of nutrients within the soil. In all practicality this is unlikely to occur because nutrients are distributed fairly evenly throughout the soil profile. This is especially true with weathered soils that contain very little organic matter.

    You also have to consider the composition of the underlying parent material. In BCP country, the parent material is granite and limestone and neither of these have the capability to supply large enough quantities of nutrients to induce periodic growth spurts.

    Also, the age of the tree has to be taken into consideration. As this link shows, the ability of an influx of nutrients to impact photosynthesis rates decreases with age. This is especially true for Nitrogen, so the Big Horn Sheep taking a leak hypothesis is not really valid.

    As far as pile of poo hypothesis, one needs to consider the amount of nutrients contained within said pile of poo. Again, there’s not enough nutrient being introduced to impact growth. Now, if you could show me where a whole herd of animals had been defecating under a tree consistently over a given period of time, I would become an instant believer. But we all know that that is not the case.

    And Paul, heat does affect growth. When the amount of water lost from the plant (through transpiration) exceeds the amount of water being taken up by the roots, growth slows. As I said before, the temperature at which this occurs is wholly dependent upon the amount of water within the soil that is available to the plant. If moisture is limited, the plant will start to shut down at a lesser temp. If moisture is adequate, shutdown will occur at a higher temp.

  52. John F. Pittman
    Posted Mar 19, 2008 at 3:46 PM | Permalink | Reply

    From Geographical Ecology: Patterns is Distribution of Species:

    p 128, “Plants have moisture limits, too…”

    p 129 “Examples of simultaneous control of plant distribution by temperature and rainfall (or other factors acting simultaneously) are common but less well worked out.

    Interestingly p 129 for a Juncus species “Perhaps at very rare intervals, during an exceptioanlly warm summer, seeds are produced at higher elevations and the long-lived plants are survirors of this occurrence.” Such a consideration per BCP’s might be interesting.

  53. ferris
    Posted Mar 19, 2008 at 4:08 PM | Permalink | Reply

    Re: #47 and #48 Paul, Mark

    I appreciate what you are saying but in order for root growth to have a sudden impact on growth rates, one would have to assume that the roots would encounter varying concentrations of nutrients within the soil. In all practicality this is unlikely to occur because nutrients are distributed fairly evenly throughout the soil profile. This is especially true with weathered soils that contain very little organic matter.

    You also need to consider the composition of the underlying parent material. In BCP country, the parent material is granite and limestone and neither of these have the capability to supply large enough quantities of nutrients to induce periodic growth spurts.

    Also, the age of the tree has to be taken into consideration. As this link shows, the ability of an influx of nutrients to impact photosynthesis rates decreases with age. This is especially true for Nitrogen, so the Big Horn Sheep taking a leak hypothesis is not really valid.

    As far as pile of poo hypothesis, one needs to consider the amount of nutrients contained within said pile of poo. Again, there’s not enough nutrient being introduced to impact growth. Now, if you could show me where a whole herd of animals had been defecating under a tree consistently over a given period of time, I would become an instant believer. But we all know that that is not the case.

    And Paul, heat does affect growth. When the amount of water lost from the plant (through transpiration) exceeds the amount of water being taken up by the roots, growth slows. As I said before, the temperature at which this occurs is wholly dependent upon the amount of water within the soil that is available to the plant. If moisture is limited, growth will start to slow at a lesser temperature. If moisture is adequate, the slowdown of growth will occur at a higher temperature.

  54. George M
    Posted Mar 19, 2008 at 6:47 PM | Permalink | Reply

    Will someone, anyone give me a rational discussion of why historical tree ring data is considered gold, but in recent centuries, where it could be directly compared to temperature records it is not? Or did I just answer my own question? I have struggled for years, long before AGW, to believe that anything useful could be teased out of tree ring data, no matter how powerful the mathematics which is applied to the analyses.

  55. MrPete
    Posted Mar 19, 2008 at 7:04 PM | Permalink | Reply

    (If the spam filter eats a post… be patient: it’s just been held, not gone… at least most of the time :)

    I just got a bit more info from Leslie H (my sweetie, who knows 1000x more about living things than I do).

    Another interesting aspect of high temp and growth (vs water). Plant also need various hormones and enzymes. Some of these break down at higher temps. So even with plenty of water, get any living thing far enough outside its preferred climate and it’s… gotta say it ;)… toast.

    Some other aspects on nutrients. Again, speculation based on hanging out with the BCP’s for a few days… clearly, storm activity has a lot of impact on these trees. Some storm-related impacts that could clearly augment or deplete nutruients for a time:

    * Gullywashers and blowdown can change the nearby landscape in several ways: exposing or covering root systems, rearranging drainage patterns, and adding or removing significant amounts of mulch and fertilizer material.

    * Just one nearby blowdown, as it “melts down” so to speak, can impact a nearby tree for quite a while.

    There’s much evidence of this kind of activity on Almagre. I’m sure it is not a unique location in that regard.

  56. ferris
    Posted Mar 19, 2008 at 10:20 PM | Permalink | Reply

    Re: 56

    MrPete, thanks for the advice on the spam filter. I’ll definitely be more patient the next time it happens.

    Some other aspects on nutrients. Again, speculation based on hanging out with the BCP’s for a few days… clearly, storm activity has a lot of impact on these trees. Some storm-related impacts that could clearly augment or deplete nutruients for a time:

    * Gullywashers and blowdown can change the nearby landscape in several ways: exposing or covering root systems, rearranging drainage patterns, and adding or removing significant amounts of mulch and fertilizer material.

    * Just one nearby blowdown, as it “melts down” so to speak, can impact a nearby tree for quite a while.

    There’s much evidence of this kind of activity on Almagre. I’m sure it is not a unique location in that regard.

    I guess I should clarify my previous comments. It’s not that I don’t believe that nutrients won’t have an impact on growth rates. Nutrients do have an impact but it’s small in comparison to moisture.

    A person needs to think in terms of the most limiting factor. IMO, based upon my knowledge of soils and plant growth, the most limiting factor in this case is moisture, so it’s reasonable to assume that moisture is going to have the greatest impact on ring widths. Heat is also important but it’s secondary to moisture.

    I also think that if a person who is on the skeptical side of the global warming debate has any desire to have his arguments taken seriously, that person is going to have to ditch these far fetched theories. Like I said above, it’s not that nutrients don’t have an impact, they do, but it’s so minimal that it should not enter into any serious conversation about ring widths. To belabor the point only serves to marginalize the skeptic.

  57. MrPete
    Posted Mar 20, 2008 at 1:55 PM | Permalink | Reply

    ferris, thanks for your thoughts. To my wife, an experienced field bio person, it is not obvious whether moisture or nutrients are more limiting up there. I’m in no position to argue either way. We took soil samples specifically because we were curious about nutrient effects. I may get a surprise, but I think the soil sample aspect never was handled due to the intensity of the tree ring work (and the samples would be useless after all this time.) Of course, one could go back up there and do soil sampling, etc… no permits needed for that!

  58. Sam Urbinto
    Posted Mar 20, 2008 at 5:07 PM | Permalink | Reply

    Some might find this interesting on the subject of PCA

    http://www.umetrics.com/default.asp/pagename/methods_MVA_intro/c/1

    Multivariate data analysis is about separating the signal from the noise in data with many variables and presenting the results as easily interpretable plots. Any large complex table of data can easily be transformed into intuitive plots summarizing the essential information. The following methods are all based on mathematical projection, but have evolved to meet different needs.

  59. Paul Foote
    Posted Mar 20, 2008 at 6:09 PM | Permalink | Reply

    Thank you Mr. Pete, Ferris, John Pittman, for addressing my points about temperature, root growth and nutrient sources. It was very interesting reading your responses. You have convinced me that temperature can have an effect on trees, but I think it is overshadowed by moisture and precipitation, and I’m going to have to go with #54 George M. on this.

    If I don’t respond to any other posts on this it is not that I am not reading them or being rude, I just don’t want to push this thread off topic or intrude more than I have.

  60. ferris
    Posted Mar 20, 2008 at 6:45 PM | Permalink | Reply

    Re: #57 MrPete

    It’s not clear from your post, but am I to understand that took some soil samples from Almegre? If you did, they would not be useless! You could still run a basic analysis for pH, P, K, OM, CEC, Base Saturation, and Micronutrients. An analysis for nitrogen content would be out of the question because it has been far too long since taking the samples, but you could definitely get some good information!

    If you do have samples, I would be more than willing to submit them for you to the lab that I use. I run an annual volume of approximately 10,000 samples through this lab so I get a very good price. In fact, I would absorb the cost if you ship them to me. If your interested, post your e-mail address and I’ll send you my contact info.

  61. MrPete
    Posted Mar 20, 2008 at 7:50 PM | Permalink | Reply

    ferris — connect with SteveM on that. All the physical material is with him right now.

  62. ferris
    Posted Mar 20, 2008 at 10:00 PM | Permalink | Reply

    Of course my offer is extended to SteveM also. :-)

Post a Comment

Required fields are marked *

*
*

Follow

Get every new post delivered to your Inbox.

Join 3,111 other followers

%d bloggers like this: