New Online Resources

The American Meteorological Society has recently placed all but their most articles online here. These include many important publications.

Going from one extreme to another, the Tree Ring Society has also placed their archives online here,

114 Comments

  1. Dave Dardinger
    Posted May 17, 2006 at 4:26 PM | Permalink

    I like the first volume of the Tree Ring Society where there’s a article called, “Necessary Information on Tree-Ring Specimens from Living Trees” by Waldo S Glock. If only the info available on modern cores had this information and it was available.

  2. Willis Eschenbach
    Posted May 17, 2006 at 9:57 PM | Permalink

    Steve, many thanks for the resources. Since several people on this blog have doubted whether the “upside-down quatratic” response to temperature was real, I looked at the Tree Ring Society records to see how long this fact has been known. I found a fifty year old paper:

    THE RELATION OF GROWTH RING WIDTHS IN AMERICAN BEECH AND WHITE OAK TO VARIATIONS IN CLIMATE

    Harold C. Fritts

    TREE RING BULLETIN, Vol. 25, December 1962, Nos. 1-2,

    http://www.treeringsociety.org/TRBTRR/TRBvol25_1-2.pdf

    ABSTRACT

    An analysis is made of beech from Ohio and white oak from Illinois using a stepwise multiple regression technique to evaluate ring growth and climatic relationships. Ring widths for beech are directly related to the moisture supply during August and to temperatures for May-July of the preceding year. They are equally related to moisture during June-August of the current year and somewhat dependent upon the precipitation of the previous winter. The earlywood width of white oak is directly related to available moisture during the preceding September and to moisture during the current June, and inversely related in slight degree to the temperature of April. Latewood width of oak is primarily dependent upon the availability of moisture during the current June and July but is somewhat related to the moisture during May of the previous year. Serial correlation is prominent from growth layer to growth layer in both species. The physiological relationships which may produce these results are discussed. Some relationships involving considerable lag in the growth response are possible controls of bud formation and food accumulation, while others involving more immediate response are primarily the effect of moisture stress within the tree.

    Several things are of note here. First, the autocorrelation of the ring widths is noted. Ring widths are highly dependent on what happened the previous year.

    Second, none of the relationships relate to annual average temperature or annual average rainfall. All are month by month, which makes sense, as heat in winter or cold in summer will make a larger difference than heat in summer and cold in winter.

    Some other relevant quotes from the paper are:

    More often the interaction of environmental factors (Billings 1952) complicates growth relationships making it difficult to analyze tree ring width in terms of a single climatic variable.

    Note that, as I have stated repeatedly, the temperature and water factors interact. We cannot analyse for just temperature.

    The average temperature in June was found to be inversely related to growth and more highly correlated than precipitation (r. as high as 0.80). However, Fritts (1956, 1960) reported that daily growth of beech in central Ohio was directly related to temperature when the effects of atmospheric humidity and soil moisture were taken into account, and that growth was greatly reduced when humidity and soil moisture were low. He suggested that such dry periods were frequently accompanied by high temperatures, and, thus, a negative correlation between ring width and temperature would be the result.

    Here we see the difficulty caused by the interacting factors. I have mentioned before, and it is confirmed here by Fritts, that when there is not enough moisture, too much heat reduces the growth. This is the origin of the “upside-down quadratic” response of the trees to heat.

    Schumacher and Day (1939) studied the relationship of white oak growth and precipitation in the southeastern United States. They found that variation in ring width was related in part to average monthly precipitation over a 15 month period, and in part to its distribution during that period. The rainfall of June of the previous year and November-July rainfall of the current year were directly related to growth, while August-October rainfall of the previous year and August rainfall of the current year were inversely related. They found that only 9 to 12 percent of the variation in width of the annual ring could be ascribed to fluctuation in meteorological factors.

    This quote shows that moisture has a different effect on trees, positive during some some months and negative during others. It also points out that the size of the signal we are looking for is very small — in that study, only 9% to 12% of ring width variation is due to the climate.

    Kleine, Potzger, and Friesner (1936) reported that in addition to direct correlation of growth in white oak with precipitation of June-August, there was an inverse relation of growth with average temperature of June-August. They also stated that little correlation existed with spring temperature, only slight correlation with precipitation in April and May, and little correlation with droughts of the previous year.

    This quote shows a negative correlation of ring width with summer temperatures, and little correlation with spring temperatures.

    Fritts (1956, 1958) reports that radial growth of beech in central Ohio starts in late April or early May, continues rapidly during May and June, and declines during the month of July.

    Here once again, as the temperature increases, the radial growth increases for a while, and then as the temperature continues to increase, the growth declines.

    Among his conclusions are:

    Tree-ring and environmental relationships are evidently complex but appear to yield to multiple regression analysis. The relatively high multiple correlations attest to the success of this technique.

    Once again, we see that we cannot analyze for just one variable, we must consider both moisture and temperature.

    The most limiting climatic conditions concurrent with the period of tree growth are droughts and high temperatures during June-August.

    This final conclusion is quite important. Mann’s assumption is that the limiting factor is low temperatures. Fritts, on the other hand, shows clearly that high temperatures are the most limiting factor.

    In addition to Fritts’s conclusions, my overall conclusion is that, because at certain times of the year high temperature increases growth and at other times of the year it decreases growth, there is no way to determine annual temperatures from an examination of tree rings.

    w.

  3. John Sully
    Posted May 17, 2006 at 11:17 PM | Permalink

    I have a quick comment on Willis’ observations: Do you really think that the “Hockey Team” is unaware of these interactions and that they don’t have a frickin’ clue as to how to sort them out?

  4. Posted May 18, 2006 at 1:09 AM | Permalink

    WOW ! great news. So many articles now available online would be helpful in increasing knowledge and information. The first article by Tree Ring Society on ” Necessary information on Tree-Ring specimens from living trees” was quite educative and interesting.

  5. Willis Eschenbach
    Posted May 18, 2006 at 1:11 AM | Permalink

    Re 3: Regarding the upside-down quadratic response to temperature, I don’t know of anyone who knows how to “sort them out” … and given the Hockey Team’s track record in how they “sort them out”, I fear I must vote that they don’t.

    In fact, they don’t even seem to be able to do many of the simpler tasks involved, including choosing the series, revealing their methods, archiving their data, and calculating the confidence intervals.

    If you have facts to the contrary, we’d all be glad to hear them. However, your unsupported assertion to the effect that “papa knows best” doesn’t carry much weight, I’m afraid.

    w.

  6. Ed Snack
    Posted May 18, 2006 at 2:01 AM | Permalink

    A very quick and appropriate answer for John Sully (#3), Yes, and Yes.

    The hockey team record to date would suggest that those are the only possible answers, certainly they have showed no knowledge, interest, or ability with respect to any confounding factors. If you disagree, please show how the team has made some efforts to establish the relevance of the Bristlecone Pine and Foxtail Pines dendro records to local temperatures.

  7. TCO
    Posted May 18, 2006 at 6:34 AM | Permalink

    Willis,

    Have you yet found a paper showing growth retardation based on a higher ANNUAL temp? Or yet another paper showing growth retardation for short periods of time?

  8. TCO
    Posted May 18, 2006 at 6:36 AM | Permalink

    And have you found a paper showing this growth retardation for stands of trees at the high elevation (or lattitude) extent of a growth region?

  9. jae
    Posted May 18, 2006 at 10:59 AM | Permalink

    Since I have not been shot down yet, I’ll repeat my belief that it is not feasible to measure temperature from tree rings, even if there is a linear relationship between growth rate and temperature. The literature clearly shows that if the necessary moisture, nutrients, etc. are available, plant growth increases with temperature from about 5 C to 25 C. It then decreases with increasing temperature above about 25 C. Let’s assume the increase in growth between 5C and 25C is linear. Now, these guys are looking for a growth increase “signal” showing an increase in “average temperature” of 1C over 70 years. That is 1/20 = a 5% change over 70 years. Let’s say we have tree ring widths of about 10 mm. Then they are trying to measure (.05)(10mm)= 0.5 mm IN 70 YEARS! It just can’t be done reliably, on the face of it.

  10. Willis Eschenbach
    Posted May 18, 2006 at 11:22 AM | Permalink

    Re 7 & 8: TCO, thanks for your questions. I have provided a wealth of papers backing up my statement, which was that above a certain temperature, more heat does not cause more growth, but rather less growth.

    To date I have shown this to be true in Australian trees, ordinary plants, and now in a host of references from the Fritts paper. I have shown that it happens on both a daily and a monthly basis, so I’m not clear about your request that a paper showing “growth retardation for short periods of time”.

    Now, you want me to do more research. Despite showing that it is happening on a daily and monthly basis, you want me to show it is happening on a yearly basis … bro’, the yearly basis is the sum of the days and the months, so if it is happening on a daily basis, it must be happening on a yearly basis. Plus you want me to show that it is happening at high elevations.

    At this point, though, I think that the burden of proof is on you. If you think that for some reason plants at high elevations are not subject to moisture stress like all of the other plants I’ve provided references for … you need to prove that.

    Because as I have shown, with other plants, when the heat is too high for the available moisture, growth slows. If you think this is not happening at high elevations, you need to provide both data and a theoretical framework to show why this is not happening.

    Me, I grew up in the mountain forests, and I’ve seen trees at the treeline wilting and browning during hot summer days. I don’t need a study to tell me that it happens at high elevations … I’ve seen it. You want to claim it’s not happening?

    Then you provide the study that says so. Me, I’ve seen it.

    w.

  11. jae
    Posted May 18, 2006 at 12:04 PM | Permalink

    re 10. I totally agree, having also spent a lot of time at and above treeline in Colorado. Part of the problem with many trees at upper elevation is the very poor and shallow soils which do not hold moisture well (think talus and sand).

  12. Michael Jankowski
    Posted May 18, 2006 at 12:41 PM | Permalink

    I have provided a wealth of papers backing up my statement, which was that above a certain temperature, more heat does not cause more growth, but rather less growth…bro’, the yearly basis is the sum of the days and the months, so if it is happening on a daily basis, it must be happening on a yearly basis.

    Should it be that simple? If a year is much warmer than average (with a prolonged period of temperature above that “certain temperature” you speak), can’t an extended growing season/extended period at optimal temps compensate?

  13. Steve Bloom
    Posted May 18, 2006 at 2:29 PM | Permalink

    It’s way less simple for lots of reasons. For example, even if it was true that boreal forests would do better with increasing temps (which I don’t believe is the case, BTW), it turns out that are other factors that come into play.

  14. jae
    Posted May 18, 2006 at 2:47 PM | Permalink

    Oh, Gawd, Steve Bloom. You always manage to come up with some obscure justification for supporting your Sierra Club position. One can ALWAYS find something that might possibly support some position. But this one is really far out. I don’t think beetles are much of a problem for trees growing at the upper limits of their range. And if they were, treelines wouldn’t generally go up with temperature. Beetle outbreaks follow a very well-known sinusoidal curve, and I think you know that.

  15. Larry Huldén
    Posted May 18, 2006 at 3:43 PM | Permalink

    # 13 by Steve Bloom. Please don’t continue with these kind of arguments !!!!!!
    Think of the consequenses !
    If plants grow better in higher temps, then insects increase in populations because of faster development. However, there is no theoretical explanations why insects would develop faster than plants! That´s why your argument is not important in this context. I am extremely interested in some proves (references) (if they exist) of what you think of here!

  16. Dano
    Posted May 18, 2006 at 4:34 PM | Permalink

    12:

    Willis is correct.

    Basic plant physiology tells us at what temperature a plant metabolizes optimally. These growth curves are well-known outside of certain comment threads.

    Best,

    D

  17. Steve Bloom
    Posted May 18, 2006 at 5:05 PM | Permalink

    Larry, did you read the whole linked article? Apparently the beetles are propagating at a much greater rate since the cold temps that used to wipe them out every winter have ceased, plus they are able to move into extensive areas where they have never been able to survive. For more info, see Alberta’s official page on the subject. Or you can just ignore it, since as any of the Canadians here will confirm the Alberta provincial government is well known to be a nest of raving AGW loonies. :) (jae, I’ll just assume you’ve been drinking again.)

    From the linked FAQ page:

    “What is their role in the environment?

    “In their normal habitats, beetles are stand-replacing factors. Beetle outbreaks remove the over-mature pine from the stand and allow other tree species to take over. However, this beetle may not play this role in Alberta’s pine forests that have evolved to the present day without the pine beetle in their ecosystem. The mountain pine beetle will be very destructive in these environments. It may have detrimental impacts on the native fauna and flora, as well as the watersheds, soils, water quality and natural ecosystem succession.

    “Why are the mountain pine beetle infestations becoming more extensive?

    “Modern fire suppression has resulted in large areas of pine forests with over-mature trees, which are more susceptible to beetle attack. Not only are Alberta’s pine forests aging, a recent warming trend in our climate is also occurring. In the past, most of Alberta has been outside the mountain pine beetle’s normal range of distribution due to the harsh winter conditions. However, with the recent milder winters, these beetles have been extending their range of distribution northwards in the last several decades. If the current warming trend continues, this invasive forest pest will become established in Alberta.”

  18. Dano
    Posted May 18, 2006 at 5:35 PM | Permalink

    Note the terminology that SB quotes: these are the words of a forester, not a climatologist or enviro-whacko (or the marginalization term du jour). Natural Fire Return Intervals limits beetle population, as does cold. Neither are present in Rocky Mountain forests.

    (overmature my hiney).

    Best,

    D

  19. jae
    Posted May 18, 2006 at 5:57 PM | Permalink

    Hey, Dano:

    Natural Fire Return Intervals limits beetle population, as does cold.

    Logging can limit them, too.

  20. TCO
    Posted May 18, 2006 at 7:05 PM | Permalink

    Willis:

    Just say “no”. Be a man.

  21. Pat Frank
    Posted May 18, 2006 at 7:40 PM | Permalink

    #17 – Yes, Steve B., you display a very commendable concern. But the real issue is whether the current warming can be laid at the feet of human-produced CO2. It can’t be so laid. The GCMs clearly cannot validly support the attribution, and the proxy studies are almost pseudo-science. And so whlle we can all be concerned about pine beetles, we needn’t feel guilty about them, need we. Maybe you ought to be sponsoring the extensive logging of mature pines so as to deplete beetle habitat.

  22. jae
    Posted May 18, 2006 at 8:12 PM | Permalink

    Hey, Dano: Fire and cold don’t explain the southern pine beetle outbreaks. There is a clear cyclic pattern of exploding populations, followed by a decrease due to disease, predators, etc. You are an alarmist Sierra Club spokesman, all the way!

  23. Steve Bloom
    Posted May 18, 2006 at 8:13 PM | Permalink

    Re #21: TCO, you do know about Google Scholar, right? Willis’ exasperation with you is understandable since the literature is littered with studies on this subject, e.g. this one specific to the boreal forests we were discussing.

    All of this should really be pretty obvious. Any given plant species is adapted to a certain climate range and isn’t found outside of it. While for any given species there are other factors affecting distribution, the fact that no plant (no complex plant anyway) is found in all climate zones should be a hint.

  24. jae
    Posted May 18, 2006 at 8:15 PM | Permalink

    I gotta hand it to you, Bloom, you have a great reserve of obfuscatory techniques. LOL. I suppose there are classes on how to do this in the Sierra Club echelon.

  25. Greg F
    Posted May 18, 2006 at 8:54 PM | Permalink

    More information can be found at Natural Resources Canada.

    Cold Hardiness

    Cold hardiness is acquired. As winter temperatures decline, the beetle develops antifreeze in its blood. By mid-winter, most larvae can withstand -40°C for short periods. For cold temperatures to have their biggest impact on mortality, they have to come early or late in the season, when the beetle is least tolerant. Also, the cold temperatures must persist for several days because, with insulating bark and snow, it can take considerable time before under-bark temperatures decline to lethal levels.

    There is also an animation that goes from 1959 to 2002.

    And Alberta has had other outbreaks.

    Alberta has had two previous outbreaks of the mountain pine beetle. The first was in the 1940s, damaging 4,000 hectares of Banff National Park. Close to 27,000 trees were cut and burned to halt the spread of the beetle. The second outbreak in the Crowsnest area, from 1977-1986, destroyed close to 3.5 million trees. In British Columbia, more than seven million hectares are affected.

    And from News Releases on the Alberta web page. The outbreak in the 40’s must have been due to GW right Steve B?

    May 21, 2003
    Border check: No forest pests allowed

    Edmonton… Over the summer months, Alberta will be on the lookout to stop forest pests from nesting in the province. From June 1 to October 31, Sustainable Resource Development (SRD) will be monitoring the transportation of pine logs that originate from areas that are infested with mountain pine beetles.

    May 13, 2004
    Alberta stops pine beetles at the border

    Travellers also asked to check their cargo

    Edmonton… To help protect Alberta’s forests, the Province will prohibit pine logs, un-manufactured timber, and forest products with bark attached from coming into Alberta. Starting June 1, the ban covers all shipments that originate from areas that are infested with the highly destructive mountain pine beetles.

    April 29, 2005
    Albertans asked to leave pine logs at the border

    Edmonton… To help protect Alberta’s pine forests, the province is banning for the fifth year in a row the importation of pine logs, un-manufactured timber and forest products with bark attached.

    Directive No. 2006-02
    Date Jan 16, 2006
    Subject: Importation of conifer logs and forest products with bark attached
    Purpose: To protect Alberta’s forests and economy from destructive forest pests that can be unknowingly introduced with shipments of imported coniferous logs and forest products with bark attached.

    From that you might just get the impression that importation on log trucks is the main source for Alberta.

  26. Willis Eschenbach
    Posted May 18, 2006 at 8:56 PM | Permalink

    Re 12: Michael, thank you for raising an interesting question:

    If a year is much warmer than average (with a prolonged period of temperature above that “certain temperature” you speak), can’t an extended growing season/extended period at optimal temps compensate?

    The missing link in your reasoning is that we are looking at whether the response (ring width) is linear with temperature. While, as you point out, an extended growing season could “compensate” for the reduced growth during the hot times, this does not make the response lineal — it just makes it somewhat less non-lineal.

    Since the (incorrect) assumption of using trees as thermometers is that the response is linear, the “compensation” does not fix the problem. In fact, it is exactly this problem (that sometimes warmth helps and sometimes it hurts, and we can’t tell which is which from the ring widths) which, for my money, makes the use of trees as thermometers a non-starter. Even in your case, how can we tell your year from a cooler year which leads to the same ring growth as your “compensated” year?

    Also, I have found nothing in the literature that says that trees respond to average yearly temperatures. In some months, high temperatures help, and in other months they hurt. But from about November to April or so, they don’t make the slightest difference at all … how on Earth can we get an annual average temperature from a thermometer that doesn’t even work for half the year, and the other half of the year has a non-linear response?

    w.

  27. Steve Bloom
    Posted May 18, 2006 at 8:57 PM | Permalink

    Re #21: Pat, it’s clear that you are unpersuadable by facts. But as long as we’re on the subject, here you go.

  28. Greg F
    Posted May 18, 2006 at 9:09 PM | Permalink

    LOL … Bloom you continue to ignore the fact that climate changes all by itself. The article you point to hardly qualifies as proof of GW.

  29. Steve Bloom
    Posted May 18, 2006 at 10:04 PM | Permalink

    Yep, all but itself, for no discernible reason, and we should all ignore the one obvious cause. Actually, Greg, I forget: Are you a solar denialist or a natural variation denialist? Just curious.

  30. Steve Bloom
    Posted May 18, 2006 at 10:21 PM | Permalink

    Re #29: jae, if you’re going to drink and comment, please at least try to figure out the point someone was trying to make before going off on some other point they weren’t trying to make. TIA.

  31. Greg F
    Posted May 18, 2006 at 10:45 PM | Permalink

    Re #30

    Yep, all but itself, for no discernible reason, and we should all ignore the one obvious cause.

    The only thing obvious is the climate changes all by itself.

    Actually, Greg, I forget: Are you a solar denialist or a natural variation denialist? Just curious.

    Actually Bloom, your biggest problem is you see the world in overly simplistic terms. It’s always A causes B. It doesn’t appear that you ever entertain the possibility that multiple variables can be responsible for an effect. Notice you’re statement “no discernible reason”, as if there could only be one cause. You then extend that oversimplification to people.

  32. TCO
    Posted May 19, 2006 at 1:42 AM | Permalink

    Steve, I’ve got my union card. I know how to look at the literature. I’ve looked at papers that Willis has cited (in the past) and shown him how he was overstating his case and what issues he was ignoring. He persists in repeating the same point, but acting as if he has inserted something new into the equation. And does so in an extremely long-winded way. The last interchange, where he didn’t have the guts to just say “no” was characteristic.

  33. Larry Huldén
    Posted May 19, 2006 at 2:57 AM | Permalink

    # 17 by Steve Bloom. The case from Canada you mentioned is interesting, but it should to be viewed in a broader context.
    I know from Finland, which has a climate corresponding to that of Canada, that there are extremely few insect species, if any, which have a range reaching some certain temperature limits. All known changes in range can be attributed to two principal factors: 1) environmental changes (caused by human settlement increase or forest industry etc.) and 2) biased sampling. The latter one is the most profound cause of apparent (and published) shifts in distribution. Variation in temperatures (monthly, seasonal and annual) is so large that no insect species will be adapted to changes in mean temperatures, instead they are adapted to opportunistic use of occasional extremes in warm or wet (or other factors) conditions. A shift in distribution is always dependant on shifts in the environment. Changes in longterm mean temperature (within the ranges now discussed) have a minimal influence on environmental changes compared with that of direct cultural influence which is far more important. I think that it will be very difficult to extract the temperature component from the cultural component because of the increasing human population.
    Insects in temperate regions have on average (average emphasized) larger populations during warm summers than during cold summers. This factor does not affect shifts in distribution. Suitable new niches do not appear just because of warm summers. A longterm increase in mean temperature (as predicted by models) in the range of 1-3 centigrades will have some influence on vegetation zones but it will be very slow. That shift will certainly be hidden behind the cultural influence. What we can measure nowadays (in distributional shifts) has nothing to do with changes in mean temperatures.
    Most apparent shifts in distribution of butterflies and moths in Finland (about 1000 species) can statistically be attributed to biased sampling (temporal and spatial sampling biases, exponential increase in sampling during 100 years, use of UV-light from the 1950’s etc. etc.). The remaining shifts (expanding or retracting) are of very different kinds. On average they cancel each other when compared to the recent increase of summer temperatures. Thus, there is no roll for temperatures in interpreting distributional shifts in any particular species.
    Shifts in distribution of single insect species are also caused by changes within the interaction system of many species. Thus, there are allways fluctuations because of metapopulation dynamics in the fauna. If the temperature has any influence on the distributional shifts, it should be visible as a shift in the range equilibrium on FAUNAL level corresponding to the level of temperature change. This is not visible for 1000 species of insects during recent 100 years in Finland.

  34. TCO
    Posted May 19, 2006 at 3:02 AM | Permalink

    I could see tree rings (if they work) having less lag then tree lines. Thus being much more sensative to high frequency signals.

  35. beng
    Posted May 19, 2006 at 10:32 AM | Permalink

    RE# 34

    To add to Larry’s comments, it seems the increase in pioneering pines in the Pacific NW due to logging of the original forests makes the modern forests more beetle-prone w/o invoking temp changes. The forest changes also open up more “pathways” for expansion (including northward) from the beetles’ earlier ranges.

  36. Pat Frank
    Posted May 19, 2006 at 11:01 AM | Permalink

    #27 — The primary issue remains the anthropogenic CO2 part of the global warming claim, Steve B., and not the warming itself. Your linked article is irrelevant to that point. I’m persuaded by science not by an insistent polemic, even when it’s yours.

  37. Dano
    Posted May 19, 2006 at 12:04 PM | Permalink

    22:

    The sub-discussion centered around Rocky Mountain forests. You are trying too hard.

    26:

    willis,

    there is not a linear response to temp in tree rings. Lots of analysis is involved to interpret rings. One doesn’t core a tree and sit down on the forest floor and say ‘a-ha!’. Your Also, I have found nothing in the literature that says that trees respond to average yearly temperatures. is because it’s not a well-framed question. Temps vary from year to year and in nature, one doesn’t expect the same average temp, say, three years in a row. Hence the reason you don’t find it. Lastly, your how on Earth can we get an annual average temperature from a thermometer that doesn’t even work for half the year, and the other half of the year has a non-linear response? points to what the dendro folk do. If you think their work is flawed, would you like to discuss your concerns on a dendro listserv to see whether your concerns are valid?

    34:

    I’m with Larry here. In my view, it is a series of ecosystem stressors that is contributing to the beetle infestations we see in many places. These stressors being fire suppression, drought, fragmentation, invasive species, and monoculturing being high on the list. Local areas may have different stressors.

    Best,

    D

  38. Michael Jankowski
    Posted May 19, 2006 at 12:26 PM | Permalink

    If you think their work is flawed, would you like to discuss your concerns on a dendro listserv to see whether your concerns are valid?

    I find the work of astrologists and psychics flawed…I’m sure that if I shared my concerns on an astrology or psychic listserv, it wouldn’t amount to a hill of beans.

  39. Dano
    Posted May 19, 2006 at 12:45 PM | Permalink

    39:

    That’s great. Thanks for sharing.

    I’m sure willis won’t take offense at your implication that his exchange won’t amount to anything.

    Best,

    D

  40. Michael Jankowski
    Posted May 19, 2006 at 1:00 PM | Permalink

    40:

    It’s so nice of you to be looking-out for Willis’ feelings.

    Assuming my point doesn’t go over Willis’ head as it apparently did yours, he’ll take no offense to my post. But maybe the astro and psychic folks will be offended for being lumped-in with the dendro folks? I guess it’s a good thing I didn’t bring-up the voodoo list serve folks!

  41. Dano
    Posted May 19, 2006 at 1:48 PM | Permalink

    41:

    My reply consisted of two lines, with the first addressing your wishy-wish to lump astrology into the same category as dendro.

    The line was composed with a voice in my mind: to be spoken as a parent would to a small child who just said something silly. The parent is at a loss for words from being surprised at the silly comparison, yet not wanting to say something negative but conveying a message that perhaps the comparison isn’t appropriate and maybe unwanted. Hence the standard use of ‘thanks for sharing’. Trying to be a good guest, see.

    Sadly, comment text often doesn’t adequately convey meaning, now all these extra bits and pixels to ‘splain. I had the whole scene in my head, but it didn’t copy over to here.

    Best,

    D

  42. TCO
    Posted May 19, 2006 at 2:02 PM | Permalink

    I’m a member of the listserv. Anyone can join.

  43. Dano
    Posted May 19, 2006 at 2:19 PM | Permalink

    The voodoo listserv? Does it have a mojo filter to stop stray voodoo?

    D

  44. Willis Eschenbach
    Posted May 19, 2006 at 2:54 PM | Permalink

    Dano, you say:

    There is not a linear response to temp in tree rings. Lots of analysis is involved to interpret rings. One doesn’t core a tree and sit down on the forest floor and say “a-ha!’.

    Perhaps you can explain, then, why linear regression (either single or multiple) is the type of analysis used to interpret rings.

    w.

  45. jae
    Posted May 19, 2006 at 3:01 PM | Permalink

    there is not a linear response to temp in tree rings. Lots of analysis is involved to interpret rings. One doesn’t core a tree and sit down on the forest floor and say “a-ha!’. Your Also, I have found nothing in the literature that says that trees respond to average yearly temperatures. is because it’s not a well-framed question. Temps vary from year to year and in nature, one doesn’t expect the same average temp, say, three years in a row. Hence the reason you don’t find it. Lastly, your how on Earth can we get an annual average temperature from a thermometer that doesn’t even work for half the year, and the other half of the year has a non-linear response? points to what the dendro folk do. If you think their work is flawed, would you like to discuss your concerns on a dendro listserv to see whether your concerns are valid?

    Just when DO the dendro guys get to the “Aha” part, where they select those trees which in their judgement “show a temperature signal?” How do they tell a temperature signal from other signals?

    BTW, Dano, you were going to get us in touch with some of the dendro guys. What happened?

  46. Steve McIntyre
    Posted May 19, 2006 at 3:33 PM | Permalink

    #42. The problem comes when you try to place significance on results reached through potentially biased choices. For example, Jacoby’s well-known NH reconstruction was created by picking the 10 “most temperature sensitive” sites and taking an average. He got an HS-shaped series which is popular in multiproxy reconstructions. If you apply this same procedure to red noise, you get a HS-shaped series and the HS-ness of Jacoby’s result is at about the median level. Jacoby has refused even to archive the other 26 sites on the grounds that they don’t tell the “story”.

    Or consider choices made for Yamal/Polar Urals. The Briffa 1992 version had a very low 11th century based on only a couple of short cores which I believe to be either undateable or misdated. The 1998 update yielded very high MWP values. A chronology from Yamal to the east, as re-processed by Briffa in 2000, had a pronounced HS-shape. Briffa dropped the Polar Urals update and adopted the Yamal series in his reconstructions. This has been followed in every subsequent reconstruction except Esper et al 2002, who used the Polar Urals update.

    These kinds of selection protocols have nothing to do with people coring in the woods, but everything to do with how they do their statistics and keep their books. There is no fountain of statistical knowledge at dendro listserv’s.

    I happen to think that the statistical issues involved with dendro are quite interesting or I wouldn’t spend the time on it. The real issue comes with trying to make MWP-modern comparisons based on this data set. The facts that there’s so little effort to reconcile the evidence of Millar et al or Naurzbaev et al with the tree ring chronologies or that there’s so little effort to analyze potential problems with “drift” are points of concern as well.

  47. jae
    Posted May 19, 2006 at 3:35 PM | Permalink

    38. Regarding my comment in 22, Dano, you missed the point (again). Same dynamics, regardless of whether it’s the southern pine beetle, or the western pine beetle. Different bug species, but same genus, and it’s basic entymology.

  48. TCO
    Posted May 19, 2006 at 3:41 PM | Permalink

    Willis,

    DanO is right. You should learn more of what the field does, what it knows. And then you should engage with people who can better respond to your percieived criticisms. Not DanO, but people on the website who are astute. Take on the BEST proponent of the counterargument. It’s not about winning an argument, it’s about testing logic and inference. Steve M does this well.

  49. TCO
    Posted May 19, 2006 at 3:43 PM | Permalink

    by “the website”, I mean the listserv.

  50. Doug L
    Posted May 19, 2006 at 4:46 PM | Permalink

    Re #40 42 39

    [Troll observation mode on]

    D: “line was composed with a voice in my mind: to be spoken as a parent would to a small child who just said something silly. ”

    Since the statement in #39 about psychics was spoken seriously (given the content on this site etc.) Such a response by D is a “crossed-transaction”, a typical feature of game playing.

    One can learn about it here:

    http://en.wikipedia.org/wiki/Transactional_analysis#Crossed_Transactions

    One can also go to the library.

    The Book: “What Do You Say After You Say “Hello'” by Eric Berne might be a good place to start. He also wrote: “Games People Play”

    [Troll observation mode off]

  51. Willis Eschenbach
    Posted May 19, 2006 at 5:11 PM | Permalink

    Re 49: TCO, which listserve?

    Thenks,

    w.

  52. Dano
    Posted May 19, 2006 at 5:18 PM | Permalink

    45:

    I cannot explain it. I could, had I extra time, go back and read my notes from Uni. and see what they say. Or I could direct you to the folks who can explain the procedure. And, BTW, I see comment 2 is yours. I wonder if you have caused a…erm…tipping point in the science with your comment. If you would be so kind as to let us know when your paper gets accepted, I’d appreciate it.

    48:

    Of course there is periodicity to outbreaks. No one said there isn’t. The point is the scale of the current outbreaks and why the extent is so vast. And Larry outlined why.

    This is not to say that, say, there isn’t some climate change going on here, as SB is stating. My point about forest mismanagement and human fragmentation still stands. SB’s arguments that climate is a component haven’t been invalidated here, IMHO.

    Best,

    D

  53. TCO
    Posted May 19, 2006 at 7:20 PM | Permalink

    I don’t think you’re ready for that listserv yet. Steve would be great there. Or dardinger. You need to get a little grounded. Don’t want you going in there and swinging for the fences and embaressing us. I don’t say anything yet, as I’m not ready either.

  54. Willis Eschenbach
    Posted May 20, 2006 at 2:12 AM | Permalink

    This is good.

    Dano says I should ask my questions on some dendro listserve.

    If you think their work is flawed, would you like to discuss your concerns on a dendro listserv to see whether your concerns are valid?

    TCO agrees.

    DanO is right. You should learn more of what the field does, what it knows. And then you should engage with people who can better respond to your percieived criticisms. Not DanO, but people on the website who are astute.

    OK, sounds good. I ask which listserve. TCO, in his infinite wisdom, says I’m not ready yet:

    Don’t want you going in there and swinging for the fences and embaressing us. I don’t say anything yet, as I’m not ready either.

    TCO doesn’t want me embarressing [sic] him … TCO, no need for me to “embarress” you, you’re doing so well at that, I couldn’t compete. (All of which incorrectly presupposes that I could embarrass TCO, as if I have any connection with him … set your mind at ease, TCO. Just as there is nothing you could do to embarrass me, there’s nothing I could do to embarrass you. We’re each on our own in that regard.)

    I ask again … which listserve?

    And TCO, you say:

    I’ve looked at papers that Willis has cited (in the past) and shown him how he was overstating his case and what issues he was ignoring.

    Man, I hate these vague accusations. Yes, I’ve made mistakes, we all have. But where did I overstate my case regarding dendro? What dendro issues am I ignoring? TCO, about all you’ve done so far is ask me to do your research for you. OK, fair enough, I did so, to prove my point. Then you asked me again to do more research for you, and when I refused, you whined about the wording of my refusal …

    w.

    PS — I note that neither Dano nor TCO have provided any scientific rebuttal to the points I’ve raised on this thread … but they’re happy to attack me for not knowing enough. TCO, if you don’t know enough to cite a single study disagreeing with what I’m saying, and since you admit you don’t know enough to say anything on the dendro listserve … then what on earth makes you qualified to say that I don’t know enough to ask intelligent questions on that listserve?

  55. Dave Eaton
    Posted May 20, 2006 at 11:40 AM | Permalink

    I haven’t seen Dano contribute anything other than snark. He might be right, but who the hell would know? Drive by comments. That’s how he rolls.

  56. TCO
    Posted May 20, 2006 at 2:25 PM | Permalink

    Willis,

    I know that I don’t know particle physics or even density functional theory for solid state band structures, but that doesn’t mean that I can’t evaluate someone who is a civilian as well and tell that he doesn’t know what he’s talking about.

    P.s. read that entire paragraph of mine that you quoted first. Read the first sentence. Then think what implication that has wrt me saying you’re not ready. And you’re not.

    P.s.s. It’s not purely a situation of “how much you know” but of your intellectual maturity. It’s not that you’re so awful. But just that that is a professional place.

  57. TCO
    Posted May 20, 2006 at 2:38 PM | Permalink

    Err…second sentence.

    Willis, the two concepts that you did not address were:

    1. The issue that generally (and Steve has cited this concept in arguments as well) GW is expected to moderate winter and night temps more than it is to raise extremese (btw, in direct disagreement with your from the hip remark that it is likely to be the converse.)

    2. The concept that just because a tree spends a few percent of its time at a temp that is “over the hump” does not mean that a general raise in temp over the year will lead to reversal of growth effects. One would expect that a very huge raise overall would be required to drive the NET effect over the year over the hump. (If you you were thinking, you would make an argument of change of slope…and one would expect that to occur before the entire year’s impact went over the hump.)

    P.s. There’s nothing wrong with making mistakes (like you did on the accuracy kerfuffle). But you might want to reflect a bit that sometimes your perceived shortfalls may be partly a problem of your understanding. And go and do your best to understand and research your bases of argument (and with humility test them, with opponents). BTW, the lack of someone here to attend to you does NOT serve as a proof of the silliness of the dendro field. It may very well be silly. But the lack of someone here to address your points doesn’t prove it. Just as the lack of Lumo spoonfeeding me Arfkin does not invalidate string theory.

  58. Dano
    Posted May 20, 2006 at 3:44 PM | Permalink

    55:

    I note that neither Dano nor TCO have provided any scientific rebuttal to the points I’ve raised on this thread … but they’re happy to attack me for not knowing enough.

    I agreed with your temp points in 16, BTW (although I may have provided a linky above that showed a non-upside down quadratic for metabolism). I also merely advised you to share your thoughts with folk who do that kinda stuff for a living. I suppose that could be an “attack”, but still. Anyway,

    We just ended a hot spell that delayed the growth of my peas, beans and young Linum seedlings, and today it is too cool to put out my pepper and sunflower starts. So I may understand temp effects, although this board seems to be rife with non-gardeners who do not.

    As to your other point, trying to help you with tipping points at RP Sr’s place didn’t do me a whole lotta good, so instead this time I just directed you right to the folks so you can use that same arguing style there and see what that gets you. Apparently you don’t like that either.

    If you want me to moderate on that listserv I mentioned some time ago (likely different than the one TCO doesn’t want to sully), I’ll be back sometime Monday as I’m leaving town soon to go watch a bike race.

    Best,

    D

  59. TCO
    Posted May 20, 2006 at 4:12 PM | Permalink

    No one wants you to moderate. What are you? Some little kid who wants to play school crossing guard. I used to beat those types up…

  60. Kenneth Blumenfeld
    Posted May 20, 2006 at 5:26 PM | Permalink

    I think Dano is suggesting that he has access to a list that is not generally publicly available, meaning he (or someone else who offers it) would have to moderate any exchanges between the list members and the non-members. I am on a similar listserv for severe weather, and if somebody had questions for some of the real experts on there (i.e., not me), they would have to go through me, or someone else they knew who was on the list. I don’t see the problem with that, and he certainly never *needed* to offer it.

    It sounds like a good opportunity to me, for anyone who is interested, anyway.

    I didn’t know any kids who wanted to play school crossing guard (it wasn’t that difficult of a job to secure), but that offense would not have met my needs-a-beating criteria.

  61. TCO
    Posted May 20, 2006 at 7:17 PM | Permalink

    I’m on a listserv also. I will forward questions if I feel them suitable. You may need to pass the loyalty test that the Team America hero had to when he rejoined the team.

  62. Willis Eschenbach
    Posted May 20, 2006 at 8:56 PM | Permalink

    Re 59: TCO, you say the concepts that I have not addressed are:

    1. The issue that generally (and Steve has cited this concept in arguments as well) GW is expected to moderate winter and night temps more than it is to raise extremese (btw, in direct disagreement with your from the hip remark that it is likely to be the converse.)

    TCO, please refresh my memory. When did I say that GW is not likely to moderate winter and night temps? I don’t recall it, and a google search doesn’t turn up anything …

    2. The concept that just because a tree spends a few percent of its time at a temp that is “over the hump” does not mean that a general raise in temp over the year will lead to reversal of growth effects. One would expect that a very huge raise overall would be required to drive the NET effect over the year over the hump. (If you you were thinking, you would make an argument of change of slope…and one would expect that to occur before the entire year’s impact went over the hump.)

    TCO, you’re not following the plot here. Read what I said in post #26. In addition, it is far more than “a few percent of the time” when trees are limited by high temperatures. See, for example, A Chi-Square Test for the Association and Timing of Tree Ring-Daily Weather Relationships: A New Technique for Dendroclimatology, by Caprio, Fritts, et. al. For the trees they studied (Pinus arizonica Engelm.), on average narrow rings were significantly correlated with high temperatures during five months of the year. To quote from the paper, there is

    an inverse relationship between maximum temperature and ring width from March through July.

    Your “few percent” is very optimistic.

    w.

    PS – In logic, the converse of “if P then Q” is “if Q then P”. The converse of “global warming causes moderate winter and night temps” is “moderate winter and night temps cause global warming”. I think you might have meant something else.

  63. TCO
    Posted May 20, 2006 at 10:01 PM | Permalink

    Willis:

    1. You did make a remark to the effect that higher temps would be more indicated by hotter summers. If I bother I can find it.

    2. You also took forever (and until now have not directly to me) to acknowledge the basic point of issue that addresses your silly "I found a place at the hottest part of the summer where growth is limited". Even reflectling your latest remark, I bet you are dwelling on a high temp that only occurs during one part of the day and not thinking about the rest of the time during the day when the temp may be on the other side of the hump.

    3. [snip of obscene comment]

  64. TCO
    Posted May 20, 2006 at 10:11 PM | Permalink

    I’m looking for the specific instance, Willis. It’s actually kind of hard since you tend to come in with the same plaintive wail in topics that aren’t even about quadraticy.

  65. Willis Eschenbach
    Posted May 21, 2006 at 1:11 AM | Permalink

    TCO, this is nonsense. You say:

    1. You did make a remark to the effect that higher temps would be more indicated by hotter summers. If I bother I can find it.

    Let me know if you find it, we’ll take a look at the context.

    2. You also took forever (and until now have not directly to me) to acknowledge the basic point of issue that addresses your silly “I found a place at the hottest part of the summer where growth is limited”. Even reflectling your latest remark, I bet you are dwelling on a high temp that only occurs during one part of the day and not thinking about the rest of the time during the day when the temp may be on the other side of the hump.

    Clearly, in your hurry to prove me wrong, you did not bother to read the paper I quoted. You should read it, it’s a fascinating study. However, let me quote the significant part again: Quoting directly from the paper, it said there is:

    an inverse relationship between maximum temperature and ring width from March through July.

    This is not "a place at the hottest part of the summer". Note the dates … it’s not even in the summer. And it’s not "a place", what part of "five months" don’t you understand?

    Nor is it a high temperature that occurs during part of the day and is offset by the additional growth during the part of the day when the temperature is cooler. It is a an inverse relationship between maximum temperature and ring width, which means THE HOTTER THE TEMPERATURE, THE NARROWER THE RINGS! Read the paper, TCO, it is not offset by anything, otherwise the rings wouldn’t be narrower, would they? Your claim that it is somehow offset by extra growth from warmer temperatures during the part of the day when temperatures are on "the other side of the hump" just shows you haven’t done your homework.

    [snip- I've deleted Willis' reasonable response to an obscene comment since I've also deleted the obscene comment]

    w.

  66. TCO
    Posted May 21, 2006 at 7:59 AM | Permalink

    Willis, I did read your remark here about 5 months. I’m referring to early discussion when I characterize your poor logic in the discourse.

  67. Michael Jankowski
    Posted May 21, 2006 at 8:27 AM | Permalink

    My reply consisted of two lines, with the first addressing your wishy-wish to lump astrology into the same category as dendro.

    And that first line exemplified how my point went over your head. I wasn’t lumping them together in that post. Permit me to ‘splain. Even armed with the body of scientific evidence suggesting astrology and psychic readings are a farce, how far do you think someone will get on an astrology or psychic listserv expressing those concerns? Post all the “linkies” you want, but you likely won’t accomplish much. Now take a legitimate field, where folks have, as you’ve said before, dedicated their lives to their work. And you want to point-out a few flaws you see, with some evidence to suggest your concerns are legitimate. Why wouldn’t the results (or lack thereof) be the same?

    In other words, if you have >99.9% of science behind you and can’t get a message across on one listserve, why wouldn’t you be just wasting your time if you had just a few percent of science behind you on another listserve?

    Note: Percentages just used as examples. I don’t have a dog in the TCO et al-Willis tree-ring fight.

    The line was composed with a voice in my mind: to be spoken as a parent would to a small child who just said something silly.

    We really don’t need to hear about the voices in your head, nor experience the annoying and condescending manner in which you reveal them (as cute as you may think your style may be). But thanks for sharing! (trying to be a good guest, ‘see?)

  68. TCO
    Posted May 21, 2006 at 8:44 AM | Permalink

    Willis, I’m going to bother looking back through the blog for that one comment. Let me spell out a basic concept now, which you did not address sufficiently at the time and are now implying that you’ve fully discussed/understood.

    1. Tree rings (absent a few species and missing rings) are ANNUAL phenomenon. Almost by definition, they measure how good an entire growing season was.
    2. Trees have a nonlinear (roughly speaking “quadratic”) response to temeperature irt their growth. (I’ve never denied this, you don’t need to show me more examples of this basic concept while failing to address the deeper points in my discussion.)
    3. Given 2 and 3, in theory a ring that is narrower than a reference might be either from a cooler year (going back down the “right side” of the hump) or from a year that has crossed over the hump.
    4. HOWEVER, the data that shows the effect of 2 is all typically less then annual effects. It’s NOT a direct proof/observation of effect 3. One could easily imagine cases where the tree spends time in various regimes (too cold to grow at all-winter), slow growth in early spring/fall, rapid growth in late spring/fall, slow growth at peak summer. (It may not be perfectly symettrrical of course because of the need to replace leaves, reproductive efforts, etc.)
    5. Now when we imagine a year getting a bit warmer: say a degree. That effect could occur at various times. It could preferentially occur in the hottest part of the summer or in the slow growth onset time. Obviously these will have OPPOSITE effects. But let’s imagine for fairness that the effect occurs eaqually distributed over the year. The point is will this impact likely lead to a narrower ring or to a wider ring? If the time in “overtemp” is breif, it will lead to a wider ring. (Perhaps not linearly as wide as expected, but still not the dramatic reversal as expected in point 2).
    ****
    My whole point is that you need to address/think about such concerns rather then just cackling that trees have retarded growth at over 25Deg C.
    ****
    I have a concern that the posited issue in (2) has not been shown to occurr much. Don’t cite another survey showing daily growth measurement, Willis. That’s not addressing the concept of ring width and growing season overall impact. And that IS what we care about when we look at the rings as proxies.
    ****

    P.s. Please leave off with the comments about “not reading your papers”. I read that Aussie paper in full. On the others I skimmed and quickly asked you if those papers were showing an annual effect or were more of your less than annual (in one case it was even DAILY!) examples of growth retardation.

    P.s.s. There’s something a bit wrong in me, that I like to beat on you. That said, you really are not ready yet for the listserve. I think with your level of brains and knowledge (both reasonable, but not SteveMian) you COULD go on that listserv. But you would need to be a bit less judgemental/more open to at least open the discussion up. Than you are here. I don’t know if those guys have an answer or even a half answer, even interesting insights. But you won’t get it by charging in like a bull. These guys are twidgety academics. They’re not tough Navy nukes that like to piss and fight like me. Really you should not just think about “winning the argument” but about getting new insights. Even if you are right in some broader sense, you benefit when you get a new insight from an opponent. Just as you’ve benefited from my correction of your fundamental misconception that accuracy scales with time…and/or the failure to quickly/thoughtfully address the time aspect as it relates to quadraticy. If you promise to be extra polite and especially extra “open to the thoughts of others” (and you’re quite kind here compared to me…but I want EXTRA), then I will give you the listserv info. I really do want to make sure that you don’t embarress us or disrupt that forum. The place needs debate, but it should be introduced gently. They mostly spend their time asking how to buy cheaper treeboring tools and such…

  69. TCO
    Posted May 21, 2006 at 9:22 AM | Permalink

    Will,

    Here we see the difficulty caused by the interacting factors. I have mentioned before, and it is confirmed here by Fritts, that when there is not enough moisture, too much heat reduces the growth. This is the origin of the “upside-down quadratic” response of the trees to heat.

    This is yet another indication of poor logic/issue analysis by you. The interaction of temperature and moisture limitations may be what makes a quadratic IN YOUR MIND. But it is easy to prove that trees (even with excess moisture at ALL TEMPS) have quadraticy effects (they don’t grow at 100deg C, no?) that are purely a function of temp. You need to be more thoughtfull. (Dardie, back me up here….)

  70. TCO
    Posted May 21, 2006 at 9:24 AM | Permalink

    Should it be that simple? If a year is much warmer than average (with a prolonged period of temperature above that “certain temperature” you speak), can’t an extended growing season/extended period at optimal temps compensate?

    Thanks, Mike. Sheesh. I feel like I’m pounding on a rock here…

  71. TCO
    Posted May 21, 2006 at 9:33 AM | Permalink

    #59:

    a. a few percent is used as a thought example to drive home the point that this is a parameter that needs to be considered and is not in the simplistic (less than annual data discussions with no discussion of overall growting season).

    b. Yes, in 26, you did adress non-linearity. I think we could have gotten to this immediately several weeks ago. But glad to see you finally there. Maybe with enough 2 by 4’s to the head, the mule can move in the right direction…

  72. TCO
    Posted May 21, 2006 at 10:12 AM | Permalink

    I read the 1962 Fritts article. Well parts I read and parts I skimmed. Decent effort I think. Some questions that I have: how much does temp/moisture vary over the study area? (there were 2 temp locations so some info exists for this). What is the significance test for the Fritts conclusions? Was 3 cores each from 10 trees adequate sample gathering? What were the core to core variations within tree and the tree to tree variations?

  73. TCO
    Posted May 21, 2006 at 10:42 AM | Permalink

    I read the Dardie referenced article on tree site records. Makes a lot of sense. you could update it now by adding GPS and digital photographs. But still, I would like to have a sophisticated discussion about this. To what extent does the primary literature or the archives follow the practice of recording this info?

  74. Willis Eschenbach
    Posted May 21, 2006 at 11:40 AM | Permalink

    TCO, you say:

    Should it be that simple? If a year is much warmer than average (with a prolonged period of temperature above that “certain temperature” you speak), can’t an extended growing season/extended period at optimal temps compensate?

    Thanks, Mike. Sheesh. I feel like I’m pounding on a rock here…

    I’ve already responded to Mike’s comment. But for your sake, I’ll do so again.

    Yes, it could compensate. But the Fritts study shows that on average, excessive temperatures during five months of the year lead to narrower rings. Not just to slowed growth during that period, but narrower rings.

    Now, narrower rings are an annual average of tree growth over the year. And the Fritts study averaged the growth over a number of years.

    So clearly, while it is possible that higher temperatures could compensate over the year for higher temperatures during part of the year, on average that doesn’t happen. Instead, when excess heat comes during those five months the tree growth doesn’t compensate no matter what happens during the rest of the year.

    Have you considered the possibility that you feel like you’re pounding on a rock because you’re … umm … wrong?

    w.

  75. TCO
    Posted May 21, 2006 at 11:46 AM | Permalink

    Given that you just inserted the Fitts paper into the discussion, no. Given that it is not in a moisture unlimited regime, no. given that you can’t deconvolutd between temp quadraticy on it’s own and the confounding of temp and precip, no. Given that you didn’t even seem to understand the difference between raising an issue and arguing a position, no.

  76. TCO
    Posted May 21, 2006 at 1:11 PM | Permalink

    I have thought about it, TCO. Temps do not increase “a degree overall throughout the year”‘?. A much more common scenario will be something like a hot spell during the growing season.

    Here’s the quote motherf***er!

  77. TCO
    Posted May 21, 2006 at 2:46 PM | Permalink

    http://www.treeringsociety.org/TRBTRR/TRBvol24_3-4_2-10.pdf

    Very touching obituary for a giant in the field.

  78. Willis Eschenbach
    Posted May 21, 2006 at 4:44 PM | Permalink

    Re 76 , TCO, wish I could say thanks for posting, but you’re getting mean-spirited. You say:

    I have thought about it, TCO. Temps do not increase “a degree overall throughout the year”‘?. A much more common scenario will be something like a hot spell during the growing season.

    Here’s the quote motherf***er!

    To skip over the puerile language, here is what you originally claimed:

    1. The issue that generally (and Steve has cited this concept in arguments as well) GW is expected to moderate winter and night temps more than it is to raise extremese (btw, in direct disagreement with your from the hip remark that it is likely to be the converse.)

    The quote from me that you found has nothing to do with moderate winter temperatures. It has nothing to do with moderate night temps. My quote said a hot spell in the summer is a more likely scenario than a one degree rise which is evenly spread out during the year. It says nothing about GW, and had nothing to do with GW. It was about the relative likelihood of two different weather scenarios.

    I still think that that what I said is true, and I suspect if you search the records, you’ll find that a hot spell in the summer happens more often than an even one degree rise throughout the year. But so what? My quote has nothing to do with GW, it has nothing to do with moderate nights and winters, and thus it has nothing to do with your claim.

    So obviously, you haven’t yet found the quote you referred to, the one where you think I said something about moderate winter and night temps. Curiously, the quote you picked shows you doing what you had accused me of doing … you were the one pushing the scenario of “a one degree rise throughout the year”, which (if we had been talking about GW) would be in opposition to the idea that GW will moderate night and winter temps.

    Nice try, though. Keep looking, you might find something to substantiate your claim that I denied that “GW is expected to moderate winter and night temps”. That quote sure didn’t.

    w.

    PS — are you posting drunk again, or is the nasty name-calling just becoming part of your usual discourse? You claim you’re worrying about me “embaressing” [sic] the folks on this list? … puh-lease …

  79. TCO
    Posted May 21, 2006 at 5:39 PM | Permalink

    Willis, it was a pain in the ass to track that quote down. If you still don’t see the flaw in your logic, not much I can do for you.

  80. Willis Eschenbach
    Posted May 21, 2006 at 7:29 PM | Permalink

    TCO, I’m sorry it took you so long to track down a quote that had nothing to do with your claim … you probably would have felt it was worthwhile if the quote actually had some relationship to what you said.

    w.

  81. TCO
    Posted May 21, 2006 at 8:10 PM | Permalink

    What are you talking about. It is dead on. It refers to the issue of how a “hotter year” has its hotness distributed time-wise. Are you thick? Don’t you realize that warmer years being indicated by ones with more extreme summer weather is the opposite of the AGW moderating the winters thing? Sheesh!

    Are you just arguing? Tendentious or stupid?

  82. MrPete
    Posted May 21, 2006 at 8:57 PM | Permalink

    It sure is difficult for people to avoid talking past one another :(

    W claims that a shorter +1 degree period within a year is more likely than a continuous +1 period… for ANY reason. Not attributed to GW, AGW or anything else.

    TCO sees this statement as opposing a statement about AGW-moderated winters.

    I can see both sides of this.

    If W were claiming that this happens to the exclusion of AGW-moderation, then I think TCO would be correct.

    I don’t see W making such a claim.

    Can you folks get past that misunderstanding and stop bickering? To this observer, you’re wasting a ton of energy on something truly nitpicky; missing the forest for the trees as it were. At the very least, it is not helpful to a mature conversation, nor to any observer who walks in, to have to waste our time skipping past such catty conversations.

    To me, the key point is this: “…when excess heat comes during those five months the tree growth doesn’t compensate no matter what happens during the rest of the year.”

    If I’m reading this correctly, on its face it says that during a five month period each year, any excess heat can ruin the “growth curve” of a tree in a way that cannot be rescued. Excess heat during those five months will destroy the validity of the proxy for that year. On a purely time basis, this means you have 150 opportunities each year for the proxy to be invalidated. What’s the probability of that? I don’t know but it’s got to be pretty high — at random, warm days do show up pretty frequently in many parts of the planet!

    Or, am I reading too much into this? [No, I've not read the paper, and have no time to... in fact, I need to disappear again for some time. Too much going on...]

  83. Bruce
    Posted May 21, 2006 at 11:34 PM | Permalink

    Re #82: I have to agree with Mr Pete’s comment that it is preferable that contributors to this site, especially those supportive of Steve McIntyre and Climate Audit, moderate their comments. We should all aspire to high standards of professionalism in all that happens here. That applies to ensuring rigourous adhesion to sound science, avoidance of ad hominem arguments, and especially avoidance of unfortunate name calling.

    I have seen blogs in other areas lose support due to the deterioration of comments into name calling, and ad hominem attacks. It would be tragic if that were to happen here. In fact, I think that the site moderator(s) could arguably censor such posts on the basis that there are risks in allowing such practice to escalate. That is potentially a lot of work for the moderator(s) though, and it should not be necessary.

  84. ET SidViscous
    Posted May 21, 2006 at 11:59 PM | Permalink

    “I think that the site moderator(s) could arguably censor such posts on the basis that there are risks in allowing such practice to escalate.”

    The problem with that is that everyone will argue that they are being unfairly censored. Steve, has had issues on the o’Plaice (other place) censoring him, in fact they have a policy of moderating all posts, and few are the posts that are allowed through without some form of comment/rebuttal by the moderators. Steve has been, unarguably, censored unfairly there.

    As a result I believe that Steve prefers everything here to be un-moderated. sometimes it goes to far and he must. And even with how much Steve lets through here he is still accused of unfairly censoring people here. Which is ludicrous, particularly since the majority of censored posts are done by an anti-spam program, and Steve re-instates most of those posts within a day or two.

  85. TCO
    Posted May 22, 2006 at 6:28 AM | Permalink

    Mr. Pete:

    1. The 5 months paper is new in the discussion. Willis’s earlier remarks were made before it.
    2. Willis’s remark was about preferential warming during the over the hump period of the year. (Read the old thread, old salt.) This is the opposite direction of the expected skew.
    3. To talk about some skew versus no skew (your point) is silly and tendentious as what matters is average direction of skew and as Willis’s remark would have been a non-seqiuter to the discussion in that older thread.

  86. Willis Eschenbach
    Posted May 22, 2006 at 12:44 PM | Permalink

    TCO, thanks for your much more reasonable post #85. Let me see if I can make my position clear.

    1. As both I and other people have pointed out, my comment about the relative probability of a hot spell in the summer versus an even one degree temperature rise throughout the year had nothing to do with GW. I was just talking about how weather usually happens, not about GW. I said that a hot spell during the summer is more likely than an even temperature rise over the entire year. I still think that assessment is true. But I said nothing about GW, or about warmer nights and winters, as you incorrectly claimed. As you say, “Read the old thread, old salt”. And in any case, it makes no difference to my position, as stated below. This position is:

    2. Fritts’s paper makes it clear that on average, high temperatures during certain parts of the year result in narrower rings.

    3. Your contention (if I understand it correctly) is that if temperatures go up evenly during the year, the growth loss from the hotter temperatures during some part of the year may be counterbalanced by extra growth during some other part of the year. This is certainly possible, although the Fritts paper shows that on average, it doesn’t happen.

    4. The fact that it can happen, however, merely reinforces my contention that trees don’t make good thermometers because of their upside-down “U-shaped” response to temperatures. Let us suppose that during some evenly warmer year, the loss during the hottest part of the year is exactly counterbalanced by the additional growth during the rest of the year as you suggest. At the end of the year, the tree ends up with an exactly average width ring … but the temperatures have been higher than normal. Not much of a thermometer, if it records a warmer year as an average year.

    That is the problem that I have been pointing to, and that you seem to think doesn’t exist.

    I don’t think that this invalidates dendroclimatology in general, as there are techniques other than tree ring widths that can reveal past climate through plants. See, for example, the thread called “Millar et al: The Sierra Nevada MWP” on this site for one example (tried to link to it, but the link didn’t work).

    Ring widths, however, are an average of instantaneous growing conditions, moment by moment, over the year. If growth were linear with temperature, we could calculate the average temperature from this. But with the upside-down quadratic response of trees to temperature, this is not possible. The growth, as I have shown through citations, is of the form G = -aT^2 + bT + c, where G is growth , T is temperature, and a, b, and c are semiconstants. (They are semiconstants because they, in turn are affected by moisture and other factors).

    Now the ring width RW is the sum of this moment-by-moment growth at all moments during the year. With a function of that type, there is simply no way to determine an average value for T from a given RW. Mathematically, it’s just not possible, because it is a “many-to-one” type of relationship. With this type of relationship, there are an infinite number of instantaneous temperature values which will result in the exact same ring width. You could get exactly the same ring width from an average year, a cold year with just the right warmth during the critical months, a hot year couterbalanced by extra growth during the critical months, an excellent year except for late frosts, a cold year where the big tree that was shading our sample tree died, a perfect year that was too dry, or any one of a number of other scenarios.

    So that, as clear as I can make it, is my position. I’m interested to know what part of it you disagree with.

    w.

  87. Steve Sadlov
    Posted May 22, 2006 at 1:00 PM | Permalink

    RE: #10. Indeed, what Willis describes is even more true of Foxtails and Bristlecones. The areas they live in are not only near timberline, they are also at the boundary of two drastically different overal ecological systems. To their West lie the Pacific Slope forests and more humid strongly orographically determined suites. To their East lie the deserts and the correspondingly Summer convectively and periodic mega drought determined suites. When I think of the niche occupied by Foxtails and Bristlecones, I think of cactus and Joshua Trees, not Ponderosa Pines or Douglas Fir!

  88. jae
    Posted May 22, 2006 at 2:41 PM | Permalink

    Just as you’ve benefited from my correction of your fundamental misconception that accuracy scales with time…and/or the failure to quickly/thoughtfully address the time aspect as it relates to quadraticy. If you promise to be extra polite and especially extra “open to the thoughts of others” (and you’re quite kind here compared to me…but I want EXTRA), then I will give you the listserv info. I really do want to make sure that you don’t embarress us or disrupt that forum. The place needs debate, but it should be introduced gently. They mostly spend their time asking how to buy cheaper treeboring tools and such…

    TCO, you sound just like John Hunter. Games. If anyone could embarass this group on a listserv, it’s you, especially when you are drunk, which appears to be at least half the time. Sheesh!

  89. jae
    Posted May 22, 2006 at 2:53 PM | Permalink

    Ring widths, however, are an average of instantaneous growing conditions, moment by moment, over the year. If growth were linear with temperature, we could calculate the average temperature from this. But with the upside-down quadratic response of trees to temperature, this is not possible. The growth, as I have shown through citations, is of the form G = -aT^2 + bT + c, where G is growth , T is temperature, and a, b, and c are semiconstants. (They are semiconstants because they, in turn are affected by moisture and other factors).

    Now the ring width RW is the sum of this moment-by-moment growth at all moments during the year. With a function of that type, there is simply no way to determine an average value for T from a given RW. Mathematically, it’s just not possible, because it is a “many-to-one” type of relationship. With this type of relationship, there are an infinite number of instantaneous temperature values which will result in the exact same ring width. You could get exactly the same ring width from an average year, a cold year with just the right warmth during the critical months, a hot year couterbalanced by extra growth during the critical months, an excellent year except for late frosts, a cold year where the big tree that was shading our sample tree died, a perfect year that was too dry, or any one of a number of other scenarios.

    Even if there were no “hump” and growth varied in a positive linear manner with temperature, you STILL could not detect a general rise of 1 degree. See my comment # 9. The signal would be WAY too small to detect. The idea of using tree rings as a temperature proxy is simply undefendable.

  90. TCO
    Posted May 22, 2006 at 2:53 PM | Permalink

    Willis, my contention is that you need to consider the effect over a year of increased temps. That simply proving that for some fraction of time trees go over the hump does not prove that rings get smaller with increasing-temp years. I’m not trying to prove the converse but to flag the issue (which you had not addressed at the time).

    Also, the issue is not “even distribution” versus “hot summer”, but “hot summer” versus all the other possiblities. EVen versus hot is an example. You have to consider hot winter, hot fall and the like. Surprised that this needs to be spelled out to you. Also, still have not seen any data to back you up in saying that hot years tend to have hot summers. One would think just from probability that even distribution would be a good mix of the aggregate of hot summer, hot fall, hot winter, etc.

  91. ET SidViscous
    Posted May 22, 2006 at 3:39 PM | Permalink

    “I have seen blogs in other areas lose support due to the deterioration of comments into name calling, and ad hominem attacks.”

    Steve/John: What would be the possibility of getting a plug in to this wordpress thingie to allow ignoring of users.

    Think it would do a lot to reduce the pissing matches.

  92. Posted May 22, 2006 at 4:32 PM | Permalink

    *sigh*, TCO, you’re agreeing with Willis while disagreeing.

    Your point is that, depending on IN WHICH WAY a year is hotter, the rings will do different things. That’s his point too. His point is, a hot year might cause wider rings, or narrower ones, or the same width — depending on which parts of the year the extra heat occurs and how extreme it is, along with the growth response of the tree.

    So, do you agree there COULD be circumstances in which a certain temperature curve for a year, which averages higher than another year, could lead to reduced ring width? If you can agree with that then maybe you can stop arguing about it. Whether such temperature curves do exist — and Willis contents they do — can be the next issue of agreement. But for now, please indicate whether you agree or not that it’s not guaranteed that, just because a yearly average temperature is higher, that translates into increased ring width?

  93. Dano
    Posted May 22, 2006 at 4:39 PM | Permalink

    87:

    Hence their value relative to pondo or Doug-fir. They are well-adapted to their sites and their lifespan is that of three generations of Doug-fir, plus whether pondo and Doug-fir are climax species is dependent upon multiple factors so their permanance at a site is problematic.

    Best,

    D

  94. Willis Eschenbach
    Posted May 22, 2006 at 7:21 PM | Permalink

    Re 97, 87, there is an interesting paper in the Tree Ring Society archives on bristlecones (LaMarche, Jr., V.C., Stockton, C.W. 1974. Chronologies from temperature-sensitive bristlecone pines at upper treeline in western United States. Tree-Ring Bulletin 34:21-45.) There are some curiosities about the bristlecones.

    One is that, as with the Fritts study, ring growth is positive w.r.t. temperature in some months, and negative in others. This shows the same problem with bristlecones as with the pines that Fritts studied — two years could have identical average temperatures, but have different ring widths depending on the temporal distribution of the temperatures. From the abstract:

    Tree ring statistics show that crossdating is poorer, the climatic response is smaller, and the autocorrelation (a measure of year-to-year persistence) is greater in trees at upper treeline sites near the arid lower forest border. Climate response functions differ in many details but generally indicate a positive response of ring growth to warm temperatures in the previous late summer and autumn and current spring and summer. There is a negative response to warm temperatures during some winter and early spring months at several of the sites.

    The response to temperature seems to differ over time as well. Again from the abstract:

    Comparison of tree-growth fluctuations with meteorological observations at selected stations shows that a general warming trend between the periods 1901-1930 and 1931-1960 is reflected by an upward trend in tree growth. However, low rates of tree growth during an earlier warm period (1850-1869) may be due to a lag in the response of ring-width growth to climatic changes at upper treeline.

    Their explanation for the “lag” in the 30 years after 1850 is that the preceding period (1820-1849) was cold. During this time, not much photosynthetic mass was added to the trees, and thus they couldn’t respond to the warmer period later … which seems unlikely to me. If that were the case, why didn’t the cold period 1870-1899 affect the following years?

    In any case, it’s a fascinating study, and one which once again shows the many-to-one relationship between temperature and ring width.

    w.

  95. Terry
    Posted May 22, 2006 at 8:26 PM | Permalink

    This is off-topic, but there is a very interesting post over at RealClimate. It is co-authored by Mann, and it is very interesting because it is a very simple statistical test about whether recent observations at one arctic site are anomalous. It raises issues of normality and autocorrelation that you have posted about. It is simple enough that fairly low-level statistics are sufficient to clearly understand what is going on.

    FWIW, I posted up a few miscellaneous comments which I reproduce below.

    Very interesting post. I especially like the cleanness of the test.

    Some miscellaneous statistical points:

    Normality of the underlying distribution is very important when making inferences here. Assessing the likelihood of extreme observations depends critically on whether the distribution is normal versus fat-tailed. Checking for normality should be fairly easy. You have a lot of observations from a lot of different locations. Calculate the frequency with which “5-sigma” events occur and compare that to the probability assuming normality. My guess is that the distributions will tend to be significanlty fat-tailed and normality is not a good assumption. If you want to control for a general warming trend, use as your benchmark temperature observations from the fairly flat, even slightly cooling, period from 1940 to 1970. If you find a 5-sigma event in that period, then your test doesn’t prove anything at all.

    Are the temperature series you are testing i.i.d. or are they autocorrelated? For the results to be meaningful, you need them to be i.i.d. (or quite close to i.i.d.). I was under the impression that temperature series usually had significant autocorrelation.

    Are there any 5-sigma negative events anywhere at any time over any time period for any station? If there is, it would be a problem. (This could be possible if there is significant autocorrelation since autocorrelation can take a series quite a distance from its mean over a long enough time period.)

    To correct for potential cherry-picking, you really should do a Bonferoni-type joint test. You need to show that the odds of one site out of many possible sites showing this extreme a value is improbable. Saying there are a priori reasons for choosing this site isn’t very convincing because you can almost always make up a plausible-sounding reason why the one anomalous site is the “right” one ex post. Also, the set of possible permutations you could cherry-pick from is potentially much larger than just this one test performed over all weather stations, (think of all the possible statistics you could have computed for all the data sets there are out there), so the joint test I mention above doesn’t really adequately correct for potential cherry-picking.

  96. Willis Eschenbach
    Posted May 23, 2006 at 1:05 AM | Permalink

    Terry, could you post a link to the site? Google doesn’t know about it yet, I guess.

    w.

  97. JerryB
    Posted May 23, 2006 at 5:43 AM | Permalink

    Re #95 Terry,

    Temperature data from Svalbard Luft (Longyearbyen airport) begins, surprise, surprise, after the airport opened. GHCN has data from 1977, not a very long time span. If one looks at temperature data for Isfjord Radio, another short time span, but a different one, one notices that temperatures in that area are quite volatile. More cherry pie.

  98. TCO
    Posted May 23, 2006 at 6:39 AM | Permalink

    Nicolas: My point was that Willis’s initial “aha” of finding daily temps that retarded growth is not the same thing as finding rings that have been retarded based on the growth aspect. That he needed to take the time aspect into concern, since the ring width is the summation of a year’s growth. He is now starting to do so. However, I still don’t think he is really being fair or thoughtful. For instance, he’s not found well documented examples of the effect (narrower rings for warmer years) in temp chronologies. (Note that one of his recent citations was specifically designated as a precip stand and consisted of trees at the arid lower boundary.)

  99. Steve McIntyre
    Posted May 23, 2006 at 7:15 AM | Permalink

    #95. Terry, I checked over at realclimate on the Svalbard thread and couldn’t find your post. Is it somewhere else or did they censor it?

  100. Terry
    Posted May 23, 2006 at 8:18 PM | Permalink

    #99. Steve. It’s been 24 hours and it hasn’t been posted. Some of my posts make it through, so it isn’t the spam filter.

    Perhaps they are just taking a while to respond since it requires a bit of thought to respond to. I thought I was being very polite, so that can’t be the reason it hasn’t been posted.

  101. Steve McIntyre
    Posted May 23, 2006 at 9:00 PM | Permalink

    The last time they waded into i.i.d. was with Rasmus and they ended up looking like fools.  They probably don’t want to go through that again.

  102. TCO
    Posted May 23, 2006 at 9:21 PM | Permalink

    I find it interesting (and disapointing and indicative of what a bunch of sophists and intellectual cowards they are…including flunkouts from physics) that they won’t allow strong counterpoints to be registered until AFTER they have their response figured out. And sometimes they just kill the strong argument posts that they don’t like.

  103. Willis Eschenbach
    Posted May 23, 2006 at 10:12 PM | Permalink

    Re Svalbard, they seem to be playing fast and loose with the data … they say:

    Here are the recent Svalbard monthly surface temperature measurements, the long-term (1961-1990) means (“ybar”) and standard deviations (“sd”), and associated anomalies i.e., departure from average (“delta”) for Dec 2005 through April 2006 (all in degrees C):

    The problem is, the Svalbard station has only been reporting temperatures since mid 1977 … which makes the calculation of a 1961-1990 mean rather problematic.

    I suspect they have merged the data from two stations, Svalbard and Isfjord Radio. The problem here is that the Isfjord Radio station data ends in mid 1976 … so there’s no overlap to determine if the two stations are correlated.

    I have attempted to post the following question:

    According to GISS, the Svalbard station only started reporting temperatures in mid 1977. How did you determine the ” long-term (1961-1990) means (“ybar”) and standard deviations (“sd”)”?

    Thanks,

    w.

    To be continued …

    w.

  104. James Lane
    Posted May 24, 2006 at 5:33 AM | Permalink

    Real Climate have still not approved Terry’s post. I wouldn’t hold my breath for Willis’.

    So much for “full and frank discourse”. Where’s Lambert when you need him?

  105. Willis Eschenbach
    Posted May 24, 2006 at 10:28 AM | Permalink

    In addition to using a merged record between two datasets with no overlap, there’s another oddity about the RC calculations. They have compared the December 05 – April 06 data, not to the mean and standard deviation of the entire dataset, but to the 1961-1990 mean and standard deviation … oooh, tricksy hobbits …

    The difference is quite significant. Here is the number of standard deviations (sigma) away the 1961-1990 mean, and away from the dataset mean:

    Month__________ ’61-’90_______Full Dataset
    Dec 2005__________2.2__________1.9
    Jan 2006__________2.7__________2.4
    Feb 2006__________1.8__________1.3
    Mar 2006__________0.7__________0.5
    Apr 2006__________4.6__________3.5

    Every one of these is smaller.

    Then they say:

    The April mean temperature is almost 5 standard deviations above the mean, a “5 sigma event” in statistical parlance. Under the assumption of stationary ‘normal’ statistics, such an event is considered astronomically improbable (less than 1 in 10^6),

    Well, this is a little wonky as well. While a 5 sigma event is less than 1 in 10^6, they don’t show a 5 sigma event, they show a 4.6 sigma event (less than 1 in 10^5). Using the real figures, however, the number is less than 10^3. And this is using normal i.i.d. statistics.

    The lag 1 autocorrelation of the statistics is a bit hard to calculate, due to missing values. Interpolating the values is difficult as well, as an entire year is missing during the time of the changeover between the stations. However, the values do not change much from the missing data (~0.01-0.02). Here, I have used the more conservative figures. The effect of the autocorrelation is to increase the standard deviation, so the values have smaller “sigma” statistics. Here is the complete table:

    Month__________ ’61-’90____Full Dataset____Full Dataset w/autocorrelation
    Dec 2005__________2.2__________1.9__________1.5
    Jan 2006__________2.7__________2.4__________1.8
    Feb 2006__________1.8__________1.3__________0.9
    Mar 2006__________0.7__________0.5__________0.4
    Apr 2006__________4.6__________3.5__________2.7

    Note that the largest sigma in the bunch represents a one in 288 odds of occurence (less than 1 in 10^2). Since there are 1108 data points in the full dataset, we’d expect to find 3 or 4 months like April 2006.

    It is also worth noting that the December, February, and March temperatures were not the warmest in the record for those months. The March temperature has been exceeded no less than 27 times in the past …

    w.

  106. Steve Sadlov
    Posted May 24, 2006 at 10:39 AM | Permalink

    RE: #93. Which makes them more of a proxy for moisture than for temperature, sort of like Swamp Cyprus but in a totally different way.

  107. BKC
    Posted May 24, 2006 at 10:42 AM | Permalink

    Since there are 1108 data points in the full dataset, we’d expect to find 3 or 4 months like April 2006.

    Do you find 3 or 4 months like April 2006 in the dataset?

  108. Dano
    Posted May 24, 2006 at 12:51 PM | Permalink

    106:

    Let me know, willya, when your groundbreaking paper gets accepted? Thanks!

    Best,

    D

  109. Willis Eschenbach
    Posted May 24, 2006 at 2:41 PM | Permalink

    Dano, if you ever find some scientific issue to comment on, please do.

    I’d be happy to hear, for example, if you find anything wrong with my analysis.

    Thanks,

    w.

  110. Willis Eschenbach
    Posted May 24, 2006 at 2:52 PM | Permalink

    BKC, thanks for an interesting question. I hadn’t checked.

    I see my statement might be misinterpreted. I should have been more accurate. What I meant was that we would expect to find 3 or 4 months with a sigma of 2.7 or greater. Upon investigation, I find that there are in fact four such months in the dataset

    All the best,

    w.

  111. Dano
    Posted May 24, 2006 at 3:22 PM | Permalink

    109:

    Dano, if you ever find some scientific issue to comment on, please do.

    You missed it, willis.

    The scientific issue, willis, was that SS asserted that the trees upthread were a better proxy for moisture than temperature.

    Again, the issue was that the assertion was the trees upthread were a better proxy for moisture than temperature.

    To restate: the scientific issue was the value of moisture vs temp in certain tree rings. That was the scientific issue.

    I, therefore, told SS to write up his ground-breaking finding to share with the dendro community.

    I’d be happy to hear, for example, if you find anything wrong with my analysis.

    It looks like you’ve crunched some numbers and that they disagree with numbers given elsewhere. You’ve done an “audit” that you claim is a…er…tipping point. Skimming your post for 30 seconds, I have nothing to add, except that I’m not as qualified as you, apparently, to judge statistical interpretation. I presume your comment is being looked over.

    Best,

    D

  112. Willis Eschenbach
    Posted May 24, 2006 at 4:26 PM | Permalink

    Dano, thanks for the explanation. I understood what the issue was. I just didn’t see any scientific content in your comment. I still can’t make out what your stand is — is it that the trees upthread (bristlecones and foxtails) are not a better proxy for moisture than temperature?

    If so, perhaps both of you could provide some evidence for your claim? Also, to clarify the question, perhaps both of you could define what you mean by a “better proxy”? It’s a harder question than it appears, as the tree rings’ value as a proxy depends heavily on which month of the year, or the past year, we are discussing.

    Finally, the paper I have been discussing on bristlecones shows the results for all of the high altitude bristlecone sites studied. Using the “response function” as a measure of how good a proxy is, the results are mixed. At some sites, bristlecones are a better proxy for temperature than moisture, and at some sites it’s the other way around. There is no site which is what I’d call a “good” proxy site for either one.

    Go figure …

    w.

  113. Terry
    Posted May 27, 2006 at 5:45 PM | Permalink

    #105:

    Are you comparing temperatures to the mean of all the observations or just the observations for the same month? At Realclimate, they are comparing the most recent April temperature to the average of all April temperatures, not the temperatures from all months in all years. Obviously, you have to adjust for seasonal observations.

  114. Willis Eschenbach
    Posted Jun 8, 2006 at 9:30 PM | Permalink

    Re: 113, Terry, I’m doing it the way RC did it, comparing observations and averages month by month. Discussion of this topic is now at http://www.climateaudit.org/?p=678

    Thanks.

    w.

Follow

Get every new post delivered to your Inbox.

Join 3,317 other followers

%d bloggers like this: