RealClimate

Comments

RSS feed for comments on this post.

  1. thanks gavin!
    “Of the 18 original records, only 5 are potentially useful for comparing late 20th Century temperatures to medieval times, and they don’t have enough coverage to say anything significant about global trends.”

    Could you list those 5 explicitly
    And are you saying that reconstructions that use the other 13 should be re examined?

    Thanks!

    [Response: The Loehle 5 that extend beyond 1970 are #1, #3, #7, #8, #16 - grip, chesapeake, shihua, yang, ge. Reconstructions are done for many purposes - and calibrated in many ways. I laid down some criteria above for what is worth keeping in or not, and it's up to other people to decide whether that's appropriate. Low-resolution cores with age errors of 100's of years are not useful IMO for millennial reconstructions. We discussed Moberg et al when it came out, and the problem there is the same as that highlighted above - how to you calibrate low-resolution data? That is still an open question. - gavin]

    Comment by Steven mosher — 7 Dec 2007 @ 1:10 PM

  2. Do you know if anyone is working on taking a core sample from Lake Tanganyika? I have read that such a thing could be very useful.

    Comment by Milan — 7 Dec 2007 @ 3:04 PM

  3. This one actually sounds like you could use the old Extended Kalman Filter to assimilate the various series; the individual natural or artificial smoothings peculiar to each series being the nonlinear parameters that the EKF would identify. The point of that is then you could have a single probability model with which all the individual series treatments were consistent.

    The reason the EKF could be expected to work well is that the proxy series appear all to have at most weakly nonlinear processing that needs to be “undone”.

    I don’t know the paleoclimate reconstruction literature enough to know if this has already been tried. It would be an old idea to meteorologists though.

    [Response: Francis Zwiers and a student of his have been looking at this, the tricky part is finding the right model for the temporal dependence. Mark Cane's group at Lamont Doherty/Columbia has looked at this as well, though the primary applications have involved sparse early historical data rather than proxy data, see e.g. here and here. -mike]

    Comment by Andrew — 7 Dec 2007 @ 3:08 PM

  4. What a great article, I have been looking for information on reconstructions, as for the low resolution data – I know core reading can sometime be inaccurate. I work for a core drilling company, and we see it all the time – inconsistancies.

    Comment by Effect of global warming — 7 Dec 2007 @ 3:11 PM

  5. Thank you for the great article. The reason I think the MWP and the LIA are important is the calibration of solar effects. Even if TSI varies only slightly, the climate response to them may be large. If we knew for certain, which we don’t, that the climate varied little when solar varied to its maximum extent possible, we could rule out solar effects. As it stands, we cannot. For instance, one has this study which shows large sensitivity to variations in solar TSI, at least in the sub arctic.

    Cyclic Variation and Solar Forcing of Holocene Climate in the Alaskan Subarctic

    http://www.sciencemag.org/cgi/content/abstract/301/5641/1890

    Comment by yorick — 7 Dec 2007 @ 3:49 PM

  6. Re 3

    That was the first thing that jumped to my mind as well. Seems like it would also provide a natural way to handle the covariance of series sampled from nearby sites.

    [Response: Actually, currently used methods in proxy-based climate field reconstruction such as RegEM are designed to deal with these precise issues. -mike]

    Comment by Tom Fiddaman — 7 Dec 2007 @ 3:51 PM

  7. “evidence for current and future climate change.”

    How can their be any evidence for future climate change?

    [Response: ummm.... evidence that supports projections of future climate change? - gavin]

    Comment by Red Etin — 7 Dec 2007 @ 4:02 PM

  8. Gavin,

    Could you please clarify this statement: “This is a function of the increase in fractionation as water vapour is continually removed from the air.”

    It is my understanding that fractionation decreases with increasing temperature, because the difference in the zero point energies of ‘light’ and ‘heavy’ isotopologues of water becomes less important as thermal energy is added to the system. In other words, both kinetic fractionation processes (i.e. evaporation) and equilibrium fractionation process (i.e. condensation) tend to discriminate against d18O more heavily at lower temperatures than at higher temperatures.

    The end result is still the same — in general, there is a positive correlation between d18O and temperature at high latitudes — but the explanation is somewhat different. Perhaps you are thinking of a different mechanism, however.

    [Response: I agree I could have been clearer. The change in the fractionation coefficient with temperature is a minor issue. Rather it is simply that the colder/drier air has had more rain out, and so the isotopic composition of the remaining vapour (and next amount of precip) has become more depleted. - gavin]

    Comment by Bruno — 7 Dec 2007 @ 4:17 PM

  9. I realize that Gavin is writing for fellow climate researchers rather than such as I who only have a medical doctorate, but surely there is some language, Gavin, that could more clearly, in plain English, describe what his objections, in the main, are, to those who raise some doubts as to the long-term climate record and what may have caused previous warmings. That previous warmings did occur is not in doubt and that basic fact caused humankind and our hominid ancestors to lose much of their body-hair. This fact is reassuring as Canadians try to accomodate to the current very frigid ( unseasonably so) temperatures right across this country. One awakes with frozen water pipes and wonders whether or not one can wait for warming to occur and the sooner the better.

    Comment by Vern Johnson — 7 Dec 2007 @ 6:01 PM

  10. Robert Strom, professor emeritus at Arizona State has a new book out entitled Hot House which seems worth reading. He has a chapter on Holocene climate that addresses this topic – here is a quote:

    “The climate reconstructions for the past 2,000 years have led to a simplistic picture of a Medieval Warm Period and a Little Ice Age. Instead, the records of climate variability indicate much more complex patterns of past regional variations that rarely coincide with the actual patterns of hemispheric or global average variations (Mann and Jones 2003, Jones and Mann 2004). They are probably biased due to emphasis on one part of the world such as the North Atlantic/Europe region. . . It is probably better to view the climate changes during the last 2,000 years in terms of cool and warm centuries in various parts of the world. For example, the early 19th century was cool in North America. In Europe the 16th, 17th, and 19th centuries were cool, but the 18th century was warm. Eastern Asia had a cool 19th century, and there was a cool period in the tropics from 1650 to 1750. During the Little Ice Age there was a discernible warm period and during the Medieval Warm Period there was a cool period. In other words, there was considerable climate variability throughout the past 2,000 years, but most of the variability appears to have occurred regionally in the Northern Hemisphere.”

    I suppose the reason this came up at all is that Steve McIntyre has been trumpeting it over at ClimateAudit. He also claims that the failure of Loehle to describe the uncertainties laid out in this RC post are not important becuase, as he says “. . . in my opinion, uncertainties are not appropriately discussed in any proxy reconstruction article.” So much for all of paleoclimatology.

    Reading the scientific-sounding but content-free word salad of McIntyre and comparing it to the above post is pretty revealing. For example, Mcintyre: “Loehle’s network is the first network to be constructed using series in which every proxy as input to the network has already been calibrated to temperature in a peer reviewed article. This is pretty amazing when you think about it. It’s actually breathtaking.” Really?

    Congress requested that the NRC look into the temperature reconstructions, and they produced this 2006 Report from the National Research Council. As they point out, “Collecting additional proxy data, especially for years before 1600 and for areas where the current data are relatively sparse, would increase our understanding of temperature variations over the last 2,000 years.” Thus, if Loehle actually wanted to make a useful contribution, he should have gone out and collected some new data in the field, rather than reworking a selected subset of proxy records.

    Comment by Ike Solem — 7 Dec 2007 @ 6:11 PM

  11. Re #9: [That previous warmings did occur is not in doubt and that basic fact caused humankind and our hominid ancestors to lose much of their body-hair.]

    Somehow I doubt that the conclusion re warming causing loss of body hair is at all warranted, since we don’t see other animals in similar habitat losing theirs.

    Comment by James — 7 Dec 2007 @ 6:22 PM

  12. I believe we lost our body hair in Africa. We never had hair that would be suitable for survival outdoors in a cold climate. Humans started showing skin for some other reason.

    Comment by J.C.H. — 7 Dec 2007 @ 6:33 PM

  13. Thanks gavin.

    I am having troubles reconciling the numbering scheme you use for Loehle proxies.
    Did you get this numbering scheme from some other place? If so, you should cite the source.

    [Response: .. and butter wouldn't melt in your mouth..... ;) . But if you want to know, the numbering scheme is that of the individual data files in the bundle that Eric Swanson sent me, who got it from Loehle. The # match to location is as follows: 1: Greenland (Dahl-Jensen), 2: Conroy Lk (Gajewski), 3: Chesapeake (Cronin), 4: Sargasso (Keigwin), 5: Caribbean (Nyberg), 6: Lk. Tsuolbmajavri (Korhola),7: Shihua (Tan), 8: China composite (Yang), 9: Spannagel (Mangini), 10: East Atl. (deMenocal), 11: Cold Air Cave (Holmgren), 12: Norwegian Sea (Calvo), 13: N. America pollen (Viau), 14: W. Trop. Pac (MD982181, Stott), 15: W. Trop. Pac (MD982176, Stott), 16: China phenology (Ge), 17: S. Atl. (Farmer), 18: N. Pacific (Kim). - gavin]

    Comment by Steven mosher — 7 Dec 2007 @ 6:40 PM

  14. Sorry, 9 Vern Johnson, but probability and statistics is a laboratory course that changes you into a different person. Once you are changed, you can’t go back. You had best be a math or physics student before attempting to take it. It is the course that separates those who will be scientists from those who won’t. Most physics majors change majors during their first Prob&Stat course. There is no royal road to mathematics, and the road to Prob&Stat is probably the toughest undergrad course there is. As with all physics/math courses, it only gets far, far harder in graduate school. There is definitely NOT any way to express it in English, plain or otherwise. If there were, you still wouldn’t get it unless you had already taken as many statistics and math courses as Gavin has. Gavin did as well as possible. Just be glad he didn’t give you the real thing. The best you can do is to believe what Gavin tells you.

    Comment by Edward Greisch — 8 Dec 2007 @ 2:50 AM

  15. That previous warmings did occur is not in doubt and that basic fact caused humankind and our hominid ancestors to lose much of their body-hair.

    I was taught that it was due to the invention and adoption of clothes/bedclothes and fire. No other animal wears clothes, no other animal makes fire. Were my teachers wrong?

    Comment by Martin Vermeer — 8 Dec 2007 @ 5:14 AM

  16. #9, 11, 12

    “…previous warmings…caused humankind and our hominid ancestors to lose much of their body-hair”

    Other species didn’t lose their fur and feathers. If this had been an evolutionary response, it should have been evident in other species. Seek ye another reason for loss of body hair in humans!

    Comment by Red Etin — 8 Dec 2007 @ 6:53 AM

  17. Humans lost their hair through neoteny. Due to the increasing size of human heads, birth had to take place earlier and earlier in development as time went on, in order to get a baby through the pelvic canal. Thus adult humans retain many juvenile traits, such as hairlessness.

    Comment by Barton Paul Levenson — 8 Dec 2007 @ 7:44 AM

  18. RE 13 Thanks gavin, Craig Loehle has thanked Eric for finding the dating errors as well.

    Comment by Steven mosher — 8 Dec 2007 @ 9:02 AM


  19. I was taught that it was due to the invention and adoption of clothes/bedclothes and fire. No other animal wears clothes, no other animal makes fire. Were my teachers wrong? …”

    Barton’s explanation is probably better and correct, but I think you’ll find a more interesting answer arise by reading an obscure anthropology journal called Playboy.

    As for any Canadian having frozen water pipes, the pelvic canal was definitely always big enough for that one.

    Comment by J.C.H. — 8 Dec 2007 @ 9:34 AM

  20. It is impressive that Loehe’s reconstruction is not dependant on any one time series. He does the tests overlaying all the n-1 and 18 of the n-4 subsets of data. Will the hockey stick tree-ring reconstructions pass that thest? If not, this reconstruction should be considered more believable.

    [Response: This kind of robustness is a standard test - see Osborn and Briffa, or the Wahl and Ammann papers linked above. I would argue that validation is a more stringent test. - gavin]

    Comment by Buck Smith — 8 Dec 2007 @ 9:36 AM

  21. Vern, Personally, I think Gavin’s post is pretty clear, but to translate a bit for the nonscientist:
    1)Dating–The question here is whether the dataset is appropriate for the period you are interested and whether it has enough resolution that you can make meaningful statements.
    2)Fidelity–Are you really measuring what you think you are, or are compounding effects producing spurious correlations.
    3)Calibration–How do you turn your measurements of your proxies into estimates of what you are interested in?
    4)Compositing–How do you combine your data–with different errors, resolution, temporal and spatial coverage, etc.–so that you can make meaningful statements about the questions you’d like to answer–e.g. global temperatures.
    5)Validation–How do you demonstrate that what you have done actually works? Usually, you apply it to a dataset where you know the answer already by some accepted, independent methodology.
    Hopefully that helps. However, it should have been clear from what Gavin wrote that the objections he raised were to methodology, not to the conclusions. The deficiencies in methodology more than account for the questionable conclusions.

    [Response: Thanks! Maybe I should employ you to write abstracts for me... - gavin]

    Comment by Ray Ladbury — 8 Dec 2007 @ 10:54 AM

  22. 7. [Response: ummm…. evidence that supports projections of future climate change? - gavin]

    Gavin,
    “evidence that supports projections of future..”

    Please consider what you are saying. The “evidence” that you refer to will only be available
    and testable in the future. The projections will only ever be projections until tested at a point
    in the future – Red.

    [Response: No. Projections are always for the future of course, but the confidence that are placed in those projections is based on evidence that has been accrued to date. Therefore evidence that models and theories have performed well in comparison to observations of the past, and that previous projections were borne out, do affect our confidence in future projections. Only in the case when we know absolutely nothing, do all projections have equal standing. -gavin]

    Comment by Red Etin — 8 Dec 2007 @ 11:10 AM

  23. Humans lost their hair through neoteny. Due to the increasing size of human heads, birth had to take place earlier and earlier in development as time went on, in order to get a baby through the pelvic canal. Thus adult humans retain many juvenile traits, such as hairlessness.

    That is not an ‘explanation’ in my book. Surely furs can grow after birth if there is a good enough evolutionary motive. A real explanation identifies the evolutionary pressures that favour hairlessness, as the one presented by me did.

    Comment by Martin Vermeer — 8 Dec 2007 @ 12:24 PM

  24. Vern, Edward is right in replying above to you, where he wrote:

    “… probability and statistics is a laboratory course that changes you into a different person.”

    Those like me who made it through the basic grad level statistics class decades ago for the biological sciences are perhaps only half-changed — I learned:
    – to get help from a statistician at the beginning, and
    – to keep on after I found what I wanted, until I’d found everything I’d planned to collect, and
    – to do the statistical test planned from the beginning, not to massage the data or change the test, and
    – to get help from a statistician at the end.

    Read this article, it makes the same point in detail:
    http://www.amstat.org/publications/jse/v3n1/konold.html
    ——-excerpt——
    “… implications for assessment of intuitive understanding. …
    (1) students come into our courses with some strongly-held yet basically incorrect intuitions,
    (2) these intuitions prove extremely difficult to alter,
    and
    (3) altering them is complicated by the fact that a student can hold multiple and often contradictory beliefs about a particular situation. …
    … we want to affect how students think (as opposed to how they respond on exams).”
    ——-end excerpt——–

    Statisticians want to affect how people think, as opthamologists want to affect how people see; they know we can improve.

    Comment by Hank Roberts — 8 Dec 2007 @ 12:41 PM

  25. An amateur wonders – During and since the International Geophysical Year, there has been an enormous increase in the number and precision of climate data recording devices. How does this change in data collection have an impact upon the way we perceive and interpret the results?

    Comment by Jack Garman — 8 Dec 2007 @ 1:18 PM

  26. #21
    Ray, Nicely put. This is similar to most of the points made on CA by JEG and others. But don’t they all also apply to Moberg? And if “The deficiencies in methodology more than account for the questionable conclusions.” doesn’t that also apply to Moberg?

    [Response: These things aren't equivalent. Moberg et al was an attempt to incorporate lower resolution data as well as high resolution data using a new methodology based on wavelet analysis - that is why it was interesting. The specific records they used were not so much the point, it was more a proof of concept. If they were to update it, they would likely have a larger sample and leave out the least well-dated records. More to the point, they seem to be aware of what they were doing and what they were dealing with. With Loehle, there is no new methodology to speak of, and so everything depends on the records and their treatment - you get that wrong, there's not much left. - gavin]

    Comment by bjc — 8 Dec 2007 @ 1:47 PM

  27. Jack, short answer: dramatically.

    Longer answer: I copied and pasted unedited words from your posting into Google, and found a remarkable collection of answers, thus:

    http://www.google.com/search?q=number+and+precision+of+climate+data+recording+devices.+How+does+this+change+in+data+collection+have+an+impact

    With a bit of editing to refine the question, you’ll find more.

    As Coby reminds us, there’s no “Wisdom” button — the search does bring up a lot of chaff — but just reading the first few dozen hits will get you several good articles about climate research relevant to your question. And some skeptics and irrelevancies of course.

    Comment by Hank Roberts — 8 Dec 2007 @ 2:12 PM

  28. And here’s a very specific answer (new, abstract only, but hunt around and you may find the paper or one like it available):

    Isolating the signal of ocean global warming
    Hadley Centre for Climate Change, Met Office, Exeter, UK

    “… We present a new analysis of millions of ocean temperature profiles designed to filter out local dynamical changes to give a more consistent view of the underlying warming. Time series of temperature anomaly for all waters warmer than 14°C show large reductions in interannual to inter-decadal variability and a more spatially uniform upper ocean warming trend (0.12 Wm−2 on average) than previous results. This new measure of ocean warming is also more robust to some sources of error in the ocean observing system. Our new analysis provides a useful addition for evaluation of coupled climate models …”

    http://www.agu.org/pubs/crossref/2007/2007GL031712.shtml

    Comment by Hank Roberts — 8 Dec 2007 @ 2:15 PM

  29. gavin> The simple question of whether the medieval period was warm or cold is not particularly interesting – given the uncertainty in the forcings (solar and volcanic) and climate sensitivity, any conceivable temperature anomaly (which remember is being measured in tenths of a degree) is unlikely to constrain anything.

    That may be true if you are talking about climate models, but in determining the impact of higher temperatures on ecosystems and agriculture, knowledge about the MWP and other past temperature extremes is likely very interesting.

    [Response: That's more of a regional issue, and of course, the regional patterns of change - whether forced or internal to the system -are of great interest for understanding climate dynamics. -gavin]

    Comment by Steve Reynolds — 8 Dec 2007 @ 3:52 PM

  30. Contributing to the hairlessness debate – surely the most off-topic branch in any RC thread – most people I ever examined have hair: who are these hairless humans you write of? There appears to be a significant range of extent of hairiness among individual members of the species, and no obvious reason to assume that that has not always been the case.
    As for the bigger-brain-less-hair evolutionary argument, a number of not notably intellectual species – field mice, for example – are born hairless and helpless but manage to become hairy adults.

    Comment by george — 8 Dec 2007 @ 4:51 PM

  31. George, maybe your version of heck has furry women, mine scorns them:

    http://query.nytimes.com/gst/fullpage.html?sec=health&res=9C03E0DE1030F93AA2575BC0A9659C8B63&fta=y

    Comment by J.C.H. — 8 Dec 2007 @ 5:31 PM

  32. I have a few questions, some methodological, others paleoclimatological. However, I am not sure to what extent some of them at least are directly relevant to current discussion, so please feel free to skip those which you regard as more tangential.

    *

    You state that in weighting various lines of evidence (e.g., different proxies for different subregions), you consider both the range of error (with a larger range of error resulting in a given proxy receiving less weight) and the area (with a larger area resulting in the proxy receiving more weight). Likewise, you point out that the larger the area, the lower the range of error — given the law of large numbers. As such these are two different aspects of the same issue and can’t actually be treated separately, I presume.

    But furthermore, when one combines such multiple lines of evidence, the range of error that results should be narrower than the range of error associated with any subset of those lines of evidence — although this wouldn’t necessarily be the case — if the different lines of evidence diverged to an extent that was considerably greater than their purported range of error. In this latter case, it would be necessary to re-evaluate their ranges of error. Something which occasionally happens I presume.

    But assuming there is no such conflict between the different lines of evidence, the range of error should be narrower. In that case, how often is a Bayesian-type approach employed for combining the different lines of evidence in order to arrive at that narrower range of error? Are the issues typically amenable to that sort of an approach? What alternatives are there to the Bayesian and to what extent are they in use?

    *

    One issue that I have wondered about for some time is to what extent the paleoclimate record supports the distinction between slow-feedback and fast-feedback climate sensitivity. Does it support the view that the slow-feedback sensitivity is double that of the fast feedback? How confident are we with regard to this?

    *

    It appears that we have been raising the level of certain greenhouse gases for the past five thousand years. To what extent does this appear to have affected the long-term behavior of the Holocene?

    *

    At a more abstract level, to what extent does the speed with which the climate system responds to a greenhouse gas forcing appear to be dependent upon the magnitude of the forcing? To what degree does the magnitude of the response appear to be dependent upon the speed with which the forcing is applied? Or are we even at the point that we can arrive at tentative answers to such questions? Frame them in a form that we may either analytically or empirically address them?

    More concretely, do we have reason to expect the speed of slow-feedbacks (e.g., the cryosphere) to be greater when the magnitude of the forcing is greater? Do we have reason to expect the magnitude of the feedback from the carbon cycle to be greater when the duration over which a forcing is raised to a higher level is shorter, e.g., that various carbon sinks can be overloaded by the rate of change such that they become less effective not simply in the short-term, but long-term? More prone to become net emitters?

    Comment by Timothy Chase — 8 Dec 2007 @ 5:42 PM

  33. So where are the diagrams?

    It’s pretty hard, and dull, to follow the discussion without pretty pictures :-)

    But more seriously, seeing the reconstruction in graphical form does give us a better idea of the scope of the paper.

    thanks,
    Robert

    Comment by Robert — 8 Dec 2007 @ 6:57 PM

  34. I greatly appreciate the recent critiques that have been posted on RC (in addition to being enlightening about climate, they show what scepticism in science is really about), but I wonder if there’s any chance that some of the contributors can choose a recent paper they like, and perform a similar dissection, showing what’s good, new, and significant about it. It need not be anything truly ground-breaking – it’s just that plenty of positive examples must be out there, and they, too, can be enlightening. Naturally, beggars can’t be choosers and all that; it’s just a thought. If buying gavin lunch is a necessary bribe, I could probably manage it, but I suspect raypierre’s tastes are a bit rich for my wallet.

    Comment by S. Molnar — 8 Dec 2007 @ 6:57 PM

  35. Hank,
    Thanks for the response. I read some of the links served up by Google. Beyond my grasp, but I see how there is some effort to tease apart the possible misconstructions that come from such a dramatic shift in data collection methods.

    Comment by Jack Garman — 8 Dec 2007 @ 8:08 PM

  36. Hank Roberts (#28) wrote:

    And here’s a very specific answer (new, abstract only, but hunt around and you may find the paper or one like it available):

    Isolating the signal of ocean global warming
    Hadley Centre for Climate Change, Met Office, Exeter, UK

    Same authors, same topic, I believe for public consumption — this gives the charts and an overview of what they are doing….

    A new perspective on warming of the global ocean
    Matt Palmer, Keith Haines, Simon Tett and Tara Ansell
    http://hadobs.metoffice.com/hadgoa/RAPID07_Palmer_et_al_poster.pdf

    Comment by Timothy Chase — 8 Dec 2007 @ 8:59 PM

  37. Gavin:
    I have no idea how you can argue that Moberg is somehow exempt from the methodological points that Ray raised. Whether Moberg is illustrating a new approach or an established process is irrelevant to the need to address these 5 points. Loehle has to address them. Moberg should have addressed them. They are too basic to try to rationalize why one guy didn’t incorporate them.

    [Response: If Moberg et al didn't know what 'BP' meant, or what the difference between a Mg/Ca ratio and temperature was, I'd criticise them too. Just as I would if they used inappropriately tuned records or inconsistently treated equally low-resolution records or didn't do any validation. Except that they did none of those things. Loehle did. Fix all that, and then we can talk. They are not equal just because they have 8 series (out of 18) in common. - gavin]

    Comment by bjc — 9 Dec 2007 @ 12:01 AM

  38. Martin Vermeer writes:

    [[

    Humans lost their hair through neoteny. Due to the increasing size of human heads, birth had to take place earlier and earlier in development as time went on, in order to get a baby through the pelvic canal. Thus adult humans retain many juvenile traits, such as hairlessness.

    That is not an ‘explanation’ in my book. Surely furs can grow after birth if there is a good enough evolutionary motive. A real explanation identifies the evolutionary pressures that favour hairlessness, as the one presented by me did.]]

    Your statement would be true only if hyperadaptationism of the sort favored by Richard Dawkins were true, and the vast majority of biologists don’t think it is. A lot of traits have no immediate effect on differential survival, and will survive simply because there is no evolutionary pressure against them. Not every trait is an adaptation.

    Comment by Barton Paul Levenson — 9 Dec 2007 @ 5:49 AM

  39. george writes:

    [[As for the bigger-brain-less-hair evolutionary argument, a number of not notably intellectual species - field mice, for example - are born hairless and helpless but manage to become hairy adults.]]

    You missed my point. Mice don’t retain juvenile characteristics. Humans do. Google “neoteny” to learn about what I’m talking about.

    Comment by Barton Paul Levenson — 9 Dec 2007 @ 5:52 AM

  40. Great post Gavin :)

    Could you please make a link to the study, can’t find it anywhere. Providing a link would make it much easier for us lay persons to see what this study is all about

    [Response: It is already linked. Click on Loehle's name above (para 4). - gavin]

    Comment by Strilen — 9 Dec 2007 @ 7:02 AM

  41. Gavin, I agree with your points. However, as a paleoclimatologist/paleoceanographer I would hope that you would get away from the ‘high’ and ‘low’ resolution description of climate records. Those of us in the field now describe records as annual, decadal, millennial, and milankovitch to avoid total confusion. When we work in deep-time, anything resolving Milankovitch is considered high-resolution.

    The big problem with Loehle’s paper starts at the beginning–he never shares the records with us but only the composite. If he was limited by E&E on publishing his original records and his sorting criteria, he should have published elsewhere.

    It only took a quick look at one of the papers (Kim et al., 2004), to see that Loehle was ‘cherry picking’ his data. Of the three records with marginally sufficient age control, two were not added to the composite. These two records,from the California margin, had warming trends.

    Furthermore, the data in the composite is the most mixed grab bag of data I have ever seen–one cannot make any climatological sense for the choices, so one must assume that the choices were based on some other, unstated criterion.

    We need honest help to understand how climate works; eventually real analysis will drive out the bad composites.

    [Response: Good point. I've actually made it myself when trying to translate between paleo people and physical oceanographers - they have very different conceptions of the term 'high resolution'! So to be specific, I use 'low resolution' here to imply multi-decadal and longer. Sufficiently high resolution to be useful in this kind of exercise is decadal (with similar sized age model errors). Finally, I agree, real analyses will win out in the end. I'm hopeful that isn't too far off. - gavin]

    Comment by Mitch Lyle — 9 Dec 2007 @ 8:51 AM

  42. From Gavin’s link above, you can download the data “presented in Figure 1″ at http://www.ncasi.org/programs/areas/climate/LoehleE&E2007.csv.

    It only goes up to 1980, but it’s interesting to compare it with GISTEMP. According to Loehle, 1966 was the warmest year in the last century – and since then we’ve been cooling at about 0.2 degrees per decade. :)

    Comment by S2 — 9 Dec 2007 @ 9:18 AM

  43. When trying to model the future from the past, how important is change in luminosity of the sun? Is is important, or do the GHG concentrations completely dominate?

    [Response: It's important, but more so in the pre-industrial period. For the last few decades, GHG concentrations are indeed dominant (by about a factor of 10). - gavin]

    Comment by Alex Tolley — 9 Dec 2007 @ 9:35 AM

  44. A lot of traits have no immediate effect on differential survival, and will survive simply because there is no evolutionary pressure against them. Not every trait is an adaptation.

    You’re pulling my leg, right? Are you telling me that furs have no usefulness to those animals having them?

    Comment by Martin Vermeer — 9 Dec 2007 @ 10:35 AM

  45. Re # 43 Martin Vermeer: “Are you telling me that furs have no usefulness to those animals having them?

    Try explaining the adaptive value of the coats of various breeds of dogs, such as a Besenji, a dalmation, an Alaskan malamute, and a collie.

    The term “adaptation” is used in various ways by various people. Evolutionary biologists tend to define adaptation as a trait that arose through selection for its current function. Many traits arise as a serendipitous consequence (an “accident”) of some other trait, or traits, that may be adaptive; those serendipitous traits may or may not have adaptive value, but since their function was not a result of selection, they would not be considered adaptations. This concept was articulated very nicely by Stephen J. Gould and Richard Lewontin in the their classic “Spandrels of San Marcos” essay:
    Gould, S.J. & Lewontin, R.C. (1979). The spandrels of San Marcos and the Panglossian program: A critique of the adaptationist programme. Proceedings of the Royal Society of London, 205, 581-598.

    [Note that some architects have criticised Gould and Lewontin for their ignorance of the architectural design of cathedrals and the function of spandrels. Thus, while their main point, that not all traits are adaptive, is valid, their example may not be.]

    For a more thorough analysis of the concept of “adaptation”, you might consult one of the many books on the subject, such as

    Rose, M.R. and G.V. Lauder (eds) (1996) Adaptation. Academic Press, San Diego. 511 pp.

    Comment by Chuck Booth — 9 Dec 2007 @ 12:45 PM

  46. Try explaining the adaptive value of the coats of various breeds of dogs, such as a Besenji, a dalmation, an Alaskan malamute, and a collie.

    Those were bred by humans, so I don’t see the relevance to my point, which is about the selective advantage (not: adaptive value; an interesting but subtly different subject) of having a (dense, thermally insulating) coat vs. not having one.

    Anyway I should apologize for what is arguably the most off-topic thread in ages. I suppose it has to do with a passion for figuring out how things really are — about Canadian water pipes or other things :-)

    Comment by Martin Vermeer — 9 Dec 2007 @ 2:39 PM

  47. Re #44 Cuck Both “The term “adaptation” is used in various ways by various people. Evolutionary biologists tend to define adaptation as a trait that arose through selection for its current function. Many traits arise as a serendipitous consequence (an “accident”) of some other trait, or traits, that may be adaptive; those serendipitous traits may or may not have adaptive value, but since their function was not a result of selection, they would not be considered adaptations.”

    Does this mean that some traits, like for instance a ‘timing of hair growth’ knock out mice or cave dwelling albino insects should not be considered an adaptation, since the ‘lack of a trait’ trait often is a question of an single accidental mutation and accidents happen all the time :-). While an population of mice that change from thick white in the winter to normal brown summertime, that most certainly an adaptive feature to cold climates.

    Comment by per — 9 Dec 2007 @ 2:44 PM

  48. Chuck Booth (#44) wrote:

    Try explaining the adaptive value of the coats of various breeds of dogs, such as a Besenji, a dalmation, an Alaskan malamute, and a collie.

    Chuck,

    I might agree with the point you are making, but I am not sure that this is a good example. Breeds of domesticated dogs (which technically still wolves) were subject to a great deal of artificial selection in recent millenia. What they are adapted to are their specific uses by humans, including the aesthetic value of their appearance.

    You might like the following.

    Just as the DNA transposons known as MITEs have made the domestication and adaption of rice to our needs possible at a greatly enhanced speed, dogs have been particularly plastic in large part due to the existence of tandem repeats in the regulatory protein coding areas.

    Please see:

    Molecular origins of rapid and continuous morphological evolution John W. Fondon III, and Harold R. Garner
    PNAS | December 28, 2004 | vol. 101 | no. 52 | 18058-18063
    http://www.pnas.org/cgi/content/full/101/52/18058

    All but the shorter tandem repeats are thought to have originated by way of the poly tails of LINEs and SINEs – in multicellular eukaryotes. Retroelements, genetic relics left behind from endogenized retroviruses. And interestingly, today I just ran into something on self-synthesizing transposons which appear to be descended from plasmids which acquired a retroviral integrase perhaps a billion years ago.

    Not sure if any of this might be of interest to you, but if so:

    timothy [no spaces] chase [at] g mail [dot] com

    Its been a bit of an obsession for me, the role of viruses and virus-like elements (e.g., phages) in the evolution of life.

    Comment by Timothy Chase — 9 Dec 2007 @ 3:18 PM

  49. Re
    Timothy’s interest in viruses and evolution in 47, you’ve probably read this, but just in case:

    Villarreal,
    Viruses and the Evolution of Life

    I found it to be astonishing – it looks to me like Villarreal has cracked the nut of evolution.

    Here’s his site:

    Center for Viral Research
    http://cvr.bio.uci.edu/

    Comment by Jim Galasyn — 9 Dec 2007 @ 3:31 PM

  50. Re # 45, 46, 47 Responses to my response to Martin Vermeer’s response to Barton’s off-topic post about adaptation.

    Had Martin specified wild animals with fur coats, your questioning of my canine example would be fully justified…but, he didn’t. My real point was that Barton was correct – not every trait is adaptive. While the term “adaptation” is easy to use, trying to figure out which traits are, and which are not, adaptations is difficult (in fact, when you consider that many so-called “traits” involve one or more organs carrying out multiple functions and composed of multiple tissues constructed from the products of multiple genes, it often becomes difficult to identify precisely what is a trait, let alone figure out how that trait evolved.)

    I will leave it at that – as Martin has noted, this thread has strayed well off topic (Unless one sees a parallel between the difficulty of reconstructing past climates and the difficulty of recontructing the evolutionary history of current biological traits)

    Comment by Chuck Booth — 9 Dec 2007 @ 5:14 PM

  51. Would a climate scientist here please provide a link for me showing when these climate change induced predictions were first published:

    Melting arctic
    Rising sealevels
    More Droughts
    (ie desertificated areas doubled in size from 1970s to 2003 – had anyone predicted increasing drought before the 1970s?)
    http://enviro.org.au/drought.asp
    More Floods
    More intense Hurricanes (and surely wouldn’t that include Cyclones/Typhoons outside the US?)
    New insect pests destroying forests as habitat moves North

    There seems a very entrenched bunch of deniers over at the Wall Street Journal and although I try to put them right when possible, I would like some of the above facts for ammunition:
    http://blogs.wsj.com/energy/2007/12/07/beat-the-beetles/

    Comment by Susan K — 9 Dec 2007 @ 7:45 PM

  52. “Finally, I agree, real analyses will win out in the end. I’m hopeful that is too far off. – gavin]”

    Never was a truer word said!!!

    [Response: fixed. thanks -gavin]

    Comment by Armagh Geddon — 10 Dec 2007 @ 5:15 AM

  53. Re # 50 Susan –

    Try the AIP The Discovery of Global Warming site – link in the orange box on the right side of the RC home page, or:

    http://www.aip.org/history/climate/index.html

    Comment by Chuck Booth — 10 Dec 2007 @ 10:01 AM

  54. Jim Galasyn (#48) wrote:

    Villarreal,
    Viruses and the Evolution of Life [book]

    I have known about him for years, but unfortunately haven’t been able to afford the book.

    Jim Galasyn (#48) wrote:

    I found it to be astonishing – it looks to me like Villarreal has cracked the nut of evolution.

    Here’s his site:

    Center for Viral Research
    http://cvr.bio.uci.edu/

    He seems to have been one of the pioneers. Obviously there has been a great deal going on. For example, it would appear that our spliceosomal introns and retroviruses are both descendants of early self-splicing Type II introns found in archaea and prokaryotes, where Type II introns are themselves regarded as retroelements since they are mobile. Likewise, the catalytic core of telomerase with which the telomeres at the ends of the chromosomes becomes extended is a reverse transcriptase which appears related to the reverse transcriptase found in retroviruses. Likewise, the adaptive immune system has relics of retroelements (a LINE and SINE) which make possible lymphocyte rearrangements. Viruses appear to be taking center stage along with cells in the evolution of life.

    However, one of the biggest ways in which viruses (primarily retroviruses) have contributed to the evolution of life would appear to be in the generation and dissemination of tandem repeats — which are subject to hypermutation and appear to have been subject to “indirect selection” for those regions where a higher rate of mutation might prove beneficial in dealing with a changing environment. Protein coding regions, introns and promoters. Point mutations may serve to regulate the mutability of such tandem repeats – by breaking up the tandem repeats, rendering them subject to reduced levels of hypermutation. Thus, for example, many proteins have cryptic repeats where the codons will code for the same sets of amino acids as an ordinary tandem repeat, but there will be substitutions in the individual letters of the triplet.

    Anyway, feel free to email me at the address I gave above — I can put together a list of papers on a few topics in this area (viruses, repeats, phages, etc.) along with links to some of the technical articles, where possible — if you are interested. Might take me a day or so, though.

    But it should be handled outside of Real Climate – as it is off-topic, as Charles Booth has indicated.

    Comment by Timothy Chase — 10 Dec 2007 @ 12:14 PM

  55. Re Jim Galasyn (#48) on Villarreal

    I’ve known about him for years, but I haven’t been able to purchase the book as of yet, but he seems to have been one of the pioneers. Obviously there has been a great deal going on.

    However, I have been able to find a fair amount of the literature: viruses, phages, hypermutable tandem repeats (in protein coding regions, introns and promoters) which largely appear to be relics of endogenous retroviruses, indirect selection for mutability, cryptic repeats in protein coding regions, somatic rearrangements in the adaptive immune system, telomerase, etc.. Viruses appear to be taking center stage along with cells in the evolution of life.

    Charles Booth is right about the fact that this is off-topic, but feel free to email me at the address I gave above — I can put together a list of papers on a few topics in this area along with links to some of the technical articles which are open access, where possible — if you are interested. Might take me a day or so, though.

    Comment by Timothy Chase — 10 Dec 2007 @ 12:47 PM

  56. Gavin, I would like to know a little more about your Loehle #12 reference, in which you state that it “doesn’t start until 1440 CE.” Furthering that thought you note that it is a surprise that Calvo 2002 is even used. In fact a quick look at the link that you have aptly provided suggests that the Calvo 2002 actually ends in 1440 CE.—–From Calvo 2002—Minimum Age: 0.510 kyr BP * Maximum Age: 8.490 kyr BP—
    This minimum age of 0.510 kyr (or 510 years) subtracted from 1950 CE yields 1440 CE.
    Have I missed something?

    [Response: No. Paleo conventions in time often go backwards. Therefore when using 'start', I implied the first data point from the present (think of the first data point in depth in the core). I'm not sure that semantic arguments are really very constructive though - the main point is that this record adds nothing to estimates of medieval to current (or even LIA) temperatures. - gavin]

    Comment by Gaelan Clark — 10 Dec 2007 @ 4:35 PM

  57. #49 Chuck:

    Had Martin specified wild animals with fur coats, your questioning of my canine example would be fully justified

    That you didn’t get that from the context means that you missed everything about my original post. So once again, con amore: when humans learned the arts of dressing in clothes and making fire, the selective advantage of having a natural warm furry coat turned into a disadvantage — growing fur costs energy. So the reverse adaptation took place. Simple, direct, obvious, no subtle mechanisms involved or needed in this context. OK?

    Comment by Martin Vermeer — 10 Dec 2007 @ 6:29 PM

  58. Re #50 Martin Vermeer:
    I attempted to respond privately, but couldn’t locate your email address. I hope the RC moderators will endure one more post on this topic:
    I interpreted “those animals having them” (# 43) to not be limited to humans. I don’t question your explanation of why humans are relatively hairless. But, I’m not up on the research on that subject, and one needs to be careful about evolutionary story telling (a major point of Gould and Lewonton). Hair is one trait that has numerous functions: It can keep animals warm in cold climates (by reducing convection across the skin), it can keep them cool in hot climates (by reflecting visible light, or absorbing visible light and IR but not conducting the heat to the animal’s skin), reduce water loss (again, by reducing convection), and protect the skin from invasion by parasites, bacteria, and viruses.
    In explaining why humans have lost most of the hair, you really need to consider all likely functions (and selection pressures). That was my point about domestestic dogs: As Timothy noted, their coats are, in many cases, the result of selective breeding for esthetics rather than thermoregulation. Artificial selection in this case is analogous to sexual selection (instead of attracting mates, the dogs attract breeders bearing mates). In nature, sexual selection can result in seemingly maladaptive traits (as Darwin noted). And selection pressures for multiple traits (heat retention, heat resistance, protection against invaders) may result in evolutionary compromises – the end result may not be optimal for any one trait, but instead may be the optimal solution for several traits together.

    Comment by Chuck Booth — 10 Dec 2007 @ 8:28 PM

  59. Re Jim Galasyn (#48)

    As Chuck Booth says, the biology is off topic. But feel free to email me and I can give you a list of related topics, check off the ones you are interested in, and I can give you a list of papers in the primary literature, links to most — and if you are interested, pdfs where the links aren’t available. As I said, an obsession, roughly for three years.

    PS Don’t have the book you mentioned. Used it is still well over a hundred. Sorry I didn’t respond sooner. Actually I did respond, but things seem to have been a little glitchy past few days.

    Comment by Timothy Chase — 10 Dec 2007 @ 8:57 PM

  60. thanks, chuck

    Comment by Susan K — 10 Dec 2007 @ 9:52 PM

  61. Re #55 Chuck Booth: OK, point taken.

    In nature, sexual selection can result in seemingly maladaptive traits (as Darwin noted)

    Yes, it often does. And not only sexual selection. It is the perfect illustration of evolution often seeking out local rather than global optima — not a very “intelligent” process :-)

    Off-topic or not, I like noticing the very broad scientific appetite of this readership.

    Comment by Martin Vermeer — 11 Dec 2007 @ 5:08 AM

  62. [[when humans learned the arts of dressing in clothes and making fire, the selective advantage of having a natural warm furry coat turned into a disadvantage — growing fur costs energy. So the reverse adaptation took place. ]]

    Wrong. Humans already lacked fur, which is why they put on the clothing. You’ve got the causality backwards.

    Comment by Barton Paul Levenson — 11 Dec 2007 @ 7:50 AM

  63. #48:

    http://cvr.bio.uci.edu/

    Thanks Jim. This is fantastic reading.

    Comment by Martin Vermeer — 11 Dec 2007 @ 8:35 AM

  64. Vern Johnson:

    The main objection, and it’s been pointed out here repeatedly, seems to be that those “raising some doubts” are simply doing that – raising doubts. Not trying to improve the actual knowledge of anything, just raising doubts like a trial attorney. When confronted with that, the doubt-raisers switch to another line of attack. They also wait a few years to recycle thoroughly debunked lines of attack. The process is simply a drive to create a fake controversy and tie down the time of scientists and science advocates.

    On the other hand, you’re obviously, to me, confusing for some reason disagreement with the conclusions of those analyses which are scientific in intent and practice, with objection to them. Scientists don’t see it that way.

    Until, and unless, you correct your false prior assumption that the disagreements here with flawed analysis, faulty research, or simply unlikely interpretations of data are objections to someone for their temerity in questioning some cabalistically approved research and analysis, which is garbage in, you’ll get garbage out – not understanding. The use of language by RC people is completely irrelevant. In general, when something is both completely out of the mainstream (and hence, in contradiction to the work of the majority of careful, hard-working scientists) AND has serious flaws, most people in science will at least conjecture that the two are related. Caring more about the lost sheep than the ninety and nine might be biblical, but it’s not how science tends to work.

    Comment by Marion Delgado — 12 Dec 2007 @ 6:31 AM

  65. Minimal fur let our ancestors cool off or not overheat when they ran to catch food. Clothing with a serious survival value I think got adopted as later ancestors moved away from the Rift Valley and then out of Africa. The hair on top could show a selective advantage in various ways.

    Comment by Adrian Midgley — 12 Dec 2007 @ 6:31 PM

  66. There is something truly sad in all this.

    I am sure that the individual researchers and their teams that gather and elusidate the raw data are most often doing their level best to try and see some light through a darkened lens.

    I am sure that often they must know their individual speciality inside out. In particular where all the bodies (uncertainties and artifacts) are buried. In particular the strengths and weaknesses of their research. That is the nature of research, more understanding leads to more questions.

    Given access to their results (often as opposed to their hard earned data) anyone with a PC and some spare time can combine their efforts, as if they were ingredients, to make a meal. I.E make a meal of them.

    “Gisser job”, “I can do that”.

    There is a pride in doing something well that seems sadly out of fashion.

    I take it on faith that the chronological inaccuracies (interpretation of BP etc.) as described by Gavin have occurred. Given that, it is truly remarkable that circumspection failed to wag its tail, jump up, and lick the author in the face, to warn that something is wrong. Given that the watch dog never stirred I have to wonder how come. By this I mean that when a serious error is not salient what other errors must their be to cover it up.

    What a shame!

    Ho Hum!

    Alexander Harvey

    Comment by Alexander Harvey — 13 Dec 2007 @ 12:55 PM

  67. On Peer Review and Openness.

    These are a qualification and a quality that many do and would welcome. They are also deeply intertwined.

    The paper under discussion is a case in question.

    If Gavin had been requested to give his opinion prior to publication one would presume that it would have made a material difference. Also given that he has acknowledge his thanks to others who have contributed materially to his critique, there was not sufficient background available with the paper to adequately review it.

    Personally I am all for openness. There is no reason why, in this day and age, that a paper should not be backed up by all the decision making process and most importantly all the data both initial and throughout each stage of its manipulation.

    Or is there? In the best of all worlds that might be the case but in reality it is a can of worms. Unless it was mandated that nothing could be published without a forensic chain of evidence then we will have the noblest and the best facing detailed scrutiny, and the illusive and illusory covering thier tracks.

    Also publish and be damned does not bother those of a speculative disposition. With little or no relevant reputation to protect they can piss and run with little fear.

    Occassionally a paper does emerge that is backed up by copious background data sufficient for one to check it for rigour. I took the time to check one of these. As it happens it has its problems or at least significant points that require clarification. Now, what should one do about this?

    Personally I feel that to undermine such an open source for being “open” to extended critism from all and sundry would be curlish and counter-productive. Better to sit on one’s hands.

    Perhaps a little myopic but should one be shooting a goose in clear view when there is more deserving prey in the bushes.

    To be frank my stance bothers me and I wish there would be a way for independent reviewers to pool their concerns so that a considered opinion could be related to the authors.

    I am sorry to say that blogs like this do not do the job. They are simply too confrontational. By this I mean, if your intention is to improve the quality and the standing of science simultaneously then rubbishing well considered but flawed work in public achieves neither.

    Alexander Harvey

    Comment by Alexander Harvey — 13 Dec 2007 @ 2:27 PM

  68. Gavin et al,

    I should like to know about peer review in this field.

    I have discussed this with workers in other disciplines and it comes done to the old anti-maxim “If it is worth doing, it is worth doing well”. By which I mean, if it is important and in your competence you agree and take great pains (it is after all your reputation on the line) or you pass. By great pains I mean perhaps a week perhaps more.

    Gavin, you have obviously (and thankfully) expended a good deal of effort on this paper and so I presume would have done as much or more if it had been referred to you pre-publication.

    Given that could you please give me/us a critique of what “peer review” means (in detail) at best and at worse.

    I would be most grateful if you could reply.

    Alexander Harvey

    [Response: Good papers are easy to review. Bad papers can be extremely time-consuming. Mediocre papers are more variable - it depends how invested you are in the subject. Therefore people tend to accept assignments for good, and not for bad. There is only so much time in the day and so trying to find competent reviewers is difficult. Personally, I do maybe 2 a month and refuse an equal number (sometimes more). Good peer review can elevate a mediocre paper substantially (but that takes work). Finding the key flaw in a bad paper and documenting it's effect is difficult - reviewers often don't bother. But that can lead to problems too since the other reviewers or editor might not see it. What more do you want to know? -gavin]

    Comment by Alexander Harvey — 13 Dec 2007 @ 2:49 PM

  69. Dear Gavin,

    First thanks for your reply.

    By good papers, I presume you mean clearly described ones, ? Good, as in stricking papers, when they occur, must also cause real effort.

    I am surprised, (I may be being naive) that you do two a month. If that is common and there are three reviewers per paper. That would seem to imply that the publishing rate is one every six weeks for each “peer”. Surely this is the fast end of scientific output.

    Personally I think this is extreme.

    I hope you realise that in your sentence: “Finding the key flaw in a bad paper and documenting it’s effect is difficult – reviewers often don’t bother.”, you are begging a question that gets to the heart of the problem. Given that circumstance what happens? You respond that it is too flawed to be deserving of comment or you sigh and let it proceed.

    I am not being critical of you or anyone in particular but papers that are of dubious scientific content do get published and it would seem to be a matter of great concern as to how to curtail this.

    Now, I may not be able to halt climatic change or hold back the tide but if I can do anything at all to improve the contect of scientific papers I will.

    I shall write more when I can but for now I would welcome any dialogue on this issue which I think goes to the root of all poorly reviewed yet published work.

    Alexander Harvey

    Comment by Alexander Harvey — 13 Dec 2007 @ 5:07 PM

  70. Hi, just out of interest, is there any work that compares proxies to the instrumental temperature record for the last century? All the work I could find in an admittedly non-exhaustive search was understandably focussed on reconstructions, rather than what we already know. I’ve heard skeptics banging on about how badly proxies reflect actual climate, and would like to know how well they’ve proved their worth against instrumental data.

    Comment by George Darroch — 13 Dec 2007 @ 6:06 PM

  71. Alexander Harvey, While it is true you don’t wan’t too many bad papers to get through, think for a moment how the average reader will react to a bad paper–namely by saying, “This is crap,” and moving on. We are talking about experts in the field here, not amateurs. Yes, the occasional grad student might get very excited about a bad paper, but he will be corrected quickly by his adviser if his adviser is worth anything. The ultimate test of a paper is how much it is cited in future work. Peer review is just a threshold, and it’s purpose is more to make good papers better rather than reject bad papers.

    Comment by Ray Ladbury — 13 Dec 2007 @ 6:09 PM

  72. Re # 66 Alexander Harvey

    As the RC moderators and others (such as Ray Ladbury) have pointed out repeatedly, peer review is merely a ticket that allows entrance into the marketplace of scientific ideas. Once in the marketplace, it is survival of the fittest – the best papers will be cited the most often and will gain the most credibility. But, again, as has been noted, a paper might have, simultaneously, a great idea or two (or valuable data) and a serious flaw – as a result, that paper might get cited for the former, but not for the latter.

    Comment by Chuck Booth — 13 Dec 2007 @ 6:46 PM

  73. Ray & Chuck,

    I am trying to get to the bottom of the realities of the review process. Given that many papers are in essence desk-top activities that are easy to produce and require little depth of knowledge of the subtleties of the original work (as evidenced in the paper being tested here). The shear effort of getting them adequately reviewed seems intolerable and the amount of time available must surely be very limitted.

    Unfortunately these type of papers are obscure in that short of retrieving the original data and reworking it, I doubt that more than an impression can be formed in the time available.

    Also a paper of this nature as opposed to original research is rarely duplicated so that layer of verification is missing. They would become simply noise were it not for there value to contrarians. Now I think this is important.

    Personally, I wish that the “hockey stick” had not been given the prominence it received as it has become a bit of an Aunt Sally. But it does exist and as long as it does papers like this one will recieve prominence well beyond their merit.

    People do, whether rightly or wrongly, put great store on peer reviewed work.

    I presume this work was reviewed and if so it does seem to make a bit of a mockery of the process.

    It would seem that if the time and effort that was put in by Gavin had been done by reviewers pre-publication it is likely that we would have been saved the necessity of reading it and he of having to debunk it.

    So not only does the process take up a good deal of time (of those that reviewed it) but it fails to prevent even more time being wasted post publication.

    That alone would seem to be important.

    Surely the review process is meant to be for the protection of good science from bad. Not just a tick in a box. If the latter is becoming the case perhaps the whoe process should come under review.

    Best Wishes

    Alexander Harvey

    Comment by Alexander Harvey — 14 Dec 2007 @ 9:49 AM

  74. I am not a climatologist, but someone studying climate effects on ecosystems. I have a question regarding past reconstructions: A paper from 1992, Jawarowski et al. (Sci Tot Environ 114:227-284) criticizes the value of ice core CO2 data. I found some discussion of Jawarowski and his criticism at realclimate.org and elsewhere, but nothing explicit on the following issue: The paper is said to contain a graph (fig. 10) which has been reproduced and highlighted very recently by Swedish climate skeptics, and which claims that the ice data has been manipulated by moving data 80 years along the time scale to make the Mauna Loa time series fit with data from an ice core reflecting the last couple of centuries (Siple). I can only access more recent volumes of this journal, and have only seen this graph out of context. The paper has been cited only 7 times so far (according to ISI), and is apparently taken rather lightly by the research community. I would like a concise explanation (and maybe some good links/references) explaining this purported mismatch between ice data and modern measurements. Were data shifted for a good scientific reason, is the claim by Jawarowski that they shifted at all incorrect, or is there no good explanation for the mismatch?

    Comment by Lars Tranvik — 14 Dec 2007 @ 10:34 AM

  75. Hank Roberts and Timothy Chase have directed me to documents that seem to say the following:

    ‘The great fluctuations in ocean temperature are found to be less dramatic when data is more carefully assessed, although there is still a clear trend upward. However, the presence of this trend is not a stark indication of human impact because the useful data only covers four decades of change in systems that certainly operate over massively longer periods of time.’

    Have I gotten on the right track here?

    Comment by Jack Garman - Amateur — 16 Dec 2007 @ 4:17 PM

Sorry, the comment form is closed at this time.

Close this window.

0.394 Powered by WordPress