Miranda Devine vs The Hockey Stick

According to this profile, Miranda Devine (last seen making stuff up in an attempt to debunk the Lancet study), once worked for the textile physics division of CSIRO. So she should know that one purpose of peer review is to weed out scientific papers that are inaccurate or where the conclusions are not properly supported by the evidence offered. She went on to write an opinion column where accuracy and supporting your claims are not important, so perhaps that explains why in her latest screed she seems to believe that peer review is a tool to silence dissent. Devine takes on the hockey stick:

The so-called “hockey stick graph” published in 1998 by University of Massachusetts geoscientist Michael Mann became an article of faith, and underpinned the Kyoto Protocol. It purported to plot average surface temperatures of the Northern Hemisphere for the previous 1000 years. Temperatures appear to remain constant until 100 years ago when the graph takes a sudden upturn, supposedly showing the Earth heating up as the industrial revolution kicks in.

i-7e14a971cb7ae71548a77c9a0f20df85-300px-1000_Year_Temperature_Comparison.png

OK, except that Mann is at the University of Virginia and the 1998 paper containing the “hockey stick graph”, entitled “Global-Scale Temperature Patterns and Climate Forcing Over the Past Six Centuries“, only plotted temperatures for the past 600 years (the title is a clue). And the Kyoto Protocol dates from 1997. Which is before Mann (and Bradley and Hughes)’s paper was published. And the graph was not an “article of faith”—there have been many other temperature reconstructions published (see graph on right).

Devine continues:

It took six years and several sacked scientific journal editors before doubt was thrown on the hockey stick.

Actually, what happened was that several journal editors resigned because the journal published a flawed paper that disagreed with the hockey stick. Not because the paper disagreed, but because the authors had failed to properly support their conclusions. And study the graph of the other reconstructions again.

Devine continues:

Last year Canadian scientists Stephen McIntyre and Ross McKitrick discovered a fundamental flaw in the computer program which produces the hockey stick. It seemed, whatever random data it was fed, the program almost always produced a hockey stick.

This is a red herring. If you do a linear regression on random data, you’ll produce a straight line. Does that mean that linear regression is invalid because it always produces straight lines when there are none in the data? Of course not. What is important is whether the result of the regression is statistically significant—for random data it won’t be. William Connolley did some experiments and reported:

What that appears to demonstrate is that M&M are right about one thing: it often does lead to a “hockey stick” shape in random data. But the problem is that the variance-explained of the PC1 done this way is tiny: the first eigenvalue is about 0.03. Whereas when you run it on real data the first eigenvalue is about 0.55 (back to 1000) or 0.38 (back to 1400). Which means the two problems are very different.

Back to Devine:

The Canadians couldn’t get their work published by a scientific journal but they put it on the web for all to see.

Actually, they got it published in Geophysical Research Letters.

Devine:

The voices of dissenters have been silenced for too long so all power to Crichton for bypassing the gatekeepers and going straight to the people.

Yes, it’s time that the elitist notion that what gets published in scientific journals be accurate is done away with! Let a thousand falsehoods bloom!

Oh yeah, Devine also drags out the Oregon Petition:

When Dr Fred Seitz, the past president of the US National Academy of Sciences, organised a petition opposing the Kyoto Protocol, he had 20,000 signatories – 17,000 with scientific degrees, including physicists, geo-physicists, climatologists, meteorologists, oceanographers and environmental scientists. But for simply stating there was no convincing scientific evidence that human release of carbon dioxide or other greenhouse gases was causing catastrophic heating of the Earth’s atmosphere and disruption of climate, they were dubbed “The Carbon Club, foot soldiers for the fossil-fuel industries”.

No they weren’t. Devine is just making things up again. Jeremy Leggett dubbed the lobbyists who work for the coal and oil industries the “foot soldiers for the fossil-fuel industries”, not the folks tricked into signing the Oregon Petition.

Comments

  1. #1 Gavin
    February 27, 2005

    Small correction. Mike was actually at UMass Amherst when MBH98 was published (look at the paper itself).

  2. #2 Eli Rabett
    February 28, 2005

    ….and he was a post-doc working with Bradley. The two most attacked papers in climate science are Rasool and Schneider in Science for their paper on aerosol effects on climate, and Mann, Bradley and Hughes in Nature.

    William Connolley has much more information on R&S in his web page on whether an immenent ice age was predicted in the scientific literature of the 1970s http://www.wmconnolley.org.uk/sci/iceage/

    In both cases for some odd reason it is the post-docs who have become most closely associated with the papers (ever hear anyone bleat that Rasool predicted another ice age?). At least Mann was the first author.

    All of which leads to a perfect description of being ignored while your postdoc grabs the glory – being Rasooled.

  3. #3 Helen
    February 28, 2005

    So Miranda worked for the Textile Physics division at CSIRO?
    Does that explain her woolly thinking?

  4. #4 gringo
    February 28, 2005

    It’s all about ambition – she’s just fishing for a spot on the ABC board.

  5. #5 Lars
    February 28, 2005

    Another small and probably niggling correction – Ross McKittrick is an economist at the University of Guelph, while Steve McIntyre is or was an analyst of some sort in the mining industry. Neither of them could be called “scientists” by profession, unless you consider economics to be a science.

  6. #6 tony
    February 28, 2005

    Thanks for the appraisal of Divine’s work.

    Problem is, of course, that most people will read her article and get the impression that the science is doubtful. Which is the aim, of course.

  7. #7 mick
    February 28, 2005

    Hi, I stumbled across your post. Nice work, Miranda Devine really went overboard for this article. Check out:

    http://www.zwire.com/site/news.cfm?newsid=14033600&BRD=1675&PAG=461&dept_id=18168&rfi=6

    If you follow this link you will find an article written on the 25th of Feb where Devine seems to have found all her information except for the hockey stick stuff. Nothing like a bit of cut n paste action….

  8. #8 nicolo
    March 1, 2005

    Great work Tim. Unlike the unwashed neonitwit hordes of her journalistic fellow travellers (unfortunately), Miranda I believe remains capable of feeling shame. So I hope that she’ll have been sent your excellent work by now and will be responding to you on your blog. She ought to: for a journalist of integrity, as she’d certainly want to consider herself to be, that’s an exposure of cringing intellectual bankruptcy in her article that you’ve produced. Carry on :)

  9. #9 Nabakov
    March 1, 2005

    But Miranda, like ole Louie “the Fly” Hissink, is not is the business of debating scientific arguements, but rather in sowing doubt in laypeople’s minds about whether there is any valid consensus on this issue.

  10. #10 John Finn
    March 2, 2005

    From above, you say

    “This is a red herring. If you do a linear regression on random data, you’ll produce a straight line. Does that mean that linear regression is invalid because it always produces straight lines when there are none in the data?”

    Can’t say I understand your point here. You perform a linear regression by choice. This doesn’t imply anything about the true behaviour of the data. Are you suggesting MBH chose to fit their data to a hockey-stick shape. The point M&M make is that the use of non-centred PCA as performed by MBH ‘mines’ for H-S shaped data series – and promotes them to be the dominant PCs. MM used computer-simulated ‘red noise’ (artificial tree networks) with the MBH method and found that it produced a hockey-stick curve in 99% of the cases. There doesn’t seem to be much doubt about this. Independent experts have confirmed that H-S curves will dominate using this method.

    Another point to bear in mind is that there is some controversy over one particular data series, i.e. the bristelcone pine series. (PC1 in MBH; PC4 in MM). It seems to be fairly widely accepted that the bristlecone pines experienced an “anomalous growth spurt” during the 20th century – which was NOT related to temperature/climate. Without the bristlecone pines – there is no hockey-stick. We, therefore, appear to be basing the entire NH climate for the past 1000 years on a few trees in North America and some unknown fertilising agent.

  11. #11 Tim Lambert
    March 2, 2005

    John F, you might just as well say that the Fourier transform “mines” for sine curve shapes. That does not mean that the Fourier transform is invalid. It is quite simply wrong to claim that the hockey stick is the product of a mathematical error. See the dummies guide to the dispute. It doesn’t matter whether you use centred or non-centred PCA — you only get MM-type results if you throw out a significant component, which is wrong.

    Maybe there is something wrong with the bristlecone proxies — I don’t know. But this is not a calculation error. M&M’s claims are misleading.

  12. #12 Eli Rabett
    March 2, 2005

    John F has obviously never taught an introductory lab class where students produce straight line fits to sine curves, hockey sticks, boomerangs, blobs of jelly and worse.

    However to follow a poster named Epicurius (a sock puppet perhaps) on Quark Soup http://davidappell.com/archives/00000625.htm If you look at PC1 of the bristlecone pine series, it does not produce a hockey stick shape, but is rather flatter. The hockey stick shape is much better matched by the north american tree line series. (See Fig 1 and 3 of http://www.ncdc.noaa.gov/paleo/pubs/mann_99.html , even better read the discussion of the bristlecone pine series in the paper, better yet read the entire paper…)

  13. #13 John Finn
    March 2, 2005

    Tim

    I’m not suggesting there was an error. Let’s just say that, bearing in mind the physical nature of the problem, the use of non-centred PCA was perhaps not appropriate in this case. Whatever – the point is the MBH analysis is not robust as it is highly sensitive to the addition/removal of one component, i.e the bristlecone series. Professor Mann and his team are aware of this – since they performed the analysis without the bristlecone series .

    Actually, Tim, as a computer scientist, you may be able to offer Professor Mann some assistance. I believe he may be having trouble setting the permissions to directories on his ftp site. If it’s a unix server, I assume something like

    chmod 700 BACKTO_1400_CENSORED

    would do the trick.

  14. #14 John Finn
    March 2, 2005

    Tim/Eli

    Let’s forget all the mathematical and statistical arguments for a minute and concentrate on ‘gut feeling’
    What’s your ‘gut feeling’ on the hockey-stick? Mine is this. The H-S reconstruction is NOT an accurate representation NH climate over the past 600 years. The following is just one reason for this view.

    On the realclimate site there is an article
    here.

    which includes 3 graphs. The first of these shows a comparison between temperature variation due to natural forcings ONLY (modelled) and actual observations since 1850. Models and observations are clearly in agreement up until around the 1970s. In other words early 20th century warming is entirely natural.

    Now look at the hockey-stick on the IPCC site here.

    Note the large upturn JUST AFTER 1900. Why? This is completely out of character with the previous
    900 years and yet all fluctuations – at least up until the mid 20th century – are supposedly due to natural variability. Not proof – I know but doesn’t it suggest just a hint of doubt?

  15. #15 Dano
    March 3, 2005

    Excuse me for interrupting, this hits me personally…

    What’s your ‘gut feeling’ on the hockey-stick?

    Why would you presume to have a gut feeling about the knowledge of events on a time scale that is ~9 times longer than your likely lifespan?

    Gosh…maybe you’ve been reincarnated?! Can you cast back into your memory and find what happened to my great-gamma’s lost love? His name was Seamus something. She never got over it…her story still affects me to this day, and if you could help, John, I’d sure appreciate it.

    D

  16. #16 John Finn
    March 3, 2005

    D

    Seamus always intended to get back in touch but … well with one thing and another he never got round to it and the rest,as they say is history. However, your ancestor’s love life does side-track us slightly from the point of the original post which is that there are 2 graphical representations of climate on the realclimate site which are inconsistent with each other.

    Also, ‘gut feeling’ is an important instinct. Don’t knock it – it’s won me a fair bit of money at the bookies over the years.

    PS Tim – I’ve thought about your F-S analogy and it’s not a good example.

  17. #17 Eli Rabett
    March 3, 2005

    Inconsistent in what way? Remember that the uncertainty limits don’t show on the graph above.

  18. #18 Thomas Palm
    March 4, 2005

    John has a point. If the warming up to 1970 was natural then the upturn in 1900 in the MBH curve is odd. It’s unlikely but it could be a coincidence or it could be something more fundamental. The MBH curve is science in progress, not the final truth, so it isn’t all that surprising that there are inconsistencies, but that just makes it more interesting.

  19. #19 Eli Rabett
    March 4, 2005

    MBH is a reconstruction of temperatures from proxy records. The “upturn” in 1900 is in the instrumental records and is not part of the proxy reconstruction, but rather is used to calibrate them. This upturn is seen in all surface temperature reconds. It has been reproduced by GCMs using measured forcings both natural and anthropic.

    Actually this is the best measured and best understood part of the curve.

  20. #20 Tim Lambert
    March 4, 2005

    I agree that John has a point, but I don’t think it is that strong. The upturn in the MBH reconstruction from 1900-1970 is about 0.4C. Earlier in the reconstruction there are ups and downs of about 0.3C over similar time scales, so this is a little anomalous. I don’t think you should call it completely out of character, though.

    More compelling are the other reconstructions that show more variation. I don’t know which ones are closer to the truth, but they all seem to suggest that current temperatures are unnaturally warm.

  21. #21 Ken Miles
    March 4, 2005

    You may also be seeing some enhanced solar effects in that period.

  22. #22 Louis Hissink
    March 4, 2005

    Tim, ups and downs of o.3 to 0.4 deg Celsius are historically unsupportable.

    You guys are arguing about how may angels could be fitted on top of a pin head.

  23. #23 John Finn
    March 4, 2005

    Thanks to everyone for the comments. I’ve just got a couple of quick responses

    Tim

    Fair point – it isn’t “that strong”, but, for me, at least, it introduced enough initial doubt to give the M&M and von Storch studies some credibility.

    Eli

    I don’t think the 20th century part of the reconstruction are direct temperature measurements – if that is what you’re saying.
    If they are – I’m not sure this is acceptable, as you’re then comparing apples with oranges. Thermometers measure temperature only; whereas tree rings give an indication of favourable (or otherwise) climatic conditions of which temperature is a part. Now I don’t know anything about tree cultivation (so bear with me), but I assume that moisture is also a factor. How might tree growth respond, say, to a very hot, but possibly very dry period compared to a warm, but wetter period. I’m sure it probably depends on tree type, but I hope you get my drift that there’s not necessarily a simple linear relationship between tree ring growth and temperature. I’ve also read a suggestion that tree ring growth may ‘plateau’, e.g. a linear relationship might exist between a certain range of temperatures but after that there is no increased effect. I can understand this. Up to around 80 F, I can detect quite subtle temperature changes. Any higher and it’s just bloody hot as far as I’m concerned (Australians may need to add 10 degrees to appreciate what I’m saying). But, if this is the case, it’s possible that tree ring data may not pick up the extreme temperature fluctuations which would have the effect of ‘flattening’ or ‘damping’ variability.

    What’s really needed is a comprehensive, up-to-date network of proxy data (not just tree ring) which can then be compared and correlated with the known temperature record. That way, each data series can provide some sort of mutual validation for the other(s). I find it surprising that this isn’t the case and I did get into a ‘discussion’ on realclimate about it. It seems that, after 1980, there is very little tree ring data available.

  24. #24 Eli Rabett
    March 5, 2005

    John, how do you think you get temperature anomolies out of proxy records, you have to compare the proxy to an instrumental temperature record. If you go to http://en.wikipedia.org/wiki/Image:1000_Year_Temperature_Comparison.png where the figure above came from you will find that the black line is labelled as the instrumental temperature record (that it stops in 1860 is also a big hint).

    Don’t worry though, AFAIK huge sets of proxy records are being created daily, for example, the tree outside my house is still growing. Without being snarky, you are barking up a strange tree there. Proxy records are already collected into databases, a huge one is http://www.ncdc.noaa.gov/paleo/paleo.html which actually has most of the data used in MBH98 and MBH99.

    Keep in mind before you go off on a tangent thinking that you are thinking original thoughts, that Hughes is one of the world’s experts on dendrology, and Bradley is a key person in putting together instrumental records from as far back as possible (Jones, as in Mann and Jones is the other one). When a non-obsessed starts looking at an area of research, he or she usually thinks of three or four things that obviously should be done, and IF he or she goes looking before stating the obvious, almost without exception, gee, it has been done. Go look.

  25. #25 Scott Church
    March 5, 2005

    All,

    The first graph John refers to, at RealClimate, is based on Stott et al. (Science, No. 15, 2000) and Tett et al. (Hadley Centre Tech Note 19, 2000). It shows modelled vs. observed warming since about 1860 based on runs of Hadley Center general circulation models (HadCM3 maybe, but I’m not sure). The second is based on Folland et al. (Geophys. Res. Lett., 2001).

    These two graphs do not disagree. It merely appears that way because they are using different abscissa scales that make the Hockey Stick look compressed and the anomalies are normalized to different time periods – the first (Stott et al; Tett et al.) was normalized to a period of 1880-1920 and the latter (Folland et al.) to 1961-1990. In each case though an observed, and largely out of the ordinary warming begins around 1900 and increases by roughly 0.6 +/- 0.2 deg. C. to the present. Discussions regarding what is known about the forcings behind this variability are in IPCC (2001) Chaps. 2, 6, and 12. The punch line is that from roughly 1910 to just after 1940 large, but natural fluctuations in solar activity combined with gradually increasing responses to greenhouse gases drove an increase. Between 1940 and 1970 decreasing solar activity ate into this increase and caused things to cool off some but with diminishing returns because greenhouse gas responses were picking up speed. The large increases after 1970 are due to the greenhouse gas forcings taking off much faster and running away from waning solar activity. This picture is, of course, very simplified but that is the overall history.

    There is no mystery here. Nor is there any inconsistency with the Hockey Stick.

  26. #26 Scott Church
    March 6, 2005

    One minor addition/clarification to my last post. Regarding the first graph John referred to (the one he got from RealClimate based on model runs that I cited to Stott et al., 2000) it was the blue curve in those graphs that were derived from model runs. The observed temperature records these were compared to (in red) were based on other data (I don’t know which because I haven’t yet gotten a copy of the paper). Either way, nothing that has been discussed here so far contradicts any aspect of the Hockey Stick.

  27. #27 John Finn
    March 8, 2005

    I think this issue, particularly relating to my initial point, has become unneccessarily complicated.

    Eli –

    It doesn’t really matter how the H-S reconstruction was done or what proxies were used. It’s a separate argument. The issue is whether the H-S is an accurate representation of climate over the past 1000 years.

    Scott –

    The same goes for the scales of the 2 graphs. I’m not trying to overlay one on top of the other for comparsion. I’m simply drawing a conclusion from one and looking to see if this matched the behavious of the other.

    From the natural forcings/observations graph we can conclude that any warming between 1900 and 1950 is mostly natural.
    If so, then the period between 1900 and 1950 should look pretty much the same as the period between 1000 and 1900, i.e. the curves and bumps representing natural variability should be roughly similar. The sharp increase just after 1900 on the hockey-stick graph is more than would be expected and, in my opinion, this is enough to throw some doubt on it’s accuracy.

    This is not intended to be a rigorous scientific proof, but a sort of ‘quick and dirty’ validation. The H-S may well be ‘correct’ after 1900 – BUT – if it is – then I suspect that the Pre-1900 period is not correct and that climate variability is greater than that which is represented by the hockey-stick.

    Eli –

    Re: lack of proxy data

    See http://www.realclimate.org/index.php?p=11#comments

    and read comments (& responses) #4, #5, #6

    When I questioned Professor Mann about the fact that proxies are unable to reproduce surface temperature warming, he replied

    “This is not correct. Most reconstructions only extend through about 1980 because the vast majority of tree-ring, coral, and ice core records currently available in the public domain do not extend into the most recent decades.”
    Actually he’s only partly correct. It’s true there is a scarcity of proxy data post-1980. But there is some available and what there is underestimates surface temps by about 0.3 deg C.

  28. #28 John Finn
    March 8, 2005

    Scott

    I’m not sure that the solar forcings over the past century agree with your description of events. But, even if they did, It’s still not really the point. There must have been changes in solar forcings between 1000 AD and 1900 so why are these not evident (or not as evident) in the hockey-stick. Solar activity is not a 20th century phenomenon.

  29. #29 Eli Rabett
    March 8, 2005

    John you are going to have to be much more specific about which post-1980 data does not follow surface temperatures. You obviously mined this little tidbit somewheres, so where does it come from.

  30. #30 John Finn
    March 8, 2005

    See the link I provided on realclimate. Go to Comment #4 then the 3rd response by Mike Mann where he says

    “however, anomalous behavior in recent decades apparently related to non-climatic (or, at least, non-temperature related) influences on tree growth does indeed compromise the use of these data in reconstructing temperature changes over the past several decades ”

    i.e. Mann himself acknowledges there is “anomalous behaviour related … to non-climatic influences”. Not sure how he knows that as they don’t know what is causing it.
    Also Cook et al 2004 and Briffa (can’t remember) who are cited on realclimate both discuss “proxy reconstructions underestimating actual observations”. In fact I quote from the Cook study on realclimate.

  31. #31 Eli Rabett
    March 9, 2005

    John, the link you provided was http://www.realclimate.org/index.php?p=100#more-100, the quote comes from http://www.realclimate.org/index.php?p=11. You also managed to miss the prolog:

    “There is no discrepancy in general between the 20th century trends indicating by the instrumental record and by most proxy reconstructions. In some tree-ring reconstructions (particularly, those based on tree-ring density information, and emphasising higher latitudes), however, …..”

    and the finale:

    ” These factors, however, do not appear to be relevant to other, multiple-proxy based reconstructions such as those referred to above. -mike]”

    BTW, he also answered your question as to why many records and reconstructions end in 1980:

    “Most reconstructions only extend through about 1980 because the vast majority of tree-ring, coral, and ice core records currently available in the public domain do not extend into the most recent decades. While paleoclimatologists are attempting to update many important proxy records to the present, this is a costly, and labor-intensive activity, often requiring expensive field campaigns that involve traveling with heavy equipment to difficult-to-reach locations (such as high-elevation or remote polar sites). For historical reasons, many of the important records were obtained in the 1970s and 1980s and have yet to be updated.”

    Go read the whole thing again.

  32. #32 John Finn
    March 9, 2005

    Eli

    I don’t need to read it again – I’ve read it. I asked the questions and – in many people’s opinions – didn’t get satisfactory answers. At one point he (Mike) tried to cut short the discussion. Professor Mann’s debating skills can’t hide 2 important facts

    <li>
    1. Proxy records are scarce after 1980.
    <li>2. Those that are available disagree with the late 20th century warming.

    Mann says “There is no discrepancy in general between the 20th century trends indicated by the instrumental record and by most proxy reconstructions”

    Clever wording – ‘in general’ – there’s no problem. Most of the proxies agree with most of the 20th century temperature record (up to around 1975 anyway) – BUT the proxies don’t pick up the ‘unprecedented’ warming from 1975 onwards.

    And why haven’t the proxy records been updated – on such an important, crucial issue? – because they won’t get the results they want it’s as simple as that

    At the end of the day, though, it’s all down to how you interpret the facts and I suspect that it’s unlikely our interpretations will agree.

    PS I provided 2 links you went to the earlier one.

  33. #33 Eli Rabett
    March 10, 2005

    John, are you claiming ALL proxy records available after 1980 disagree with the instrumental record of 20th century warming? If so you need to point at some source that has studied this and come to that conclusion. Mann clearly said that there are types of tree ring records that diverge, but that is not all proxy records by far. Moreover, he pointed out that dendrologists are aware of this problem and that it does not extend to all tree ring series. But that says nothing about boreholes, ice corse, coral, etc.

    Second, given the fact that proxy records are local by nature, one would not necessarily expect that any particular proxy record would follow the global trend by itself, however one should be able to test a proxy record against local instrumental records and I am aware of cases where this has been done and there is no disagreement within experimental limits.

    Since I am aware of proxy records that show unprecedented warming from 1975 onwards, your statement that they don’t is simply wrong. Where did you get that idea.

    As to why proxy series have not been updated, that is simply a red herring. In the first place, a direct instrumental record is always superior to a proxy record, simply because it is direct and precise. A thermometer reading is better than determining the width of a tree ring for measuring temperature. The major point of taking proxy’s is to extend measurements back into the past where no instrumental record is available. Given the choice of obtaining a proxy from an unsampled area, in the southern hemisphere or resampling a heavily studied area, guess which has the highest priority.

    The other possibility, of course, is that you are claiming that no proxy record can be trusted or that no instrumental record can be trusted.

  34. #34 Dano
    March 10, 2005

    1. The borehole record agrees fairly well with the sfc temp record.

    2. There are a number of papers that discuss the issues surrounding the tree-ring record for the latter part of the last century. I don’t have the time today to dig some up, but will soon if someone asks.

    3. I completely agree with Eli’s As to why proxy series have not been updated, that is simply a red herring. In the first place…

    My experience is that when folks start quibbling over semantics and ‘satisfactory answers’, etc then there’s no there there.

    D

  35. #35 John Finn
    March 10, 2005


    But that says nothing about boreholes, ice corse, coral, etc.

    Possibly. But many of these other proxy records also show strong medieval warming and even more pronounced cooling during the LIA. So what are you saying – that the proxies are only used for the periods where the results agree with the “consensus” opinion.


    Second, given the fact that proxy records are local by nature, one would not necessarily expect that any particular proxy record would follow the global trend by itself

    Agree – but an amalgamation of records representing the NH, say, should agree with any fluctuatuions in NH climate. Thats what the hockey-stick attempts to do. That’s what studies by Cook and Briffa also try to do but both note large disagreement in the late 20th century.


    As to why proxy series have not been updated, that is simply a red herring. In the first place, a direct instrumental record is always superior to a proxy record, simply because it is direct and precise. A thermometer reading is better than determining the width of a tree ring for measuring temperature.

    Yes – and balloon pressure transducers, balloon thermistors and satellite MSUs are all superior to a mish-mash of thermometer readings – many of which are in areas of the world where there is questionable maintenance. Also the thermometer readings don’t, of course, adequately represent 71% of the world that is covered by the oceans. As you know the airborne readings show nowhere near the warming shown by the thermometers. The satellite readings have come under massive scrutiny – how come the surface temperature record has escaped investigation. An up-to-date tree ring record might, at least, provide some validation for the thermometer record. I would have thought it’s fairly important to determine whether the measured temperature rise over the past century is ‘real’ – or simply the result of the world’s population QUADRUPLING since 1900.

    Note we are moving away from the original discussion which concerns the validity of the hockey-stick and into a whole range of controversial issues. I don’t mind continuing – as long as Tim Lambert (the “host”) has no objections.

  36. #36 Dano
    March 10, 2005

    Note we are moving away from the original discussion which concerns the validity of the hockey-stick and into a whole range of controversial issues.

    Pardon me, but this sounds like a cut/paste from GES or see-oh-too.

    The sat record, you’ll recall because you’ve read the relevant literature (you HAVE read the literature, right? That’s why you are speaking to the issue, right? Right?) is less than 30 years old. You’ll also recall the T tests and r^2 of the papers that looked at GSTs. And you’ll recall that, despite what the IndyFunded websites say (they really DO need to be updated) that the current analyses of the MSU data are very close to the GST record over that time.

    Oh, and of couse you recall from your research that the sfc temp record has not escaped investigation and there are plenty of papers in your research pile that investigate the record.

    The effect that you imply has been found to be minimal in diverse places like China, South Africa, North America, Meditteranean Europe…

    oh, and you’ll recall that there are papers in your pile that look at the recent tree ring data and find it has been affected by human emissions, and many researchers find the recent tree ring record to be problematic.

    I realize that’s a lotta stuff to remember, John, but you are speaking authoritatively to the issue, so you should ensure your recall faculties recall the research you’ve done on this subject.

    HTH,

    D

  37. #37 Scott Church
    March 11, 2005

    John, With all due respect, you have been shown the research regarding the Hockey Stick and related subjects, yet you keep repeating the same claims over and over again without getting to the point of where the literature is actually flawed. If you have read it, that is not at all apparent from your statements here.

    “Scott – The same goes for the scales of the 2 graphs. I’m not trying to overlay one on top of the other for comparsion. I’m simply drawing a conclusion from one and looking to see if this matched the behavious of the other… From the natural forcings/observations graph we can conclude that any warming between 1900 and 1950 is mostly natural. If so, then the period between 1900 and 1950 should look pretty much the same as the period between 1000 and 1900, i.e. the curves and bumps representing natural variability should be roughly similar. The sharp increase just after 1900 on the hockey-stick graph is more than would be expected and, in my opinion, this is enough to throw some doubt on it’s accuracy.”

    Mostly natural, but not entirely, and no, the period between 1900 and 1950 does not have to look similar to the last millenia. The sources I cited above regarding solar forcing in the last century (IPCC, 2001, Chap.s 2, 6, and 12) speak to the strength and frequency of solar fluctuation in the 20th century and the previous millenium, as characterized by direct observation and proxy records. Solar fluctuations occur on several time scales and can be either small or large. There is nothing about the observed level or frequency of solar fluctuations during the last century or the last millenia that is inconsistent with the Hockey Stick. Simply repeating the contrary ad-infinitum does not change this. You need to show where the literature on the subject is actually flawed. You have not done this, despite being shown where it is.

    “but an amalgamation of records representing the NH, say, should agree with any fluctuatuions in NH climate. Thats what the hockey-stick attempts to do. That’s what studies by Cook and Briffa also try to do but both note large disagreement in the late 20th century.”

    They do no such thing, and you have been shown this. Exactly where are the disagreements you keep referring to?

    “and balloon pressure transducers, balloon thermistors and satellite MSUs are all superior to a mish-mash of thermometer readings – many of which are in areas of the world where there is questionable maintenance. Also the thermometer readings don’t, of course, adequately represent 71% of the world that is covered by the oceans. As you know the airborne readings show nowhere near the warming shown by the thermometers.”

    Every word of this is false, yet the myth keeps being repeated. The satellites are not any more rleiable than the alleged “mish-mash” of surface temperatures. If anything, they are far less so, and the radiosonde record is even less well characterized. Furthermore, even if these did disagree with the surface record, this would prove nothing regarding the surface record, no matter how many times climate change skeptics shout otherwise. I have written a paper on this very issue which I recommend you read. It is at http://www.scottchurchimages.com/enviro/docs/MSU-Troposphere-Review01.pdf.

    As for getting off the subject, I agree that this is a good thing, as long as it is productive. But the continual repeating of discredited statements without a demonstration of their flaws, as if stidency trumps data, is not productive.

  38. #38 Dano
    March 11, 2005

    …as if stridency trumps data…

    Good ‘un Scott. There’s a tactic right there. BTW, I haven’t fergotten.

    D

  39. #39 John Finn
    March 11, 2005

    Dear Scott (and Dano)

    As I’ve stated previously I think we are trying to deal with too many issues at once, so I propose we restrict the ‘debate’ to one issue at a time. Let’s try and deal with the hockey-stick first, since that is the main subject matter for this thread. Meanwhile, I promise I’ll read your paper, Scott.

    On to your first point, where you say solar forcing is responsible for the increase in global temperature between 1910 and 1940. This is from your previous post

    The punch line is that from roughly 1910 to just after 1940 large, but natural fluctuations in solar activity combined with gradually increasing responses to greenhouse gases drove an increase. Between 1940 and 1970 decreasing solar activity ate into this increase and caused things to cool off some but with diminishing returns because greenhouse gas responses were picking up speed. The large increases after 1970 are due to the greenhouse gas forcings taking off much faster and running away from waning solar activity.

    A casual observation might be that it seems a bit of a coincidence that this increase in solar forcing – unseen in the previous 900 years – happens at the very starting point of the period for which Professor Mann bases his PCA, but we’ll let that pass. Let’s look at solar irradiance over the past century or so to see if it agrees with your (or the IPCC’s) version of events.

    Gavin Schmidt (on realclimate) provided me with this link to a pair of graphs

    http://www.giss.nasa.gov/data/simodel/F_line.gif

    The top graph shows all the individual climate forcings since 1850. The one we are interested in is the orange line (i.e. the solar irradiance). It’s a bit difficult to make out but I think you can see that for the latter part of the 20th century solar irradiance is POSITIVE. In fact, this agrees with studies that claim solar activity – since 1940 – is the highest it’s been for a millennium. Now, if you look closely at the period in question (i.e. 1910 to 1940) you should see it is NEGATIVE compared to the long-term average. I have a much clearer blown-up version which makes it easier to see. But, the evidence from the NASA graph is directly opposite to what you say above.

    In fact, it’s hard to see how any increase (let alone a large one) could have possibly occurred in 1910 as all forcings (see bottom graph) are negative leading up to 1910. Notice the big dips for the volcanic eruptions (e.g. Krakatoa in 1883). So we are left with the conclusion that natural variability – other than known forcings – is (or can be) quite considerable.

    OK – on to your next point where you question the lack of aqreement between tree ring data and the late 20th century warming as measured by the surface temperature record. Well, it’s all here and on realclimate. I can only repeat the quotes I have already previously referenced

    Mike Mann (see http://www.realclimate.org/index.php?p=11#more-11) says “however, anomalous behavior in recent decades apparently related to non-climatic (or, at least, non-temperature related) influences on tree growth does indeed compromise the use of these data in reconstructing temperature changes over the past several decades [see e.g. the discussions by Briffa et al (1998)]

    So Mike accepts there is “anomalous behaviour” I’ll leave you to check out the Briffa paper, Scott.
    But also referenced in realclimate (same article) is the Cook et al paper in which the comments on the “.under-estimation of actual warming since 1982″. Read both Briffa and Cook they both contain graphs which show various reconstructions compared with the surface temp record. By the end of the 20th century, the ST record is 0.3 degrees higher than ANY of the proxy reconstructions.

    But the real clincher for me is an article by realclimate called “what if the hockey-stick were wrong” which was released just a few days before the MM05 study was published. Seems they think it’s a possibility.

    The next stage should be where we go on to actually look at what the studies by MM and van Storch are actually saying – and more importantly what independent experts are saying about the studies, but I’ll wait for you to respond to the other stuff first.

  40. #40 Eli Rabett
    March 12, 2005

    Take a look at http://www.grida.no/climate/ipcc_tar/wg1/fig6-8.htm

    Solar forcing definitely increased from 1900 to 1950 (about 0.3 W/m^2 according to Lean, et al. and 0.4 according to Hoyt and Schatten) and has oscillated since then according to the solar cycle with maybe a small systematic increase. The change in the greenhouse gas forcings are about ~0.4 W/m^2 btw 1900 and 1950 eyeballing the curves. All of this is compressed in the figure Schmidt sent you to because of the huge jump since 1950 in GHG forcing.

  41. #41 Dano
    March 12, 2005

    John, the solar influence on warming has already been quantified; you’ll want to focus your efforts directly at that, not at meandering, non-focused assertions about vague things that require more backing and elucidation on your part that what you provided.

    Oh, and using the analyses of amateurs who have already been shown to…well. Tim’s got a section on that in the left column. But that’s a clue for us all, I’d say.

    In your pile of research papers you accumulated on your various trips to the library, you’ll likely find the quantification of solar influence on GST. Here’s a hint on where to look in your pile, since you’re a Googler, right?

    HTH

    D

  42. #42 John Finn
    March 12, 2005

    Eli

    You are a star. I couldn’t remember where I’d seen that solar forcing diagram before. This agrees with what I said above about solar forcing (i.e negative 1900-1940; positive 1940-to date)

    On GHGs – yes slightly up but this is offset by sulphate aerosols so net effect is zilch. You also have a problem with a time delay (decades – apparently) because of radiative equilibrium, so any increase doesn’t take effect for several years. This is a topic on it’s own and I’d prefer to keep my powder dry on this for a bit.

    cheers again for the link.

  43. #43 John Finn
    March 12, 2005

    you’ll likely find the quantification of solar influence on GST ..

    Dano, I’m not disputing solar influence. I’m disputing (with Scott) the periods for which solar activity was at it’s strongest.

    Scott says solar activity was strong between 1900-1950 and weak between 1950-2000. I say the exact opposite. Eli has provided a link which supports my view. Go to his link and look at diagram C.

  44. #44 Dano
    March 12, 2005

    Well, John, the thread is meandering all over the place. A couplea posts ago you posted a buncha misinformation that, corrected, will make your overarching assertions tighter. Your overarching job, if I understand your intent, is to show natural variability is at work here.

    You need to show how the people who study this for a living (as opposed to Googlers) are incorrect. The best way to do this is to present evidence – empirical evidence – that shows that the small influence of solar is stronger than the strong influence of CO2. Your posts on RC show me you don’t know how to read a paper (the lake sediment thingy), so you have some catching up to do.

    HTH

    D

  45. #45 Eli Rabett
    March 12, 2005

    Well John, what do you think a negative forcing means, and more to the point, what is it negative with respect to.

  46. #46 Yelling
    March 12, 2005

    John Finn, I don’t quite follow your point, but I think the graph supports Scott’s position.

  47. #47 Yelling
    March 12, 2005

    John Finn, I don’t quite follow your point, but I think the graph supports Scott’s position.

  48. #48 John Finn
    March 14, 2005

    John Finn, I don’t quite follow your point, but I think the graph supports Scott’s position.

    Please explain – how?

    Scott says high solar activity between 1900 and 1950 – both links (mine and Eli’s – Eli’s is much clearer) show LOW solar forcing between 1900 and 1950.

    Scott says “waning solar activity” in late 20th century – both links show HIGH solar forcing.

    The graphs show the exact opposite to what Scott has said.

    Eli

    Well John, what do you think a negative forcing means, and more to the point, what is it negative with respect to.

    Actually this is a good question. A change in forcing will produce a cooling (negative change) or warming (positive change). e.g. an increase in aerosols tends to reflect solar radiation so is a NEGATIVE forcing; an increase in GH gases will prevent IR radiation leaving the atmosphere causing it to warm – so is a POSITIVE forcing.

    The forcings on the graphs (I believe) are measurements relative to some long-term base-line figure. So solar forcing, for example, is relative to the long-term average of 1367 W/m2 (which used to be thought of as a constant). GH gas forcings are relative to the forcings given by the pre-industrial levels of the various gases.

    Basically the graphs are representing the change in forcings over the past century or more. It’s the changes we are interested in – more that the actual values.

    Dano

    I’m struggling to understand your point. My original argument is that the warming between 1910 and 1940 is natural. Climate models suggest it is natural. Scott says this is due to enhanced solar activity, but from the graphs we (apart from yelling) can see that solar activity between 1910 and 1940 was BELOW the long-term average.

    In other words the sudden temp increase in 1910 is due to TOTALLY natural variabilty – but the hockey-stick shows the previous 900 years (pre-1900) to be much ‘flatter’ – in general – which would suggest the pre-1900 variability represented by the H-S is wrong.

  49. #49 Yelling
    March 15, 2005

    John Finn: Looking at Scott’s comments we find:

    “The punch line is that from roughly 1910 to just after 1940 large, but natural fluctuations in solar activity combined with gradually increasing responses to greenhouse gases drove an increase. ”

    From Eli’s link we can see that the solar activity is increasing which will drive an increase. Point to Scott.

    “Between 1940 and 1970 decreasing solar activity ate into this increase and caused things to cool off some but with diminishing returns because greenhouse gas responses were picking up speed. ”

    The only possible thing you could disagree with on this is whether you use Lean or Hoyt. With Hoyt it is 1940 to 1970, with Lean it is 1950 to 1970. Point to Scott.

    “The large increases after 1970 are due to the greenhouse gas forcings taking off much faster and running away from waning solar activity. ”

    After 1970 the links shows solar activity relatively steady except for sunspot cycle. Game to Scott.

    Regards,
    Y.

  50. #50 John Finn
    March 15, 2005

    Dear Yelling

    A simple question. Is solar activity higher between 1940 to 2000 or between 1900 to 1940.

    Let’s ignore issues like radiative equlibrium and ocean heat uptake for the moment and assume that the warming response to any increase in solar forcing is immediate.
    So where are the big temperature rises and falls on the hockey-stick which should have occurred earlier. There should be much wilder fluctuations in temperature between 1750 and 1850 than ever there was between 1900 and 1940. Scott’s theory relies on solar activity remaining fairly constant AND below the long-term average for the previous 900 years in order to be consistent with the hockey-stick. Remember this is what the original discussion was about.

    Whether there is an increase between 1900 and 1940 is not that relevant (the increase should be before). The question we need to ask is the solar forcing SUFFICIENT to produce an increase in temperature which is GREATER THAN THAT IN THE PREVIOUS 900 YEARS. The rises (and falls) between 1900 and 1940 are simply fluctuations around the long-term average – but with a slightly rising trend. In actual fact, global temperatures are FALLING at the ‘local’ peak around 1950.

    Another problem is that temperatures begin to rise around 1910 – when solar forcing is still NEGATIVE. A temperature rise cannot anticipate an increase in solar forcing it must follow it.

    But having said all that, perhaps, we should go with Scott’s theory as this would almost certainly explain the late 20th century warming. Since 1940 solar activity has been constantly above the long-term average which will have the affect warming the earth’s atmosphere – but more importantly perhaps it will gradually warm the oceans. The total energy emmitted and absorbed is far greater that in any similar period. Yep – I think we’ve cracked it.

    All 20th century temperature changes can be explained by changes in solar forcing.

  51. #51 Jeff Harvey
    March 15, 2005

    John,

    Yup, you’ve cracked it alright. Now you’ve stumbled onto this solar forcing bombshell that has somehow been overlooked by thousands of independent climate researchers around the world, I am sure they’ll be eternally grateful for your input.

  52. #52 Carl Jarrett
    March 15, 2005

    McKitrick and McIntyre fool another columnist – see Margaret Wente’s column in March 15th Globe and Mail (www.globeandmail.ca)

    Margaret Wente
    From Tuesday’s Globe and Mail
    Tuesday, March 15, 2005

    Steve McIntyre is Paul Martin’s worst nightmare. A couple of years ago, the Toronto mining consultant got interested in the science behind global warming. Mr. McIntyre is not a scientist. He’s just a curious citizen with a first-rate mathematical mind who was intrigued by the biggest public policy issue of the age. “It started as a hobby,” he says.

    But soon his research on climate change took over his life – at some cost, he says ruefully, to his business.

  53. #53 Eli Rabett
    March 15, 2005

    There is an interesting 1997 article on tree rings and how they are (or are not) as it may be affected by carbon dioxide and climatic change. http://www.nap.edu/openbook/0309058767/html/78.html

    The bottom line is that, at least at that time, CO2 availability was not the limiting factor in most tree ring samples, and therefore the rings could be used as temperature proxys throughout the 20th century. There were some cases however, where this was not the case and one therefore had to be cautious.

  54. #54 John Finn
    March 15, 2005

    Before we draw any more conclusions from the IPCC solar forcing diagram, it perhaps ought to be verified. I’m not sure the vertical axis on the diagram is correct.

    The axis label says “Radiative Forcing (W/m-2)” but the large increments are only in steps of 0.2. Changes in forcing on this scale are negligible and extremely difficult to detect. So it’s possible the vertical axis should be the temperature response. This might make a slight difference to the above argument – but how much and in which way I’m not sure.

  55. #55 Dano
    March 16, 2005

    Eli’s on it again.

    The Italians are having trouble with recent tree rings (did I mention that here, or elsewhere?).

    Anyway, the Italians: Dongarra G, Varrica D 2002. delta C-13 variations in tree rings as an indication of severe changes in the urban air quality. Atmospheric Environment 36:39-40 pp. 5887-5896 [Although the present survey is restricted to a limited zone within the urban area and it is containing some uncertainties due to the high degree of spatial and temporal variability of the urban CO2 distribution, we feel that the results of this numerical deconvolution of small delta, 13C in tree rings document the large perturbation of the local atmosphere by a cumulative emission of CO2 from fossil fuel burning and, at the same time, the accelerating rates of anthropogenic pollution. These trees have experienced, during the last 50 years, a continuous increase of the local content of atmospheric CO2, which has reached the average values of about 390-400 ppmV. This is to be compared with the 368 ppmV of global atmosphere. ]

    I came across this paper while doing research on another topic, but it appears to be robust, viz:

    Rakowski A, Nakamura T, Pazdur A 2004. Changes in radiocarbon concentration in modern wood from Nagoya, central Japan. Nuclear Instruments & Methods in Physics Research Section B-Beam Interactions With Materials And Atoms 223-24 pp. 507-510. [not something the average reader would read...]

    And this compilation:

    Martinelli N 2004. Climate from dendrochronology: latest developments and results. Global and Planetary Change 40:1-2 pp. 129-139. [On a global scale, this supposed positive influence on wood production and forest regeneration was thought to have the possibility of balancing the CO2 increase by carbon sequestering through photosynthesis. On the contrary, tree-ring records indicate that, under recent climate warming, drought may have been an important factor in limiting carbon uptake in a large portion of the boreal forest, one of the planet's major potential carbon sinks. If this limitation in growth due to drought stress is sustained, the future capacity of northern latitudes to sequester carbon may be less than currently expected.]

    D

  56. #56 Dano
    March 16, 2005

    Before we draw any more conclusions from the IPCC solar forcing diagram, it perhaps ought to be verified. I’m not sure the vertical axis on the diagram is correct.

    That’s what’s great about Murrica: a Galileo can arise out of our midst and smite down hundreds of scientists in their hubris.

    Let us know, John, when your paper gets published.

    D

  57. #57 scott church
    March 16, 2005

    John Finn, Thank you for supplying the links and the information about solar forcings.

    They are most interesting. But you don’t seem to have understood me. Perhaps my earlier
    remarks about solar vs. greenhouse forcing could have been worded better. The absolute
    strength of either is not the issue – it’s the relative
    contributions of variations in each
    , with respect to each other, to the overall forcing during this period.

    Furthermore, it does not appear
    that you understand your own data either, or that you have considered its place in the larger picture.

    The link you provided for solar forcing is not actually solar forcing data as such – it’s the

    solar forcing input that is used in collective natural/anthropogenic forcing inputs to the GISS

    SI2000
    General Circulation Model (Hansen et al., 2002; 2005). The actual solar forcing data from

    which this curve was derived is from
    Lean et al. (1995). That data is in the IPCC 2001 report, Chap. 6, this link – Fig. 6.5 alongside of

    data from Hoyt and Schatten (1993), Solanki and Fligge (1998), and Lockwood and Stamper
    (1999). Click Fig. 6.5 to enlarge it. Note that all these datasets show clear increases in
    total solar irradiance (TSI) from the late 19th century until roughly the middle of the 20th,

    followed by a decrease from then until the 70′s, and a upturn after that. The mid to late
    20th century fluctuation is least obvious in the data your SI2000 forcing curve was based on,

    but it is more clear in the others, especially in Hoyt and Schatten (that data has been
    updated to 1999). Overall, estimates of overall solar forcing since 1750 range from 0.1 to
    0.5 W/m^2 (rms value of 0.3), and 0.6 to 0.7 Wm-2 since the Maunder Minimum of which roughly
    half has occurred since 1900 (Hoyt and Schatten, 1993; Lockwood and Stamper, 1999;

    Crowley, 2000;
    IPCC, 2001, Chap. 2). You can even see this in the curves linked above brom IPCC (2001,
    Chap. 6) which clearly shows that the observed levels of solar forcing have not been equaled
    since at least 1600. Note however that the level observed in 1900 – right when the Hockey
    Stick takes off – had been duplicated at least 2 or 3 times previous. Overall, the direct
    effect of solar forcing during the 20th century was about 20 to 25% of that due to well-mixed

    greenhouse gases (IPCC, 2001, Chap. 12.2.3.1;).

    So all in all, this pattern of solar forcing was, as you say, smaller than the corresponding
    greenhouse forcing since 1900, and it was much more variable and less certain. But what is
    different here is that from roughly 1900 to 1950 it had a larger contribution to the overall
    forcing with respect to the corresponding greenhouse gas forcing at that time. From
    1950 to 1970, we see in the data cited above that solar activity waned while anthropogenic
    greenhouse gases increases but the ratio was still high enough for the relative contribution
    to show up in overall trends as a decrease. From the 70′s on greenhouse gases dominate the
    forcing, and continue to do so to this day. Furthermore, volcanic activity, which has a
    significant moderating effect on solar forcing, was low during the first half of the 20th
    century, allowing full solar impact then. During the latter portion it was considerably
    higher and damped much of the increase in solar forcing since then shown in the data above
    (IPCC, 2001, Chaps. 6; 12.2.3.1). Well mixed greenhouse gases, including the sizeable

    anthropogenic contribution, dominate the climate forcing of the latter 20th century by a much

    larger fraction than they did at the beginning (Crowley, 2000; IPCC, 2001, Chap. 2). Today

    John, overall solar contributions are peanuts by comparison.

    So you see, my “theory” does not require “solar activity remaining fairly constant AND below the long-term average for the previous 900 years”. Even if it did, the data cited above shows the level at 1900 to have at least 2 to 3 rough precedents during the previous 4 centuries, and none for the period after. Depending on how they are smoothed and trended, the data cited above show solar increases beginning anywhere from 1890 to 1910, and the temperature response still has an increasing anthropogenic component then anyway, so this solar piece is not the whole picture during this period (I never said it was).

    Which brings us right back to what has been repeated here from Day One – the Hockey Stick
    stands
    . The satellite measured troposphere record does not change that, and neither does

    any and all hand waving about generalized “natural variations”. The observed global
    temperature increases since the late 20th century are the largest since at least 400 years
    ago, and very likely since 1000 AD.

    And we’re doing it.

    REFERENCES

    Crowley, T.J., 2000: Causes of Climate Change Over the Past 1000 Years. Science, 289,
    270-277.

    Hansen, J., M. Sato, L. Nazarenko, R. Ruedy, A. Lacis, and 23 other co-authors 2002. Climate
    forcings in GISS SI2000 simulations. J. Geophys. Res. 107.

    Hansen, J., M. Sato, R. Ruedy, L. Nazarenko, A. Lacis, G.A. Schmidt, G. Russell, and 38 other

    co-authors 2005. Efficacy of Climate Forcings. Submitted to J. Geophys. Res.

    Hoyt, D.V. and K.H. Schatten, 1993: A discussion of plausible solar irradiance variations,
    1700-1992. J. Geophys. Res., 98, 18895-18906.

    Lean, J., J. Beer and R.S. Bradley, 1995: Reconstruction of solar irradiance since 1610:
    Implications for climate change. Geophys. Res. Lett., 22, 3195-3198.

    Lockwood, M. and R. Stamper, 1999: Long-term drift of the coronal source magnetic flux and
    the total solar irradiance. Geophys. Res. Lett., 26, 2461-2464.

    Solanki, S. K. and M. Fligge, 1998: Solar irradiance since 1874 revisited. Geophys. Res.
    Lett., 25, 341-344.

  58. #58 John Finn
    March 16, 2005

    Scott

    Can I just remind you how this all began. One of the reasons I had doubts about the hockey-stick was not it’s representation after 1900. It was the lack of variability before 1900 – relative to the post-1900 period.

    You said that solar forcing (a natural forcing) could explain the sharp increase (relative to the earlier period) between 1910 and 1940. I’m not convinced that solar forcing is strong enough in that period – relative to previous periods – to explain it. Eli’s link extends back to 1750 for Lean and H&S. Both show plenty of variability throughout the whole period. There is nothing exceptional about 1910-1940. If you read Chap 6 of the IPCC report you’ll notice that Douglas Hoyt’s (H&S) reconstruction is extended back to include the Maunder Minimum in the late 17th century. This shows a ‘huge’ increase over the following 70 years (apologies – iv’e just mnoticed you have read it). A time, incidentally, which saw the Central England Temperature (CET) – the only thermometer record at the time – measure annual temperature increases of 3 DEGREES CELSIUS within 40 years. The hockey-stick shows none of this.

    Now 2 separate studies have shown the statistical reasons why the hockey-stick may be flawed. The MM05 study has generated a lot of discussion and argument about whether or not there are errors in the MBH methodology but this acts as a smoke-screen to the real issue. Right or Wrong -


    The use of non-centred PCA was highly inappropriate in this case.

    But don’t take my word for it. Here is a selection of comments from leading experts (at least two of whom are supporters of AGW position)

    Professor Hans von Storch, an IPCC Contributing Author and internationally-renowned expert in climate statistics at the Center for Coastal Research in Geesthacht, Germany, is quoted as saying that MM criticism is “entirely valid.”

    Dr Mia Hubert, a statistician (specialising in robust statistics) at the Katholieke Universiteit Leuven in Belgium, examined the materials at the request of NWT and agreed with the MM05 conclusions, saying: “Tree rings with a hockey stick shape dominate the PCA with this method”

    In an essay published in the MIT Technology Review Professor Richard Muller of the University of California at Berkeley said “the findings hit me like a bombshell, and I suspect it is having the same effect on many others. Suddenly the hockey stick, the poster-child of the global warming community, turns out to be an artifact of poor mathematics.”

    Dr. Rob van Dorland, IPCC Lead Author and climate scientist at the Dutch National Meteorological Agency is quoted saying it will “seriously damage the image of the IPCC ……It is strange that the climate reconstruction of Mann has passed both peer review rounds of the IPCC without anyone ever really having checked it. I think this issue will be on the agenda of the next IPCC meeting in Peking this May.”

    Add to this the dominant influence of the controversial bristlecone pine series and it’s clear that the hockey-stick is skating on pretty thin ice. Perhaps it’s a ice-hockey stick. (note: this is a joke but not a very good one)

    Finally – none of this proves the existence or otherwise of anthropogenic global warming. I think it exists – as it always has – but I don’t think it is as dominant or threatening as is being portrayed. We are under-estimating past climate variability and over-estimating climate sensitivity (but that’s another topic)

  59. #59 John Finn
    March 16, 2005

    Dano


    Before we draw any more conclusions from the IPCC solar forcing diagram, it perhaps ought to be verified. I’m not sure the vertical axis on the diagram is correct.


    Let us know, John, when your paper gets published

    It’s ok, Dano, I made a mistake here. Please ignore that earlier post about solar forcing diagram.

    If you’re interested – I was comparing the small changes in the diagram to to TOTAL solar irradiance (TSI) of 1367 W/m2. Obviously, the changes as a percentage of TSI are miniscule.

    However, the earth only receives, on average, about 1/4 of that – of which a further 30%-ish is reflected, so the actual base line average is only around 240W/m2.

    Sorry about that, Dano. But it’s good to know the peer review process is operating efficiently on this site.

  60. #60 Yelling
    March 16, 2005

    John Finn:

    You said that “The use of non-centred PCA was highly inappropriate in this case. ” and backed it up with a couple of quotes, one of which was from von Storch “Professor Hans von Storch, an IPCC Contributing Author and internationally-renowned expert in climate statistics at the Center for Coastal Research in Geesthacht, Germany, is quoted as saying that MM criticism is “entirely valid.” ”

    Do you have a reference for this? I was aware of Dr. von Storch’s comments regarding the fact that the methodology produces hockeysticks, but I was not aware that they said the methodology was inappropriate or that all of MM criticism is “entirely valid.”

    Regards,

    Y

  61. #61 Eli Rabett
    March 16, 2005

    Before John hops onto the Hoyt and Schatten or Lean bandwagons he should remember that they are proxy reconstructions, and as such <nonesense> should not be trusted in the 20th century</nonesense>

  62. #62 scott church
    March 17, 2005

    John, Per your comments,

    “One of the reasons I had doubts about the hockey-stick was not it’s representation after

    1900. It was the lack of variability before 1900 – relative to the post-1900 period.”

    My last post shows the variability directly, and provides clear citations to the data.

    IPCC (2001), Fig. 6.5 linked there shows 3

    distinct peaks in solar forcing prior to 1900 that are nearly equal in magnitude to the solar

    in that year. The solar forcing between 1900 and 1940 is clearly stronger, and this will

    have been supplemented by an increase in well-mixed greenhouse gases that was not

    present before. Remember, it’s the two together that makes the overall difference,

    though either or both may impact whether the trend at any given time is monotonic or not

    depending on what percentage of the joint forcing they are contributing. The overall solar

    forcing of the last century is roughly one fourth that of well-mixed greenhouse gases with

    most of this contributed earlier in the century rather than later (IPCC, 2001, Chap.

    12.2.3.1). In addition, we expect a somewhat delayed response from the greenhouse gas

    component due to the nature of its forcing in conjunction with oceanic thermal uptake, so

    it’s not surprising that its contributions to the actual change would be delayed somewhat

    compared to the solar component.

    “You said that solar forcing (a natural forcing) could explain the sharp increase

    (relative to the earlier period) between 1910 and 1940. I’m not convinced that solar forcing

    is strong enough in that period – relative to previous periods – to explain it.”

    I said no such thing. The combination of well-mixed greenhouse gases and solar together

    drives the overall trend, though at any given time, one or the other may dominate the

    magnitude, the variability, or both. Per the data I’ve already provided, solar forcing

    between 1900 and 1940 is equal to or greater than the observed solar forcing of at least the

    previous 4 centuries. Solar plus greenhouse gas forcing is clearly greatest in the

    20th century. I also pointed out that volcanic activity was much stronger since 1940-1950,

    and this masks much of the solar forcing since then.

    As for the MM study, it was literally riddled with basic mathematical and data errors and has

    been soundly refuted elsewhere, so I won’t re-invent the wheel here. For instance, see Tim’s

    other posts about MicKitrick’s work on the Hockey Stick here, and here, or this write-up at RealClimate.

    As for all the quotes from “leading experts”, these are arguments from authority, not data –

    a well known logical fallacy. Particularly in this case as every one was presented without

    the context in which the statement was made or a citation that

    would allow for the determination of it. As for Von Storch, his criticisms of the original

    Mann et al. work were largely in regard to where the confidence intervals came from (I

    believe). Assuming I understand it correctly, if he is right there will be a possibility

    that the

    Northern Hemisphere might – might – have seen higher temperatures once or twice in the

    last 1000 years. That is hardly the same thing as saying that the Hockey Stick is invalid

    and everything we’ve been seeing for the last 50 years is as normal as Christmas pudding. It

    is also noteworthy that the overall Hockey Stick picture has been reproduced by Moberg et al.

    (2005) using different methods than those criticised by your cited authorities. The chief

    differences in their results alter the picture at certain points, but leave the overall

    “hockey stick” effect intact – and robust.

    Furthermore, claiming Von Storch as an ally is a rather dangerous tactic for enemies of the

    Hockey Stick. His criticisms of it are rather specific and limited to method, and he’s not

    exactly a big enemy of the

    overall Hockey Stick picture or the idea that human activity is causing dangerous climate

    change – all the more reason to be careful about quoting him without

    context. You will recall that it was he who resigned in disgust from

    his position as Editor-in-Chief of Climate Research over their decision to publish the

    Soon & Baliunas (2003) paper on the Hockey Stick that argued many of the same points you have

    been trying to
    defend throughout this post.

    REFERENCES

    Moberg, A., D.M. Sonechkin, K. Holmgren, N.M. Datsenko, and W. Karlen, 2005. High variable

    Northern Hemisphere temperatures reconstructed from low- and high-resolution proxy data.

    Nature, 433, 613-617.

    Soon, W. & S. Baliunas. 2003. Proxy climatic and environmental changes of the past 1000 years. Climate Research (23), 89-110.

  63. #63 Yelling
    March 17, 2005

    Scott: I think your comments about von Storch are good and valid. If I can add the following quote by von Storch.

    [I]Is the hockey stick curve crucially affected by Mann’s shoddiness? We tested it by way of a one-thousand-year simulation with a climatic model and found that the effect wasn’t significant. The error is real, but probably not far-reaching. [/i]

  64. #64 Dano
    March 17, 2005

    If I may – and I’m traveling soon so I don’t want this to seem like a hit and run – but I think it’s obvious someone on this thread does not understand the papers they read.

    Someone on this thread needs to get a degree in the natural sciences, during which they will read lots of papers, and do lab work such that they will write lots of papers (or at least be graded upon how well they discipline themselves in the disciplinolatry).

    After that degree is achieved, said person will look back upon these posts sheepishly.

    D

  65. #65 John Finn
    March 17, 2005

    Ok, guys, I’m not going to deal with all your points – there are other things in life, but I’ll try and respond to a few.

    Dano

    Throughout this ‘debate’ (and others) you’ve consistently attacked me or my credentials. It would have been – oh so easy – for me to respond in kind. But, just for the record I have a degree in Mathematics. I was considered a decent statistician some years ago. Good enough, at least, to take on – and beat – a large utility company in the UK on what, I suppose, could be loosely described as a compensation claim. Admittedly, some of my knowledge is a bit rusty now, but with a bit of reading I can usually get up to speed on most climate change issues.

    Eli, you say


    Before John hops onto the Hoyt and Schatten or Lean bandwagons he should remember that they are proxy reconstructions,

    What is the point in YOU providing a link and then dismissing it’s importance. Yes – all the early solar forcing data is based on a variety of proxies depending on which study you use. But that’s all we’ve got to go on. All the early hockey stick data is based on proxy records. But what else is there? You’re right to be cautious about the results, but as long as the correct levels of confidence are stated there’s not much else can be done.

    Scott/Yelling

    There are 2 ways you can interpret von Storch’s slight reticence in fully endorsing the MM05 findings. One is he genuinely thinks that MM05 has only partially refuted MBH. The other is, that it diminishes the role of his own study which actually questions past climate variability as represented by the H-S. If MM05 is wholly correct then von Storch’s work is interesting, but not ground-breaking.

    Scott

    You’ve moved slightly on the 1910-1940 period. We are now talking about solar forcings + ghg forcings. Ok –
    <li> At the start of the discussion I referenced a graph
    which showed modelled natural (including solar) forcings in agreement with observations. Basically, this broadly suggested that temperature increases were due to natural causes. </li>
    <li> It is true there would have been an increase in atmospheric ghg levels in 1910 – but the effect of these was offset by an increase in aerosols. </li>
    <li> Even without the aersosols, it’s doubtful if any warming due to ghg concentrations would have been noticeable. One – there simply wasn’t enough. There were approx 295 ppm in 1900; 315 ppm in 1958. Even with linear growth you’d be hard-pressed to reach 300 ppm by 1910. Secondly – a warming response will not show up immediately. There are discussions on the realclimate site which explain why warming is less that it should be – if the IPCC assumptions about climate sensitivity (with feedbacks) is correct. You must have heard about the 0.5 deg rise that’s still in the pipeline. It’s all down to ocean heat uptake and radiative equilibrium. Ocean heat uptake is fairly self-explanatory; radiative equilibrium is achieved when incoming solar radiation equals outgoing (infrared) radiation. An increase in ghg concentration will absorb more of the outgoing IR radiation than previous. This is re-radiated back to the surface – the surface warms and radiates more IR (see Stefan-Boltzmann Law). Eventually, in theory, equilibrium is achieved. But it doesn’t happen overnight – it’s more like several decades according to my ‘quick and dirty’ calculations on current forcings
    </li>

    To conclude, I have a problem with the ghg contribution before around 1950/60. And, if you read the realclimate discussions carefully – I think they do as well.

    Anyway thanks for the discussion – also thanks to Tim for ‘hosting’ it.

  66. #66 Scott Church
    March 17, 2005

    Thanks as well John! An interesting and thought provoking discussion. All the best.

  67. #67 Dano
    March 19, 2005

    John,

    the tactics I use probably inhibit clarity, but my contention is the persona adopted is unsupportable. A Googler doesn’t have sufficient information to make such proclamations.

    Nonetheless, nothing personal and all the best.

    D

The site is currently under maintenance and will be back shortly. New comments have been disabled during this time, please check back soon.