RealClimate logo

Curve manipulation: lesson 2

Filed under: — stefan @ 14 June 2007

Two weeks ago, we published the first lesson in curve manipulation taught by German school teacher and would-be scientist E.G. Beck: How to make it appear as if the Medieval times were warmer than today, even if all scientific studies come to the opposite conclusion. Today we publish curve manipulation, lesson 2: How to make it appear as if 20th Century warming fits into a 1500-year cycle. This gem is again brought to us by E.G. Beck. In a recent article (in German), he published the following graph:

Notice how temperature goes up and down in beautifully regular cycles since 800 B.C.? At the bottom, they are labelled “Dansgaard-Oeschger cycles” – this refers to the Dansgaard-Oeschger events found in Greenland ice cores during the last Ice Age (but not during the last 10,000 years), about which there is a serious scientific discussion whether they are paced by a 1500-year cycle (see my paper in GRL). Beck’s curve shows a warm phase 400 BC and the next one 1200 AD – that’s 1600 years difference, so it just about fits. (I’m not endorsing his curve, by the way, I have no idea where it comes from – I’m just playing along with it for the sake of the argument). So the next warm phase should be in the year – oooops… 2700 or 2800? Hang on, how come it looks like the current warmth fits so nicely into the cycle? Shouldn’t we be right in the coldest phase? Now I see it… two little lines across the x-axis indicate that the axis has been broken there – tick-marks after the break are in 200-year intervals and before the break in 400-year intervals, and there’s also a gap of 200 missing years there. So that’s how we make the current global warming fit past climate cycles – it’s so easy!

p.s. Beck appeared on German TV last Monday, after the “Swindle” film was shown, and he is announced to appear on the program “Report München” in the first channel of public German TV next Monday (18 June), to educate the viewers about another of his fantasy graphs, namely his CO2 curve. It promises to be a must-see for friends of the unintentionally farcical.

346 Responses to “Curve manipulation: lesson 2”

  1. 301
    Edward H. Moran says:


  2. 302
    Edward H. Moran says:

    Did everyone go to sleep?

  3. 303
    Edward H. Moran says:

    I believe we’re still on topic according to #256.

    We’re trying to establish what is a significant regression fit from environmental data as reported in….33..545W, which reports a relation between Magnetic intensity and ambient temperature as our paper did?

  4. 304
    ray ladbury says:

    Might I suggest that we take discussion of the paper by Tindall and Moran offline and provide comments to the authors directly via the email address provided by Dr. Tindall.

  5. 305
    J.S. McIntyre says:

    re 299

    I apologize in advance for venturing further off topic (though, as I’ve noted elsewhere, issues of biodiversity and man’s footprint upon the natural world are tied to Global Warming). What we are witnessing now in terms of species and habitat loss is unprecedented. One of the most obvious has been the collapse of fisheries. For example, Cod have all but disappeared as a viable food source from the North Atlantic.

    And this is not a problem confined to the North Atlantic:

    Fishing Down Aquatic Food Webs;jsessionid=baa9

    Ecologically Sustainable Yield

    It is not a “few” species, it is not the “business as usual” species coming-and-going you imply, but instead a very critical situation where we are witnessing large holes being punched in the food chain/web that will have invariably negative consequences. Put another way, we are entering uncharted territory, and the only comparable examples we have to draw upon are the five previous species die-offs found in the geological record. Wikipedia has a decent summary of the current event, if you are interested, though a Google search will provide a wide range of information:

    It is not a sudden occurrence, but something that has been a growing phenomenon, as Niles Eldredge discusses here:

    The point being is the extinctions are happening at an ever increasing rate, and will likely reach a tipping point in the conceivable future.

    It’s estimated we’ve already lost half the forests, half the amphibians, and the number of species now considered endangered continues to grow at an unprecedented rate. Once a species enters endangered status, the prospects are grim.

    It gets worse when you consider the effects of deforestation and habitat loss. Not only do we lose species, we lose important elements of sustainability such as water filtration and oxygen generation. In ‘Planet Earth: Saving Species’ biologist E.O. Wilson estimates that the ‘value’ we receive from the natural world in terms of sustenance, fresh water and soil, air and a plethora of other ‘services’ exceeds 30 trillion dollars annually. We get this ‘value’ at no cost to us.

    We could spend hours calling up all the literature that is available on these and other related threats to conserving sustainability, but what’s the point? If you want to believe there isn’t a problem, that is your right. You can pretend this is not a problem, that everything is, as I characterized, a case of ‘business as usual’, but you would be wrong.

  6. 306
    spilgard says:

    Sorry for contributing to the off-topic drift, but since I raised the question I feel obligated to respond. Henceforth, I promise to cease and desist.

    Re #295 – The linked article contains nothing about a connection between temperature and geomagnetism; rather, it argues that analytical tools which model phenomena in the earth’s atmosphere are also applicable to the study of magnetic phenomena in the earth’s dynamo.

    Re #301 – This paper examines only the 40-year interval 1935-1975. Again, a bit of support from paleomagnetic records of the earth’s magnetic field intensity is in order. Given that the field intensity has exhibited past variations which dwarf the present variation, it’s to be expected that a mechanism linking magnetic intensity to global temperature should leave a clear signal over geological time.

  7. 307
    Chuck Booth says:

    Re # 299 Rod B. saving a human vs. saving a bird/fish/insect/etc

    That is really a false dichotomy — can you offer an actual case in which such a decision was made that resulted in real harm to a person (as opposed to the person just being inconvenienced by, say, a requirement of the Endangered Species Act)?

    Re # 302 Did everyone go to sleep?
    Comment by Edward H. Moran â?? 1 Jul 2007 @ 3:21 am

    Hmmm…I don’t know what time zone the RC time stamp represents, but I suspect many North American RC readers were indeed asleep at 3:21 am in any NA time zone.

    BTW: Edward – I think it might reduce your frustrations a bit to realize that while the RC threads are started by the climatologists whose names are listed as moderators, most of the people posting questions and responses here are not climatologists, and don’t pretend to be. Thus, with the exception of an occasional in-line response from an RC moderator (usually in green type), the responses you receive (or don’t receive) to your posts, and the criticism of your paper, reflect the personal views and interests of those responders, and in no way represent the views of the community of scientists who consider themselves climatologists. If you want feedback from climatologists, you need to publish your paper in a journal read by climatologists.
    Finally, as someone has already pointed out, it often takes several hours for a posted message to appear on a thread, and sometimes messages are posted out of sequence, at least temporarily, until a queue of messages is cleared up. There is no full time RC moderator approving messages 24/7.

  8. 308
    Timothy Chase says:

    Edward H. Moran (#301) wrote:



    This is fairly old, but slightly better.

    I have a better suggestion, something much more recent:

    Can we use the aa geomagnetic activity index to predict partially the variability in global mean temperatures?
    M.A. El-Borie and S.S. Al-Thoyaib
    International Journal of Physical Sciences Vol. 1 (2), pp. 067-074, October, 2006

  9. 309
    Hank Roberts says:

    The Wolin et al. paper from 1981, if compared to the recent one, might be grist for the pertinent topic for curve manipulation.

    I have only the abstract of the former, which says in part:

    “Lag correlations are established for 27 of 29 pairs of Northern Hemisphere magnetic observatory and
    weather station data sets concerning magnetic intensity and temperature over the period 1935-1975.
    When plotted, horizontal component and total intensity component values did not correlate with
    temperature curves. The lag in temperature trends against magnetic variation ranges from one to three years ….”

    The abstract says Wolin et al. suggested possible mechanisms; I don’t have the full text. Did they justify leaving out two pairs of station data?
    (They didn’t rely upon data from CO2Science’s website as a data source, of course.)

    One wonders if the Wolin et al. “one to three years” offset is related to their proposed mechanisms, proposing a reason the “one to three years” offset made physical sense. And of course how their statistic would look if they included the omitted two pairs of station data, or why they were dropped. Wolin et al. didn’t rely on the website “CO2Science” to draw their data from, and had a lot fewer pairs of numbers.

    If an offset can be varied over time to keep the curves matching up, that sounds like Beck’s manipulation, which would be troublesome.
    Beck also had matching curves — his also got disjunct over time —- Beck moved one of the lines over to obscure his problem.

    My statistics teacher insisted:
    — first you define the statistic to be used,
    — _then_ you collect the data;
    — you must use all the data you collect without losing any that are inconvenient;
    — you should have hired the statistician to tell you your statistical design is going to be of any use _before_, yes, before collecting any data.

    Seems Beck failed to pose a question that could be usefully answered and is just painting lines on the page to claim they prove something.
    And that after chopping out bits, and after pushing them around til bits and pieces of them line up. And ignoring what else changed over the time span.

  10. 310

    [[re 287: while protecting the environment and species is generally a good thing, I don’t think it rises to the criticality or even significance you imply. Species come and go by the thousands over years/decades/centuries. No need to get in a dither over a loss of one or two (or a hundred…) ]]

    The ecosystem is all interconnected and we don’t know what the crucial species are. We can’t say a species is expendable until it goes and we either survive or don’t survive. We do know that we are presently undergoing a mass extinction. That’s not a good thing.

    Maybe we can compensate for the loss of the species we’re wiping out and maybe we can’t, but as Paul Ehrlich put it, “The first rule of intelligent tinkering is to save all the parts.”

  11. 311

    [[You can trash me from now until the end of the millennium if you wish, but as a scientist, I would invite you to “prove us wrong”]]

    We don’t have to prove you wrong. You have to prove your proposal is right. Your challenge is scientifically equivalent to the UFO believer’s “prove that UFOs aren’t alien starships!” It’s not our burden of proof, the burden of proof is on the affirmative.

    Nonetheless, I’ll demonstrate why you’re almost certainly wrong.

    You draw a correlation, using an unnecessarily elaborate function, between two time series, attribute causality from one to the other, and that’s pretty much your whole paper.

    When two time series are increasing at the same time, they may appear to be causally linked when they’re really not. This is known as “the spurious regression problem.” It can occur when time series are “integrated,” meaning that they don’t have a time-stationary mean or variance. Correlations found in that case are meaningless. The rare cases where the two variables in question are really linked can be identified by finding them to be “cointegrated,” i.e., they have a linear combination of the two which is stationary in a statistical sense.

    I see no indication that you applied the standard tests for integration and cointegration to your time series. Not having done that, your statistical correlation is meaningless, especially considering how unnecessarily elaborate your function is. Curve-fitting proves exactly nothing without some kind of theory behind it. I spent a great deal of time as a teenager finding “better” expressions of Bode’s Law, until some people who knew better kindly pointed out to me why the elaborate functions I was drawing up were meaningless.

    Your paper would not have passed peer review at a climatology journal, or a geology journal, or even at an economics journal, economists having been the first to identify the spurious regression problem. You are working out of your field and you made a beginner’s mistake which invalidates your whole paper.

  12. 312
    Rod B says:

    re 307/299: A little off topic, but… the loggers and fishermen losing (a major part of) their line of work and income, the Spanish-Indians of New Mexico losing their ability to gather their wood for fuel (cooking and heating), the western farmers unable to irrigate their fields, or the forest firefighters unable to load water on their helicopters — I suppose you might call that “inconvenience” but I think harm is more appropriate.
    [edit – no DDT discussions here – take it to Deltoid if you must]

  13. 313
    Rod B says:

    re 310: “…The ecosystem is all interconnected and we don’t know what the crucial species are. We can’t say a species is expendable until it goes and we either survive or don’t survive…..”

    That’s reasonable enough. Conceivably there might be “critical” species that we ought to worry about, if only we knew what they were. Nor does general concern over species/environment/etc. bother me as long as it is done in context.

    Shoot! I was almost on your side and then you had to go and quote Paul Ehrlich and blow your cover!! [;-)

  14. 314
    Rick Brown says:

    re 313 Rod B “Shoot! I was almost on your side and then you had to go and quote Paul Ehrlich and blow your cover!! [;-)”

    I don’t suppose it helps that the original quote is from Aldo Leopold: “To keep every cog and wheel is the first precaution of intelligent tinkering.” (A Sand County Almanac, and Sketches Here and There)

  15. 315
    Hank Roberts says:

    Precaution = ….?

    Rod, have you had a basic ecology course yet? At least you should read the Leopold books and the recent studies along the same lines.
    Or rent the movie ‘Never Cry Wolf’ — Mowat also deals a bit with bogus studies set up to decide what species are ‘important’ — ecology is real. Read some.
    Don’t confuse ecologists with nonscientists who don’t know ecology — eschew bogosity wherever it’s found, it’s the true believers who don’t read the science who lose the important bits without realizing they matter.

  16. 316
    dhogaza says:

    re 307/299: A little off topic, but… the loggers and fishermen losing (a major part of) their line of work and income

    Yes, this always seems to happen when they don’t listen to biologists and the resource disappears.

    My guess is that you’re not aware that the old-growth wars in the Pacific Northwest weren’t about allowing to continue harvesting old-growth indefinitely. It wasn’t, it was whether or not to allow them to liquidate the remaining 5% left, which would’ve taken them about 20 years when the fight began, perhaps 10 when it was over.

    You’re a real pave-the-earth kinda guy at heart, eh?

  17. 317
    Michael Peterson says:

    Earth first.

    We’ll pave the other planets, later.

  18. 318
    Rod B says:

    “You’re a real pave-the-earth kinda guy at heart, eh? ”

    No, but stopping development in the NW or AZ because a spotted owl might be somewhere in the vicinity, or doing the same in Austin TX because the micro-ranging cave dwelling pseudo-scorpian (not even a real one at that!) or the little golden-cheeked warbler might want to nest near by is nothing short of stupid. Texas even planned a mass execution of cowbirds to save the GC warbler (eggs).

    There’s nothing wrong with ecology and environmentalism (or anything for that matter) as long as it is kept in mind that there are two ways of doing anything: smart and stupid.

  19. 319
    Hank Roberts says:

    You confuse ecology (the science) with “environmentalism” (whatever that is, it’s not science).
    That’s the point people are trying to make, and it’s where the manipulation of curves and figures by politically bent people to misrepresent scientific information is so very common.

    “Biologically rational decisions may not be politically possible once investment has occurred.

    “… A conservative (precautionary) TAC, leaving a safety margin for natural fluctuations and unanticipated food web interactions, is needed to prevent overfishing and overinvestment. For adaptive management, data on ecosystem status, indicator species’ populations over time, and food web interactions are needed to build quantitative understanding and to inform future management decisions.”

    Illustrating what you’re not yet understanding, that the choice at the present rate of extinction will usually be the third way — the ‘we don’t know what would be smart, but we sure do know what is likely to be stupid” approach.

  20. 320
    Chuck Booth says:

    Re 318 smart vs stupid

    Golden-cheeked Warblers are an endangered species ( Brown-headed Cowbirds, on the other hand, are not endangered – indeed, they have expanded their range in North America ( and parasitize the nests of at least 144 bird species. In other words, they have become a nuisance in many parts of the U.S., so extermination may not be unwarranted, just as deer are often exterminated in areas where their population is out of control.

    I assume your silly remark about an endangered pseudoscorpion in Texas caves was facetious, yes? In point of fact, despite its name, it is another endangered species that has just as much right to exist as any “real” scorpion, which is why it is protected by Federal and state Endangered Species Acts ( ; According to the Texas Parks and Wildlife, “The primary threat to the Tooth Cave pseudoscorpion and other endangered cave invertebrates is the loss of habitat due to urban development. Many caves have been paved over or filled in. Other caves have been altered so that they no longer provide the stable temperatures and high humidity needed to support these animals. Contamination by pollutants is also a threat to their survival.” In the case to which you are probably referring, there are actually a number of rare or endangered cave-dwelling species (Bee Creek Cave Harvestman, Bone Cave Harvestman, Tooth Cave Pseudoscorpion, Tooth Cave Spider, Tooth Cave Ground Beetle and the Krestschmarr Cave Mold Beetle)whose protection (in accordance with Texas and Federal Law) is holding up development of a 215-acre parcel near that city with an estimated property value of $60 million ( I guess if you feel strongly that people have a right to get rich by exploiting the natural environment without concern for the plants and animals that are integral components of natural ecosystems this case seems rather ridiculous. But, a lot of people would disagree with you.

    Sorry for drifting off topic again.

  21. 321
    Rod B says:

    re 319: Hank says, “That’s the point “people” are trying to make”

    I don’t think so, though it may be your point. I pretty much agree what you say here — it clearly smacks of an intelligent approach to ecology and environmentalism (which many/most improperly don’t make much of a distinction…)

  22. 322
    Rod B says:

    re 320: “Cowbirds….have become a nuisance [to endangered species]….so extermination may not be unwarranted,….”

    and this is not upsetting the natural balance because….?

    “…silly remark about…pseudoscorpion…that has just as much right to exist…”

    Yes, I was poking fun at its “pseudo-” name. Relax.
    …and it got its inalienable right from where again????? Or is it just as much right to exist as the developer???

    “…you feel…people have a right to…[exploit] the natural environment without concern for the plants and animals that are integral [to…] natural ecosystems… — is ridiculous…”

    I think people definitely have a right, bestowed by nature, to exploit the environment with reasonable and intelligent concern for the other stuff integral to…and the long-term consequences to ecosystems. Not doing so would be ridiculous.

    I feel this is relevant to the thread, but would take way to much time to make my case.

  23. 323
    Nick Gotts says:

    Re #322 [re 320: “Cowbirds….have become a nuisance [to endangered species]….so extermination may not be unwarranted,….”

    and this is not upsetting the natural balance because….?]

    …there isn’t one. If there ever was (most ecologists would recognise that change is the rule rather than the exception at all temporal scales) there are now practically no ecosystems that have not been fairly fundamentally altered by human action. That doesn’t undermine the grounds for attempting to slow the processes of change, and in particular to protect vulnerable species (there is no doubt whatever that human action has enormously increased the rate of species extinction). The grounds are both practical (we don’t know what use to us a particular species is now, or may be in future), and moral, esthetic and scientific. If you can’t see it’s wrong to reduce the beauty and variety of the world, or the chance to learn more about it, without a very good reason, I doubt I can convince you; I can only promise to oppose you.

  24. 324
    Niemand says:

    Looking forward to your commentary on Hansen & Siddall 2007, Phil. Trans. Royal Soc.
    Not because of the mixed axes labels. What’s new in it? Does it provide new arguments or evidence that Antarctic and/or Greenland ice will melt more and/or faster than IPCC expects? Or is the high uncertainty around that unchanged, so we can hold to the lower IPCC guesstimates for this century?

  25. 325
    Hank Roberts says:

    The full text is online;
    Climate change and trace gases
    Phil. Trans. R. Soc. A (2007) 365, 1925-1954
    doi:10.1098/rsta.2007.2052 Published online 18 May 2007

    On p. 1935 you’ll find your question answered.

    Off topic here, though.

  26. 326

    [[re 320: “Cowbirds….have become a nuisance [to endangered species]….so extermination may not be unwarranted,….”
    and this is not upsetting the natural balance because….?

    Because the expansion of the cowbirds is not their natural population level. That’s why they have become a menace to the endangered species.

  27. 327
    Burkart says:

    Hi Stefan,

    I don’t know if you’ve noticed, but Beck has given the source of his curve. It is in fact a sine curve, regular by definition. Beck argues that this is an acceptable simplification, as the curve was meant for demonstration purposes only.

    Quite frightening, actually, that he attained such media attention in such a short time with such obvious nonsense.

  28. 328
    Hank Roberts says:

    You can look this stuff up, rather than divert the topic. It’d lead you to questioning the wisdom of common practices.
    “Cowbird parasitism probably is not responsible for the continent-wide declines of many North American songbird species…. cowbird control is a short-term solution that ignores the real problem of habitat degradation as a result of agriculture, grazing and development.”

  29. 329
    Nick Gotts says:

    Re #328. is actually the article – the URL given just gives you a small image. Here’s another quote:

    “Scientists caution against diverting limited human and financial resources to cowbirds and neglecting the root causes of why species are at-risk. Scientists, however, support limited control to help restore local populations of Threatened or Endangered species.”

  30. 330
    Chuck Booth says:

    Re 328 Hank, I think the following link has the information to which you were referring:

    I will no longer post on this topic – I promise. I’m much more interested in Beck’s attempts to mislead his followers.

  31. 331
    Mike says:

    What about and its projections of CO2 induced warming?
    What about solar cycle 24? It seems to be some uncertainly yet…

  32. 332
    Chuck Booth says:

    Re 331 Link to conference presentation by David Archibald
    He states:
    “The 100 ppm carbon dioxide increase since the beginning of
    industrialisation has been responsible for an average increase in plant growth rate of 15% odd. [sic] The 50% increase in plant growth rate due to a 300 ppm increase in atmospheric carbon dioxide can be expected about the middle of the next century. What a wonderful time that will be.”

    What a wonderful time, indeed- I can only hope I’ll be around to expericence it!

    By the way, who is David Archibald?

  33. 333
  34. 334
    Jim Eager says:

    Re 331 Mike: “What about and its projections of CO2 induced warming?”

    What about it?
    Read through it and it’s clearly just a rehash of the same old nonsense, complete with graphs similar to Beck’s:
    * medieval warm period was far warmer than today
    * sun spot cycle & solar forcing are driving the current warming
    * speculation that the next two solar cycles will lead to cooling
    * understates the role of CO2
    * no mention what so ever of H2O, methane, albedo change or other amplifiers/feedbacks
    * increased CO2 is good for plants; quote: “What a wonderful time that will be.”

    In other words, Archibald is an Australian Beck.

  35. 335
    Ryan Stephenson says:

    Very naughty. However, it seems that AGW proponents aren’t averse to bending the rules of plotting graphs to get their point across.

    The Wikipedia version:

    Real Climate version:

    Neither of these graphs show the lag of 1000years of the CO2 following the temperature. Nevertheless, it exists. As do a numbger of other anomolies that call in to question the accuracy of the AGW theory. Why is is not shown on any of these graphs? Because some sort of running averager has been applied to the data to smooth out the spikes and the time lag. That, however, doesn’t stop AGW proponents from adding a spike for the present day, despite the fact that the data for the present was collected by different means and with a resolution that is averaged over the periods for the ice core.

    Both sides are playing games with the data, which is why I remain agnostic on the issue. Why should I believe either side, when both sides are proven liars?

    Another ice-age will come and Britain will be wiped off the face of the earth. That much we KNOW. Anything else is just speculation.

    [Response: You are mistaken. Both of those graphs are on the correct timescales (within uncertainties) and if you were to blow them up at the resolution sufficient to see an 800 year lag at Termination II, they would be there. However, you do not immediately perceive it because they are graphs of 650,000 years of results – an 800 year lag is just over 1/1000th of the scale of the graph – less than the width of the lines. What it demonstrates clearly is the minor importance of the lag when looking at signals that are 10,000 year and higher oscillations. PS. Given our understanding of CFCs and their tremendous greenhouse potential, there is very little chance that a new ice age will be allowed to happen while industrial societies exist. To use that (which in any case is not due for another 30 or 50,000 years) as a reason not to be concerned about the human addition of CO2 is to completely miss the point. – gavin]

  36. 336
    Ray Ladbury says:

    Ryan, 650 thousand years are represented in about 8.5 inches if viewed at highest resolution–and you claim you could pick that out of the graph? You remain unconvinced because you remain ignorant. That’s OK. Ignorance is 100% curable. Willful ignorance is unforgivable.
    Look at who has urged action on climate change:

    Now look at the fact that the so-called skeptics are relying on obscure, dishonest highschool teachers as experts. How can you compare the two sides. One is science. The other is pure PR and obfuscation.

  37. 337
    SteveF says:

    r.e. #335

    Being smugly pedantic, Britain is unlikely to be wiped off the map by a coming ice age. Only during the last event (the Devensian) and MIS12 (Anglian) did ice reach far enough to cover most of the country. Anglian ice came furthest south, reaching North London.

    Scotland and Wales would be buggered though.

  38. 338

    I didnt notice that tic mark at first.

    Its ashame how people want to fool the public into thinking that the way we are living right now is just great for the environment. It reminds me of the doctors promoting smoking back in the 70s.

    Lets all just do our part and Stop Global Warming!

  39. 339
    Ryan Stephenson says:

    It is not the point that the graphs are not correct – it is the point that they are misleading at first sight. Even to the point where the vertical scale does not start at zero – hence at first sight the current CO2 levels are 100% higher than previous peaks when in fact it is 30%.

    A less misleading graph of the EPICA data, showing the lags, various points of inflexion in the T curve that are subsequently shown in the CO2 curve, CO2 rising with no observable impact on T, linear increase of T when exponential rise of T would be expected if CO2 were feeding back into the climate, can be found here:-

    [Response: Actually I disagree. The age scales for the Vostok and Epica are not aligned – therefore the apparent difference between T and CO2 is an artifact. Leads and lags can only be determined on equivalent timescales, which due to the uncertainties in the ice age- gas age difference, is actually very hard to do. Caillion et al (the paper from where the ‘800 year’ number comes from) show that process clearly. It is not derivable from a naive visual inspection. Thus your linked graph is fundamentally more misleading. Sorry. – gavin]

  40. 340
    Ryan Stephenson says:

    “Being smugly pedantic, Britain is unlikely to be wiped off the map by a coming ice age[…..] Scotland and Wales would be buggered though”

    Being equally smugly pedantic, Finland struggles to support a population of some 4million people. Britain currently has a population of 60million. We produce 60% of our own food in the UK, the rest we import. Any ideas how we could feed 60million people in a country knee deep in snow that refuses to grow any crops that we have today and where the rest of the world is struggling with the same problems?

    Vast areas of land in continental Europe, Russia and North America will become unproductive in the next ice age.

    Another ice-age will definitely be more difficult to cope with than global warming. Guaranteed. And you only need to look at the EPICA data to see it could happen tomorrow.

  41. 341
    Ryan Stephenson says:

    Really Gavin? That would mean that the temperature data from EPICA was shifted by thousands of years relative to the Vostok data. Look at the peaks between CO2 and T between the two graphs – in the EPICA data they are not coincident but in the Wikipedia Vosto graph they ARE coincident.

    So what you are saying is that the timescale for the two graphs cannot be relied upon to give coincident readings because they are two different sources of data. That would mean an inaccuracy in the case of Vostok vs EPICA at 320000 years of a staggering 10,000 years. Well, if it was that inaccurate it wouldn’t be much help to anyone would it? Do you really believe what you have just told me?

    Well, if you do, it is time to set you straight. You see the graphs from Wikipedia are sourced from data held at the NOAA, and we can take a look at what really happens to the maxima for the two charts. The Vostok deuterium (temperature) data is here:

    And it shows a deuterium (temperature) maxima at 333,000 years

    The Vostok CO2 data is here:

    And it shows a CO2 maxima at 323,000 years. So the Vostok data IS NOT different from the EPICA temperature data. The CO2 spike really is 10,000 years AFTER the T spike. The Wikipedia graph has been massaged so this difference does not show, and is NOT a plot of the raw data, although that is the clear implication by linking the graph to its sources.

    The graph here reflects the raw data more accurately:

    But I can quite understand why proponents of global warming theory would prefer this graph to go away. I notice that the Wikipedia chart has drawn some criticism on the discussion page, which is tainted by the politics of AGW. But the science should stick to the facts.

    [Response: The relevant science is this: timescales on ice cores are uncertain – at least +/- 10,000 years as you go back to 400,000 BC. Therefore independent timescales (as plotted in your favorite graph) cannot be used to resolve leads and lags of under a thousand years. If you plot the Vostok and EPICA deuterium on the same timescales as above, they would indeed be offset by thousands of years which is clearly not correct. So to determine leads and lags you need to align the timescales (using CH4, 10Be, volcanic markers etc.) before you start. Absolute timescales are extremely difficult with this age of ice. – gavin]

  42. 342

    [[Another ice-age will definitely be more difficult to cope with than global warming. Guaranteed. And you only need to look at the EPICA data to see it could happen tomorrow. ]]

    It could not happen tomorrow. Ice ages are governed by the Milankovic cycles of Earth’s eccentricity, axial tilt, and precession. By those cycles, we are indeed descending into another ice age, but not for another 20,000 years. It could NOT happen tomorrow.

    And the global warming so far has probably prevented it from happening again for far longer than that. Remember that ice ages are not a constant feature of Earth history. They only happen in a few eras (the Huronian and Vendian glaciations, and the Pleistocene/Holocene).

  43. 343
    Ryan Stephenson says:

    I cannot find any science paper on the web that points to an error of more than 5000 years. These two maxima are 10,000 years apart.

    An attempt was made recently to claim that the ice in the more recent ice-cores is actually 800 years in error, thus making the curves coincident and disposing of the claim that there was an 800 year lag. But here we have a 10,000 year lag. Now you are telling me that when it suits you there is a 10,000 year error.

    This isn’t science. This is cherry picking data to fit a theory. The error becomes “whatever error correction is needed to get the curve to fit the theory IS the required error correction”.

    How about this for a theory:

    When we were in the middle of an ice age, 2/3rds more of the planet was covered with ice than today. There was little opportunity for CO2 to be present in the atmosphere. Due to changes in Earths orbit, the ice retreated. But not right away, because it takes ENORMOUS amounts of energy to turn ice into water. Thus it was a long time before fresh land was exposed and released methane which broke down to CO2. That delay is what we see in the ice core record. CO2 follows T and there is no evidence of CO2 forcing T

    Can you really argue with that, just based on the ice-core data? I suspect not. Because even from your position, there is so much possibiity of error in the timescales that you really can’t find evindence of ANYTHING from these ice cores. They are meaningless. Thus any attempt to suggest ice-core data is evidence of AGW is false. There isn’t enough accuracy to permit such a conclusion. It doesn’t stop proponents using the ice-core data for their own arguments though.

    Thus I no longer believe in AGW. I used to be a firm believer, but I have become very skeptical about the way proponents of AGW use the data they obtain. They don’t even question the data unless someone attacks what they are saying. That is bad science. And thats before you get to the environmentalists telling me I mustn’t use my car to take my kids to school when domestic car use in the UK contributes just 5% of CO2 emissions.

    Pscychologists have this mantra: “An escalating commitment to a particular (failed) course of action”. This is what I smell here.

    [Response: I’m extremely puzzled as to what point you trying to make. I am not disputing that Milankovitch forcing is the driver for the ice ages and CO2 follows along. I would add that CO2 is a greenhouse gas and therefore feeds back on the temperature changes but, as is often said, correlation doesn’t causation, so that isn’t determinable from the ice core data alone. However, when people talk about leads and lags they are discussing relative dating, not absolute dates. Thus in the work of Caillon et al (2003) – please read it, you’ll learn a lot – they compare records that are both on the gas age timescale so that there aren’t any ambiguities of absolute dating. The error bars they find are still large – a few 100 years, and so the lag is only barely detectable. But none of this is detectable in the data you point too because the absolute errors are much larger than any possible lag. To get a sense for what goes into making the age model, read this recent paper: and for trying to pin down the lag even further read this one: . PS. None of this has any particular relevance for global warming. – gavin]

  44. 344
    Ryan Stephenson says:

    Gavin- it doesn’t have relevance for AGW other than ice core data is used to support global warming theory.

    The errors estimated for the period of interest are 4K years, so that still doesn’t get the peaks coincidentv (but then you have a problem, since most of the chart will happily be coincidnet with only 1000 years of shifting, bearing in mind that Milankovich forcing must be the motivation behind temperature change and CO2 can only follow it or be coincident with it). But even if we played fast and lose with the graphs and shifted the two arbitrarily so the maxima of T and CO2 at 333K years WERE coincident, we would still have a problem – because it would be apparent that the Milankovitch forcing cares not one jot about the extra CO2 in the atmosphere and forces an ice-age regardless of its presence.

    It would be difficult, therefore, to insist that excess CO2 in the atmosphere has any impact on the climate, at least from these ice core charts.

    [Response: Ice age data is indeed used to support GW theory, because you can show that the degree of cooling seen is completely consistent with the standard climate sensitivity, the changes in GHGs (CO2, CH4, N2O) and the change in albedo from ice sheets and vegetation changes. But why do you think that standard theory implies that something else should be seen in the ice core results? The driving forces (Milankovitch) have time scales much longer than the response time of CO2 to climate or vice versa and even the simplest of toy models will tell you that you expect a very strong correlation (as seen), regardless of the value of climate sensitivity to increased CO2. The amplitude of changes however, does depend strongly on the sensitivity and that is what is used to validate ideas on sensitivity (see: for a discussion). – gavin]

  45. 345
    Hank Roberts says:

    Gavin, the latter two 2007 papers are very helpful.
    (Ryan, you understand _why_ the ice core data isn’t immediately relevant? What humans are causing now is not what happened previously.)

    Ryan, to encourage you to read the papers:
    This is from the latter paper Gavin pointed to; the mirror site will have the image for 24 hours.

    See the green line comparing the apparent age found in the two interpretations? Note “today” is at the left side; the more recent data is less variable.

    Ryan, you _really_ ought to read this — if your question is based on your own understanding of the science these articles will improve it.

    If you’re bringing the question here because you’re reading it elsewhere, and someone you trust and believe is saying it’s a good question, tell us where you’re getting it so we can look at _your_ sources, eh?

  46. 346
    David B. Benson says:

    Ryan — The essential fact is that carbon dioxide is an important so-called greenhouse gas. (See, for example, the link on the sidebar to the AIP history of climaltology.) Humans have been adding excess, fossil carbon to the atmosphere for the past 250 years or more. (Read Ruddiman’s popular book about the ‘or more’.) Humans have especially added lots of excess carbon to the atmosphere for the past 50 years.

    Therefore it is too warm and will get worse. Irrespective of the ice records.