Why Did CO2 Lag Temperature During Ice Ages?

Another favourite argument of global warming sceptics is that increases of carbon dioxide (CO2) lagged increases of temperature during the ice ages, so CO2 can’t be the cause of global warming.  At first sight this seems like a very reasonable argument.  However climate scientists are not as stupid as some make them out to be.

Climate scientists have not been arguing that the ice ages were driven just by CO2. They argue that the triggers for ice age fluctuations are small changes in the heat from the sun, due to small changes in the Earth’s orbit, called Milankovic cycles.  By themselves, the small changes in solar heating would not cause very large changes in temperature.  However CO2 is released into the atmosphere from the oceans and from soils as the Earth warms.  The greenhouse effect of this extra CO2 then adds to the increase in solar heating and amplifies it – a lot.

In other words, CO2 was not the driver of the ice ages, but it was an important amplifier.  This does not mean CO2 could not also drive global warming, if the level of CO2 in the atmosphere were to change for some other reason – such as from people burning fossil fuels.

This has been understood by climate scientists for some time, but that understanding tends to be buried in complex scientific papers about complicated climate models.  Recently my colleague Andy Hogg at ANU published some simplified modelling that left out a lot of the detail in order to highlight the amplifying effect of CO2.  He also did a calculation (unpublished) of the present situation, i.e. CO2 increasing for an extraneous reason.  The calculations show CO2 can be both the follower in the ice ages and the driver in the present global warming.  There is no contradiction.

Therefore the fact that CO2 lagged temperature in the ice ages does not “prove” climate scientists are wrong and people are not causing global warming.  On the contrary, it shows that our understanding of the effect of CO2 can explain its role during the ice ages as well as its present role in global warming.

That is the story in words.  Now I’ll show it in graphs, for those who would like to see it in more detail.  You’ll need to follow carefully, there are quite a few details.

Ice Age Calculation

Here is a graph of changes of temperature (blue) and CO2 (red).  There have been about 6 ice ages over the past 600,000 years.  Typically an ice age comes on slowly, with a lot of fluctuations, but it ends “suddenly”, in only a thousand years or so (much more quickly than the 50,000 – 100,000 years it took to develop).


Ice age temperature and CO2

You can see that CO2 also went up and down in the same way and around the same times – slowly down, with fluctuations, and then suddenly up.  You can barely see the contentious lag in this plot.  If you look closely around 130,000 years ago you can see the CO2 seems to lag the temperature rise a little bit.  You have to be careful though, because the vertical scales on the two superimposed plots are a bit arbitrary.  You have to use statistical correlation methods to do a rigorous comparison.  Such careful comparisons show that the rise in CO2 tends to lag the rise in temperature by as much as 800 years (a very small amount on the scale of this plot).

Now here is Andy’s calculation, replicating the main features of the observations.  Both the temperature (black) and the CO2 (red) go through the same kind of variations – a slow, fluctuating decline and a sudden rise.  There aren’t as many wiggles in his plot because he’s left out minor details to make the main pattern clearer.  (The plot is from A. M. Hogg, Glacial cycles and carbon dioxide, a conceptual model, Geophysical Research Letters, Vol. 35, doi:10.1029/2007GL032071, 5 pp, 2008)

Hogg Fig 2 cut

Calculated ice age variations

(The scale on Andy’s plot goes from 0 to 500,000 years because he’s calculating forward in time. The scale on the observations, above, goes from 450,000 years ago up to the present – in other words it shows years before now, or age.)

The blue curve is the solar heating.  You can see that mostly the temperature isn’t affected very much.  However the big jumps in temperature always occur when the solar heating is rising.  This occurs when a temperature rise happens to trigger the CO2 feedback.

In the bottom panel the changes between 405,000 years and 413,000 years are shown on an expanded scale.  You may be able to see that the CO2 goes up just slightly later than the temperature.  This is as it should be, because in this calculation the initial rise in temperature is driven by the increase in solar heating (blue), which starts well before the temperature actually starts to go up.  The temperature rise takes less than a thousand years.

This calculation was done in order to understand the ice ages better.  (The point was to highlight how the “asymmetry” of the temperature fluctuations (meaning the slow declines and sudden rises) can be explained by the strong amplifying effect of CO2)

Fossil Fuel Calculation

Now let’s look at a calculation that simulates the present situation.  In the plot below, the CO2 was suddenly increased at about 284,000 years, which is near the end of a simulated ice age, as the temperature is slowly rising.  In the top plot, the CO2 (red) spikes up to double its value (from about 280 ppm to about 560 ppm, a good simulation of what we are doing now).  The CO2 drops back quickly and then declines more slowly over the next 50,000 years.

Hogg global warming

Simulated global warming

The whole thing is clearer in the expanded scale of the lower panel.  There you see the CO2 rise rapidly within a few hundred years, then drop back over the next thousand years or so.  In response, the temperature rises by more than 2 degrees, but it does so only after the CO2 has jumped.  Thus in this case the CO2 leads and the temperature follows.  However essentially the same physics is operating as in the ice age simulation shown first.

The Lessons

People who claim to be serious sceptics need to do their homework.  This example is a little more complicated than interpreting the recent global temperature record (see earlier post), and there are traps for the unwary.  When you look carefully at what climate scientists have been saying (ice ages are triggered by solar heating, not by CO2) and what they have come to understand (CO2 can still cause global warming), you see there is no contradiction.

Miranda Devine quotes Ralph Alexander, scientist and author of Global Warming, False Alarm. Alexander has not done his homework.

Miranda Devine is not interested in what scientists really understand, she just wants to rouse the rabble, which she does very effectively.  I don’t know if she believes what she claims, but I do know her proprietor makes money from her rabble rousing.  I also know she takes a highly partisan position in a complicated subject in which she has no qualifications at all.  Why does she choose to believe a minority of scientists rather than the majority?  It’s not because she understands the science.

18 Jan 10:  Here’s an excellent YouTube link that explains it all very clearly, including taking you back to original sources.


12 thoughts on “Why Did CO2 Lag Temperature During Ice Ages?

  1. Ralph Alexander

    You’re quite wrong – I have indeed done my homework, and lots of it. This is all covered in Chapter 5 of my book (pages 67 to 71). The big problem comes at the end of the global warming spell following an ice age. There’s too much to repeat from the book as a comment here, but I suggest you buy it and read what I’ve said.


  2. Andy Hogg

    Ralph, of course, we can’t comment on your “big” problem unless you let us know what it is. But Geoff’s point is that Devine quotes you as saying: “Either temperature and CO2 go up and down at the same time or they don’t … You can’t have it one way during the ice ages and another way today.” The model which produces the above data (and this is a transparently simple, 2-equation model) shows that, in fact, depending upon the situation you can indeed have CO2 either lagging or leading global temperature.


  3. omni

    I am a poet. A poem sent to you all with love. May climate change be wrong would be my wish, yet in my heart i know.


    Gaia is turning,
    The world is learning,
    A sacred balance,
    That is not cost accounting,
    But accounting for our actions,
    For the summit is not a talk fest of hot air,
    But the summit of human potential,
    That is the potential of humanity to care,
    For itself.

    For one must not dam any Franklin,
    For wild rivers are free,
    As the flow of truth can not be diverted,
    By self interest and denial,
    As the water course changes direction,
    Meandering becomes a dry creek bed,
    For we have made our bed,
    Do we sleep in it?
    Or do we awaken to a new earth.

    Ocean outfall or ocean current?
    Mount Sinai or Mountain shadow?
    River cruises or refreshed water ways?
    Natural resources or natural bounty?
    Carbon dioxide or free oxygen?
    A carbon sink or rising atmospheric tempers?
    Is carbon clean or pristine?
    What is our Cole inquiry into this question?
    Is it a natural formation or corrupt in-formation?
    A renewable resource or ancient shroud?

    The sedimentary rock of beliefs is under pressure,
    As plant deposits submerged in acidic waters,
    A natural antiseptic combating microorganisms is a natural healer,
    A pristine coal protecting against oxidization,
    Containing the imbalance of atmospheric carbon,
    Serves a higher purpose than burning the fossil record,
    For this is unprecedented as the CO2 graph ventures into unchartered waters,
    For the new world order is in disarray,
    And perhaps this is the first rays of a new world out of control.

    It is the squeaky wheel that requires no oil,
    This is the thorn that is immovable,
    For truth and action cannot be censored,
    And we are sensing the times where all must speak up and be counted on,
    For the world is in dire straits,
    This is a Bermuda triangle where most have lost direction,
    Yet when one finds magnetic poles have shifted in 2012,
    There is a monumental shift,
    For consciousness is a rising see level,
    That’s line of sight is the edge of the cliff,
    For we are close to midnight,
    There is no midnight oil to fabricate light in the darkness of ignorance as buying time.

    We must be the light we wish to see in the world,
    For Gandhi envisaged self-sufficiency, self-reliance and independence,
    Truth and love set the navigation points,
    The spinning wheel as the inner and outer wheel of life,
    A karmic wheel of good fortune,
    For it spins a new dream weaver,
    A kinetic energy of perpetual e-motion towards self service and self love.

    The spinning wheel threads a society together into one social fabric,
    Enriching natural resources as a talent quest,
    Educating children in simplicity, purity, self development and self respect,
    For this is the social change that yields no results only progress,
    In a school with no wall or winners,
    For to work and play together is a given,
    In an enlightened society that values best interest over self interest,
    And from this web of life a new symmetry manifests its destiny,
    As the love of your life
    Embracing all life as your own true nature


  4. Geoff Davies Post author


    I’ve been away over our weekend.

    What am I wrong about? That you didn’t do your homework, or that that CO2 can lead or lag, depending on the circumstances.

    If the former, it’s not enough to do a lot of stuff if you miss the important issues. Lomborg and Ian Plimer, for example, cite lots of references, but they still fail to come to grips with the key points.


  5. Ralph Alexander


    I haven’t checked this site again until now. But to respond to your post, I agree with you and Andy Hogg that CO2 can either lead or lag. What you’re wrong about is that I really did do my homework.

    The “big problem” I referred to comes at the end of ice ages. Yes, the slight initial warming from the Milankovic effect could not have continued unless the temperature rise was amplified by positive feedbacks, including CO2 feedback. But my contention is that when the ice age was finally over, the temperature suddenly stopped going up and leveled out, yet CO2 continued to rise for another 600 or 800 years before plateauing.

    This behavior is well-documented in the historical record and represents a major stumbling block for the CO2 hypothesis of global warming. How can rising CO2 in the atmosphere be the cause of escalating temperatures today, but not at ice-age terminations, when the mercury stood still as the CO2 level kept ascending?

    Andy’s model calculations, I recognize, show something different. Which is that rising CO2 flattens out first at the end of an ice age, followed by the temperature. In fact, his graph above appears to show that CO2 leads temperature even at the onset of post ice-age warming (contrary to your statement in the blog)- just as it does in his modern-day fossil fuel calculation.

    However, to me all this is moot, since we’re talking about a MODEL anyway. I have a whole chapter in my book devoted to the shortcomings of computer climate models for the present day, and do not believe that models of the paleoclimate are any better.

    And Julie, no I’m not a geometer but a physicist. My professional work includes researching and writing reports on complex technical issues (such as European chemical legislation).


  6. Geoff Davies Post author


    I’ve been away, and then there was Christmas …

    You raise a valid question, but it is not one that automatically negates CO2 as the main present agent of global warming. It would if CO2 were claimed to be the only influence, but climate scientists have not been arguing that. Until a good understanding is found, the question doesn’t argue either way.

    As most people on either side of this debate seem to agree, there are many influences, large or small, and they may differ in their importance in different situations. For example, in the ice ages, solar heating seems initially to dominate, but CO2 then magnifies the effect greatly. I don’t know the answer to your point about the end of warming, and it deserves to be investigated. Perhaps clouds became more important, for example.

    Also, the evidence for CO2 as a primary present agent rests on much more than computer models. Paleoclimatic evidence shows clear associations, and Andy’s analysis implicates CO2 in the highly asymmetric temperature pattern through the ice-age cycles – a major observation that would otherwise be unexplained.

    Finally, the graph may not be clear that CO2 lags, but it is clear in the way the calculation is set up that insolation is the initial driver. I would agree the CO2 lag is not as pronounced as in the observations. This is a highly simplified model meant to illustrate an important aspect of the problem, not to be the last word in realism.

    As I have noted in my post For Global Warming Sceptics, there is of course much detail that is still being debated. That does not alter the majority judgement that human-generated emissions are the main agent of the present global warming. Until such time as someone demonstrates an alternative that persuades substantial numbers of climate scientists, that remains the best guide to policy.


  7. Ralph Alexander


    Well, I think we should agree to disagree.

    You say that the question I raise about the behavior of CO2 during past global warming periods (at the end of ice ages) doesn’t automatically mean CO2 isn’t causing present global warming. That may be so – but the ice age problem, taken together with other problems for the CO2 hypothesis, casts serious doubt on the whole CO2 theory of global warming.

    In my book I refer to these problems as red flags. Briefly, the three red flags are:

    1. The CO2 steady level problem. By this I mean that if CO2 is truly the dominant source of global warming, then past temperatures should mimic the CO2 level over long periods of time, both in and out of ice ages. The problem is that temperature does NOT track CO2 for thousands of years, starting at the end of the last ice age and lasting until about 1850. Although the CO2 concentration was rock steady over this interval, the temperature was all over the place. As I’m sure you know, this mismatch is what led to the infamous and erroneous hockey stick graph – in an attempt to make temperature and CO2 records conform during that inconvenient period.

    2. The CO2 amplification problem. While extra CO2 in the atmosphere does indeed cause the temperature to rise, the effect is very small unless an amplification mechanism is at work. You state above that “CO2…magnifies the effect greatly”, but there is no good observational evidence for this. The truth is that we don’t know how big or small the amplification is, or even if it’s an amplification and not a diminution. Recent satellite data suggest it’s the latter (negative feedback from clouds) and not the former that dominates.

    3. The CO2 lag problem, which we’ve been discussing. As far as I’m concerned, there’s no particularly strong evidence that CO2 feedback was operating at all in glacial times. The rising CO2 level at the end of an ice age could simply be due to the oceans releasing CO2 as they warmed, and the feedback required for amplification could have come from other sources such as snow and ice. If it was CO2, then why was the temperature descent at ice age onset so much slower than its rapid climb at the end?
    You say that Andy’s analysis implicates CO2 in the asymmetric ice-age temperature pattern, but this just tells me that CO2 prolonged the ice ages (by holding the temperature down for so long)!

    And that’s just on CO2. There are plenty of other problems discussed in my book that reinforce the case against present global warming being man-made. For example, data manipulation and other deceptions (as revealed in the recent Climategate emails),
    computer model shortcomings, and the obsession with CO2 to the exclusion of a host of possible natural causes.

    It may be the “majority judgement” of climate scientists that humans are responsible for warming, but I have two things to say. First, many of those climate scientists have been unduly influenced by the likes of Phil Jones and Michael Mann, whose integrity is now in question. And second, science is not a democracy. As Einstein liked to say, a single contrary observation is enough to bring down a whole theory.

    Therefore, policy based on the CO2 theory is very shaky to say the least. I’m all for trying to mitigate the effects of climate change, though not by spending trillions of dollars trying to deal with what may well be an imaginary problem with CO2.

    By the way, I was once at the ANU (as a postdoc in nuclear physics). That’s when the world was equally obsessed with global cooling.


  8. Geoff Davies Post author


    Yes, I think we’re at the point of agreeing to disagree, but there are three things I want to clarify.

    First and most fundamentally, I invoked the “majority judgement” in the context of policy, not as a way to “decide” the science. This is a crucial distinction that much contrarian writing confuses or confounds. See several of my posts for more on this. I wish you would grasp this distinction and not mis-portray the role of IPCC in assessing the majority view, as is so often done.

    Second, Andy’s model was specifically intended to address the claim made by sceptics that the CO2 lag during the ice ages invalidates CO2 as a major driver of the present global warming. It shows that claim is false – CO2 can play different roles in different contexts. It was not intended to “prove” that the CO2 feedback was the mechanism, but the success of even such a simple model in explaining the asymmetry during ice ages certainly counts as evidence in its favour. You say you want more evidence of CO2 feedback, but the evidence is there that CO2 did rise (plausibly from warming oceans, warming soils etc). You could read his paper to see why the decline is so slow.

    If you say “It’s only a model”, well science does not deal in “proof”, it only deals in models, be they conceptual and qualitative or highly detailed and quantitative. The criterion is whether the model (/hypothesis/theory) is a useful guide to observations of the world. See other posts on this site. Of course more observational tests are always welcome.

    Third, if done sensibly the reduction of CO2 emissions could bring many other benefits as well, and not cost a great deal. See my post “Cut carbon emissions unilaterally, and boost the economy”. Of course most politicians, under the influence of fossil fuel lobbies, are parroting that it will be hideously expensive, as a way to scare us all into doing nothing. This is the second great and (still) hidden issue of this debate. You could inform yourself on this, instead of just making throw-away comments like “spending trillions of dollars”.

    Finally, to reiterate the central point, Andy’s demonstration is an example of how there need NOT be a one-to-one correspondence between CO2 and temperature. You go astray at the beginning (your point 1 above) by assuming

    “if CO2 is truly the dominant source of global warming, then past temperatures should mimic the CO2 level over long periods of time, both in and out of ice ages”.

    That is a simplistic straw man. Climate scientists have been investigating the many influences on temperature, and the relative importance of the various influences can be different in different circumstances. Too bad you’re not taking an equally open-minded view.

    It is clear you are sceptical/hostile to the majority climate view, and it’s fine if you try to disprove that view, but I think on this issue you fall short. Your assessment of Phil Jones and Michael Mann is symptomatic of hostility in my view. It should be clear by now to readers where I disagree with your more detailed statements.


  9. Ralph Alexander


    This will be my last post (although of course you’re welcome to respond) – but I’ve enjoyed the debate!

    Without getting into detail again, let me just say this about CO2 and global warming. I disagree that my previous assertion about temperature not tracking CO2 is a straw man argument. You say, in discussing influences on global temperatures, that

    “…the relative importance of the various influences can be different in different circumstances.”

    I would agree with that, but if indeed this is true, then we can infer very little about the present day climate from the past. Paleoclimatology can teach us things about similar circumstances today, but beyond that you enter the realm of speculation.

    On the cost of cutting CO2 emissions, my comment about trillions of dollars is hardly a “throw-away comment”, as you put it. The trillions estimate (over 8 years)as quoted in my book came from Congressional staff briefed by the White House on President Obama’s intended cap-and-trade plan. Your own ideas on CO2 and the economy are interesting, but I’m not convinced they’re all practical.

    In the US, the most impartial financial analyses of proposed legislation come from the Congressional Budget Office, which has repeatedly estimated that cutting CO2 emissions will cost the average US household several percent of its annual income – not a big deal for the well-off, but a burden for the poor and those on fixed incomes. For poorer nations, the cost becomes a significant issue as we all know.

    And like others on your side of the debate, you link opposition to regulating CO2 with the fossil fuel lobby. What you may not know is that today it is the fossil fuel suppliers and utility companies who are backing cap-and-trade systems and other measures, since as European experience has shown, it is they stand to profit the most from such schemes.

    Lastly, I don’t feel hostile toward Phil Jones, Michael Mann etc, but I do feel strongly that their approach to the global warming issue has tarnished the reputation of science. If you read just a few hundred of the hacked emails in detail, you’ll find it hard not to come to the same conclusion. It’s a conclusion reached by quite a few commentators on your side as well.

    And I don’t apologize for making the IPCC out to be the villain. While most people who contribute to IPCC reports are well-meaning climate researchers genuinely trying to piece together a coherent picture of the Earth’s climate system, the reality is that only a small number (maybe 60)actually assess the science and formulate recommendations. What this means is that the IPCC is dominated by scientists and bureaucrats who manipulate the data and the reports for their own ends.


  10. Geoff Davies Post author


    OK, three brief points.

    Just because there are multiple possible influences on climate doesn’t mean there’s nothing to learn from the past. That’s exactly what all the climate science (paleo and present) is trying to unravel. This brings us back to our disagreement about assessing the state of the evidence I think.

    Regarding fossil fuel lobbies, they are backing so-called cap-and-trade schemes with such huge loopholes, and subsidies to themselves, they’ll have very little effect on fossil fuel sales. That’s why the proposed cap-and-trade schemes are turning out so piss-weak they’re worse than nothing, because they’ll just entrench our addiction for longer.

    Many scientists, including me, feel the IPCC was unduly cautious and too influenced by political vetos from the US, China, et al. So I agree it’s imperfect, but by being too conservative, not too “alarmist”.


  11. Pingback: Climate science and Hansen misrepresented – aerosols disguise real global warming | Better Nature: commentary by Geoff Davies

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s