Climate Science Glossary

Term Lookup

Enter a term in the search box to find its definition.


Use the controls in the far right panel to increase or decrease the number of terms automatically displayed (or to completely turn that feature off).

Term Lookup


All IPCC definitions taken from Climate Change 2007: The Physical Science Basis. Working Group I Contribution to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Annex I, Glossary, pp. 941-954. Cambridge University Press.

Home Arguments Software Resources Comments The Consensus Project Translations About Support

Twitter Facebook YouTube Mastodon MeWe

RSS Posts RSS Comments Email Subscribe

Climate's changed before
It's the sun
It's not bad
There is no consensus
It's cooling
Models are unreliable
Temp record is unreliable
Animals and plants can adapt
It hasn't warmed since 1998
Antarctica is gaining ice
View All Arguments...

New? Register here
Forgot your password?

Latest Posts


CO2 lags temperature - what does it mean?

What the science says...

Select a level... Basic Intermediate Advanced

CO2 didn't initiate warming from past ice ages but it did amplify the warming.  In fact, about 90% of the global warming followed the CO2 increase.

Climate Myth...

CO2 lags temperature

"An article in Science magazine illustrated that a rise in carbon dioxide did not precede a rise in temperatures, but actually lagged behind temperature rises by 200 to 1000 years.  A rise in carbon dioxide levels could not have caused a rise in temperature if it followed the temperature." (Joe Barton, US House of Representatives (Texas) 1985-2019) - Full Statement

Earth’s climate has varied widely over its history, from ice ages characterised by large ice sheets covering many land areas, to warm periods with no ice at the poles. Several factors have affected past climate change, including solar variability, volcanic activity and changes in the composition of the atmosphere. Data from Antarctic ice cores reveals an interesting story for the past 400,000 years. During this period, CO2 and temperatures are closely correlated, which means they rise and fall together. However, based on Antarctic ice core data, changes in CO2 follow changes in temperatures by about 600 to 1000 years, as illustrated in Figure 1 below. This has led some to conclude that CO2 simply cannot be responsible for current global warming.

Figure 1: Vostok ice core records for carbon dioxide concentration and temperature change.

This statement does not tell the whole story. The initial changes in temperature during this period are explained by changes in the Earth’s orbit around the sun, which affects the amount of seasonal sunlight reaching the Earth’s surface. In the case of warming, the lag between temperature and CO2 is explained as follows: as ocean temperatures rise, oceans release CO2 into the atmosphere. In turn, this release amplifies the warming trend, leading to yet more CO2 being released. In other words, increasing CO2 levels become both the cause and effect of further warming. This positive feedback is necessary to trigger the shifts between glacials and interglacials as the effect of orbital changes is too weak to cause such variation. Additional positive feedbacks which play an important role in this process include other greenhouse gases, and changes in ice sheet cover and vegetation patterns.

A 2012 study by Shakun et al. looked at temperature changes 20,000 years ago (the last glacial-interglacial transition) from around the world and added more detail to our understanding of the CO2-temperature change relationship.  They found that:

  • The Earth's orbital cycles triggered warming in the Arctic approximately 19,000 years ago, causing large amounts of ice to melt, flooding the oceans with fresh water. 
  • This influx of fresh water then disrupted ocean current circulation, in turn causing a seesawing of heat between the hemispheres.
  • The Southern Hemisphere and its oceans warmed first, starting about 18,000 years ago.  As the Southern Ocean warms, the solubility of CO2 in water falls.  This causes the oceans to give up more CO2, releasing it into the atmosphere.

While the orbital cycles triggered the initial warming, overall, more than 90% of the glacial-interglacial warming occurred after that atmospheric CO2 increase (Figure 2).

Shakun Fig 2a 

Figure 2: Average global temperature (blue), Antarctic temperature (red), and atmospheric CO2 concentration (yellow dots).  Source.

Last updated on 21 April 2021 by eckahle. View Archives

Printable Version  |  Offline PDF Version  |  Link to this page

Argument Feedback

Please use this form to let us know about suggested updates to this rebuttal.

Related Arguments

Further reading

That CO2 lags and amplifies temperature was actually predicted in 1990 in a paper The ice-core record: climate sensitivity and future greenhouse warming by Claude Lorius (co-authored by James Hansen):

"Changes in the CO2 and CH4 content have played a significant part in the glacial-interglacial climate changes by amplifying, together with the growth and decay of the Northern Hemisphere ice sheets, the relatively weak orbital forcing"

The paper also notes that orbital changes are one initial cause for ice ages. This was published over a decade before ice core records were accurate enough to confirm a CO2 lag (thanks to John Mashey for the tip).

Also, gotta love this quote from Deltoid in answer to the CO2 lag argument: See also my forthcoming paper: "Chickens do not lay eggs, because they have been observed to hatch from them".

Further viewing

Myth Deconstruction

Related resource: Myth Deconstruction as animated GIF

MD Lag

Please check the related blog post for background information about this graphics resource.


Prev  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  

Comments 626 to 641 out of 641:

  1. Michael sweet this is why jumping multiple pages for a flowing conversation causes issues.  I did respond to your graph and then in this comment section I had it pointed out when the timing for ice ages changed from 41000 years to 100000 years, which reinforces your post and made it more revalant to my discussion again but was on a different page.  I do appreciate the answers you are giving me and am not trying to be rude.

    Looking at your graph yet again, I also see that there does appear to be a very slight cooling trend over the last ~8000 years, and before that a fairly steady temperature rise over about a 4k year period.  This is what lead me to ask why the cooling trend had occurred when that chart appears to be showing temperature rise to be expected, and why it is alarming for the temperature rise to resume when it would be expected for the trend to continue until the buildup of the next iceage ~80000 years from now?

  2. Map,

    The length of the cycle is 100,000. We reached the interglacial stage about 12,000 years ago (consider it the end of the cycle, or beginning, I'm not sure it matters), and what is known as the Holocene climatic optimum is centered around 7-8,000 years ago, depending on proxies.

    Plenty of references on that page for data and publications.

  3. Philippe chantreau thank you, that link will answer many of the questions that this discussion would have led me to ask. 

  4. This has been interetsing, and explains a lot about the confusion in the general public because of prior missives that CO2-Ice core data were proof of a causive link, when it was more likely a result. NASA covers this in the section on Milankovich theory, but fail to actually split the response from CO2, which is easy to do, since the greenhouse gas response is easily modelled with the equation T=3.2563ln(C)-3.0323, where T is the Earth's average temperature in degC and C is the CO2 concentration in ppm. From the ice cores, the change from 190 to 280ppm yields a contribution of 0.95 degC, so minor compared to the overall change of 7-8 degrees.

  5. typo corrections: Milankovitch; interesting

  6. brneilsen @629,
    I'm curious as to the origin of your 'greenhouse gas response' equation T=3.2563ln(C)-3.0323. And if there were any merit in such an equation, I'd be interested to learn how it 'yields' a "0.95 degC" boost to global ice age temperatures resulting from a 190ppm to 280ppm rise in CO2. My abacus (which I would be the first to admit is not always reliable) 'yields' +1.26ºC using this bizarre equation.

    The usual calculation of CO2 forcing is ΔF = 5.35 x ln(CO2[1]/CO2[0]) which gives a forcing of +2.07Wm^-2 and a thus 'direct' impact on global temperature of+0.56ºC which would cause climate feedbacks that would perhaps triple this value to +1.7ºC.
    The global average temperature rise out of an ice age is usually reckoned at +5ºC to +6ºC so this calculated CO2 forcing would perhaps be responsible for a third of this temperature rise. And this result fits with assessments which find the contributions to deglaciation warming to be roughly 50% surface albedo, 37% GHGs (of which CO2 is the major player) and 13% atmospheric albedo.

  7. I don't have much of a problem with CO2 lagging behind temperature at the end of a glacial period. It isn't all that surprising and the explanations are convincing.

    That CO2 still lags behind temperature (though apparently by only a matter of months rather than hundreds of years,"Changes in carbon dioxide content lag those in temperature by five months."[1} is more of a puzzle to me and I havn't seen an explanation for it yet.

    I'm sure there is one that wont require denial of recent anthropogenic climate change, but I don't know what it is.

    This enquiry arose because I'm teaching using an online resource that includes the following question for students
    5. Using Models To Make Predictions

    "Why is there a lag between changes in CO2 levels and temperature?
    (Hint: Remember that there are many reservoirs for carbon dioxide. Where can carbon dioxide be stored when the temperature is low?)"

    I find this question unclear (Do they mean changes in atmospheric CO2, in which case the "hint" doesn't appear to make any sense?) and it implies that CO2 leads temperature change, which wasn't the case in 1990 

    {1} Kuo, C., Lindberg, C. & Thomson, D. Coherence established between atmospheric carbon dioxide and global temperature. Nature 343, 709–714 (1990).


    [BL] Links activated.

    The web software here does not automatically create links. You can do this when posting a comment by selecting the "insert" tab, selecting the text you want to use for the link, and clicking on the icon that looks like a chain link. Add the URL in the dialog box.

  8. Ducked:

    The quote/question about time lags is actually on the page following the one you linked to in your first link (reached by clicking "next"). The two pages do not appear to have different URLs, though, so readers will have to navigate to the next page after following the initial link.

    The quote about 5 months time lag (from the second link, which only gets me to the abstract) appears to come from a study that looks at the past 30 years of data. The article is paywalled (I might be able to get to it via work), but my guess is that if they are looking at only 30 years of data, then they are looking at monthly or more frequent readings.

    Atmospheric CO2 measurements from locations such as Mauna Loa show strong seasonal variation. This short-term variation is driven by different fluxes and reservoirs from long-term changes. This will affect any calculations of time lag, when compared to long-term patterns on the century scale (fossil fuel combustion) or millenial and longer scales (geologic processes). Geological records of temperature and CO2 will not be capable of resolving monthly values, so a 5-month lag in seasonal patterns is not visible.

  9. Ducked @632,

    There is indeed a wobble in  the MLO CO2 record that matches a preceding wobble in the global temperature record of a few months earlier, both of these wobbles matching an even earlier wobble in ENSO. The actual thing wobbling CO2 is rainfall which promotes/reduces CO2 absorption by the likes of the Amazon rain forest. (See map from this NOAA webpage below.) There is quite a literature covering this phenomenon (eg the likes of HERE or HERE) but be warned - there are a few papers by swivel-eyed denialists lurking within the proper work.

    ENSO impact on climate map

  10. article: CO2 lags temperature - what does it mean? LINK
    "This positive feedback is necessary to trigger the shifts between glacials and interglacials as the effect of orbital changes is too weak to cause such variation.
    While the orbital cycles triggered the initial warming, overall, more than 90% of the glacial-interglacial warming occured after that atmospheric CO2 increase"
    - Orbital changes alone are too weak to cause such variation.
    - 90% of the glacial-interglacial warming occured after that atmospheric CO2 increase
    Question: What caused the cooling? Certainly not orbital changes, since they are too weak as pointed out above.
    With CO2 causing 90% of the warming and, at the end of the warming trend the atmosphere having very elevated levels of CO2, we should have a runaway warming trend that is unstoppable, but clearly that was not the case as the next ice age approached.
    There seems to be a serious logical flaw in the argumentation of that article.


    [DB] Shortened and activated link.

  11. Yoshi @635,

    I fear you misinterpret the 90% figure. As described by Skakun et al (2012) (& discussed in this SkS post), the 90% is not the percentage of warming coming out of an ice age that is caused by CO2. It is the percentage when increases in CO2 occur prior to increases in global temperature.

    The actual post-ice-age warming resulting from increased CO2 is a portion of the GHG warming (which also includes methane). The GHG warming is given as 37% of the total in this CarbonBrief explainer. (The remainder is given as 50% ice albedo & 13% dust & aerosols.) The actual CO2 forcing is about 2.5Wm^-2.

    The cooling of the world that leads to a glacial maximum is much slower than the warming of the world that leads to an interglacial. The cooling begins with increased albedo in high northern latitudes as they lose sunlight through the orbital wobbles.

    The warming is quicker because it takes less time to melt down an ice sheet than it does to build it up. As with the warming, CO2 reacts to this cooling and increases the effect.

  12. It redirects focus over to like the greater part of the compelling anticipated from a multiplying of CO2 ought to have as of now occured except that is unrepresented. For this reason, assuming there was a space. Without it, however, input numbers or moderately.

  13. Welcome to Skeptical Science, Fya.

    It is difficult to know what you are responding to here. It helps if you refer to the comment number, or the portion of the post you are responding to.

  14. The time lag between CO2 and temperature is due to the time offset between warming oceans and continued ocean CO2 emissions. with this cumulative effect Carbon dioxide, therefore it will becomes the main driver of temperature during glacial and interglacial warming. Shakun et al 2012 paper showed that warming was indeed triggered by the Milankovitch cycles and that a small amount of orbital cycle-caused warming eventually triggered the CO2 release, which caused most of the glacial-interglacial warming. So while CO2 did lag behind a small initial temperature change, it led and was the primary driver behind most of the glacial-interglacial warming. According to the Shakun data, approximately 7% of the overall glacial-interglacial global temperature increase occurred before the CO2 rise, whereas 93% of the global warming followed the CO2 increase.

  15. Figure 2a in the Shakun paper seems to confirm that the very minor warming due to higher CO2 after the peak temperature is reached following deglaciation is substantially smaller than the effects of the Milankovitch cycles, which clearly dominate earth's climate. Is that not apparent from the graph?

  16. The other thing apparent from that graph is that temperature climb rate early in the post-glacial period seriously flattens at the peak. That seems to imply a variation in sensitivity that is not mentioned in the Shakun paper. If that were not the case, the temperature would continue to climb at the faster rate.

Prev  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  

Post a Comment

Political, off-topic or ad hominem comments will be deleted. Comments Policy...

You need to be logged in to post a comment. Login via the left margin or if you're new, register here.

Link to this page

The Consensus Project Website


(free to republish)

© Copyright 2023 John Cook
Home | Translations | About Us | Privacy | Contact Us