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 Donate

Twitter Facebook YouTube Pinterest

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...

Keep me logged in
New? Register here
Forgot your password?

Latest Posts


Climate Hustle

CO2 Currently Rising Faster Than The PETM Extinction Event

Posted on 17 June 2011 by Rob Painting

The Paleocene-Eocene Thermal Maximum (PETM) was a period of natural global warming that took place almost 56 million years ago.  It came at a time when the atmospheric concentration of CO2 was already higher than today, and global temperatures also much warmer. The PETM warming was a roughly 200,000-year long event where global temperatures rose by a further 6–8°C, and is thought to have been caused by a massive injection of CO2 into the atmosphere.

The PETM: A great carbon mystery 

The existence of the PETM was first noticed when analysis of marine sediment cores, retrieved near Antarctica, revealed an abrupt change in the ratios of carbon-12 & carbon-13 isotopes. Land and marine plants discriminate against the heavier carbon-13 isotope when they are photosynthesizing, and because the sediments are depleted in that isotope, the carbon isotope ratios indicate that the CO2 was organic in origin. This abrupt PETM change is shown in the right-hand dip (@55.5 million years ago) in Figure 1 below: 

Figure 1  PETM changes in the carbon isotopic composition of carbonate. The upper records are from Maud Rise in the southern Atlantic Ocean (Kennett and Stott 1991). The lower record is from the South Island, New Zealand (Nicolo 2007). BFEE = benthic foraminifera extinction events discussed in Nicolo (2007). Figure from (Dickens 2009)

A truly colossal amount of CO2 must have found its way into the atmosphere to create the observed fall in carbon-13 isotopes (called a negative carbon isotope excursion, or CIE, in the literature). Since the initial discovery, the CIE has been found in paleo records from around the world (the smaller CIE's in figure 1, show abrupt warming periods during the Eocene).  

The PETM pulse of CO2 has been linked with acidification of the deep ocean and the extinction of tiny marine life called forams (foraminifera), and proved to be a difficult time for coral reefs. It was also a time of rapid change in land plants and animals, with a quick turnover of species and large migrations, although extinctions were limited.

No smoking gun 

Hundreds of scientific papers have been published on the PETM, but because of the scarcity of paleo-data from this time, there has been no clear scientific agreement over what initiated this warming, or where all the CO2 came from.  

A number of researchers have converged on methane clathrates in deep sea sediments as a possible culprit. Methane clathrates are molecules of methane frozen in a cage of water molecules, which are buried in sediments at the bottom of the ocean. If heated, or de-pressurized, they can quickly break down (oxidize) to CO2, either in the ocean or in the air. An attraction of methane hydrates as the source of CO2, is that methane is highly depleted of carbon-13, and therefore much less of it (than CO2) need be released to match the carbon isotopes shifts observed.

A number of other sources have also been examined too, such as volcanic activity disrupting organic sediments, and methane from permafrost on Antarctica (when it was much warmer), but the case is very much still open.

Abrupt is how long?

Examination of the sedimentary records shows an abrupt increase in CO2, followed by a rapid drawdown (perhaps a rise in silicate weathering), then a long tail of recovery as the PETM warming interval comes to an end. The deep sea cores are affected by the ocean acidification event, which dissolved carbonates in the sediment cores, so it has been difficult to tie down exactly how long this release of CO2 lasted. Previous estimates have been between 10,000–20,000 years.

A recently published study, Cui 2011, indicates that this splurge of CO2 was at the longer end of estimates - around 19,000 years. Cui and co-authors loooked at a sediment core that had been drilled at Spitsbergen, Norway, in a shallow coastal marine environment. Unlike other cores which can be highly condensed, the section from Spitsbergen, covering the initial PETM warming, is some 70 metres long, which provided a more detailed analysis.

Cui 2011 fed the data, collected from the sediment core, into a bare-bones climate model (GENIE), and then ran the model under two scenarios: one with organic carbon as the source, and one with methane, to find a best fit with the core data. They found that the maximum rates of CO2 release that match the PETM data are much lower than present rates observed today from fossil fuel burning.

Figure 2  Model results of the PETM carbon release rate and cumulative amount of carbon added versus time from the onset of the CIE  (age model is from Charles 2011). a. carbon-13 atm that we used to force GENIE. b. model results of the PETM carbon release rate. c. Model results of the cumulative amount of carbon added. d. Model results of the PETM atmospheric pCO2. e. Model results of the PETM global average temperature (°C). The two best-fit simulations are shown in b–e: (1) Methane (CH4) simulation (black solid line); (2) Organic carbon simulation (red dotted line).

The authors find that the maximum PETM rate of emission for organic carbon as the source is equivalent to 6.2 billion tonnes of CO2 per year, and for methane as the source, 1.1 billion tonnes of CO2 per year. For comparison: 2010 human-carbon emissions were 30.6 billion tonnes. So if organic carbon was the source, current emissions are almost 5 times faster than the PETM, and if methane, current emissions are rising 27 times faster.  

With the available data, it wasn't possible for the authors to determine which scenario was the more likely; however, many lines of research indicate that the massive pulse of CO2 was likely to have come from multiple sources. Therefore the above rates can be thought of as the most likely range of values. 

Lessons from a previous global warming event

The PETM took place at a point in Earth's development when the climate was very different than today. It's important to stress that none of the preceding discussion implies that direct and complete comparisons can be made between the Earth climate of today, and the Earth climate of 56 million years ago. Much has changed since then, such as the layout of the continents, and the development of major mountain chains such at the Himalayas and Andes, the growth of major ice sheets, major cooling of the deep ocean and the poles, slight warming of the Sun, and changing Earth-Sun orbital characteristics, all of which greatly alter global circulations and therefore climate. 

But now that we humans have embarked on a global warming experiment, there are some useful lessons from the past:

  • Ocean acidification (of the deep sea at least) can occur even under conditions of CO2 release much slower than today.

Whether the plants and animals upon which humans depend can survive the present rapidly changing environment remains to be seen.

Recommended reading: Sks post - Wakening The Kraken 

0 0

Bookmark and Share Printable Version  |  Link to this page


1  2  Next

Comments 1 to 50 out of 65:

  1. Nice post, Rob!

    Interested readers may want to follow on over to the Wakening the Kraken thread for even more information.
    0 0
  2. lovely! Just lovely.
    0 0
  3. Great read, Rob. And good link, Daniel. All pretty worrying, to put it mildly.
    0 0
  4. It is very frightening, I wonder why the American media ignores this kind of information, but gladly posts the information about a possible sunspot disappearance.

    When one thinks we could inject the atmosphere with as much C02 and Ch4 in 100 years, as was done 56 million years ago in 20,000 years, it makes me realize how truly bankrupt this consumption driven capitalist society has become, when no one seems to see the dire situation we are in.
    0 0
  5. Serendipity at work!

    The July issue of Scientific America features an article by Lee R Krump...

    "The Last Great Global Warming"

    Surprising new evidence suggests the pace of the earth's most abrupt prehistoric warm-up paled in comparison to what we face today. The episode has lessons for our future.
    0 0
  6. Hooray!

    I finally posted a hyperlink that works!
    0 0
  7. DB @ 1 - an oversight, now fixed. Thanks.
    0 0
  8. Hej,

    I guess my first thought is ... wow ... it's great to see people interested in the PETM. It's a truly wonderful topic of current scientific interest, in part because we still cannot explain the event. How can Earth's surface warm and receive massive amounts of carbon in a geological instant 55 million years ago?

    The event was discovered 20 years ago; a wide array of data across the event shows basic signatures predicted by climate models for our future; it defies a satisfactory explanation within any accepted model for global climate or carbon cycling. Welcome to a really interesting puzzle!

    Thought might be good to pass along the text to Figure 1 (and the background to the PETM):

    0 0
  9. Jerryd- a hyper-link to the full Dickens focus article is provided under Figure 1. Are you he?
    0 0
  10. PETM "explained" at
    0 0
  11. when you say "The rapid pulse of PETM CO2 followed by rapid warming (figure 2e) indicates high climate sensitivity. ", what is the delay between CO2 and temperature ? I thought that more recent glacial-interglacial transitions showed that CO2 lagged the temperature variations. Can it not be explained the other way round?
    0 0
  12. The glacial-interglacial shifts are caused by Milankovitch Cycles which isn't the case with the PETM. Nor were there any major ice sheets back then. What exactly are you proposing?
    0 0
  13. Hi Rob,

    An excellent article, but I would criticise the repeated use of "CO2" in the first few paragraphs.

    As you go on to state later on, it is very possible/quite probable that the PETM was caused by methane (CH4), not CO2.

    Looking at the records of atmospheric composition, it would seem that CO2 has tended to undulate up and down, whereas the level of CH4 looks like a saw blade, with sudden spikes which then gradually fall back towards normal.

    I would speculate that this is an indication that, on many separate occasions, events such as the Storegga landslide of c.6100 BCE have occured; i.e. a submerged gasfield has collapsed and vented directly into the atmosphere.

    Methane in the atmosphere acting as a greenhouse gas, this should cause further warming, destabilising the methane clathrate layer, and causing further warming from this.

    Furthermore, the methane clathrate layer is solid; as it decomposes, it will liquefy, and possibly allow the uncapping of further gaseous deposits of methane.

    Unless we can find evidence that the dinosaur car use was about 10% of current human car use, I'm not sure we have any other good explanation of the CO2 spike at the PETM.

    In conclusion, I stongly suspect that the CO2 spike is actually decayed methane. I should round off though, by saying that this is largely based on a layman's (my own) imperfect understanding.

    This may well be quite wrong; but if it is right then your first several paragraphs make too many references to CO2, IMO.
    0 0
  14. okatiniko - different carbon characteristics. d13C ratio is way depressed for PETM.
    0 0
  15. @idunno: CH4 reacts rapidly to CO2 (oxidation with OH). Possibly the oxidation happend already in the sea water or later in the atmosphere (CH4 has a short lifetime compared to CO2). The CH4/CO2 differentation is important to explain the negative carbon isotope excurison. CH4 has an extremly negative deltaC13 ratio of about -50 per mille, therefore you need less CH4 than CO2 to explain the excursion.
    0 0
  16. There is another temperature spike in the middle of the Eocene around 44 Mya. Has this been studied as much as the PETM? Any information on it?


    0 0
  17. Tony - they are the Eocene hyperthermals (although I don't see where you get 44 mya in the figures above). They seem to be driven by orbital changes . See the hyper-link above for Dickens 2009 and also:

    Sexton 2011

    Galeotti 2010
    0 0
  18. Rob

    Thanks. I was not referring to the figure in your post. The 44 Mya spike reference comes form a paper by Hansen and Sato [1] and I don't know if it has been published yet, but Hansen has used similar figures in several of his papers. All of Hansen's stuff is available on his web site. I appologize for making you do extra work but I assume you are more familiar with Hansen's work than I am. As I look at the figure, the spike may be closer to 42 Mya. Anyway, this more recent spike actually looks stronger than PETM and I'd always been curious about it. Again appologies for the confusion.


    [1] James E. Hansen and Makiko Sato, Paleoclimate Implications for Human-Made Climate Change, NASA Goddard Institute for Space Studies and Columbia University Earth Institute, New York
    0 0
  19. Tony, that may be the Middle Eocene Climatic Optimum. Not something I've read much about.
    0 0
  20. No Ice then either and having a white ocean go black is surely going to increase the climate sensitivity a little further.

    From Pliocene data 350ppm by 2100 means ~2C by 2100 and to get to 350ppm by 2100 means no CO2 emissions from 2017 and then some major carbon sequestration.

    So what will happen above 2C?

    Maybe its time everyone who actual feels global warming is real to stop using CO2 themselves!!!?
    0 0
  21. Rob#12, Scaddenp#14 , what is your point ? I didn't doubt that CO2 could be produced by the decay of biomass , I just asked why it was an evidence for a high sensitivity - it is not because you haven't thought of any other explanation that the one you imagine is true. It must be proved by some evidence. Why not for instance some bifurcation of oceanic circulation that could have produced independently a shift of the temperature and a release of CO2 ? I have no evidence for that of course, but not for other explanations as well.
    0 0
  22. 21, okatiniko,

    Did you follow and read the link in the original post, at the statement on high climate sensitivity (in particular, the reference to and explanation of Hansen and Sato 2011 in that referenced post)?

    The short answer to your question, however, is that no matter what mechanism caused the increase in CO2, that mechanism was driven by a change in temperature, and an increase in CO2 increases temperature. This implies (regardless of the mechanism involved) that climate sensitivity is high, i.e. that an increase in temperature will trigger a corresponding increase in CO2 which will exacerbate the temperature increase.
    0 0
  23. okatiniko - unfortunately I don't have access to my papers at home but I think it is Zeebe, R. E., Zachos, J. C., and G. R. Dickens, 2009. There is an issue with the accounting. The upper limit on the carbon release consistent with the record is problematic for producing the temperatures reaching if sensitivity is only in the 2-3 range. Show me how ANY model for ocean currents can produce that much global change. However, there is a lot of work going on (my own sedimentary basin models are being pressed into service) so watch this space. I doubt the answer will prove that exotic.
    0 0
  24. Hej Rob,

    In response to the query, yes, I am “he” … who also happened to make the figure with ridiculously bright colours! [After giving about 20 lectures on the topic across North America courtesy of AAPG in 2009, I was asked to write an article on the topic by CSEG but with a special note that the graphics should be vivid. I responded accordingly, but I think, in retrospect, I went a bit silly].

    I can send you my latest views on the PETM if you think interesting and worth discussion.

    0 0
    Moderator Response: (DB) By all means, please do share.
  25. okatiniko at 15:29 PM on 17 June, 2011
    okatiniko at 02:38 AM on 18 June, 2011

    The comparison between PETM and glacial-interglacial transitions you mention in post # 11 is instructive in addressing the source of [CO2] and its temporal relationship to warming.

    During Milankovitch-driven glacial-interglacial transitions, the Earth warms by around 5 oC globally-averaged, and this is associated with a rise in [CO2] from around 190 ppm to 280 ppm. Since all of this CO2 results from ocean warming and circulation changes, we can assess the [CO2] response to global temperature rise. It's about 18 ppm [CO2] per degree of global temperature rise [(280-190)/5)].

    Since the PETM warming was associated with an increase in [CO2] of around 700 ppm, resulting (with some unknown positive feedbacks possibly involving methane clathrate release), to give a global temperature rise of 5-9 oC, it's inconceivable that the huge rise in [CO2] could be a response to warming. We could only account for perhaps 90-150 ppm's "worth" in this way.

    There's some compelling evidence that the greenhouse gas release at the PETM was a result of the tectonic events associated with opening up of the N. Atlantic at the nascent plate boundary:

    M. Storey et al. (2007)Paleocene-Eocene Thermal Maximum and the Opening of the Northeast Atlantic Science 316, 587 - 589
    link to paper
    0 0
  26. 24, jerryd,

    If you don't mind my asking, at the very end of your paper, you referred to: metamorphism of a large petroleum system in the northern Atlantic Ocean.
    Can you elaborate very briefly on what this actually means?
    0 0
  27. I suspect (but I would love to here more) that he means what happens when you put an intrusion complex into a petroleum system. Our modelling software has been used to look at this for intrusions in the Taranaki basin of NZ. Sedimentary basin contain very large amounts of carbon so I certainly think we should be looking at them. There is a paper relevant to this which be published soon. Subducting such carbon reservoirs has also been postulated. Getting meaningful numbers to put into such models is a difficult process and I suspect years away.
    0 0
  28. Dr. Hansen feels that C02 was released from rich carbon beds in what is now the Indian Ocean during the PETM.

    India was still not attached to the Asian continent- and was moving rapidly through this ocean before its collision with the Asian continent. This carbon rich area was an area where many rivers emptied their debris.

    At least that is how Dr. Hansen perceives the PETM from a paleo climate and geological perspective in his book 'Storms of my Grandchildren'.

    Once India collided with Asia, C02 levels began to drop.
    0 0
  29. The most fundamental problem regarding our understanding of the PETM (and other "hyperthermal" events) is one of carbon mass balance. All evidence consistently points to the idea that at least 2000-3000 Gigatonnes of carbon rapidly entered the exogenic carbon cycle (the combined ocean-atmosphere-biosphere).

    This is different than during glacial-interglacial cycles, when carbon was being shuttled between the ocean, atmosphere, and biosphere. It is, however, very analogous to what we are doing currently through carbon emissions (and why so many people have become fascinated with the PETM).

    To conceptualize the differences, I often give students a picture of a nightclub with three rooms: the big dance hall (ocean), a modest bar (biosphere), and a small restroom (atmosphere). As a spectator, there are two general views for checking the flow of people. One is watching people shuffle between the dance hall, the bar and the restroom, and sometimes there are more people at the restroom (this is the cycling of carbon within the exogenic carbon cycle, and how we think of things during glacial-interglacial cycles). The second is watching people enter and leave the nightclub, and sometimes there are more people in the nightclub, and consequently a more crowded dance floor, a packed bar, and longer queues at the restrooom (this is the cycling of carbon to and from the exogenic carbon cycle, and how we think of things during the PETM and in our future).

    It is easy to understand how and why the nightclub/exogenic carbon cycle is gaining mass at present-day. It’s 22:00 on a Saturday night, the door is open, and we are adding an excess of about 8 Gt C/yr through combustion of coal, oil and natural gas. However, it is not so easy to conceptualize why this happened rapidly in the past. To follow the analogy, albeit somewhat awkwardly, the PETM is a bit like finding the nightclub packed at 13:00 on a Tuesday afternoon. How can massive amounts of carbon suddenly enter the exogenic carbon cycle ~55 million years ago?

    Numerous explanations for the PETM carbon mass balance problem have been given. At present, only three seem viable -- intrusive volcanism (Svensen et al., Nature, 2004), burning/oxidation of peat (Kurtz et al., Paleoceanography, 2003), dissociation of gas hydrate in marine sediment (Dickens et al., Paleoceanography, 1995). None are compatible with current views for how carbon cycles on Earth’s surface.

    It really is an interesting puzzle … on multiple levels. On the one hand, there is the obvious tendency to make comparisons between the PETM and future climate predictions. In general, model simulations for a world perturbed by a rapid and massive input of carbon nicely explains many of the observations in sediments spanning the PETM. On the other hand, the models are based on a framework in which the PETM cannot have occurred … but it did.
    0 0
  30. To follow with specific points:

    Chris: This is not the correct way to think about the problem. The glacial-interglacial changes in pCO2 likely involve carbon redistribution within the exogenic carbon cycle; changes in the PETM (and in our future) likely involve carbon inputs and outputs to the exogenic carbon cycle. In any case, all indications are that massive carbon input during the PETM was a response to external forcing. (And this also hits at the root problem because there is no way to explain this with conventional thinking as to how Earth works).

    Sphaerica and scaddenp: The idea here comes from Svensen et al. (Nature, 2004). They documented, using seismic techniques, thousands of fluid escape structures in the North Atlantic, which appear to have occurred near the onset of the PETM. Thus, they suggested that instrusive sills converted large amounts of organic carbon to methane, which then escaped from the seafloor. It is a very interesting idea and explains several observations; however, it invokes catastrophism (i.e., essentially an equivalent to all the world’s oil, gas and coal were formed and released within <50,000 years). It also fails to explain the other hyperthermal events following the PETM.

    Newcrusader: With all deference to Dr. Hansen, this idea makes no sense given the timing. The massive carbon injection at the onset of the PETM happened within a maximum of 60,000 years, and probably less.
    0 0
  31. Jerryd - thanks for the input. Very much appreciated. I noticed in Zeebe 2009, of which you were one of the co-authors, a mention of trace greenhouses gases in the concluding remarks. Have you read Beerling 2011? Any comments?
    0 0
  32. jerryd at 20:33 PM on 18 June, 2011

    Yes that was the point I was making with respect to okatiniko's suggestion (his post #11) that the raised atmospheric [CO2] in the PETM might be a response to raised temperatures (by analogy with the raised atmospheric [CO2] during glacial-interglacial cycles). The numbers (massive amount of raised [CO2] during PETM and the delta 13C excursion) simply rule out that as a significant contribution. The carbon must have come from "without" the exogenic carbon cycle!

    P.S. By "exogenic carbon cycle" I assume that you mean the "accessible" carbon within the carbon cycle that involves the carbon in the atmosphere, oceans and living things.
    0 0
  33. newcrusader at 19:15 PM on 18 June, 2011

    I wonder whether Dr. Hansen was referring to the raised atmospheric [CO2] levels throughout the Early/Mid Cenozoic (65 MYA to around 40 MYA) onto which the PETM "piggybacked" around 55.5 MYA. There is a very nice theory that this was indeed the result of the remorseless drifting of the Indian subcontinent (to be!) into sub-Asia and the subduction of carbonate-loaded plate beneath the ever-narrowing Tethys sea. The carbonates were driven off as CO2, maintaining a steady high level of [CO2] during this period and warm earth conditions.

    Once India had "squeezed out" the Tethys sea (around 50 MYA), the "CO2 factory" ceased and was overtaken by enhanced weathering of the Deccan Traps (formed near the end-Cretaceous) as these moved (with the Indian sub-continent) into the warm moist tropical humid belt where basalt weathering was very efficient. By around 33 MYA atmospheric [CO2] thresholds had dropped towards the threshold that allowed polar continental ice sheet formation...

    Can't find a downloadable version of this fascinating paper:

    D. V. Kent and G. Muttoni (2008) Equatorial convergence of India and early Cenozoic climate trends PNAS 105:16065-16070

    However there is a “Commentary” accompanying their article that summarises their proposal quite nicely here.
    0 0
  34. jerryd,

    Two probably bizarre questions, but:

    1) Is it possible that carbon existed in some form prior to the PETM which we no longer see in noticeable quantities today? That is, could there have been some sort of release mechanism which entirely escapes our imagination, exactly because it all "let loose" during the PETM? I have no idea what such a form would be, although I imagine such insights would come from a study of the biosphere in any of the millions of years prior to the PETM (e.g. could a massive die off of plant matter during a previous extinction, combined with a certain predominant climate [dry, wet, hot, cold] have produced a "high layer" of what amounts to fossil fuels, something like peat, but with different properties from peat, and yet more accessible than the coal and petroleum with which we are familiar)?

    2) Has anyone ever done a sort of "carbon accounting through the ages" to try to add up and track how much carbon has entered the system at various stages of the earth's existence, and in what quantities in what forms it has existed in the various stages of the earth's existence (atmosphere, living biosphere, decaying plant matter, various sequestered carbon forms, etc.)?
    0 0
  35. Chris

    the peak in global temperatures coincides with India colliding with Asia 50 million years ago- The PETM occurred 56 million years ago. See Hansen's book page 153.

    There was another spike in temperatures about 42 million years ago- not as severe as the PETM.

    Temperatures and C02 fell for the next several million years until global temperatures reached 3 degrees C above what they where circa 1900. This was about 34.5 million years ago, when C02 fell to 450-500ppm.
    0 0
  36. Chris – Yes, by the exogenic carbon cycle, I mean pretty much everything on Earth’s surface within a few meters of sedimentary depth. To be honest, it’s a hard concept to describe rigorously, as I found out on NPR about 15 years ago. When asked what the exogenic carbon cycle is, I think I said something like “well, it’s like if you took all the carbon in the ocean, the atmosphere, and on land including soil, trees, plants, animals and newts and put them in a blender.” [I have no clue why I said newts and blender, but it went across the airwaves, and several colleagues continually poke me on this comment]. Basically, it is all carbon that can exchange on geologically short time scales.

    The idea of slow (million-year) carbon changes caused by tectonism is certainly plausible. I have not yet read the paper by Kent and Muttoni, but will do so shortly.

    Rob – Thanks for passing along the Beerling reference. I have not yet seen or read this paper either. It certainly looks very interesting.

    [Here I think an interesting aside: I know both Dennis Kent and Dave Beerling personally, and appreciate their work very much. They are both top-notch scientists. That I have not read these works, I think, signals the state of the field – it is moving really fast. It was easy to keep up with things regarding early Paleogene climates 5-10 years ago, but numerous papers are now coming out every week. I probably need to do some more reading rather than blogging!].

    Anyway, there are several major problems with understanding “early Paleogene” climates (here loosely meaning the time from about 62 to 45 million years ago).

    One is the “hyperthermal problem” noted in previous posts. How do massive amounts of carbon enter the exogenic carbon cycle quickly (and according to recent papers) repeatedly?

    Another is the “equator-to-pole gradient problem”. All data consistently suggests that the equator-to-pole temperature gradient was much lower in the early Paleogene relative to present-day. That is, the equator was ~5 °C warmer than today but the poles were ~25 °C warmer than today. Some of this problem is solved by the removal of ice (the albedo effect); some of this problem may lie in the use of proxies for past temperature, namely that many records generated at high latitudes may be biased toward summer temperatures. Even after this accounting, the poles seem too warm (or the equator seems too cold) in coupled climate models for the early Paleogene. Matt Huber has a current paper discussing the problem ( This is where I think Dave Beerling might be correct: trace gases may explain the discrepancy.

    Sphaerica – Great questions.

    1/ All sorts of things are conceivable. However, I think that the geological record is telling us that we do not have an appropriate view for how carbon cycles on Earth’s surface.

    2/ Several people have done this (notably Bob Berner at Yale). But, a roadblock hits at the PETM, because it is impossible to explain this event within the context of conventional modeling of carbon cycling. Basically, one needs to invoke an "ad hoc" mechanism for carbon transfer. This may be correct, if for example, the source of carbon came via intrusive volcanism.

    I am 99% sure, however, that we need to rethink how the global carbon cycle operates over geological time.
    0 0
  37. Chris @ 33

    I found downloadable Kent and Muttoni PDF some weeks ago by googling it.

    I was very skeptical of this notion at first because in classic plate tectonic theory carbonate sediments are too light to be subducted and are scraped off at the continent/arc margin.

    It now appears that within the last decade or so various isotopic analyses indicate that there are both low and high sediment flux trenches.I have not yet found any remotely articulate explanation for what feaures (angle?)might account for the difference.

    Tethys is thought to have been shallow and at first glance seems an excellent candidate for substantial carbonate sediment buildup. Closer analysis reveals that the Tethys ocean floor was extremely active, with two spreading centers through most of the Mesozoic. Much carbonate sedment may have already beern subducted or scraped off prior to India's excursion.

    Unfortunately there is no modern tectonic analogue.

    There is, however, active rifting going on today in the Red Sea and the Gulf of California.If Lee Kumps "baking carbon rich sediments and perhaps even some coal and oil near the surface."is to be taken seriously, we should see carbon anomolies in these areas.

    Apparent polar wander paths indicate that the rifting of Greenland from Europe and America was well under way 80mya in the Cretaceous.
    0 0
  38. I just read the Kent and Muttoni paper (PNAS, 2008) ...

    It's interesting. However, I think the hypothesis is demonstrably wrong.

    Here's why:

    1/ The long-term maximum in Earth surface temperature over the last 80 million years or so occurred at about 50-52 Ma. For better or worse, we call this the Early Eocene Climatic Optimum (EECO).

    2/ I entirely agree with the paper that this peak in warming was associated with high amounts of carbon in the exogenic carbon cycle. There are several good arguments to support this notion.

    3/ However, the d13C of carbonate is around 0 to +2 per mil (depending on age). (For those that I have lost here, see below).

    4/ This means that, if the temperature rise was driven by CO2 from carbonate, there should be no significant change in the d13C of the exogenic carbon cycle. (The ocean having a d13C composition of nominally +1 per mil).

    5/ In fact, the d13C of the exogenic carbon cycle reaches a prominent low at 50-51 Ma. This implies that the source of carbon was very depleted in 13C; that is, most likely related to organic carbon.

    Basically, there is no way to add massive amounts of carbon to the exogenic carbon cycle from carbonate and have a prominent long-term d13C excursion.


    After numerous scribblings on wet napkins, and trying to explain carbon isotopes and the carbon cycle, here's my best effort.

    Carbon has two stable isotopes, 12C and 13C. The ratio of these two isotopes is typically expressed in terms of d13C, with carbonate close to zero. Various reservoirs (ocean, biosphere, atmosphere) have different d13C.

    For whatever reason, it seems much easier for most people to think of things where carbon reservoirs are pools, and carbon isotopes are white (12C) and red (13C) paint.

    Consider the ocean as very light pink. Carbonate is white; organic carbon is red.

    The Kent and Muttoni paper suggests that massive amounts of white are added to a very light pink reservoir. It implies that the pool should become lighter. The geological record shows that, instead, the pool becomes a deep crimson.
    0 0
  39. jerry - what about multiple event scenario? Elevated carbon in atmosphere, then volcanism in sedimentary basin releasing a spike big enough to trigger hydrate release at poles. With low pole - to equator gradient, the feedbacks release hydrate world wide. On longer term, the elevated temperatures speeded up methane production from sedimentary reservoirs (which are huge by comparison to surface carbon reserves). (This latter is what colleague is pursuing).
    0 0
  40. jerryd at 07:49 AM on 19 June, 2011

    O.K. that's very interesting jerryd. I haven't looked at the d13C data associated with the entire Paleocene/Eocene period with consistently raised [CO2] (rising to a maximum at the EECO). Do you have a paper/link where the long term d13C data are compiled?

    I guess the implication is that at least a good chunk of the increased [CO2] during this period was biogenic.

    And a couple of questions arise. Why were [CO2] levels maintained so high (2500-3000 ppm apparently) during this period? Was weathering particularly inefficient for some reason? And does the low values for d13C apply to the full period (from around 65 MYA to 50 MYA) or does it drop particularly during the apparent rise of [CO2] during the period ~ 55-50 MYA?

    There's no question that the d13C data are a fantastic bonus with which to supplement proxy data for absolute [CO2]. I have to say the Kent and Muttoni hypothesis is rather appealing since it does seem to provide a coherent explanation for post-Cretaceous [CO2] data, but if it's not consistent with the d13C data then that's not too good for the theory.

    Sounds like a promising area for further thought!
    0 0
  41. Further thought - Kent and Muttoni are assuming carbon from carbonate, but could also be subducting a pile of carbon-rich sediment, pushing them rapidly through the oil/gas window.
    0 0
  42. Sphaerica at 23:51 PM on 18 June, 2011

    "Is it possible that carbon existed in some form prior to the PETM which we no longer see in noticeable quantities today? That is, could there have been some sort of release mechanism which entirely escapes our imagination, exactly because it all "let loose" during the PETM?"

    There is a possible answer that was reported at the American Geophysical Union in December which fits your question. Quoting from a write-up in Science by Richard Kerr (see page 142) reporting on a couple of talks on the PETM:

    "In the next talk, paleoceanographer Robert DeConto of the University of Massachusetts, Amherst, and his colleagues claimed that a larger source of carbon dioxide in Eocene times would have been the permanently frozen ground of polar regions. No ice sheets covered polar ground in Eocene times, and the land area of Antarctica was far larger than it is today, DeConto said. That would have made for widespread permafrost, storing organic matter that—if thawed and decomposed—would yield large amounts of carbon dioxide. In their model, extremes of orbital variations caused permafrost to melt in summer and release its carbon dioxide. Enough would have come out fast enough to drive the observed warming, DeConto said."
    That would fit your scenario...except perhaps the bit about "entirely escap(ing) our imagination"!

    Another group (A. Ridgewell et al) also suggest a role for Milankovitch cycles (earth orbital varibility) in the PETM:

    "Ridgwell and his colleagues included orbital variations in a computer model of early Eocene climate, when the world was slowly warming under a strengthening greenhouse. When the orbital variations in the model combined to make an extreme contrast between summer and winter temperatures, the model's ocean circulation would slow and stagnate. That allowed the deep ocean to warm, warming the sea floor. And that's where natural methane hydrates reside—the “ice that burns.” Warm them and they decompose into water and methane. That methane can escape the sea floor, oxidize to carbon dioxide, and warm the world."
    The role of orbital cycles as a factor in the PETM is quite appealing since the timescales (few thousands of years) are appropriate.
    0 0
  43. jerryd @38, I do not follow your reasoning.

    First, you state that the d13C of Cabonate is from 0 to 2. It is not possible that carbonate on the floor of the Tethys was closer to 0 per mil and hence depleted in C13. It seems that given the range of d13C concentrations you give for Carbonate, you cannot run the argument that you run against Kent and Muttoni.

    Second, given that the floor of the Tethys was carbon rich because it was a shallow, life rich sea, does this not suggest that Tethys carbonate is likely to have been C13 depleted relative to other carbonates?
    0 0
  44. jerryd

    Wouldn't pelagic carbonate(shallow photosynthesizing silicic slab carbonates)be your red and benthic clay carbonates (marls)be your white?

    It seems photosynthesis is the isotopic filter.

    Getting way out there even for me, cosmic rays turn nitrogen into carbon 14, which becomes co2 for a half life of 5700 years. After a burst ceases it all decays innocently back to nitrogen (who me?)with paleoclimatologists none the wiser.
    0 0
  45. truckmonkey, and that explains the dC13 how? Also to create significant amount of C14 would suggest radiating the earth in a way that would be somewhat deadly for all land fauna.
    0 0
  46. scaddenp

    Doesn't explain dC13. Not a really great idea. Pretty much untestable even if it were. Just threw it out in the spirit with several other folks thinking outside the box.
    0 0
  47. The long-term benthic foraminifera carbon isotope record for the Cenozoic is published in Zachos et al. (Science, 2001). An updated benthic foraminifera oxygen isotope record was published in Zachos et al. (Nature, 2008) but without the paired carbon isotope record; however, Jim has put the data on his web site (go to, and then to 2008 publications).

    Interestingly, The first really nice long-term carbon isotope record for the Early Cenozoic was published by Shackleton and Hall (DSDP, 84, 1984). It can be downloaded at:

    The age axis is bit off in the above works, but not by too much.

    In any case, you will see that, from about 62 to 58 Ma, the d13C of the ocean (~exogenic carbon cycle) increased by ~1 per mil. Then, from about 57 to 52 Ma, the d13C of the ocean decreased by about 2 per mil.

    The peak in long-term temperature (the EECO) occurs at the end of the drop.

    Superimposed on the fall in d13C are the carbon injections related to the hyperthermals (i.e., the PETM occurred about half way along the drop).

    Assuming that the long-term and short-term d13C excursions are related, there is really only one way to explain things. Massive amounts of organic carbon were stored somewhere in the Late Paleocene (this preferentially removes 12C, increasing the d13C of the ocean); this carbon returned to the ocean, sometimes sporadically, during the latest Paleocene-early Eocene.

    The ideas in both Rob's and Andy's abstracts are appealing in this respect. Store a bunch of peat or gas hydrate, and then return it. However, I think the peat idea has some serious mass balance problems.
    0 0
  48. Tom -- it all comes down to mass balance.

    There is no way to change the isotopic composition of the huge mass of carbon in the ocean by adding something of similar composition. This gets back to the pool and paint analogy. You have a large swimming pool of very light pink colour (d13C of ~1 per mil). You can add pretty much as much white or light pink as you want (d13C of 0-2 per mil), and the pool gets more full, but the colour does not change very much. You need to add something very red (d13C of -25 to -80 per mil), and still quite a bit, to get the pool to change colour.

    Following the previous post, Kurtz et al. (Paleoceanography, 2003) did the carbon mass balance exercise for the early Paleogene. They tried to drive the late Paleocene carbon isotope high and the early Eocene low as well as the PETM through formation and burning of peat. It's a very cool idea ... except there is a problem when one looks carefully at the masses involved.

    The +1 d13C excursion necessitates the storage of about 60,000 Gigatonnes of carbon in peat during the late Paleocene. (Note, this would be storage in a transient surface reservoir -- think of this as a hot tub on the side of the pool. It would not be burial, because the carbon needs to return to the exogenic carbon cycle in the early Eocene to drive PETM, other hyperthermals, and the long-term drop in d13C). The problem concerns the mass. The entire terrestrial biosphere, including soil, humus, peat, plants (and newts) at present-day is on the order of 2000-3000 Gt C. So, if this is the explanation, then one needs to imagine a terrestrial carbon reservoir radically different than present-day (more specifically, at least 5-10 times larger).

    I think the press releases for the forthcoming paper by Cui et al. (Nature Geoscience, 2011) initiated this thread. If you read the paper, you will see that they touch on this problem toward the end. More specifically, the authors suggest an input of 12,000 Gt C from land to cause the d13C excursion across the PETM; they then note that this is tremendously large compared to terrestrial pools at present-day.

    I have hassled Lee Kump and Rob DeConto about this mass balance problem (both are friends). With Lee, a classic exchange: "Lee, this is really hard to imagine; Jerry, you need to imagine harder."
    0 0
  49. scaddenp --

    jerry - what about multiple event scenario? Elevated carbon in atmosphere, then volcanism in sedimentary basin releasing a spike big enough to trigger hydrate release at poles. With low pole - to equator gradient, the feedbacks release hydrate world wide. On longer term, the elevated temperatures speeded up methane production from sedimentary reservoirs (which are huge by comparison to surface carbon reserves). (This latter is what colleague is pursuing).

    This is, I think, close to the correct answer. I have made a figure showing how I think things work in the Early Paleogene. It's from a paper that should come out in the next month or so in Climates of the Past. Maybe there is a way I can send the figure and paper to this site? (I am traveling through Europe right now so cannot post on my home site easily). Then, you all can have fun finding the holes and problems in my logic.

    0 0
  50. Jerryd - "I think the press releases for the forthcoming paper by Cui et al. (Nature Geoscience, 2011) initiated this thread."

    Yup, but I have read the paper. Cui 2011 "imagine" that the Late Paleocene was a 'coal giant'. Maybe you need some of this Jerry? The attraction of the paper was that it was the latest attempt to constrain the rates of carbon release during the PETM. I kinda preferred the organic carbon scenario (warts n'all) because it implied our current emissions weren't quite so bad.

    And yes, we'd be very interested in your latest paper. John Cook has sent you an e-mail I believe.
    0 0

1  2  Next

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

The Consensus Project Website


(free to republish)

© Copyright 2019 John Cook
Home | Links | Translations | About Us | Privacy | Contact Us