RealClimate logo

Methane game upgrade

Filed under: — david @ 14 June 2012

Walter Anthony et al (2012) have made a major contribution to the picture of methane emissions from thawing Arctic regions. Not a game-changer exactly, but definitely a graphics upgrade, bringing the game to life in stunningly higher resolution (/joke).

Katey Walter Anthony draws upon her previous field findings that methane emissions from the Arctic landscape tend to be focused at the intersection between frozen and thawed, in particular in rings around a peripheries of lakes. She also knew what a methane seep looks like in that landscape, leaving visible bubbles frozen into the ice or maintaining an unfrozen hole in the ice. Now she takes to the skies to produce an aerial survey of the Alaskan landscape, data that is so much more voluminous than before that it becomes different in kind.

The methane emission fluxes are higher than previous estimates, but that’s not really the most important point, because emissions from the Arctic are small relative to low-latitude wetlands, and doubling or even nearly quadrupling the Arctic fluxes (in one of their analyzed regions), they would still be small in terms of global climate forcing. And the lifetime of methane in the atmosphere is short, about 10 years, so methane doesn’t build up like CO2, SF6, and to a lesser extent N2O do.

The really interesting take-away from the new paper is how it shows that the near-surface geology and freezing state conspire to control the venting of accumulated gas dribbling up from below, and the decompostion of frozen soil carbon. They have so many methane seep observations that they are able to correlate them with (1) currently thawing permafrost, which allows fossil soil carbon deposits from the last ice age called Yedoma to decompose (Zimov et al 2006) and (2) melting ice sheets and glaciers “un-crunching” the landscape as they fade away, making cracks that vent methane from deep thermal sources. Glaciers that melted long ago no longer vent methane, showing that the methane is transiently venting from built-up pools of gas.

What these results do not do is fundamentally change the game, in my opinion. We can now see more clearly that most of the methane flux from the Arctic today are of types of emission that will respond to climate warming. But the general response time of the system is slow, decades to centuries, rather than potentially poised to release a huge pulse of methane within a few years. Earthquakes and submarine landslides are sudden events, but small individually in terms of potential methane release. The new data do not change that. Walter Anthony et al. compare an estimate the amount of methane in the Arctic, 1200 Gton C, with the 5 Gton C of methane in the atmosphere. That’s the nightmare comparison, but it’s only really relevant if the methane comes out all at once. (The Arctic estimate is for methane itself and is mostly methane hydrate, but keep in mind that there is also a comparable amount of decomposable soil carbon.)

In my opinion, the largest impact of all this methane will probably be to the long-term future evolution of climate. Avoiding a peak warming of 2 degrees C or more requires keeping the total emission of carbon down to less than about 1000 Gton C (Allen et al 2009). We have already burned about 300 Gton C, and cut about 200 Gton C. So maybe we’re 1/2 of the way there, say 500 Gton C left to go. The 1200 Gton C of Arctic methane hydrates and the permafrost carbon stack up pretty menacingly against our 500 Gton left to go, and the comparison is relevant even if the carbon is emitted slowly, or as CO2 rather than methane, or even if it is released into the ocean rather than into the air (it will still equilibrate with the atmosphere, after a few centuries, converging to the same “long tail” CO2 trajectory that would have resulted from atmospheric release).

Arctic methane, and all that frozen soil carbon, could easily play a huge role, not so much in the near-term evolution of Earth’s climate, but in the long tail of the global warming climate event.

Allen M.R., D.J. Frame, C. Huntingford, C.D. Jones, J.A. Lowe, M. Meinshausen & N. Meinshausen (2009) Warming caused by cumulative carbon emissions towards the trillionths tonne. Nature 458 doi:10.1038/nature08019

Shuur E. et al (2008) Vulnerability of Permafrost Carbon to Climate Change: Implications for the Global Carbon Cycle. BioScience, 58(8):701-714.

Walter Anthony, K.M., P. Anthony, G. Grosse, & J. Chanton (2012) Geologic methane seeps along boundaries of Arctic permafrost thaw and melting glaciers. Nature Geoscience doi:10.1038/NGEO1480

Zimov, S.A., Schuur, E.A.G, and F. Stuart Chapin III, F. (2006) Permafrost and the Global
Carbon Budget. Science 312: 1612-1613.

177 Responses to “Methane game upgrade”

  1. 1

    So if thawing Arctic regions releasing methane and frozen soil carbon are the future “long tail” what in your opinion are the current “head terms” (borrowing from SEM terminology)?

    [Response:Not sure if this is what you’re asking, but I think of the CO2 trajectory as a peak (which we are embarking on now and which will persist for some centuries until the atmosphere equilibrates with the ocean) and a long tail. David]

  2. 2
    MMM says:

    “and to a lesser extent NO2”

    I’m guessing you mean “N2O”. =)

    [Response:ah, yeah… fixed. David]

  3. 3
    Pierre Allemand says:

    A simple glance to the GISS data ( indicates that the main origin of atmospheric methane is human action.
    Simply speaking, methane production is more or less linked to food production and so is more or less linearly linked to the number of human beings on earth.
    So, studying arctic methane production and its influence on climate is, certainly a good idea, but money spent for that could probably find a better use elsewhere. (This is, of course, a personal opinion).

    [Response:Well, all of the anthropogenic methane comes from human activity, but that’s only about half of the total methane in the atmosphere today. This is great work well worth the money, though, even if the direct link to climate change today is probably small, to understand what will happen in the long term, how it worked in the past (paleoclimatology), and to make real sure we’re wrong about the short-term climate impacts being small. David]

  4. 4
    sidd says:

    Fig 2 in Allen(2009) suggests that the 1200 Gt (methane) + 1200 Gt(yedoma) + 1000Gt human = 3400Gt would result in peak warming north of 3C. Is it also not true that none of the models used in the paper incorporate Hansen’s ‘slow’ feedbacks (which double the sensitivity,) and which seem to be a lot faster than hoped…


    [Response:Yes, that’s true, and it takes 2 C as a target, which is no picnic, so I’ll agree with the drift of your comment and the next several that the 1000 Gton target is, what’s the opposite of the word conservative, used in this scientific sense? An overestimate of a real “safety” boundary. David]

  5. 5
    MARodger says:

    The “maybe we’re 1/3 of the way there” appears to underestimate the situation rather too much.
    Allen et al 2009 are explicitly talking of “emmissions” not atmospheric values.

    [Response:Ya, me too. David]

    Total anthropogenic emissions of one trillion tonnes of carbon (3.67 trillion tonnes of CO2), about half of which has already been emitted since industrialization began, results in a most likely peak carbon-dioxide-induced warming of 2 °C above pre-industrial temperatures, with a 5–95% confidence interval of 1.3–3.9 °C.
    And CDIAC give emissions (ex land use) of 340 GtC to 2007 to which 10% can be added to 2011.

    I would thus suggest we are past halfway. And I’m not sure the what we’re halfway towards isn’t also an underestimation by Allen et al.

    [Response:Agreed, in that 2 C is too warm, in my personal opinion. David]

  6. 6
    michael sweet says:


    How would you know if the Arctic methane is a problem or not if you did not study it? Sidds estimates of total carbon from the Arctic look pretty scary to me. If we do not know all the natural reserves we will not be able to estimate the response.

  7. 7
    Ron Broberg says:

    I recently looked at warming scenarios with and without methane feedbacks assuming a poisson distribution of equilibrium climate sensitivity (IPPC [2,3,4.5]) and a gaussian distribution around possible future emissions from fossil fuel sources of 6100 Gt of CO2 (burn everything we can extract). (most recent on top)

  8. 8
    sidd says:

    I asked this in the unforced variations thread already, perhaps the moderators will allow a repeat of the question:

    We see more methane coming out of the arctic. But methane concentrations are not rising quickly. So the methane is being efficiently oxidized. This should lead to decrease in OH- concentrations. Is the decrease visible ?


    [Response:Best I understand it the estimates of global OH inventory, all indirect like abundance of trace organics in the atmosphere, has been thought to have held relatively steady through the transition from preanthropogenic to industrial atmospheres. Increasing load on the OH due to emission of CO and CH4 has been balanced, more or less, as far as they can tell, by increasing production of OH due to NOx emission. Factors of two cancelling each other out by luck, apparently. David]

  9. 9
    Uli says:

    I have a question to the methane release using natural gas. I have found that 0.3 per cent to over 4 per cent of the methane in natural gas is released to the atmosphere, in different sources, a rather large span. This influences the impact of fuel change from coal to natural gas on climate.
    Are there now better/narrower estimates for this value? Does it depend on the way natural gas is produced (conventional vs. unconventinal) or transported (pipline vs. LNG)?

  10. 10
    Pete Best says:


    If 300 Gtonnes of C is what we have released and 1000 Gtonnes C is the limit for 2C and we are burning 10 Gtonne C per year then thats 100 Gtonne every decade and hence the alarmist future is 70 years away – surely we can avoid a 2C peak and above if we start combating emissions today?

    [Response:I think technologically it would be a piece of cake, it’s just hard to make the decision to leave the coal in the ground. If there were no more coal, we’d figure a way. It would be some decades before we’d have to quit cold turkey, but realistically emission would probably have to glide down some exponential decay, cuts of x% per year. Beginning now, cuts of 2-3% per year would ramp down to a total burn of 1000 Gton C, but if we wait 10 years, maybe 5% per year would be required. The steeper the cutbacks, the more new stuff has to be built every year, the more it will cost. David]

  11. 11

    David, are you measuring the total 1200 Gton C of methane as CO2-equivalent, or actual C?

    [Response:Carbon, just mass of carbon. David]

    Something BTW I was interested to read in Raypierre’s book is that methane’s apparently high greenhouse effect is largely an artifact of the fact that it exists in relatively low concentration, i.e., the more absorbing parts of the spectrum haven’t yet been saturated. See my earlier comment on methane.

    [Response:If fact, I think if CO2 and CH4 were at equal concentrations, CO2 would be stronger since it’s in the middle of the Earth’s IR spectrum and methane is on the fringe. David]

  12. 12


    Pete, not to rain on your parade, but why on Earth would you assume a fixed annual increase? Sadly, that seems like unbelievably good results from a case where “we start combating emissions today.”

    Though Heaven knows that a serious start (as opposed to the current fumbling one) is highly indicated.

    [Response:It’s true that business-as-usual has probably 15-18 Gton by 2070, 20 by 2100. David]

  13. 13
    Jim Eager says:

    “why on Earth would you assume a fixed annual increase?”

    Kevin, Pete is also ignoring CO2-equivalent emissions.

    [Response:Ah, but if the methane comes out slowly, it will probably have its largest impact as the accumulating CO2 rather than as the increase in methane concentration itself in the atmosphere. The equivalent-CO2 thing only applies if the methane were all released to the atmosphere at once. David]

    Plus, that “300 Gt C” figure has been in use for at least as long as I’ve been following the science, which is well over 5 years now, in fact, IIRC, CDIAC was using 305 Gt C that long ago. It’s kind of like the “BP” date notation that keeps receding.

  14. 14
    Ron Broberg says:

    Re “combating emissions”:

    US CO2 emissions are projected to remain flat or slighty decling over the next three years. See EIA June 2012 Short Term Energy Outlook page 42

    I suspect the same is true for the EU – for much the same reason.

    [Response:The global trend is still resoundingly upward. David]

  15. 15
    Hank Roberts says:

    Methane leakage from the distribution system is considerable but not well measured; up to five percent maybe:
    Leak detection technology is changing improving, new info ought to come in on this generally. I’ve seen studies done driving around old cities mapping areas of high methane background from leaking distribution systems.

    Handheld leak detectors are fairly cheap, I just put one in the earthquake kit in case I’m foolish enough to go crawl under the house after the next big quake to check for little leaks in the old pipes.

  16. 16
    Guy Schiavone says:

    AGW is having its most extreme impact in the polar regions, which in turn will have large positive feedbacks in the form of decreased albedo and increased atmospheric water vapor coming from an open Arctic Ocean, among others…not to mention that a large sea-level rise will occur with the rapid melting of the Greenland Icecap…yet, “methane emission fluxes…higher than previous estimates…would still be small in terms of global climate forcing”.

    Non Sequitur, your facts are uncoordinated.

    [Response:2% of global methane emissions going up to 5% is a small relative change. Sequitur. David]

    I have been hearing this refrain often lately in the face of countervailing evidence, could it be that our current models are very deficient in this particular area?
    To go even further, are we seeing a social phenomena similar to that displayed by the paleontologists who believed that “mammals ate the baby dinosaurs” in their reaction to the Alvarez father and son team’s discoveries?

  17. 17
    MARodger says:

    Re Response @5.
    Are you saying you are still happy with “1/3 the way there“?

    [Response:No, didn’t say that at all. David]

    According to CDIAC, the emissions since pre-industrial times are well past halfway to 1,000 GtC.
    The land-use emissions 1850-2005 total 156 GtC. Add 37 GtC for the preceeding century plus 8 GtC for 2006-11 yeilds emissions of 201 GtC.
    The FF & cement emissions 1751-2008 total 347 GtC. Add 27 GtC for the last 3 years to give 374 GtC.
    By that count, the grand total emissions works out at 575 GtC.

    [Response:The issue is how to count the land use carbon and the land uptake carbon. David]

    And let’s not forget that CO2 is not the only agent behind AGW. Some are negative forcings but I would argue that the net effect of non-CO2 forcings will be positive in coming decades, adding to the urgency of responding to AGW.

  18. 18
    Dan Riseborough says:

    A small semantic point about the transformation of permafrost as the climate warms: Strictly speaking, when permafrost warms to the melting point of ice, it is the ice in it that melts, and not the earth material. Permafrost scientists use the term “thawing permafrost”. Consider frozen hamburger meat: the ice in the meat can melt or thaw, but we thaw the meat.

    See also: wikipedia and
    the editorial in this issue of the International Permafrost Association bulletin

    [Response:Along the same lines, methane hydrates don’t melt, according to a persnickety reviewer of one of my papers; they decompose. David]

  19. 19
    Chris Colose says:


    Thanks for this great post. I always like learning more about this topic, and somehow it’s one I don’t often find the time to investigate very much.

    I think this perspective is a spot-on and consistent with the 2011 Climate Stabilization Targets: Emissions, Concentrations, and Impacts over Decades to Millennia report some people may find useful. (On a more personal note, it is a bit of a pet peeve of mine to see the “methane catastrophe” and “runaway greenhouse” stuff get the attention it does in many popular discussions of carbon-based feedbacks; it actually gets quite annoying).

    The way I like to think about this problem (concerning the methane impact on the short and long term) is to decompose the problem into being a function of the size and duration of the release. Suppose we released 200 GtC into the atmosphere, all in the form of methane. That corresponds to about 90 ppm of methane. That reduces the OLR in the neighborhood of 8 W/m2, or a quadrupling CO2. Using a transient climate response of ~1.5 C/2xCO2, that gives you maybe 3 degrees of rather rapid warming. But the methane dissipates on timescales of a decade or so, and assuming it all stays in the atmosphere as CO2, then the corresponding radiative forcing is ~1 W/m2 (assuming the 90 ppm is being added to a background of 400 ppm CO2). Given the slow removal time of CO2, that can affect the long-term response; assuming an equilibrium sensitivity of ~3 C/2xCO2, that adds on another 1 C or so to the tail end.

    If the methane release is on the order of 1000 GtC, then there is a rather catastrophic instantaneous response, but this is not very realistic. However, even for a slow release, that can fully double the amount of CO2 in the entire atmosphere and affect the long-term warming. Moreover, with a slow release, there will be a small but persistent CH4 radiative forcing on top of that.

    P.S. Philip (#11) is absolutely right that methane only “looks like” a stronger greenhouse gas than CO2 because its local slope (on a OLR vs. concentration plot) is larger than the local slope for CO2. This is just because of the difference in background concentration. A proper comparison of, say, 100 ppm CO2 vs. 100 ppm CH4 would show that CO2 is a better greenhouse gas on Earth. This might seem like an esoteric point, but it’s one that people have messed up even in the academic literature.

  20. 20

    Thanks for an illuminating comment, Chris C–as usual.

    But the methane dissipates on timescales of a decade or so, and assuming it all stays in the atmosphere as CO2, then the corresponding radiative forcing is ~1 W/m2 (assuming the 90 ppm is being added to a background of 400 ppm CO2).

    Which raises the question: what would the feedbacks during that decade do? I’m guessing it would be
    “bye-bye Arctic summer sea ice” for sure, and that we’d get a secondary pulse of methane too–but how big? And what else could be expected?

    Perhaps this would be ‘non-catastrophic’ (whatever that means in this context), but ‘non-negligible’ seems like a decent bet, OTTOMH.

    [Response:That stuff is happening mostly because of radiative forcing from CO2. If the atmospheric methane concentration went up a few % due to the Arctic (a few % of total emissions, doubled, say), then the radiative forcing from that methane would be a few % of the total anthropogenic forcing. Worry about CO2! David]

  21. 21
    Jim Eager says:

    “A proper comparison of, say, 100 ppm CO2 vs. 100 ppm CH4 would show that CO2 is a better greenhouse gas on Earth.”

    Indeed, nothing currently in the atmosphere takes a deeper bite than CO2, just look at the depth of the trough it makes in the outgoing LW plot. H2O may absorb over a far wider range of wavelengths, which is why it accounts for ~80% of the greenhouse effect, but nowhere does it absorb anywhere nearly as completely as CO2 does at 14 microns near the peak of the curve.

    Not bad for a well mixed “trace” gas, eh?

  22. 22
    Chris Colose says:

    Kevin- I would expect that the typical “fast feedbacks” of interest (water vapor, lapse rate, etc) would be of most importance on this decadal timescale. You could reduce the Arctic sea ice rapidly, but this is only a very small contribution to the total planetary energy budget. As an aside, unlike with the Greenland ice sheet, there is evidence that the Arctic sea ice loss is reversible, so I’d expect its long-term fate to be determined by the gradual tail. Dr. Archer would be much better suited to answer the other part about further methane feedback.

    Jim- Indeed. You need to weight the relative absorption strength with where the Earth emits, and CO2 is a very good absorber right near the peak of Earth’s Planck emission spectrum. Methane is toward the fringe as David mentioned, but it’s still significant for Earth temperatures, and the absorption by water vapor and CO2 is not strong in that region. There’s also the vertical profile to consider. To have a greenhouse effect, you need enough absorber at altitude where the emission temperature is lower than the surface value, and water vapor is not as abundant at these altitudes.

    By the way, water vapor isn’t quite 80%, not even in clear sky conditions. It’s closer to half of the greenhouse effect, and most of the rest is due to CO2 and clouds. Methane isn’t much of a contribution, although its slope (i.e, its potential radiative forcing) can be relatively large on a molecule-for-molecule basis, just like with CFC’s or a number of other gases. By the way, N2 and O2 actually contribute to about half a percent of the greenhouse effect directly (even though they are symmetric, diatomic molecules)- and in cold, dry regions like Antarctica the contribution from those gases is comparable to methane.

  23. 23
    Jim Eager says:

    “By the way, water vapor isn’t quite 80%, not even in clear sky conditions. It’s closer to half of the greenhouse effect, and most of the rest is due to CO2 and clouds.”

    Chris, which is why I wrote “H2O” instead “water vapour”; i.e. water vapour plus liquid water in the form of cloud droplets accounts for ~80%.

    “To have a greenhouse effect, you need enough absorber at altitude where the emission temperature is lower than the surface value, and water vapor is not as abundant at these altitudes.”

    That’s almost an understatement. IIRC, on average water vapour drops to ~300 ppmv between 6-8 km altitude–less than CO2, and to only 3-4 ppmv by 10-12 km.

  24. 24
    Pete Best says:

    Re #10 – Thanks for the reply but as the west outsources more and more of its low profit manufacturing to the east and as they grow in prosperity (non OECD) we can see that emissions are actually rising globally even though the west appears to some degree to be cutting back on emissions to some degree. 2010/12 emissions have increaed globally and we are in a period of slow growth globally as the recession is still with us.

    Secondly is the annual growth issue which in order to have high employment and more prosperity growth needs to be at 2-3% er year which means more energy usage of around the same increases. Therefore the cut backs of 2-3% per annum in Carbon emissions need to also take into account the annual emissions of 2-3% as well making cuts of 4-6% year more likely and as you say the more we delay the cuts needs to be 5-8% per annum to offset growth as well.

    I dont doubt we have the technology but its always the same in our political partisan world. Those for the status quo and those for some change and thoise for radical change. Its a monstrous undertaking.

  25. 25
    wili says:

    Could someone with access to the article explain exactly what mechanisms “conspire to” control the flow of gas into the atmosphere. Are these mechanisms completely consistent and reliable in all places and in all scenarios?

    [Response:What I meant was, methane comes out when the permafrost melts thaws, and when mountain ice caps melt. No, not completely consistent in all places and all scenarios, but a statistically significant tendency in the data. David]

    The stunning thing about this article to many of us is that it shows that there are vast stores of FREE methane ready to erupt into the atmosphere through just a crack in the rapidly thawing permafrost and ice cap. And the area if warming incredibly fast and getting faster. Shrubs widely distributed across the Arctic, as it turns out, grow directly into trees under warming conditions. This is happening now. So we don’t need to wait for forests to slowly migrate up from the south–they area already ready to go right there. And of course the new Arctic forests will not reflect sunlight the way the unblemished snow surface would, so the regions will warm yet faster, leading to more forest, leading to more warming…

    [Response:Yes on the basic albedo change concept but it’s more complex than that; there are forest and forest type definitions, hydrological and other considerations. The study (and Supplemental) referred to is here and here, and a similar recent one is here and here. I recommend reading them both.–Jim]

    And that feeds into thawing areas any one of which could have a vast pool of methane ready to burst into the atmosphere. I find nothing comforting about the abstract, but more info on the control mechanisms would be appreciated.

  26. 26
    Hank Roberts says:

    What he read:

    “… most of the methane flux from the Arctic today are of types of emission that will respond to climate warming. But the general response time of the system is slow, decades to centuries, rather than potentially poised to release a huge pulse of methane within a few years.”

    What he says he read:

    “… this article … shows that there are vast stores of FREE methane ready to erupt into the atmosphere through just a crack …. a vast pool of methane ready to burst into the atmosphere ….”

  27. 27
    Mike Roddy says:

    It appears to me that you are letting the past govern the future a bit too much here. Given the large quantity of methane in the Arctic, why should we assume gradual and orderly releases? And if 2x or 4x more is still not a big deal, shouldn’t the fact that the intersection of frozen and thawing sources lead to an expectation of much larger bursts? Isn’t it possible that forcings from 1-2C warming could result in a major permafrost and subsea methane escape event, driving concentrations upward and possibly triggering both local and global feedbacks? Recent paleo studies indicate that these events can occur with great suddenness.

    When I have posed this question here before, the answer was something like “we don’t know what would trigger major short term releases”. I see no reason to find that point comforting.

    [Response:Not really my aim to offer comfort I’m afraid, but there were no methane spikes during the last interglacial, and I’m not a believer in the Paleocene Eocene as resulting from hydrate release (see Pagani et al., Science 314: 1556-1557 (2006) for the reason why — the C and O isotopes don’t fit methane). David]

  28. 28
    Jim Eager says:

    Mike Roddy, while there is a large frozen-in methane reservoir in the Arctic, and thus the potential for a large influx into the atmosphere, the likely hood of release is tempered by the fact that there was no such release during previous interglacials that were a degree or two warmer than present. However, when—not if—we exceed those slightly warmer temperatures we will be in uncharted territory, and we won’t be able to take so much comfort in the lack of past large scale injections of methane from permafrost and seabed clathrates.

    captcha sums it up: tisfores uneasy

  29. 29
    Hank Roberts says:

    > why should we assume gradual and orderly releases?

    Nobody’s asking you to assume.

    I’d ask you to assess the likelihood;

    what choices made now would increase this feedback, compared to others?.

    Then we avoid the steps that make rapid release begin to seem likely.

    For those arguing that rapid release is imminent — whose plan do you like?

    I read the “methane emergency” folks as favoring “depressurizing” (aka drill and sell) and rolling back the surface sulfate air pollution regulations.

    Seriously — nobody’s asking anyone to assume.

    It would be wise to try to figure out the probabilities.
    Most of what we know is from previous rapid warmings in the paleo record.

    Fire will burn peat, that’s a concern. How much?
    Burned areas will melt permafrost faster, that’s a concern. How much?
    Seabed warming will melt clathrates, that’s known to be a very slow process.
    There’s lots of natural gas easy to drill around the Arctic. Who benefits?

  30. 30
    Geoff Beacon says:

    Surely Allen et. al underestimate the problems we face because of missing feedbacks in the climate models they used. Here is a quote from EU climate policy badly out of date?:

    There are clearly feedbacks missing from the current generation of climate models. Professor John Mitchell has given me 2 examples which are not included in current models (indicating that although these things may be important, they are not always easy to quantify, model, initialize and validate)

    “melting permafrost – we don’t have [CO2 and CH4 emissions from permafrost] in the GCM [global climate model], but have some simple modelling of. Too early to show any results yet, but we plan to publish later this year. Bottom line is that both CH4 and CO2 will be released as permafrost thaws. The magnitude is uncertain, but likely to be significant.”


    “more forest fires – We don’t do yet, but could be important for changing ecosystems response to climate.”

    Now this is 2012. I don’t see how the models used for Allen et. al. could, in 2009, have accounted for these feedbacks and probably others.

    Am I wrong?

    [Response:These are carbon feedbacks, amplifying possibly the amount of carbon that we release. For climate modelers these usually get folded into the uncertainty in our own emissions, which are one of the largest uncertainties in the forecast. Most climate models runs specify the CO2 concentration then predict climate from there. Just as terrestrial uptake can either be subtracted from the overall source, as I did but some of the commenters didn’t, so too can future feedback emissions be sort of folded into the 1000 Gton C. But keep in mind also that today, the natural world is absorbing our fossil carbon, acting as a negative feedback on our climate provocation. It could reverse, turning into a positive feedback on longer time scales, like it did during the glacial / interglacial cycles. David]

  31. 31
    wili says:

    David, thanks for the clarification.
    Jim, thanks for the links.
    Hank, for the record, I am opposed to drilling/selling and to any kind of deregulation. Please do not make assumptions about motives.

  32. 32
    Richard Giles says:

    Sorry but I’m confused by your response to #17. If, for example, we stop all land use changes, how many GtC of CO2 can be emitted while staying within the 1000 GtC? Is it about 700 GtC, which I understand from your article, or as I thought and which #17 estimates, about 425 GtC?

    BTW thanks to you and Ray Pierrehumbert for your excellent textbooks.

    [Response:You are most welcome! If we stopped deforestation, the land surface would become a stronger net sink than it is now. See Tans, Oceanography 22(4) 26-35, 2009 for a excellent review of historical carbon fluxes. I’d guess that any net uptake would be subtracted from Allen’s 1000 Gton C. David]

  33. 33
    wili says:

    What the abstract says:

    “Methane, a potent greenhouse gas, accumulates in subsurface hydrocarbon reservoirs, such as coal beds and natural gas deposits. In the Arctic, permafrost and glaciers form a ‘cryosphere cap’ that traps gas leaking from these reservoirs, restricting flow to the atmosphere. With a carbon store of over 1,200 Pg, the Arctic geologic methane reservoir is large when compared with the global atmospheric methane pool of around 5 Pg. As such, the Earth’s climate is sensitive to the escape of even a small fraction of this methane. Here, we document the release of 14C-depleted methane to the atmosphere from abundant gas seeps concentrated along boundaries of permafrost thaw and receding glaciers…

    Our findings imply that in a warming climate, disintegration of permafrost, glaciers and parts of the polar ice sheets could facilitate the transient expulsion of 14C-depleted methane trapped by the cryosphere cap.”

    None of which I find comforting.

    By the way, I oppose drilling in the Arctic and rolling back any pollution regulations.

    (reCaptcha: man, tontask)

  34. 34
    Ron Manley says:

    From what I have read methane acts a GHG over a narrow wavelength band in the 6 micron range which it shares with nitrous oxide. I know that, relative to CO2, methane is a powerful GHG but surely its narrow wavelenrth band limits its impact. Any clarification would be appreciated.

    [Response:You can see a plot of the outgoing IR light from the Earth, as it gets modified by the greenhouse gases in the atmosphere, here. Methane absorbs off in the fringe, at 1300 cm-1 (wavenumbers, cycles/cm), whereas the peak of Earth’s IR emission is closer to 700 cm-1, where CO2 absorbs. David]

  35. 35
    Geoff Beacon says:

    Will earthquakes increase in frequency and release greenhouse gasses?

    The melting of the Greenland and Antarctic ice-sheets will have two effects:

    1. The weight of the ice will be removed from the poles – and melt water will distribute itself round the world increasing sea-level.

    2. The masses of ice will no longer be pulling up the seas around them by their gravitational pull: The ice-sheets are so massive that they has a significant gravitational pull on the seas around them. See Rising sea levels – but where?

    Both effects mean that weight will be moved from the poles to the equator. More earthquakes will happen as the Earth is squeezed round the equator and released at the poles. Wikipedia has a piece Post glacial rebound

    Post-glacial rebound … produces measurable effects on vertical crustal motion, global sea levels, horizontal crustal motion, gravity field, Earth’s rotational motion and state of stress and earthquakes.

    Could post glacial rebound earthquakes liberate greenhouse gasses?

  36. 36

    #34–Ron M., there is discussion of this point upthread–perhaps you missed it?

    IMO, the most central point is that methane is still relatively scarce. Since the efficacy of any GHG is logarithmic–usually specified as temperature change *per doubling*–the lower the concentration of a given GHG, the less additional gas is required to achieve a specified amount of warming.

    A comparative example: to double CO2 would require the addition of nearly 395 ppm; but methane concentration is on the order of 1 ppm. So 100 GT of the former would have much less effect than 100 of the latter.

    (To be sure, there is still the question of what ‘bite’ each takes out of the spectrum. Dr. Pierrehumbert has previously written on RC that in this regard CO2 is the ‘better’ of the two GHGs, as your comment implies.)

  37. 37
    Geoff Beacon says:


    Thank yu for your comments on my #30.

    Would you care to give your estimate of the “future feedback emissions be sort of folded into the 1000 Gton C”? What is it now and has it changed over the past few years?

    [Response:Today it’s negative, taking up carbon. I don’t know if the forecast for the future has changed, it’s unreliable anyway.]

    Is 1000 Gton C and overestimate or an undersestimate to keep within the 2 degrees C increse?

    [Response:Allen et al say that there is a significant risk of exceeding 20 C if more than 1000 Gton C is released. I don’t have a strong hunch if that’s high or low.]

    I thought one reason for the 2 degrees C limit was to avoid the feedbacks that would cause dangerous climate change. Are you sill happy that 2 degrees C limit or do you think there are multiple feedbacks that might interact to be “dangerous climate change before” that limit?

    [Response:No, profoundly unhappy. :(]

    Have you quantified the chances that the negative carbon feedback “turning into a positive feedback on longer time scales, like it did during the glacial / interglacial cycles”?

    [Response:Nope. Don’t understand the glacial cycles in CO2, and anyway warming is different than cooling. Loose cannon.]

    Should we expect any of these “longer time scales” to be shorter under the current man-made forcing?

    [Response:Get out the magic 8-ball, I’ll tell you. David]

  38. 38
    t_p_hamilton says:

    Ron M.,

    The width of the IR band of methane would spread if there was enough methane to saturate absorption on the center of the band. More importantly it is off the emission maximum, as is the asymmetric CO2 *stretch* ( a very strong absorber/emitter itself). The CO2 *bend* is square in the middle of the IR emission peak of earth.

  39. 39
    Hank Roberts says:

    wili — _what_else_ besides “depressurizing” could you do different to say you’re doing something about a “methane emergency”?

    There are lots of horrible possibilities down several different future paths.

    But the path that _gets_ to where those are more probable — is burning carbon.

    Whatever particular feedback out of the huge list rocks your personal boat — they’re all consequences of burning more carbon.

    If you have no _alternative_behavior_ besides saying it’s going to be bad, what’s the point?

    Use the effort to push away from the notions that lead to paths investing money and time in burning more carbon.

  40. 40
    Hank Roberts says:

    Wili, don’t take this personally. I’m not saying you’re from the “Methane Emergency” group. We had one of their people here earlier, the guy arguing that sulfate pollution was a little local problem, just before the regulations restricting sulfate pollution were tightened.

    I’m just saying — there are lots of emergencies out there in the possible futures, and all of them are feedbacks.

    No matter which feedback event worries you — burning carbon is the path leading to making those feedbacks more likely.

    If you answer to solving the “methane emergency is: stop burning carbon

    Then I agree.

    The answer to many possible problems is: stop burning carbon.
    The answer to how to stop burning carbon is:

    … well? What’s the path that leads toward that future?

    Sketch out what choices need to be made soonest, and what choices made now have the best longterm consequences.

    Those of us in the currently older current generation can die rich and fat leaving the pain to your grandchildren — or we can give up the benefits we got by burning so much carbon so fast and make things better.

    And we’ve got to be wary of the greenwashing crap that says, let’s solve the methane emergency by — burning more carbon immediately.

  41. 41
    Ron Manley says:

    #34 Ron
    Thanks to David and t_p_hamilton.

    So, if I interpret you both correctly, methane does at present occupy a narrow band. How would this band grow in width as it reached saturation? I thought the absorbtion frequency was gas dependant?

    Another question. Why is ozone never, or at least rarely, mentioned? It does not appear to be saturated and is fairly close to the emission maximum?

  42. 42
    wili says:

    “If you answer to solving the “methane emergency is: stop burning carbon

    Then I agree.”

    Yes, we agree on this.

    I got the impression that this site was primarily focused on the science of accurately describing what is happening in the world that relates to global climate. I don’t think pointing out how bad things may be requires me or anyone else to have a nifty answer to getting out of the predicament we are in.

    Generally I get the impression that the hosts do not want long digressions into the politics of nuclear power, for example, here. That can be done on other fora.

    What’s the point? Perhaps it is just my academic bent, but I tend to think of knowledge as its own point.

    Beyond trying to understand our situation as fully as accurately as my small brain can handle, I try to “be the difference I want to see in the world” and through giving up on meat eating, flying, most motorized travel, and a few other life-style choices, I am down to about “one earth” on the site. I also am active in my neighborhood, workplace, municipal, state, national and global levels in raising GW and related concerns.

    All this at considerable risk and detriment to many of my closest relationships and to my professional career.

    So, yes, whatever the situation is, we all need to rapidly move to a much-lower-impact way of living while moving rapidly to a build-out of renewables.

    Yet perhaps it is a personal perversion, but I still want to know the best and most accurate information on where exactly we are, without sugar coating, as things continue to disintegrate, even if I don’t have any easy-to-implement silver bullets up my sleeve to magically ‘solve’ them.
    Part of how we get to the truth is by asking hard questions and probing into what may be the worst consequences, even those that many may be not emotionally equipped to face.

    If nothing else, I think it is important to demonstrate that uncertainties are not always our friend. The general public gets the impression that the ‘debate’ is between those who are in total denial that GW is happening and some watered down, MSM presentation of the already-outdated-at-the-time-of-its-publication last IPCC report.

    People need to know that the actual scientific debate ranges from the more-dire-than-the-latest-IPCC-report to scenarios that are far more catastrophic on a number of fronts.

    Anyway, this is exactly the conversation that strikes me as being considered rather off topic on this site.

    Again, thank you, hank for your many informative comments and links, and thanks again to all for responses to my impertinent questions.

    (And again, for the record, I see pretty much any kind of ‘geo-engineering’ as further evidence of our utter insanity as a species and as a culture–proof that we think of ourselves as gods that can micromanage complex systems we don’t fully understand to derive maximal benefit to ourselves.)

  43. 43
    Chris Colose says:


    It’s a pretty common occurrence that greenhouse gases absorb very strongly in the center of a band, with opacity decaying toward the “wings” of an absorption region.

    If you look at David Archer’s MODTRAN model, you can see this with CO2. Throw 1000 ppm into the model. Notice that there is a huge reduction in outgoing radiation between roughly 600 and 800 cm^-1; but in the very center (say 670 cm^-1), there is a tiny blip where emission increases as CO2 concentration goes up (at fixed temperature). This is even more pronounced if you do 10,000 ppm CO2. This arises because the absorption is so strong in this region that you are seeing emission from the stratosphere, where temperature increases with height (This increase in emission primarily acts to cool the stratosphere). However, toward the edge, the opacity is weaker and emission comes primarily from the troposphere. Continuing to add CO2 will begin to saturate the opaque centers but there will still be room for the wings to eat up radiation. They can gradually become saturated too, but not until concentrations well in excess of anything relevant to Earth. And insofar as temperatures continue to drop with height, adding more GHGs will still warm the surface even in saturated regions.

    Methane works in a similar way, except it is much weaker than CO2. It also has a weak absorption feature shortward of 500 cm^-1 (which I’m not sure is in MODTRAN). I’d expect methane’s peak absorption to begin to saturate near 10 ppm or so based on the model, with plenty of wiggle room to make things warmer after that, but its efficiency does decay pretty rapidly (see Figure 1d of Haqq-Misra et al (2008, Astrobiology) for example. This enters into deep-time climate problems such as the pre-oxygenated Earth or Neoproterozoic, where there is speculation that methane could have been much higher than today, but it gets far too much attention as a “sexy” greenhouse gas that could cause all sorts of catastrophe…in reality it can’t build up to significant concentrations in today’s oxygenated atmosphere, and it doesn’t actually solve much in deep-time either. In anoxic conditions, it’s even worse because photochemical models tend to develop haze layers once the CH4/CO2 concentrations become comparable; this acts like an anti-greenhouse effect and cools the surface (described in my post). Just another reason why I think Dr. Archer is right in that the main thing to consider with methane feedbacks is its influence on the long-term radiative forcing of CO2.

    Ozone isn’t a huge greenhouse gas either on Earth, though it plays a special role in absorbing incoming UV radiation and a lot of astronomy people like it as a possible biosignature when we get spectra of extrasolar planets. By the way, the spike in the ozone band actually works a bit different than that of CO2. Opacity is a minimum here, and the “warm blip” in the emission spectrum is radiation coming from the surface, not from in the stratosphere.

  44. 44
    sidd says:

    Prof. Archer seems to agree that a substantial fraction of Arctic fossil carbon might be released into the atmosphere over a millenium or two. I find it quite unsettling that we may not be in a position, even if actions are taken today, to prevent this. In which case we are committed to 1400 Gton extra fossil carbon into the atmosphere. I would like to sharpen this question, can we put some numbers to the likelihood of such release, _even if all possible action_ is taken today ? This is quite important, for if we cannot prevent, we are looking at almost certainly exceeding 2C rise.

    The difficulty is that arctic methane destabilization is just one of the feedbacks that might result in a worse prognosis than IPCC estimates.


  45. 45
    Unsettled Scientist says:


    I have learned a lot from your course materials and videos online. Unfortunately my humble financial situation has prevented me from picking up the texts on Amazon. But I find your posts and responses to comments illuminating and wanted to say thanks for all your hard work to make this stuff accessible.


  46. 46
    Geoff Beacon says:


    It’s not often I get the urge to argue with you but I’m irritated with sites like because they don’t explain enough about how carbon footprinting should be done so are of limited use in adjusting lifestyles to be “low-carbon” (a frequently misused term).

    I have a stalled project, The Green Ration Book. This set up a mechanism to address the difficulties with carbon footprinting by setting up a citizen’s jury – and publishing our workings. Some of the issues that a conventional footprinting method does not address but which the informed citizen can judge on are:

    1. What is the GWP of methane. (Should the timeframe be decades, making methane 105 times CO2 or centuies making methane 32 times CO2?)
    2. How should embodied carbon be accounted for in products like cars or buildings? A new building can cause the emission of more than 1 tonne of CO2 per square metre (or if you build with purpose actually take CO2 from the atmosphere – build with sustainable wood!)
    3. How should we account for black carbon? How bad is wood smoke for the atmosphere?
    4. How bad are animal products? Which ones are worse? (Clue: don’t eat beef or lamb)
    5. How should recycled material be assessed?
    6. Etc…

    Our newsspace is full of greenwash (sustainable, low-carbon, environmentally friendly…). Sadly I have to admit that I make little progress against this – See some of my other rather sad efforts

    1. No beef
    2. No high buildings
    3. No cars
    4. No bottles
    5. No noplanes

    If anyone wants to [help with/take over] any of these, Green Ration Book included, let me know.

    I believe that informed juries should be set up by governemnts to do official but independent Green Ration Books but the best immediate action governemts can take (when we are all worried about the world economy) is to Tax carbon to create jobs

  47. 47
    Ron Manley says:

    #43 Chris Colose

    Thanks for that Chris.

  48. 48
    Eli Rabett says:

    Ozone can never really build up because it is reactive (although slowly compared to say HO or other free radicals). In the stratosphere, it is destroyed by UV photodissociation. Near the surface reactions on surfaces play an important role.

  49. 49
    D Coyne says:

    Hi David,

    You claim that about 300 billion tons of carbon have been emitted from fossil fuels and that the contribution from land use change can be ignored because of the uptake of carbon by the land and ocean. I have just re-read the paper by Allen at al (2009) and they account for carbon uptake when modeling carbon emissions. Thus the claim that land use change can be ignored is incorrect, the model considers total carbon emissions from fossil fuels, cement production and land use change. Up to 2006, about 500 billion metric tons of carbon had been emitted (this is stated in the abstract of the Allen paper).

    “Total anthropogenic emissions of one trillion tonnes of carbon (3.67 trillion tonnes of CO2), about half of which has already been emitted since industrialization began, results in a most likely peak carbon dioxide induced warming of 2 6C above pre-industrial temperatures, with a 5–95% confidence interval of 1.3–3.9 C.” (Allen, 2009)

    See also

    [Response:OK, I stand corrected. I guess since I don’t have much faith in forecasting natural land uptake in the future, I have a tendency to just leave land changes out, since they wash so far. Head in the sand I suppose. David]

  50. 50
    John E. Pearson says:

    42: Willi wrote ” I see pretty much any kind of ‘geo-engineering’ as further evidence of our utter insanity as a species and as a culture–proof that we think of ourselves as gods that can micromanage complex systems we don’t fully understand to derive maximal benefit to ourselves.”

    I see geo-engineering as the only option that we are leaving to our grandchildren. Despite your personal sacrifices to reduce your carbon footprint we are doing approximately nothing to reduce CO2 emissions. Burying CO2 strikes me as quite sane.