A Few Things Ill Considered

From up north, we have some more troubling news. Actually very troubling. Catastophic release of methane hydrates is a prime suspect in a few events dramatic enough to show in the earth’s geological records, coarse and obscured as that record may be. (Our actions today will be featured prominently in that record for anyone looking back a million years from now.) It has been a worry for many years that humanity is running the risk of triggering such a release again, which would truly pile disaster on top of calamity.

New research coming out in Science today indicates that this most dire of feedbacks may well be underway already. Below is the text of a press release I received about it last night.

Fairbanks, Alaska–A section of the Arctic Ocean seafloor that holds vast stores of frozen methane is showing signs of instability and widespread venting of the powerful greenhouse gas, according to the findings of an international research team led by University of Alaska Fairbanks scientists Natalia Shakhova and Igor Semiletov.

The research results, published in the March 5 edition of the journal Science, show that the permafrost under the East Siberian Arctic Shelf, long thought to be an impermeable barrier sealing in methane, is perforated and is leaking large amounts of methane into the atmosphere. Release of even a fraction of the methane stored in the shelf could trigger abrupt climate warming.

“The amount of methane currently coming out of the East Siberian Arctic Shelf is comparable to the amount coming out of the entire world’s oceans,” said Shakhova, a researcher at UAF’s International Arctic Research Center. “Subsea permafrost is losing its ability to be an impermeable cap.”

Methane is a greenhouse gas more than 30 times more potent than carbon dioxide. It is released from previously frozen soils in two ways. When the organic material–which contains carbon–stored in permafrost thaws, it begins to decompose and, under oxygen-free conditions, gradually release methane. Methane can also be stored in the seabed as methane gas or methane hydrates and then released as subsea permafrost thaws. These releases can be larger and more abrupt than those that result from decomposition.

The East Siberian Arctic Shelf is a methane-rich area that encompasses more than 2 million square kilometers of seafloor in the Arctic Ocean. It is more than three times as large as the nearby Siberian wetlands, which have been considered the primary Northern Hemisphere source of atmospheric methane. Shakhova’s research results show that the East Siberian Arctic Shelf is already a significant methane source: 7 teragrams yearly, which is equal to the amount of methane emitted from the rest of the ocean. A teragram is equal to about 1.1 million tons.

“Our concern is that the subsea permafrost has been showing signs of destabilization already,” she said. “If it further destabilizes, the methane emissions may not be teragrams, it would be significantly larger.”

Shakhova notes that Earth’s geological record indicates that atmospheric methane concentrations have varied between about .3 to .4 parts per million during cold periods to .6 to .7 parts per million during warm periods. Current average methane concentrations in the Arctic average about 1.85 parts per million, the highest in 400,000 years, she said. Concentrations above the East Siberian Arctic Shelf are even higher.

The East Siberian Arctic Shelf is a relative frontier in methane studies. The shelf is shallow, 50 meters or less in depth, which means it has been alternately submerged or terrestrial, depending on sea levels throughout Earth’s history. During Earth’s coldest periods, it is a frozen arctic coastal plain, and does not release methane. As the planet warms and sea levels rise, it is inundated with seawater, which is 12-15 degrees warmer than the average air temperature.

“It was thought that seawater kept the East Siberian Arctic Shelf permafrost frozen,” Shakhova said. “Nobody considered this huge area.”

Earlier studies in Siberia focused on methane escaping from thawing terrestrial permafrost. Semiletov’s work during the 1990s showed, among other things, that the amount of methane being emitted from terrestrial sources decreased at higher latitudes. But those studies stopped at the coast. Starting in the fall of 2003, Shakhova, Semiletov and the rest of their team took the studies offshore. From 2003 through 2008, they took annual research cruises throughout the shelf and sampled seawater at various depths and the air 10 meters above the ocean. In September 2006, they flew a helicopter over the same area, taking air samples at up to 2,000 meters in the atmosphere. In April 2007, they conducted a winter expedition on the sea ice.

They found that more than 80 percent of the deep water and greater than half of surface water had methane levels more than eight times that of normal seawater. In some areas, the saturation levels reached at least 250 times that of background levels in the summer and 1,400 times higher in the winter.

They found corresponding results in the air directly above the ocean surface. Methane levels were elevated overall and the seascape was dotted with more than 100 hotspots. This, combined with winter expedition results that found methane gas trapped under and in the sea ice, showed the team that the methane was not only being dissolved in the water, it was bubbling out into the atmosphere.

These findings were further confirmed when Shakhova and her colleagues sampled methane levels at higher elevations. Methane levels throughout the Arctic are usually 8 to 10 percent higher than the global baseline. When they flew over the shelf, they found methane at levels another 5 to 10 percent higher than the already elevated arctic levels.

The East Siberian Arctic Shelf, in addition to holding large stores of frozen methane, is more of a concern because it is so shallow. In deep water, methane gas oxidizes into carbon dioxide before it reaches the surface. In the shallows of the East Siberian Arctic Shelf, methane simply doesn’t have enough time to oxidize, which means more of it escapes into the atmosphere. That, combined with the sheer amount of methane in the region, could add a previously uncalculated variable to climate models.

“The release to the atmosphere of only one percent of the methane assumed to be stored in shallow hydrate deposits might alter the current atmospheric burden of methane up to 3 to 4 times,” Shakhova said. “The climatic consequences of this are hard to predict.”
Shakhova, Semiletov and collaborators from 12 institutions in five countries plan to continue their studies in the region, tracking the source of the methane emissions and drilling into the seafloor in an effort to estimate how much methane is stored there.

Shakhova and Semiletov hold joint appointments with the Pacific Oceanological Institute, part of the Far Eastern Branch of the Russian Academy of Sciences. Their collaborators on this paper include Anatoly Salyuk, Vladimir Joussupov and Denis Kosmach, all of the Pacific Oceanological Institute, and Orjan Gustafsson of Stockholm University.


  1. #1 GFW
    March 5, 2010

    Is there a graph of atmospheric methane concentration over time? I realize CH4 isn’t as well mixed as CO2, but as long as the methodology is consistent and described, it would really help understand whether the current arctic releases are “normal” or not.

  2. #2 Joseph
    March 5, 2010

    Methane has been rising recently even though for a number of years levels were stable.


    The stabilization of methane must have had some effect on temperature trends, BTW. This is seldom considered.

  3. #3 canbanjo
    March 5, 2010

    Do you know how / whether this information changes / will change the current climate model predictions? Is this really big news, or is it a rather sensationalist press release?

  4. #4 Hey! Get This . . .
    March 5, 2010

    This had to happen.

    Natural sequestration of methane would necessarily include a portion of stores held near long-term equilibrium. Nature does not employ ‘margining’ of structural & material properties as Engineers would. If it could freeze at 10milliKelvin below long-term equilibrium, it did.

    Long-term equilibrium is being re-established, including the activation of feedback mechanisms. Too bad for superstitious humans and their meager cohort of rationalists: it’s positive feedback.

  5. #5 pft
    March 6, 2010

    We have no idea where all the methane released to the atmosphere comes from. Methane oxidizes fairly quickly, within 10 years, to CO2. Doubling methane from current values of 1.8 ppm would be the equivalent of adding 60 ppm CO2 (about 0.2 deg C w/o feedbacks (which can be positive AND negative). There has been relatively little increase in atmospheric methane over 15 years. Methane started increasing in 1750 which was near the end of the Little Ice Age.

    Methane bubbles out of the surface in many locations. Before we started using oil and gas for energy, oil and gas just seeped out of the land. There is still quite a bit of seepage in the oceans, but we have no idea how much since it’s hard to measure.

    For all we know, this seepage in this small area of the Arctic (relative to global surface area) has been going on off and on for the past 10,000 -20,000 years, since the end of the last ice age. We have had warmer periods in the Arctic, including the 1930′s , MWP, etc. but it’s not clear this seepage is even due to warming.

    And what does this mean anyways “”It was thought that seawater kept the East Siberian Arctic Shelf permafrost frozen,” Shakhova said. “Nobody considered this huge area.”

    What keeps the permafrost/methane frozen is low temperatures below freezing and/or high pressure which causes methane to form hydrates even at warmer temperatures (up to 18 deg C). However, 40-50 meters is not enough pressure for methane to form hydrates at warmer temperatures. Permafrost and hydrated sediment does have a low thermal conductivity so it probably takes a long time for any warming to penetrate. If there are hydrates, and hydrates are melting, the reaction is endothermic, so this will help slow any warming as well. However, since the end of the last ice age, we know land that was previously under glaciers has been rising since it is free from the weight of the glaciers, this would have the effect of reducing the depth (and pressure) on any hydrates formed long ago.

    Any methane being released at such shallow areas may very well be seepage (nothing in the article rules this out). It is not like there is not a lot of oil and gas in the Arctic. Russia has no clear idea as to the reserve potential of East Siberia as it has been relatively unexplored, but since the Western Siberian fields are being depleted they will be looking East.

  6. #6 Joseph
    March 6, 2010

    Methane started increasing in 1750 which was near the end of the Little Ice Age.

    It actually started to increase abruptly in the 1790s, when the industrial revolution broke. I don’t have a graph of it online, but I have a graph of CO2 (here) that is similar. Imagine it like that, but add a slight upward slope.

    It’s incomprehensible to me that people can look at the exponential rise in methane in CO2 that occurred during the industrial era, and still deny we are responsible for that rise.

  7. #8 yogi-one
    March 6, 2010

    This news doesn’t bode well.

    It would have been nice not to put humanity through a mass extinction event in the 21st century, but we lost that opportunity already. It’s all damage control now, baby.

    The only question left is how bad are we going to let it get before we a. admit that it’s happening, and b. decide to co-operate globally to do something to mitigate potentially catastrophic consequences.

    This will be the 21st century equivalent of WWII in terms of the impact it has on global humanity, and just as in WWII, horrors resulting in hundreds of millions (possibly billions) of deaths and unprecented environmental effects will leave a lasting imprint on civilization’s psyche.

    Just as in WWII, we will be forced to recognize the ways in which we kill ourselves, and be forced, eventually, to grapple with the new level of responsibility it brings us.

    God didn’t cause this and God won’t fix it. We’re in the driver’s seat, drunk on oil and thrilled by our speed.

    The hangover is going to be a bitch.

  8. #9 maxwell
    March 7, 2010


    maybe you could put in an update with the link that dhogaza has provided. If RealClimate.org is the authority on these matters, then I think you ought to try to include the analysis of why they think this isn’t a big deal.

  9. #10 skip
    March 7, 2010

    That was interesting. Thanks, Dho.

  10. #11 Joseph
    March 7, 2010

    What RealClimate.org are arguing is that this is like — to use a different analogy to the one they used — having cancer and then getting AIDS on top of that.

  11. #12 maxwell
    March 7, 2010


    from the RealClimate.org post

    ‘Is now the time to get frightened?



    ‘Methane sells newspapers, but it’s not the big story…’

    and, most importantly,

    ‘What’s missing from these studies themselves is evidence that the Siberian shelf degassing is new, a climate feedback, rather than simply nature-as-usual, driven by the retreat of submerged permafrost left over from the last ice age.’

    I think the analogy of cancer on top of AIDS has to do with a catastrophic methane, which they seem to believe is not what is happening in the situation in this paper. In fact, David clearly states,

    ‘The Siberian Margin might rival the whole rest of the world ocean as a methane source, but the ocean source overall is much smaller than the land source…The ocean is small potatoes, and there is enough uncertainty in the methane budget to accommodate adjustments in the sources without too much overturning of apple carts.’

    It’s important to focus on the heart of the argument rather than taking the most extreme portions of it. David is clearly making the argument that this paper is not evidence for some catastrophic event that is unfolding before us. Using your analogy, we still just have AIDS.

  12. #13 Jack Savage
    March 7, 2010

    I could not figure out from the Press Release whether it was thought this was a new process or if it has been occurring all the/for some time.
    Surely that must be established before we start worrying too much about this?

  13. #14 mandas
    March 8, 2010


    I absolutely LOVE the selectivity and sheer hypocrisy of the deniliast movement when it comes to constructing an argument. Firstly, to attempt to appeal to RealClimate (post #9) as the experts on an issue, because you think they support your denialist position, is absolutely priceless, because we all know RealClimate is well known for debunking denialist propoganda, and to accept them as the ‘experts’ on a climate change is strange in the extreme. Please tell me that you accept them as the experts on all things to do with climate change, and not just when it appears that they support your position? Are you going to quote them on every issue?

    And I am totally confused about when and where it is appropriate to ‘appeal to authority’. In post #9, you appear to be appealing to the authority of RealClimate to support your view, as follows:

    “….maybe you could put in an update with the link that dhogaza has provided. If RealClimate.org is the authority on these matters, then I think you ought to try to include the analysis of why they think this isn’t a big deal….”

    However, in post # 46 of ‘A chilling effect on a warming theory’, you state:

    “…. The point is that making an argument pandering to the fact that ‘experts’ said such and such, therefore it must be true is first rate example of how not to make an argument. It is known as argument from authority or argumentum ad verecundiam. Don’t you love Latin?…”

    So I am confused. Should I appeal to authority – or as you say in Latin argumentum ad verecundiam (argument to respect) – to support my argument, or should I only do that when the authority says something that supports my own world view?

  14. #15 mandas
    March 8, 2010


    Oh – and just a little tip. Try to remember who’s blog you are posting on. ‘James’ is over at “The Island of Doubt”. Our host here is coby. It’s a bit like calling out the wrong woman’s name when you are cumming. Never a good thing.

  15. #16 Jack Savage
    March 8, 2010

    I am not sure Maxwell is the only one tying himself in knots. Realclimate has become a bit of a propaganda machine itself lately,spending far too much time debunking the tiny minority of “deniers” or whatever you like to call them.
    It is supposed to be a reliable source of carefully researched climate science and as such can surely be drawn upon by any side of the argument as an appeal to the science rather than an appeal to authority.
    I do not see any difficulty with Maxwell agreeing with Realclimate on the issue and Mandas disagreeing (if indeed he does, he has not said!) . Mandas should no more taunt Maxwell now to agree with everything on Realclimate than if Maxwell should now taunt Mandas now to disagree with everything on Realclimate. It really does not move a debate forward.

  16. #17 mandas
    March 8, 2010

    ‘mandas’ is neither agreeing or disagreeing with what is said over at RealCimate. But he (I) do have an issue with people selectively quoting from them and holding them up as ‘experts’ on one issue, when they patently disagree with them on just about everything else. And I do have an issue with people appealing to authority when it suits them, but castigating others for similar appeals to authority.

    I don’t like hypocrites – pure and simple.

  17. #18 Dappledwater
    March 8, 2010

    ‘deniers” or whatever you like to call them” – J Savage


  18. #19 coby
    March 8, 2010

    RealClimate are being correctly cautious especially about whether this is anything new or not. So I guess it is apparent that all the accusations of “alarmists” were misguided after all.

    However, I find maxwell’s quote pretty ironic to say the least! He quoted RC:
    “Is now the time to get frightened? No.” Okay, but why no? They go on: “Imagine you are in a Toyota on the highway at 60 miles per hour approaching stopped traffic, and you find that the brake pedal is broken. This is CO2. Then you figure out that the accelerator has also jammed, so that by the time you hit the truck in front of you, you will be going 90 miles per hour instead of 60. This is methane. Is now the time to get worried? No, you should already have been worried by the broken brake pedal.”

    Seems to have just a little bit different message than maxwell would have us believe.

  19. #20 skip
    March 8, 2010

    I think, in Max’s defense, he simply meant the article denies that we have to worry about methane in especially.

  20. #21 maxwell
    March 8, 2010

    mandas, I confused Joseph with James. Sorry about that.

    As for hypocrites, I am merely pointing out to Coby that if RealClimate can be cited in the meat and potatoes of this post about the concerns over methane as a greenhouse gas, then they should be properly cited explaining why the nature of THIS PARTICULAR ARTICLE is not a sufficient reason to get all up in arms. I’m holding him to his standard. Since I have never mentioned anything disparaging about the science the RealClimate brings to the table, I really don’t know what you’re talking about when it comes to my hypocrisy.

    If you have a problem with the citations I provide, then I think you’d better provide better citations of the OVERALL ARGUMENT THAT PARTICULAR POST ON REALCLIMATE IS TRYING TO MAKE! Merely pointing and yelling is not really an argument. Calling me names is not really an argument. Pointing out that I confused two people’s names is not an argument. If you think that post on RealClimate is making another argument, please cite it.

    Coby, the same goes for you.

    It is obvious even to a child that David, in that quote, is discussing the prospects of methane released on a catastrophic scale. This is because methane only lasts in the atmosphere about a decade, as David also points out, while CO2 has a longer lifetime. But in relation to the rest of the methane release on the planet, the vent featured IN THIS ARTICLE is ‘small potatoes’.

    No amount of taking quotes out of context can dispute that this is the conclusion of that particular RealClimate piece.

    But that’s quite an argument you guys are making. Because I mix up names, hold people to their standards and actually know how to make an cogent argument means that I must be wrong. You’re really making progress.

  21. #22 mandas
    March 8, 2010


    Thanks for validating my views regarding your hypocrisy (and I was making that point about your double standards when it comes to appeals to authority, not about your citations). You point out to me that ‘calling you names is not really an argument’, then you go ahead say this about coby:

    “…Coby, the same goes for you. It is obvious even to a child…”

    So I guess it is ok to call someone names if you are doing it, but not ok if someone does it to you. I’m pretty certain that fits the definition of hypocrisy pretty well.

  22. #23 jcrabb
    March 8, 2010

    There is evidence to show the current degassing is a recent occurance, starting in 2003

    From ‘The Independant’,

    “The preliminary findings of the International Siberian Shelf Study 2008, being prepared for publication by the American Geophysical Union, are being overseen by Igor Semiletov of the Far-Eastern branch of the Russian Academy of Sciences. Since 1994, he has led about 10 expeditions in the Laptev Sea but during the 1990s he did not detect any elevated levels of methane. However, since 2003 he reported a rising number of methane “hotspots”, which have now been confirmed using more sensitive instruments on board the Jacob Smirnitskyi.”

  23. #24 Joseph
    March 8, 2010

    I’ve produced a graph of the methane concentration from 1008 to 1993, based on the Law Dome reconstruction made available by Etheridge et al. (1998). Feel free to use.

    I don’t doubt David over at RealClimate.org is right that methane does not stay in the atmosphere as long as CO2, but it’s clear that some of it stays.

    There is a mechanism to increase the methane concentration in the atmosphere: Human civilization.

  24. #25 jcrabb
    March 8, 2010

    The ESAS contains an estimated 540 billion tonnes of Methane, as it has started venting due to ‘melting’ of it’s permafrost lid, it seems reasonable to expect it to continue venting, barring refreezing, releasing billions of tonnes of Methane. Seems reasonable to expect that as warming continues, the rate of venting will increase.

    Even though Methane has a short lifespan in the atmosphere, it still degrades into CO2, so either way the result is a significant increase in GHG’s, hence I don’t quite understand the nonchalance towards this new, massive source of GHG’s that we have no control over. To me this seems more like a tipping point, permafrost collapse was not expected for some time yet, and it was posed as a tipping point.

  25. #26 Juice
    March 9, 2010

    Joseph, you might want to change the left axis to ppb.

  26. #27 Joseph
    March 9, 2010

    @Juice: Good catch. I’ve fixed it.

  27. #28 maxwell
    March 9, 2010


    how does your reconstruction take the ‘divergence problem’ into account?

  28. #29 Joseph
    March 9, 2010

    how does your reconstruction take the ‘divergence problem’ into account?

    Are you referring to the tree-ring divergence problem, maxwell? That’s for temperature reconstructions. The Etheridge et al. (1998) methane reconstruction is from gas trapped in ice-cores, obtained at the Law Dome.

    There’s also instrumental data for methane, but it doesn’t go back too far, of course.

  29. #30 maxwell
    March 9, 2010


    I saw there was a reference to ice cores in the link you had on your first comment on the reconstruction. There was also a mention of ‘firn air’ and other air born measurements, but I’m not sure what they mean.

    I was more interested in a general divergence in correlation between these other proxies and observational data for the reconstruction you’ve provided.


  30. #31 GFW
    March 9, 2010

    Joseph’s graph may be a “reconstruction”, but it’s definitely not a proxy reconstruction. The modern instrumental measurements are of actual methane in the air. The ice core measurements are of methane trapped in air bubbles in the ice. Ice core dates are solid. I would have to read the original papers to see what steps they take to avoid contamination, but ice cores from multiple locations agree on the pre-industrial value and the timing of the increase. Finally, the ice core data matches the direct instrumental data in the period of overlap. The only thing to add is that between the end of Joseph’s graph (1993) and present, the methane concentration has risen to about 1785 ppb.

    (A proxy reconstruction measures one thing as a proxy for another (like tree ring widths for temperature or other factor that limits tree growth).)

  31. #32 maxwell
    March 9, 2010

    GFW, thanks for the update. I will quibble slightly, however, over the language.

    Ice cores measurements are proxies for the amount of any gas in the atmosphere at the time of freezing. One is able to measure the actually gas concentrations in such cores, but the question remains of whether this is actually a measurement of the past atmosphere. It seems this is true for ice core measurements, but this fact does not mean they are not proxies.

  32. #33 jcrabb
    March 9, 2010

    As atmospheric Methane has not changed significantly over the last thousand years, prior to the current period, isn’t it fair to say that current temps. are unprecedented. If there were comparable temps. then similar substantial increase of Methane emissions from Wetlands etc, that are currently occuring, should be seen around the MWP, whenever that was.

  33. #34 maxwell
    March 9, 2010

    jcrabb, your conclusion seems to contradict Joeseph’s. If humans are responsible for the rise in methane we are seeing currently, then that could account for it and past warm periods may or may not have had similar amounts of methane to today. There would be no way to make a correlation because we know that the warmth is not causing it now.

    That said, I will again point out that the venting this article describes is a very small component to the overall methane budget and even if such things are common during warmer periods, they may not affect past readings for atmospheric concentrations very much at all.

  34. #35 jcrabb
    March 9, 2010

    maxwell, around 50% of the current atmospheric Methane level is due to warm wetlands emitting more Methane.

    So higher temps are proven to raise Atmospheric Methane levels.

    As there is no natural Methane sequestration process, then increased Methane emissions due to increased temps. should show up in the historical record.

  35. #36 maxwell
    March 9, 2010

    jcrabb, do you have a source for your number? This site says that a quarter of methane comes from wetlands.


    That seems like an awfully hard thing to measure, however.

    And there is a natural sequestration mechanism. It’s called oxidation chemistry.

    From Wikipedia’s entry on methane,

    ‘Methane is a relatively potent greenhouse gas with a high global warming potential of 72 (averaged over 20 years) or 25 (averaged over 100 years).[2] Methane in the atmosphere is eventually oxidized, producing carbon dioxide and water. As a result, methane in the atmosphere has a half life of seven years.’

    If reconstructions do not have sufficient time resolution of time scales smaller than 100 years, then whatever rises in concentrations that have occurred could be completely missed.

    Joeseph, do you know how the ice core data deals with this fact? It looks like the x-axis has units of 30 years or just over 4 half-lives. Is that correct? If it is, almost 95% of methane released just after one time point would have oxidized by the next time point.

  36. #37 Dappledwater
    March 9, 2010

    Maxwell, the ice core data is capable of resolution down to annual timescales due to the seasonal deposition of ice, of course not all of the ice core sample is so obliging.

  37. #38 jcrabb
    March 10, 2010

    maxwell, my mistake, your figure is the right one, this amount is still a significant amount.

    As the current warming of wetlands has created a net increase of naturally sourced Methane, seems reasonable to expect the ‘MWP’ to have a noticable increase.

  38. #39 Matt Bennett
    March 10, 2010


    I think you’re rather confused with the ice core measurements. They are decidedly not proxies as nothing is ‘standing-in’ for anything else. As has been pointed out to you, they are direct measurements of tiny ‘fossil’ samples of Earth’s past atmospheric gas concentrations (ie trapped air bubbles) – take your pick which. What needs to be adjusted for is the time taken for freshly fallen snow (mostly air) to compact into solid ice (hardly any air). The loosely packed intermediate is known as “firn” which you asked about. The time taken is anything up to 180-200 years (often much less) and the depth at which this point is reached varies quite a bit. Both are dependent on many considerations such as local snow fall, summer temps, wind patterns etc – all this having been accounted for by scientists in any given location. Once ice bubble isolation or ‘seal-off’ is complete, you can, for many tens to hundreds of metres below this level, have interannual resolution of gas concentrations with great accuracy. Lower down than this however, compaction and ice-flow dictate that modelling is needed as resolution goes out to decades/centuries – but never-the-less it can and has been done for many cores, providing definitive information on past atmospheric composition. The point being there is NO DOUBT about these measurements – the methane concentration HAS done exactly what Joseph’s graph so ably demonstrates.

    Then there’s the little matter of CO2…..

  39. #40 Joseph
    March 10, 2010

    If reconstructions do not have sufficient time resolution of time scales smaller than 100 years, then whatever rises in concentrations that have occurred could be completely missed.

    Joeseph, do you know how the ice core data deals with this fact?

    I’m quite sure that simplification is not quite right, maxwell. It depends. For example, a 400,000-year Vostok reconstruction has low resolution. This further depends on what the reconstruction is. For example, the temperature reconstruction from Petit et al. has pretty good resolution (maybe about 50 years initially, and about 500 years as it approaches 400KYR BP.) The CO2 reconstruction from Vostok I’ve looked at has lower resolution (maybe in the order of 1,000 years initially, and approaching 5,000 years toward the end.)

    A Law Dome reconstruction is a completely different thing, however. The raw CO2 data from Etheridge et al. (1998) apparently has a resolution of about 1 to 3 years in the 1960s. That increases, of course, with older data.

    Note that resolution and dating error are not the same thing, but again, more recent data will have less dating uncertainty than old data.

  40. #41 maxwell
    March 10, 2010


    how can scientists properly account for past local conditions that vary for a variety of subtly different reasons? That is, how can someone tell from looking at a particular snow pack that variations in depth are due to specific different reasons you have cited in your comment?

    I still think that in a strict sense of the word ‘proxy’, since scientists are not measuring the actual atmosphere of the past, still works. I think that the calibration techniques (accounting for local effects for example) for such measurements are pretty good indication of this fact. The language of the literature might be different. I am merely stating an opinion, however. If you disagree, that’s great.


    the context of my comment has to do with being able to resolve any kind of drastic increase in methane in the past that would have been present due to past warming. I’m not sure what CO2 reconstruction have to do with such things other than serving as a proxy measurement for methane I would assume.

    But it seems from what you are telling me is that in the past, time resolution is not on the scale able to see sharp rises in methane because the half-life is so short. I was trying to determine this because jcrabb was making an argument about how we could interpret the rise in methane you show to be correlated with a rise in temperature that is unprecedented rather than your theory of human activity.

    It seems hard to make either argument if the ability to resolve changes in methane concentrations can only capture 5-10% of the variation. I don’t know that for sure, but if it is true, then most of this is moot.

  41. #42 Joseph
    March 10, 2010

    I’m not sure what CO2 reconstruction have to do with such things other than serving as a proxy measurement for methane I would assume.

    CO2 is what I was looking at while typing. A methane reconstruction from the Law Dome will also have pretty good resolution.

    But it seems from what you are telling me is that in the past, time resolution is not on the scale able to see sharp rises in methane because the half-life is so short.

    That’s right to some extent, but we do have data with resolution below the century scale going back at least 1000 years, so we know that a doubling in one century is not normally what goes on, and we also know that the last 400,000-year concentration range for Methane is normally between 340 ppb and 700 ppb, give or take. The close-to-1800 ppb level we’re at now is most definitely not normal.

  42. #43 maxwell
    March 10, 2010


    hold on a second. I didn’t mean to upset you. I was confused about the CO2 reference.

    If we’re only talking about a factor of three or four times the average methane concentrations, that amount of methane oxidizes out of the atmosphere in 40 years, pretty much, as I have already stated. So any resolution less than that is going to make it hard to say anything about the past. Do you agree?

  43. #44 Joseph
    March 10, 2010

    @maxwell: That time I was not upset; not sure what gave that impression.

    So any resolution less than that is going to make it hard to say anything about the past. Do you agree?

    I do not. For one, a rise in the concentration is not going to be instantaneous most likely. So it’s not just the 40 years you claim it takes for levels to get back down to normal (and I can’t say whether that figure is correct.)

    Additionally, in a methane reconstruction with almost 500 data points, like the available Vostok-based reconstruction, you might expect to see at least one data point with a concentration higher than 700 ppb, if periods of high concentration like the present are not exceedingly rare. Just by chance you should be able to hit one such period of high concentration when they analyze the ice core, don’t you think?

    Finally, the Etheridge et al. Law Dome reconstruction has data with resolution below 40 years starting in 1448.

  44. #45 maxwell
    March 10, 2010


    the figure 40 years comes from the half-life I found from Wikipedia entry for methane (maybe not the best source, I’ll look for a better one).

    The half-life means the amount of time it takes for half of the molecules of methane to be oxidized to carbon dioxide and water.

    So if we start out at 1800 ppb, in 7 years the concentration will, on average, be 900 ppb. After another 7 years, the concentration will fall to 450 ppb. So actually, you only need 14 years to go from current levels to levels that might be considered ‘normal’ under average oxidizing conditions, ie normal hydroxide ion concentrations.

    I’m not sure about normal rise times for concentrations of methane, nor do I know much about fall times in terms of the processes that serve as sources for methane. They are most likely highly variable.

    As far as ice cores go, however, I would wonder what kinds of conditions lead to ice cores that are easier to read and more likely to end up in reconstructions. If the probability of an ice core being useful for a concentration reconstruction go up as the temperature goes down, they might be heading in the wrong direction to be able to sample high methane concentrations, which might go up as temperature goes up. This is mere speculation, but it seems physically intuitive to me at least.

    That may explain why the reconstructions from ice cores do not show these ‘spikes’ in concentrations, if they did indeed happen.

    For a point of reference, I agree that humans are likely the cause for the levels of methane we see today. I’m just confused about how confident you are about it.


  45. #46 crakar
    March 10, 2010

    Just a question, is Siberia in the Arctic? If so then when you consider the latest sea ice extent shows we are over the 14.5 mil sq kilometer march and cryosphere has us at only -0.278 MSK below the anomoly then why would the methane be venting? Lets say it is then what is causing it?

  46. #47 jcrabb
    March 10, 2010

    Does anybody have a possible reason why if the ‘MWP’ was a global occurance there would not be a significant rise in ‘natural’ Methane, as is occuring today? :-)

  47. #48 GFW
    March 10, 2010

    1. It’s not yet clear that the methane venting is greater than “normal” for the region.
    2. The measurements described in the paper in question were taken over a number of recent years, not last month.
    3. *If* the venting is greater than normal *and* it’s ongoing even in this season of near-normal ice extent, then I suspect the mechanism is a warming of the currents that flow into and through the arctic. Just as the deep water in the tropics is still a frigid 3C or so, the deep water in the arctic can be a bit warmer than the surface (which is clearly below zero when new ice is forming). The continental shelf isn’t really “deep” water, but it’s mostly below the depth of quickly mixed surface water. Call it “upper deep”. :-)

    1. Same point 1. as to Crakar. If the current venting is relatively normal, then it probably happened during the MWP too. So far, the total venting from the oceans does not add up to anywhere near what is produced from rice or cattle agriculture.
    2. Although there may have been localized areas (e.g. southern Greenland) that were warmer in the MWP than now, it does not appear that the northern hemisphere as a whole was warmer than now (warmer than before or after the MWP, but likely 0.5C cooler than today). So while there’s no proof, one could speculate that the deep currents didn’t warm as much.

  48. #49 Joseph
    March 10, 2010

    So if we start out at 1800 ppb, in 7 years the concentration will, on average, be 900 ppb. After another 7 years, the concentration will fall to 450 ppb.

    @maxwell: No, that’s not right. If it were, the concentration would normally tend to be zero.

    The half-life concept applies to the ‘excess’ concentration. This assumes there’s an equilibrium level, and anything above that level will tend to be reclaimed. The bigger the ‘excess’, the faster it’s reclaimed.

    (Reality probably doesn’t work quite that way in all cases.)

    For a point of reference, I agree that humans are likely the cause for the levels of methane we see today. I’m just confused about how confident you are about it.

    No need to be confused. It looks damn near certain.

  49. #50 jcrabb
    March 11, 2010

    In 2008 Shakhova wrote,

    The total area of submarine permafrost within the Siberian Arctic shelf is estimated to be more than one and half million square kilometers. Amount of methane hydrate deposited beneath and/or within submarine relic permafrost is estimated to be at least 540 Gt. Amount of free gas, accumulated beneath the hydrate deposits, is expected to be about 2/3 of the amount of hydrates or 360 Gt. Additionally as much as 500 Gt of carbon could be stored within as minimum as a 25 m-thick permafrost body of this type. The total value of ESS carbon pool is, thus, not less than 1,400 Gt of carbon. Since the area of geological disjunctives (fault zones, tectonically and seismically active areas) within the Siberian Arctic shelf composes not less than 1-2% of the total area and area of open taliks (area of melt through permafrost), acting as a pathway for methane escape within the Siberian Arctic shelf reaches up to 5-10% of the total area, we consider release of up to 50 Gt of predicted amount of hydrate storage as highly possible for abrupt release at any time. That may cause 12-times increase of modern atmospheric methane burden with consequent catastrophic greenhouse warming.

  50. #51 Dappledwater
    March 11, 2010


    Richard Alley, an ice core expert and always entertaining to watch.

  51. #52 Matt Bennett
    March 12, 2010


    The ice cores and the “closure times” of the bubbles they contain are easily calibrated and dated using various methods (you can read up on this, just Google it) Scientists can use interannual banding by physically counting layers, they can compare them with ocean core sediments from nearby, they can use the oxygen isotope ratios (16-18), they can look for confirmation markers such as volcanic residue to calibrate parts of a core to the exact year, they can radiodate gaseous inclusions, they can model ice flows quite well and on….. I’ve just looked up for you the extremes of this snow-firn-ice transformation process. In the interior of the East Antarctic ice sheet, with little accumulation and no melt, it’s 3500 years. On Alaska’s Seward glacier, with about coastal precipitation, it’s achieved in 3-5 years.

    This is all VERY well established stuff and is not where any of the uncertainty lies – as I said, if you were genuinely skeptical and interested in this, it is discoverable in about 45 seconds and is simply not a source of contention. However….

    Your statement: “I still think that in a strict sense of the word ‘proxy’, since scientists are not measuring the actual atmosphere of the past, still works” is DEAD WRONG and THAT is the main point I was trying to get across to you. You were questioning the accuracy of the methane (and other) graphs and claiming them as ‘proxies’ which is a non-trivial error. I’ll say it again, they are NOT proxies. You ARE, indeed “measuring the actual atmosphere of the past” directly. That is the exquisite beauty of this.

    As a comparison, to make it absolutely clear. Pretend you found a beautiful piece of Baltic amber while strolling along a lonely shoreline. You polished it up and found inside a perfectly preserved honey bee, pollen sacs overflowing, lying alongside a couple of prehistoric ants. You date the amber to 5.8 million years, with a known error of +/- 60,000 years. You do not turn around and say, “well, I suspect that living in the forests of this area during the very late Miocene were a number of thriving insect communities, because I have found a proxy for insects”… No. You’ve found the insects themselves! As perfectly formed as the day they were sealed off for eternity. In EXACTLY the same way, we can directly measure samples of the atmospheres of ages past using AMS and be certain of its various constituents, literally down to the atom. Clear?

  52. #53 maxwell
    March 12, 2010


    that’s a nice analogy, but I still disagree with you. I don’t think you’re wrong. I just don’t agree. I hope I am making that difference clear.

    And I’m sure these cores are calibrated to something, but how do they know the time resolution is constant looking up to down in a piece of ice? When it’s not constant, how do they decide the rate of change of time where the ice isn’t as easy to ‘read’ as other places? At some point, like in any other research endeavour, someone has to make a judgment call. Where do they do so with ice cores?


    the equilibrium amount of methane in the atmosphere is set by two constraints. First is the concentration of hydroxide ions that oxidize methane to water and CO2. Second is the sources of methane. Right now, you and I agree that it is likely that humans have thrown the sources of methane part of this equation into overdrive.

    But if we stopped producing methane today, it would be interesting to see how long it took to oxidize away the ‘excess’ methane. If there are enough hydroxide ions, then I would gamble it would take 14 years before we got to a point where the ‘natural’ sources were putting enough methane into the atmosphere to find an equilibrium.

    But without such sources, yes the concentration of methane would tend toward zero. Because it’s modeled as an exponential, it wouldn’t ever get to zero probably, but it would be exceedingly smaller than it is today, given enough water in the atmosphere to make the necessary ions to oxidize it.

    So I don’t understand your point other than maybe a physical model where methane production stops in a day may be unphysical. But even if methane productions decreases over the course of fifty years, the atmospheric concentrations are still going to fall rather rapidly assuming enough hydroxide ions, which seems like a sound assumption here. They might even rise and fall too rapidly for ice cores to resolve such ‘spikes’.

  53. #54 Joseph
    March 12, 2010

    But if we stopped producing methane today, it would be interesting to see how long it took to oxidize away the ‘excess’ methane. If there are enough hydroxide ions, then I would gamble it would take 14 years before we got to a point where the ‘natural’ sources were putting enough methane into the atmosphere to find an equilibrium.

    OK. Let’s do the math. Let’s take the Wikipedia half-life figure of 7 years as correct. Let’s also assume the ‘equilibrium’ concentration (without human-caused emissions) is 650 ppb in the current interglacial period.

    Now, the concentration has increased to 1750 ppb by some unknown mechanism. The question is: How long does it take to get back down to equilibrium? Well, in theory, it never quite reaches equilibrium, but let’s figure out when it reaches 660 ppb.

    The excess concentration at time T in years (we start out with T = 0) is given by:

    M = (1750-650) e-0.099 T

    Does that make sense? e-0.099×7 is 0.5 — that’s the half-life.

    So now we solve the following for T:

    (660-650) = (1750-650) e-0.099 T

    The result is:

    T = 47 years

    So, maxwell, I think you were right on the first estimate you gave, but not on your subsequent calculations.

  54. #55 Joseph
    March 12, 2010

    They might even rise and fall too rapidly for ice cores to resolve such ‘spikes’.

    I understand your point, but I don’t believe you’re right. Again, the Vostok methane reconstruction has 458 data points. It’s not sufficient that the hypothetical spikes are short lived relative to the resolution of the reconstruction. They would also need to be exceedingly rare.

  55. #56 maxwell
    March 12, 2010


    I think I see the point your making.

    I think you’re right. It would take about 50 years.

    Thanks for clearing that up.

  56. #57 Matt Bennett
    March 12, 2010

    Maxwell, I can’t make it ANY clearer, I don’t think you understand what a proxy is. You need to read up a bit more. It doesn’t become a proxy just because you misunderstand and decide to call it one. Scientists literally ARE measuring ancient atmospheres. And I don’t understand where your problem with dating things comes from – that is an undisputed and unremarkable part of this work. As I said, please read up so you can fill in this gap in your knowledge. Then point out to me, if you care to, where the Vostok data and dating processes (for example) demonstrates the uncertainty you allege and quantify for me their maximum error (you might also want to email the research team themselves to let them know they overlooked these glaring uncertainties)

    Otherwise, please just admit that, indeed, you didn’t quite understand where the methane graph came from but now you do and are happy to agree with it.


  57. #58 GFW
    March 12, 2010

    Upper layers of an ice core have individually countable layers, like tree rings. So that’s pretty damn accurate. I believe some of the antarctic cores have up to 160,000 discernible layers. Below that, other (typically isotopic) means of dating are used, with an accuracy usually better than 1%.