Siberian seabed methane: first numbers

The latest estimate of methane release from the shallow seas off the north coast of Russia — the East Siberian Arctic Shelf (ESAS) — suggests that around 8 teragrams per year (1Tg = 1 million tonnes) of the gas are reaching the atmosphere. This is equivalent to previous estimates of total methane release from all oceans. The study, led by Natalia Shakhova and Igor Semiletov and published in this week’s Science (Extensive Methane Venting to the Atmosphere from Sediments of the East Siberian Arctic Shelf, Science 5 March 2010, Vol. 327. no. 5970, pp. 1246 – 1250 DOI: 10.1126/science.1182221), is based on fieldwork over 2003 – 2008. Over 80% of the bottom water over the ESAS was found to be supersaturated with dissolved methane, and 50% of the surface water. More than 100 “hotspots’ were discovered, where large quantities of methane are escaping from the sea-floor. Here’s Shakhova discussing the paper’s findings in a University of Alaska Fairbanks video (press release):


Map of the study region showing fluxes of methane to the atmosphere

The CH4 emitted is about 2 per cent of global annual emissions, so it is certainly significant. Ed Dlugokencky of NOAA, who confirmed a couple of weeks ago that recent increases in atmospheric methane were continuing, tells me that the emissions estimates are reasonable, but that the global data is not yet consistent with a large and growing source of Arctic methane:

We saw an increase in atmospheric CH4 growth rate in the Arctic in 2007, but not in 2008. There were also large increases in the tropics in 2007 and 2008. I can not determine the relative amounts of the increase from the tropics vs. Arctic without a chemical transport model, but much of the increase was because of increased tropical emissions. The 2007 increase in the Arctic likely resulted from terrestrial wetland sources (i.e., not the processes Shakhova discuss), because of warmer than average conditions affecting microbial CH4 production of CH4. This is supported by measurements of CH4 isotopic composition. So the bottom line is that the atmospheric data are not consistent with a long term increase in Arctic sources.

This week’s Science podcast features Shakhova discussing the team’s findings [transcript]:

…subsea permafrost acts as a lid – the seal to prevent this methane escape. And being prevented for a period of time, being sealed for a period of time, means that this gas accumulates, and it accumulates under higher pressure – this is what we have to give an example, this is what we have, for example, this bottle of champagne. So, you have a lot of gas inside, but it’s sealed for a period of time, and when you uncork this bottle, what you can see – it’s different from a bottle of mineral water left open for period of time, it’s just little bit of different. And I think that release of methane from this kind of seabed deposits disturbed by destabilization of subsea permafrost, provides a pathway for this methane – ready to go methane – because its release does not depend on production. It’s not time-dependent, it’s not temperature-dependent, it only needs the pathway to be released.

The release pathway could be melting of the permafrost on the sea floor, or through faults, reefs and other sea-floor features. This graphic from UAF gives an idea of what’s going on:


In the podcast, Shakhova emphasises that this study establishes a baseline against which future methane fluxes can be judged. But she is in no doubt about the stakes. As she says in the press release (and video):

“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. The climatic consequences of this are hard to predict.”

Sounds like understatement to me… Shakhova and Semiletov are currently working to produce figures based on the summer 2009 fieldwork. More coverage of the paper at New Scientist, Times Online, and for Hot Topic‘s coverage of methane news over the last three years follow the methane tag.

[Update: Forgot to add that Shakhova and Semiletov’s chapter in last year’s WWF Arctic report provides an excellent overview of the ESAS methane issue.]

24 thoughts on “Siberian seabed methane: first numbers”

  1. Hello,

    sorry to be picky, but how can 1 teragram be 1.1 million tonnes. They are both metric measures, surely it should be 1 million tonnes.


    1. Thanks for the heads-up Bryan. I’ve just posted a comment over there. It’ll be interesting to see if Archer provides an in-line comment on the points I make. I have noted elsewhere that Archer generally takes a very cautious view of the risks of large methane hydrate releases. The key issue, though, is what happens in Siberia. S&S have provided a baseline (2003-8), and will soon publish data for 2009. As I say at RealClimate: one to watch…

          1. Right, well that message didn’t really come through in the article.

            And it depends really. As the article points out, this is the first measurement, so there is no way of knowing if this is the normal level of methane released from the area or if it is due to global warming. If it is stored methane being released then yes each tonne of methane will result in 2.7 tonnes of CO2. If it is a cycle then the methanogens that are creating the methane will be using CO2, so the CO2 budget will be neutral (like ruminant methane).

            Also, Rob, where did you get 12 from?

            “Stevenson et al. (2006), from 25 CTMs, calculate an ensemble mean and 1 standard deviation uncertainty in present-day CH4 global lifetime τglobal of 8.7 ± 1.3 years, which is the AR4 updated value” AR4

            1. My reference for the methane lifetime was the “Encyclopedia of World Climatology” (Springer, 2005), p.394, “Greenhouse Effect and Greenhouse Gases” by A.J. Pitman. Pitman in turn references TAR from 2001, so your figure is later and better.

            2. Oh, sorry, I am also slightly mistaken, its the difference between the average lifetime and the life time of the last (extra) emission.

              So if I put ten trout in a shark pool the average life time might be 2 minutes, but if I put an eleventh in the average lifetime may become 2 min 10 seconds, but the life time of the eleventh is two min and 30 seconds.

              AR4 calculates average life time is 8.7 years, but effective lifetime of a pulse emission is 12 years.

  2. I’ve just read the Science article and a number of web logs on the subject.It’s crisis time now!Borrowing the metaphor from James Lovelock,the methane emissions from the clathrates and tundra are akin to Hitler annexing Austria,or worse.It really is WWII time.We need to immediately ramp up (AKA ‘and create panic’) GW politics and take serious and far reaching positive action.The tragedy,however is we won’t (and can’t?).Groucho Marx’s dicta “What do I care about posterity?What’s posterity done for me?”applies,as well as a savagely cruel-to- future-generations attitude equivalent (worse?) to Hitler at his ethnic cleansing bench.This is all crazy;lunatic stuff,and I do despair of it mightyly.

    1. As discussed above, the methane may cause some warming in the next 20 years, if it is indeed new; there have been no dramatic increases in atmospheric methane concentrations that would cause your kind of alarm. However, as methane has a short life time, and climate change is a long term problem, the contribution this may have will be through the CO2 tail, if it is indeed stored methane and not a closed cycle (in which case it will have no additional effect and is not anthropogenic).

  3. “No dramatic increases in atmospheric methane concentrations…?” You haven’t been keeping up, have you? See this recent post describing a continuation of the recent increase in methane. It’s linked to in the post above, even…

    The extent to which this is a problem depends on the rate of methane release, and that’s what we have to monitor.

    There is no way that this can be considered a “closed cycle” on anything other than geologic time scales. The carbon being released has been stored for a very long time.

    1. Gareth, you are very good at taking one or two parts of a comment out of context and not addressing the rest.

      As I say in post 9, I agree it is not a closed cycle if it is stored methane that is being released, but we don’t know that. If it is the normal methane emission for the area then it is a closed cycle. Don’t take me out of context.

      Also, dramatic?? Have you done a significance test on the 7ppb variation compared to the historical standard deviation to even see how that compares against normal noise?? I don’t expect 7 out of 1750 to be significant when there is historical variation. But I haven’t done the test.

      There have been increases of the same magnitude in the past and none led to the kind of crisis PeterC is fretting about. In 1988 & 1998 levels rose by about 15 ppb. What caused that?

      1. That’s three years of 7 ppb increases — 21 ppb. The guys measuring the stuff thinks it’s significant.

        It doesn’t appear to be “normal” for the region. We don’t have definitive data, but Shakhova’s co-author, Igor Semiletov has said that when he started working in the area in the 1990s, there wasn’t as much methane being emitted.

        1. “Igor Semiletov has said that when he started working in the area in the 1990s, there wasn’t as much methane being emitted.”

          Oh in that case don’t worry about normal scientific caution

    1. Your second link is broken.

      Dappled I’m not sure of your point. The historical increase is likely caused by anthropogenic emissions (I think we agree here?). We are discussing if arctic emissions are causing a new increase. Given that historical increases and recent variations have occurred I cannot conclude that the recent increase is due to arctic release without seeing a wider study looking at sources and sinks.

      Do you disagree?

  4. Gareth,
    Thank you for this post.

    Dr. Shakhova and Dr. Semiletov have indeed done the world
    good service in this work, and are a credit to the scientific

    I shall continue to follow this work closely as I have now for
    four years. Of course, I’m well aware there are those that will
    continue disputing the details of methane release in the Arctic.
    As you are well aware, a methane burst of 50Gt or greater from
    the ESS (e.g. per their EGU paper) would resolve the matter permanently.

    As always, thank you for your website.

Leave a Reply