Large-scale methane outgassing in the Siberian Arctic: ongoing process or dangerous new development?

Part one of two: the background.

By John Mason

Reports of extensive areas of methane - a powerful greenhouse gas - bubbling up through the shallow waters of the East Siberian Arctic Shelf (ESAS) have been doing the rounds in the media recently, with some articles taking the apocalyptic approach and others the opposite. So what IS going on in the far North? In this two-part post we will first examine the data available to date and then in part two we go on to discuss the findings of 2011 with the research team who have been doing the work on the ground.

Background

To understand the goings-on up at the ESAS in context, we need to go back to the time of the last glacial maximum, some 20,000 years ago. Although the climate was cold, much of Siberia remained unglaciated for the simple reason that the climate was also extremely dry: the main area of glaciation was in the Verkhoyansk Range in the east, which rises to nearly 2500m. The low-lying plains of central Siberia saw the development of permafrost - defined as soil that remains at below freezing point for two or more years. The prolonged cold of the last glacial period saw permafrost develop to great depths - over 1000m in places.  Extensive areas of this old permafrost, albeit thinner than at the last glacial maximum, exist at the present day. On land, permafrost occurs several metres below surface, and is overlain by the so-called active layer, soil which seasonally thaws out and in which the Siberian flora grows.

Above: Bathymetric map (source - NOAA) of the Arctic with key features noted and the subject area highlighted in red.

During the last glacial maximum, global sea levels fell by over a hundred metres, with the result that the shallow seas of the ESAS became dry land, which allowed permafrost to develop there. Climate warming in the Holocene melted the big ice sheets in N America and NW Europe, leading to sea level rise and flooding of the ESAS, which once again became an extensive shelf sea, averaging some 45 metres in depth. The incoming seawater raised the temperature of the seawater-seabed interface dramatically so that it is considerably (>10C) warmer today than the annual average temperature over the adjacent land permafrost areas. This warming led to a certain amount of seabed permafrost degradation but until recently the remaining subsea permafrost layer was thought to be relatively stable, acting as a cap or lid to the methane that was expected to be present in and beneath it.

Permafrost degradation and methane release on land are things that most people will be familiar with: footage of people igniting methane on frozen Siberian lakes has been broadcast many times. This is primarily biogenic methane - formed via microbial decay of organic matter such as plant-debris. As permafrost degrades due to the warming climate, the organic matter, trapped in the frozen ground for thousands of years, is freed and bacterial decay rapidly sets in, releasing methane to the atmosphere.

At greater depths in the sedimentary column, methane may exist in a second form, trapped in clathrate molecules. A clathrate is a naturally-occurring chemical substance which consists of one type of molecule forming a cage-like crystalline lattice  - the host - which traps a second type of molecule - the guest. In the case under discussion here, the host is water and the guest is methane, hence the commonly-used term 'methane hydrate'. Methane hydrate looks just like ice: it is a white, crystalline solid but is only stable at low temperatures and/or high pressures: otherwise it decomposes, liberating its methane content.

Above: methane hydrate forming irregular white masses embedded in the marine sediment of Hydrate Ridge, in the Pacific Ocean off Oregon.

This sensitivity to temperature and pressure means that outside of very deep water environments, methane hydrate typically occurs at considerable depths in the sedimentary column (ref. 1): values of 200-500m beneath surface are commonly cited as being within the 'gas hydrate stability zone' (GHSZ). Any deeper than that and temperatures tend to be too high due to the geothermal gradient; any shallower and temperatures are again too high - except, perhaps, where the hydrates are locked-in and kept at low temperatures by extensive, bonded permafrost. Within the GHSZ, methane hydrate occurs as pore-filling cements in coarse-grained sediment such as sand; conversely, in finer-grained sediment such as mud it forms pure masses of irregular shape. Typical concentrations in sandy sediment are a few percent of pore-volume. Estimates of the total amount present globally vary: although some very high values have been suggested, more commonly-cited figures are 10,000 Gt carbon or less. This is still a substantial figure when compared to e.g. estimates of carbon in global coal reserves.

Methane hydrate has been exploited on a limited scale as a fossil fuel. At Messoyakha, in western Siberia, the Soviets extracted methane trapped beneath a dome of permafrost 450m in thickness; at least a third of the resource, exploited over 13 years, was thought to exist as hydrate which was artificially destabilised by pumping hot water and solvents into the wells in order to collect the gas.


Recent observations on the East Siberian Arctic Shelf

That the sea in this area of the Arctic has warmed up significantly should come as no surprise to anybody who has been following the unfolding reductions in sea-ice and other developments in that region. A 2011 paper (ref. 2), citing hydrographic data collected since 1920, reported a dramatic warming of the bottom water layer over the ESAS coastal zone (<10 m depth), since the mid-1980s, of 2.1°C. The warming was attributed to atmospheric changes involving enhanced summer cyclonicity, reduction in ice extent, the consequent lengthening of the summer open-water season and - consequential to that - solar heating of the water column.

Until relatively recently, the subsea permafrost of the ESAS saw little or no attention compared to the onshore permafrost: it was simply assumed that it was unlikely to be a source area for methane because it was all frozen solid. That assumption was turned on its head in 2003 when the first of a series of field expeditions by scientists from the University of Alaska at Fairbanks took place and resulted in an ominous discovery: surface and especially bottom waters were super-saturated with methane, implying that outgassing from the sea-bed was occurring. Further fieldwork went on to discover plumes of methane gas bubbling up to the surface. In deeper waters, methane does not make it all the way up to the atmosphere - it all dissolves in seawater - but over the shallower waters of the ESAS this is not the case. Air sampling surveys over the ESAS revealed great variability in methane levels: against the global background level of 1.85ppm, they were elevated by typically 5-10% up to 1800m in height, with local spikes over gas-productive areas as high as 8ppm. The researchers calculated the annual total methane flux from the ESAS to the atmosphere to be 7.98Tg C-CH4, which in plain English is 10.64 million tonnes of methane per year, a figure similar to what, up until now, was thought to be the methane emissions of the entire world's oceans (ref. 3). This figure needs to be seen in the context of other sources, however: domestic animals emit about 80 million tonnes a year, for example.

More worryingly though, the same team made estimates of the methane present as free gas and methane hydrate beneath or within the ~1.5 million sq km of the submarine permafrost of the ESAS. The total came to >1000 Gt. The area of this permafrost affected by active fault zones and by open taliks - zones of permafrost that have melted - was stated to be 1-2% and 5-10% of the total area respectively. As such zones are exactly those through which buried methane can escape from under the permafrost, they went on to suggest that up to 50Gt of methane hydrate was at risk of destabilisation leading to "abrupt release at any time" (ref. 4).

That is a colossal figure, when put against annual anthropogenic methane emissions which in 2010 were approximately 275 million tonnes (or 0.275 Gt). Methane is a far more potent greenhouse gas than carbon dioxide - by a factor of 25 (global warming potential as stated in the IPCC AR4) - so that a 50 Gt methane release would be like releasing 40 years' worth of anthropogenic carbon dioxide emissions (at 2009 emission levels) all at once. However, there are some issues with cranking atmospheric methane levels up in this drastic way.

The first problem is that in none of the glacial-interglacial transitions of the past 400,000 years has a sudden large methane-spike been recorded. Ice-core data instead reveal transitions from 0.4ppm (glacials) to 0.8ppm (interglacials) and back. Such records would tend to suggest that no such releases occurred during this period of geological time despite drastic fluctuations in climate. Does that suggest that large-scale abrupt methane outbursts are rather unlikely? Leading on from that, the second problem is finding a physical mechanism by which such an abrupt release of that magnitude could actually happen: so far, on a subshelf environment where major undersea landslides are unlikely, nobody has proposed a detailed mechanism by which that could happen.

Furthermore, a recent permafrost modeling study (ref. 2) has indicated that permafrost melting lags behind changes in surface temperature: after 25 years of the summer seafloor warming reported above, in the model the upper boundary of the subsea permafrost deepened by only a metre. This is one of the current controversies associated with ESAS research: has the model accurately depicted the actual situation? The team who did the modelling study have attibuted the observed methane outgassing to "degradation of subsea permafrost that is due to the long-lasting warming initiated by permafrost submergence about 8000 years ago rather than from those triggered by recent Arctic climate changes". Although they accept that severe subsea permafrost degradation will occur in about a thousand years' time, they see the current degassing as nothing new. Which of these views will turn out to be correct?

Controversy, however, does not invite complacency. Any increased Arctic methane flux, tapping into vast stores of steadily destabilising methane hydrate, has the potential to keep going over a considerable time-period as a response to warmer (and rising) sea temperatures. We certainly do not need any feedbacks that bring additional natural sources of powerful greenhouse gases to the table, yet that is exactly what we risk up in the Siberian Arctic. The big questions that we now need the answers to are for how long has this outgassing been going on, does it appear to be intensifying and how might a colossal and rapid outburst occur. These are among the points we will be raising with the people on the ground and the answers from our interview with Dr Natalia Shakhova, part two of this post, will soon be appearing, here on Skeptical Science. In the meantime, David Archer, who has worked extensively with gas hydrates, looks at some release scenarios over at Realclimate, here and here.

References

1. Archer, D (2006): Destabilization of methane hydrates: a risk analysis. A Report Prepared for the German Advisory Council on Global Change (40pp). PDF

2. Dmitrenko, I.A., Kirillov, S.A., Tremblay, L.B., Kassens, H., Anisimov, O.A., Lavrov, S.A., Razumov, S.O. & Grigoriev, M.N. (2011): Recent changes in shelf hydrography in the Siberian Arctic: Potential for subsea permafrost instability. Journal of Geophysical Research, 116, C10027. Abstract

3. Shakhova NE, Semiletov I, Salyuk A, Yusupov V, Kosmach D, Gustafsson O (2010): Extensive methane venting to the atmosphere from sediments of the East Siberian Arctic shelf. Science, 327:1246-1250. Abstract

4. Shakhova NE, Semiletov I, Salyuk A, Yusupov V, Kosmach D (2008). Anomalies of methane in the atmosphere over the East Siberian shelf.  Geophysical Research Abstracts 10, EGU2008-A-01526. Abstract

Good stuff, John. I'll make some detailed comments later.

Until then, I'll point everyone to a recent RealClimate article Much ado about methane

This is a good point made in a question on the RealClimate thread:

Question, if you take an exposed shelf with permafrost and hydrates and hit it with a 15F temp change at the seafloor via transgression for 10000 years…how is it even possible that another couple degrees change in air temp could have any discernable effect?

[Response:Transgression, for the non-geological reader is flooding by rising sea level. There's permafrost, meaning soil/sediment colder than the freezing point of water, on the Siberian shelf because the air was so cold when the shelf was exposed during the last ice age. Very complicated playing field, but you're right, a bit of water warming is probably a small perturbation to the huge temperature change when it flooded, from which the permafrost is still responding. There's also frozen hydrate beneath the sea floor, because the hydrate freezing temperature is somewhat higher than the pure ice freezing point, so it doesn't rely on that extreme transgression thing to drive it. Good question. David]

This is very good.

Some, perhaps pedantic, comments and suggestions:

The prolonged cold of the last glacial period saw permafrost develop to great depths - over 1000m in places - and extensive areas of this, albeit thinner, frozen ground exist at the present day. [Awkward, needs rewording]

On land, permafrost occurs some distance below surface, and is overlain by the so-called active layer, soil which seasonally thaws out and in which the Siberian flora grows. [Rather than "some distance" I'd say "a few metres"]

During the last glacial, global sea-levels fell by many tens of metres, with the result that the shallow seas of the ESAS became dry land and permafrost developed here, too. [Suggest: During the last glacial maximum, global sea levels fel by over a hundred metres, with the result that the shallow seas of the ESAS became dry land, which allowed permafrost to develop there.]

Deglaciation and Holocene climatic warming led to sea-level rise, inundating the ESAS which once again...[Climate warming in the Holocene melted the big ice sheets in N America and NW Europe, leading to sea lea level rise and flooding of the ESAS, which once again...]

This warming led to a certain amount of seabed permafrost degradation [certain amount? The sea water was probably slightly below zero, so I think the main effect was not so just melting of the permafrost at the top but raising the temperature of the entire PF section to just below zero There was actually more PF melting at its base, oddly enough, see Fig 2 here]

This is primarily biogenic methane - formed via microbial decay of organic matter such as plant-debris. As permafrost degrades due to the warming climate, this methane, trapped in the frozen ground for thousands of years, is released to the atmosphere. [I'm not sure that the methane is bubbling up in lakes is old biogenic methane that was trappped and released but is also (or instead) methane formed by current, ongoing biogenesis]

This sensitivity to temperature and pressure means that methane hydrate typically only occurs at considerable depths in the sedimentary column (ref. 1): values of 200-500m beneath surface are commonly cited as being within the 'gas hydrate stability zone' (GHSZ). [Hydrates form at very shallow depths in the sedimentary column in deep water, nearly at the seabed. Perhaps if you specified that you are just talking about hydrates on land or under shallow seas, this sentence would be OK]

...except, perhaps, where the hydrates are locked-in and kept at low temperatures by extensive, bonded permafrost [Actually I think the effect is that when impermeable permafost surrounds a pod of hydrate, the pressure can rise to lithostatic and preserve the hydrate above the normal stable depth. But it's true that without thick permafrost in the section, there would be no onshore or shallow water hydrates]

In deeper waters, methane does not make it all the way up to the atmosphere - it all dissolves in seawater - [I think that the correct chemistry is that it oxidizes and the CO2 dissolves in the water]

The researchers calculated the annual total methane flux from the ESAS to the atmosphere to be 7.98Tg C-CH4, which in plain English is 10.64 million tonnes of methane per year, a figure similar to what, up until now, was thought to be the methane emissions of the entire world's oceans (ref. 3). [I always thought the framing of this was by Shakhova was a little misleading. The numbers may be accurate but the fact is the oceans are currently a minor source of methane (cow farts are eight times bigger). See here ]

To appease some readers' minds, a short calculation of the additional warming caused by the increase of the release would be nice, but it should be noted at the same time that no one knows whether it continues or intensifies further. My very unscientific estimate would be that if the release continues as  strong as it's been between 2010-2011, the additional warming would still be at max 0,1C/decade (and that only in the Arctic), but I'm very much not sure how correct the estimate is. But of course if the release increases ten-fold (as it did during this latest increase 2007-2011) that starts to be significant even globally I guess.

Thanks! Will work the edits in during the coming days.

David's piece at RC was interesting. Seems to be two scenarios to me: either a lot of experts in the field are missing something here or Shakhova/Semiletov are being a bit maverick! With a bit of luck their responses to our questions may help thrash this out.

Cheers - John

one consequence of arctic wintertime methane release that might not be evident to the average reader that while it warms the near-surface, it cools the stratosphere. this leads to increased ozone loss so better pack some heavy duty sunscreen when travelling in spring and early summer in the high latitudes.

Andy, I've incorporated most of your edits. One I have yet to attend to is the issue of whether methane bubbles rising through the water column oxidise or dissolve: can anyone else clarify this as all accounts I've read speak of dissolution.

Cheers - John

The title of the article is way too convoluted. It needs to be simplified.  

There's also a dsocnnect between the text which says, "...the shallow waters of the East Siberian Arctic Shelf (ESAS)" and the first graphic which defines the study area to be the East Siberian Sea.

Although I am by no means on this topic, I believe "ESAS" is the sea foor. The East Siberian Sea is the water above it.  

John, I am happy to amend the graphic.

Cheers - John

John,: You are probably right on the oxidize versus dissolve issue to leave the text as it is. I have seen several publications (eg Nature) refer to escaping seabed methane dissolving in seawater. Probably both processes occur at the same time and my rudimentary knowledge of ocean chemistry isn't a reliable basis to explain what's really going on. This article in Science supports emphasizing oxidation and paragraph 17 of the Archer Buffet 2005 paper refers to oxidation.

I think the title is OK. I have come around to the idea that it's better to have a long title that describes what the article is about than a short, cryptic one. I've got a weakness for cute titles that I need to curb. I think Joe Romm advised John on this a few months ago and I think Romm is right.

There's a pdf link to the Shakhova Science paper here.

Agree there Andy. John, I think the reader needs to know exactly what the article is about from the title - although I don't think "bombshell" needs to be liberally added ;) - the title of my last piece was definitely too cryptic. Joe is good with his post titles as a rule.

I'll update with that link too. Hopefully it will remain valid for the forseeable.

Cheers - John

Interesting slides from Martin Manning on methane:

http://www.niwa.co.nz/sites/default/files/i4_-_martin_manning.pdf

at realclimate.

I've emailed David Archer for a chat about the subject. Thanks for the link Tom.

Dan - that is a fascinating presentation - thanks, too!

Cheers - John

Jeez - I have just had to send David a link by return to the ref 4 2008 abstract above - he had not heard of it!

There is some serious non-communication going on here between people who should by rights be swapping notes on an annual basis!

Cheers - John

There is some serious non-communication going on here between people who should by rights be swapping notes on an annual basis!

That's the impression I get, as well. Even within the same University at Fairbanks, there seem to be two camps researching permafrost who do not discuss each others' work, at least publicly. There's probably  more going on here than meets the eye. I might guess that the mainstream (people like Archer) are critical of what they see as attention-grabbing press releases from the ESAS workers. Perhaps S&S regard the rest as recklessly complacent about what they consider to be real and alarming observations. It seems that both sides are responsibly trying to avoid turning this into a public personalized argument. I suppose that if you had a few drinks with either side you'd hear a different story.

Just for completeness, here's what David said (he's not responded further since I sent him the link, but that might be down to the time difference):

Cheers - John

David has since posted this response on RC:

why pick 100x as the worst case, why not 1,000x? Genuine question, I have no feel for what might be possible.

[Response:The number was my own pick, as I wrote, of a blow-the-doors-off worst case. Perhaps I lack imagination, maybe it could be worse. But 200 Gton is a lot of carbon. Shakhova et al are claiming 50 Gton C. They argue that a release of that size could come out instantaneously, and if it did, I agree with them, the climate impacts would be immense. I'd written in the previous post that Arctic methane fluxes would have to increase 10 or 100 times before they would start to become significant. The calculation in the second post bore it out, that a factor of 100 would be needed to reach the climate impact of the CO2. But I don't have a strong reason to draw the limit precisely there. In that sense, in retrospect, describing it as a "worst case" was probably sloppy. Worst I can imagine, and if anyone has a reason to think it could be higher I'd be interested, astonished actually, to hear. David]

Cheers - John

Well I have the responses from Natalia. In Docx format - so I have an online converter trying to sort that out! I shall respond with a bit of IT advice.... will be back in touch once that has been completed. It can take a daft amount of time sometimes. Why Microsoft ever thought this format was a Good Idea is likely best left for them to ponder over, as a user of Open Office these days myself!

Cheers - John

John, would it help to have it as a doc-file? I can open the docx, save it as a doc and resend to you if that helps. Feel free to email me the file  (baerbel.winkler@web.de)

Thanks - I downloaded the latest version of Open Office which now handles docx format files. This is a big improvement!!

Note to all: please see http://www.skepticalscience.com/thread.php?t=4025&r=0 for the full interview, just posted-up.

Cheers - John

John Mason,

Your current titles now read:

 Arctic methane outgassing on the E Siberian Shelf part 1 - the background

Arctic methane outgassing on the E Siberian Shelf part 2 - Interview with Dr Shakhova

I recommend these be tweakied to:

Arctic methane outgassing on the East Siberian Shelf: Part 1 - Background

Arctic methane outgassing on the East Siberian Shelf: Part 2 - Interview with Dr Shakhova

Hi John,

Will do!

I'm looking into whether we can run both of these soon, given the covering email from Shakhova with the answers:

I am sorry it took little longer than I anticipated. Attached is my
answers to your questions. As for any official announcment about new
findings of summer 2011 - this takes much longer than everyone would like
to. As labor-intensive analysis should be performed after we stay in the
line waiting for instrumentation availability, I think, no sooner than
summer 2012.

I'd like to see these two posts going out in the next few weeks on that basis.

Now I have the minor task of summarising the interview... (/sarc to myself).

Cheers - John

OK this one is good to go, if people can give me a thumbs-up :)

Perhaps it can go into the schedule so that the interview (part two) can be posted quite soon afterwards - within a week?

Cheers - John

Thumb-ey

We can probably post them both next week, if the second one is ready, John M.

Good stuff. I'd just like to see what Scott Mandia can dig up WRT the last question in the interview, but have told him we could do with this within a few days. It would be useful to have an elaboration on that. This post is ready though :)

Cheers - John

It's over at Real Climate and may be worthwhile mentioning even if it does not change anything in the blogpost.

One should be clear about a timed release and a resulting steady state concentration. (also see new app at RC)

"Controversy, however, does not invite complacency. Any increased Arctic methane flux, tapping into vast stores of steadily destabilising methane hydrate, has the potential to keep going over thousands of years as a response to warmer (and rising) sea temperatures."

- but perhaps not such a high steady state concentration, as indicted in your earlier statement -

"The first problem is that in none of the glacial-interglacial transitions of the past 400,000 years has a sudden large methane-spike been recorded. Ice-core data instead reveal transitions from 0.4ppm (glacials) to 0.8ppm (interglacials) and back. Such records would tend to suggest that no such [sudden] releases occurred during this period of geological time despite drastic fluctuations in climate."

Dunk

As a result of all this good and hard work I will know much more than before about methane, and still not know what to expect.

Thanks Dunk - I think one thing we have established is that there remains a hell of a lot more work to be done in this area! I think I might dequantify my first sentence that you flagged up.

Cheers - John

I think we can publish Part 1 tomorrow if you're okay with that John M.

Yes, let's go for it :)

Cheers - John