Key Extracts from Peter Wadham's A Farewell to Ice: Extract I: Chapter 9, on Methane

Chapter 9, "Arctic Methane, a Catastrophe in the Making"

"Offshore Permafrost and Warm Water

The potentially catastrophic feedback effect that I am about to describe arises from the combination of two phenomena: sea ice retreat, and the continued existence of offshore permafrost in shallow Arctic seas.

I have already described the rapid retreat of the summer sea ice limit, which has removed the sea ice form large areas of the Arctic continental shelves, particularly the seas north of Siberia where there are wide shelves with water depths of only 50-100 metres. What happened to the water column in these newly ice-free seas?

The water structure in the deep Arctic Ocean is made up of three layers. The upper layer, called polar surface water, is about 150 metres deep and is at or near the freezing point. The layer below it, called Atlantic water, extends down to about 900 metres and contains heat which comes from warm North Atlantic water sinking at the ice edge and moving into the Arctic at middle depths. Below that is another cold layer, bottom water, extending to the ocean bed. Therefore if a continental shelf is only 50-100 metres deep there is only a single layer of polar surface water present – the warmer, deeper Atlantic water stays outside the shelf break. In the ‘old days’, before about 2005, this polar surface water layer was covered by sea ice even in summer, and this provided a kind of air-conditioning system in that incoming solar radiation could not warm the water up since its first role was to melt some of the ice. Similarly, the ice kept local air temperatures at about 0^oC. From 2005 on, with the summer sea ice completely melted away, solar radiation has been able to penetrate into this shelf water and warm it up. Instead of being held to 0^oC all summer thanks to the ice, the polar surface water is now able to rise in temperature through the ice-free summer. In the summer of 2011 a surface temperature of 7^oC was detected by a NASA satellite in the Chukchi Sea (as warm as the North Sea in winter). During a recent (August 2014) voyage in which I participated, the US Coastguard icebreaker Healy recorded extraordinary sea surface temperatures in the Chukchi Sea; on our way north towards Bering Strait, off Nome, we experienced air temperatures of 19^oC and sea surface temperatures of 17^oC. […]

Winds over these wide ice-free areas can now create significant waves, which mix the warmed water down to the bottom, so we now have water above freezing point impinging on the Arctic seabed for the very first time in several tens of thousands of years.

On the seabed the warmer water encounters the second element in the story, frozen sediments. These are relics of the last Ice Age, and represent a seaward extension of the permafrost on land. Within them is embedded methane in the form of methane hydrates or clathrates. This extraordinary solid material looks like ice but it burns. It is a compound of methane gas (CH₄) and water, with an open crystal structure which is stable only under conditions of high pressure and/or low temperature. It is found in various ocean sediments, usually in deep water where the overlying water pressure confers stability. The amount of methane stored in hydrate deposits in the entire ocean bed is estimated to be more than thirteen times the amount of carbon in the atmosphere and amounts to 10,400 gigatons (Gt). On the Arctic shelves, because of the shallow water depth, the hydrates should be unstable, but the solid frozen sediment applies sufficient overpressure to hold them in place. The warm summer water arising from the recent ice loss causes these sediments to thaw out, so they no longer provide a solid cap over the hydrates. The frozen sediments originally formed on land during the Ice Age when sea levels were lower, but then became flooded 7,000-15,000 years ago when the shallow East Siberian Sea formed in the so-called ‘Holocene transgression’ as the ice sheets melted and sea level rose. So the hydrates, which have rested for tens of thousands of years in the frozen sediment, are now disintegrating as the sediment thaws, producing pure methane gas which is emerging from the sediments and rising to the surface in great bubble plumes. Methane is oxidized in water, so if that rising plume occurs in deep water, as has been seen off the coast of Svalbard in 400 metres, it dissolves and the plume disappears before it reaches the surface. But in water only 50-100 metres deep the methane does not have time to dissolve, and it emerges almost intact from the sea surface into the atmosphere. We must remember – many scientists, alas, forget – that it is only since 2005 that substantial summer open water has existed on Arctic shelves, so we are in an entirely new situation with a new melt phenomenon taking place.

The methane emerges as huge clouds of bubbles rising through the water column, a process called ebullition. The bubbles can be recognized as a series of individual plumes, like plumes from a seabed gas-oil blowout, originating from a number of point sources on the seabed. The East Siberian Arctic Shelf is exceptionally shallow – more than 75 per cent of its entire area of 2.1 million square kilometres is shallower than 40 metres – so most of the methane gas avoids oxidation in the water column and is released into the atmosphere. Atmospheric concentrations of methane above the sea surface there have been found to be as much as four times greater than normal atmospheric levels. It has been widely assumed that no methane could be emitted from the Arctic shelf during the winter ice-covered period. However, new observational data suggest that methane ebullition and other emissions are now occurring throughout the year. Methane fluxes from European Arctic polynyas have been found to be 20 to 200 times higher in methane than the ocean average, strongly suggesting winter emissions. Methane has also been observed directly accumulating under winter ice. This all suggests that once the cap of frozen sediment has been removed by summer melt, the methane is able to escape at all seasons.

The discovery and observations of these powerful bubble plumes on the East Siberian Shelf in summer were first made by annual US-Russian expeditions led by Natalia Shakhova and Igor Semiletov, which brought back some dramatic underwater pictures. They estimate that 400 Gt of methane equivalent is held in these sediments, and that 50 Gt could be released from the uppermost tens of metres of the sediment within a very few years of this warming process gathering pace. Modellers such as Igor Dmitrenko at the University of Manitoba have looked at the sediments closest to shore, in only 10 metres of water, and have estimated that the time scales for thawing and methane release are slow, of the order of 1,000 year. But other things are going on further out to sea.

Natalia Shakhova herself, and others, have drawn attention to the role of taliks in facilitating methane release from the sediments. Taliks are irregularities in the subsea permafrost layers, caused by faults or localized irregularities, which provide a route for methane to be released from hydrates deep in the sediments and to make its way upwards towards the seabed. Shakhova found that many of the methane plumes observed in the East Siberian Sea consisted of methane being released from the top of a talik. There is a parallel here with the role of moulins on the Greenland ice sheet – they permit thermal processes to occur deeper inside the material than modellers suspected. A talik provides a route for methane molecules to escape from their hydrate cage and move up past the barrier supposedly offered by the seabed permafrost, and then be emitted. The emission rate therefore does not depend on layer after layer of sediment gradually giving up its methane as the permafrost thaws.

Methane is exceptionally powerful as a greenhouse gas. As I said in Chapter 5, it is 23 or 100 (depending how you calculate it) times as powerful per molecule as CO₂ in its warming potential. Arctic offshore emissions may well be the main cause for global atmospheric methane levels starting to rise again in 2008, after stabilizing about the year 2000 (the other likely candidate, leakage from fracking, did not begin until more recently). How much methane is waiting to be emitted in this way, and when will it emerge? What will this do to the climate? We expect that it will further speed up sea ice retreat, reducing the reflection of solar energy and hasten sea level rise as the Greenland ice sheet melt accelerates. But the ramifications of vanishing ice will also be felt far from the poles.

Global Impact of Arctic Methane Release

With two colleagues, Gail Whiteman and Chris Hope, I have modelled what a 50 Gt methane release over ten years would do to climate, both in terms of temperature and cost. Let us remember that, although this is a huge and seemingly impossible quantity to let loose on the world (out total annual release of CO₂ is only 35 Gt), it is still less than 10 per cent of the total volume of methane believed to be locked into the East Siberian Sea sediments. To quantify the effects of a major Arctic methane pulse on the global economy, we used the PAGE09 integrated assessment model, which allows the extra emissions to be traced through to changes in sea level, regional temperatures, and regional and global impacts, such as flooding, health and extreme weather, taking account of uncertainty. PAGE09 calculates the amount by which the Net Present Value (NPV) of impacts, aggregated between now and 2200, increases if one more tonne of CO₂ is emitted, or decreases if one less tonne is emitted – effectively, the social cost of CO_2. PAGE09 is the most recent version of the PAGE model, developed by Chris Hope at the Judge Institute in Cambridge and used by the Government’s Stern Review of the Economics of Climate Change to calculate the impacts of climate change. All results are based on 10,000 runs of the model, which allows a full picture of the risks to be built up in order to work out uncertainties.

We tested two standard emissions scenarios. First was the ‘business as usual’ scenario, where it is assumed that the world carries on following on its present course with increasing emissions of carbon dioxide and other greenhouse gases year on year without any mitigating actions. Secondly we ran a ‘low emissions’ case, with a 50 per cent chance of keeping the rise in global mean temperatures below 2^oC (the ‘2016r5low’ scenario from the UK Metereological Office). In each case, we superposed a decade-long pulse of 50 Gt of methane released into the atmosphere between 2015 and 2025. We also explored the impacts of later, longer lasting or smaller pulses of methane.

The extra temperature rise due to the methane by 2040 is 0.6^oC, a substantial extra contribution. This would be a catastrophe for mankind, partly because it is so quick. It would speed up all the other global warming effects and there would be nothing that we could do to shut off the methane except for cooling the water column (i.e. bringing back the sea ice), which is exceedingly difficult to envisage. Such a methane pulse would bring forward by fifteen to thirty-five years the date at which the global mean temperature rise exceeds 2^oC above pre-industrial levels – to 2035 for the ‘business as usual’ scenario or 2040 for the low emissions scenario. Note how rapidly the methane generates its climactic effect, because, although the peak of 0.6^oC reached twenty-five years after emissions begin, a rise of 0.3-0.4^oC occurs within a very few years.

Measured at present values the cost of this increase comes out as 60 trillion dollars over a century for the business-as-usual scenario. We had anticipated that the price-tag for changes to the Arctic would be steep, notwithstanding short-term economic gains for Arctic nations and some industries, but it came as a surprise to discover just how steep it might be.

 The amount is 15 per cent of the total of $400 trillion estimated by the same model as the total cost to the world of all climate change impacts over the same period. For the low emissions scenario the cost will still be an extra $37 trillion. These costs are the same irrespective of whether the methane pulse is delayed by up to twenty ears, kicking in at 2035 rather than 2015, or stretched out over two or three decades rather than one. A pulse of 25 Gt of methane has almost exactly half the extra impact of a 50-Gt pulse.

The model divides the planet into eight regions to model where change would result. Under both scenarios, the global distribution of the extra impacts closely mirrors the total impacts of climate change: 80 per cent of the extra impacts by value occur in the poorer economies of Africa, Asia and South America. Inundation of low-lying areas, extreme heat stress, droughts and storms are all magnified by the extra methane emissions. So, a purely Arctic effect, switched on by the global warming-induced phenomenon of sea ice retreat from Arctic shelves, turns out to have a global impact. And, as usual, it is the global poor who will suffer most.

Action This Day

When it comes to environmental and climactic change, two criteria are meant to be applied as the basis for action or inaction in the face of a possible threat. The precautionary principle states that we should take action to mitigate a plausible threat even if we are not sure that it has happened yet. For instance, when the first assessment of the IPCC was published in 1992, conclusive proof that we were altering the climate by our emissions was not year available, but the presumption that this was what was happening was sufficiently strong to produce a call for action. Risk analysis helps us to quantify the magnitude of the threat. Mathematically, risk can be defined simply as probability of an effect happening multiplied by the negative effect if it does [expected (dis)utility formula, the central equation of Decision Theory]. Particularly difficult to assess are risks where the probability is very small but the effects would be very large, such as the risk of Earth being hit by an asteroid. In the case of an Arctic offshore methane pulse there is no question that the risk is huge. First, the probability of this pulse happening is high, at least 50 per cent according to the analysis of sediment composition and stability by those best placed to know what is going on, Natalia Shakhova and Igor Semiletov. Moreover, if it happens the detrimental effects are gigantic, with an economic cost of $60 trillion, including high levels of human mortality. So, on any definition, the risk of an Arctic seabed methane pulse is one of the greatest immediate risks facing the human race.

Why then are we doing nothing about it? Why is this risk generally ignored by climate scientists, and scarcely mentioned in the latest IPCC assessment? I fear it is a collective failure of nerve by those whose responsibility is to speak out and advocate action. It seems to be not just climate change deniers who wish to conceal the Arctic methane threat, but also many Arctic scientists, including so-called ‘methane experts’. For some such experts, accustomed only to minor Arctic methane seeps, just one of many natural and anthropogenic sources of methane, there is some excuse. But they have not woken up to the fact that the environmental conditions now are unprecedented: it is only since 2005 that the Russian Arctic shelf waters have been routinely exposed to the atmosphere over most of their area, permitting the water temperature to rise far above melting point. It is perhaps difficult for non-Arctic scientists to grasp that this is an entirely new situation, and that previous concepts no longer fit. Some other scientists do clearly grasp what is going on, yet by a psychological process of denial prefer to try to wish it away. For example, at a Royal Society meeting on 22 September 2014, methane expert Gavin Schmidt (Director of the NASA Goddard Institute for Space Studies) publicly derided the concept that large amounts of methane could be emitted from the seabed, just when new results from the Laptev Sea were being announced that were demonstrating large increases in emissions. The integrity and accuracy of field researchers have even been called into question, with personal abuse being hurled at Shakhova and Semiletov because they are Russian and one of them is a woman. This is a pretty low point for the scientific community to have reached, but it has happened in part because the implications of this discovery are truly momentous. Even if we wish to sit on our hands and dawdle when it comes to reducing CO₂ emissions, we cannot sit quietly by while 50 Gt of methane is probably going to be emitted into the atmosphere and cause a rapid rise of 0.6^oC in global temperatures. And this is just the first instalment: much, much more methane remains in these sediments and will emerge over coming decades as the sediments continue to thaw, while terrestrial permafrost (see the next section) will add in the long term an even greater amount of methane.

What can we do? For a start, immediate research is necessary on an emergency scale, because there is still too much that we simply don’t know. It is true, and easy, to say that if we could halt and reverse global warming in some other way, for example by geoengineering, water would return to its previous 0^oC temperature level. Permafrost thawing and methane emission would cease. But, as methane plumes are coming out now, and already causing radiative forcing, it is difficult to see how we can make a heroic effort to bring down temperatures enough to prevent further methane emission. If we could, we would have already conquered climate change and would have nothing much to worry about. No, the only effective way is directly to prevent the methane from emerging from the seabed sediments under present and near-future conditions, and this is where we are short of solutions. The idea of catching methane in plastic domes or sheets and feeding it to a central flaring site has been suggested, but since the entire seabed seems to be erupting with methane the plastic sheets would have to cover the whole East Siberian seabed, which is impossible. The only suggestion that seems plausible so far comes from the oil industry itself, which is to carry out a version of fracking in which a well is drilled to below the active layer of sediment, with horizontal drilling then connecting the well to cavities created under the sediments. Methane drawn into these cavities could be pumped out and flared. Flaring makes sense, as when a molecule of methane burns it produces a molecule of CO₂ which has only one twenty-third of the heating power of methane. But if the methane can be captured and used, so much the better. This solution would require a network of wells covering the entire East Siberian Sea. Nobody has calculated exactly how many, nor what the entire venture would cost. But in the absence of any other solution, this one must be researched urgently, and if it is practicable we should be ready to implement it. It would be ironic if the oil industry saved the world through its advanced technology, but I am sure that God would raise a smile.

The Threat from Permafrost Decay on Land

The Arctic offshore is our greatest immediate threat, but the threat of methane and CO₂ emissions from decaying permafrost on land is also not only real but inexorable. We know from the careful work of Arctic biologists that as terrestrial permafrost thaws, the rotted surface vegetation which becomes unfrozen goes through a sequence of chemical and biological processes in which it ends up producing both methane and CO_2. This is different from the Arctic offshore permafrost where the methane is already there waiting to be released when the sediments thaw. Here the methane has to be generated by a longer, slower set of chemical processes, but eventually it is still produced.

Let us look at some statistics. The area of terrestrial permafrost in the world today is about 19 million km^2, including both continuous and discontinuous (that is, patchy) permafrost. It is thawing; permafrost areas have warmed by 2-3^oC since the 1980s. When it thaws, permafrost emits a mixture of methane, carbon dioxide and (some) nitrous oxide (N₂0), all of them greenhouse gases. According to the IPCC, the quantity of carbon contained in this permafrost is 1,400-1,700 Gt. It is estimated that 110-230 Gt will be lost (as both CO₂ and CH₄) by 2040, and 800-1,400 Gt by 2100, at a rate of loss of 4-8 Gt per year before 2040, rising to 10-16 Gt per year afterwards.

Please note these figures. What this means is that, by the end of the century, the quantity of carbon that will have been emitted from thawing permafrost on land is some thirty times the 50 Gt offshore methane pulse which we fear in the next decade. It is unclear how much of this carbon is in the form of hyperactive methane, but it is probably substantial. So a major climate warming boost from methane is inevitable – it may be fast, due to the thawing of the offshore permafrost releasing trapped methane; it may be slow, due to methane creation by terrestrial permafrost; or it may be both fast and slow, a pulse from the offshore permafrost followed by a slower but larger released from onshore permafrost. But we are certain to receive this extra boost to warming by the end of the century at the latest.

Once again, an extraordinary aspect of the 2013 IPCC assessment is that these figures on methane emissions from terrestrial permafrost are quoted, but the implications for accelerated climate warming are not pursued, although the implications are as bad as, or worse than, the implications from offshore release.

The Area Widens

Following the discoveries of Shakhova and Semiletov, the pace of exploration of the Arctic shelves has intensified, yielding further discoveries of warm offshore water and methane production in shelf areas other than the East Siberian Sea.

Semiletov and Shakhova expanded their area of operations out of the East Siberian Sea by joining the Swedish icebreaker Oden for the ‘SWERUS-C3’ cruise to the Laptev Sea in the summer of 2014. The ship found a zone several kilometres across on the outer shelf at a water depth of 200-500 metres where large quantities of methane bubbles were being emitted, while closer to shore they found 100 methane sources on the seabed at depths of 60-70 metres, including one intense methane outbreak at 62 metres which the Chief Scientist, Orjan Gustafsson, termed a ‘mega methane flare’. This large emission was discovered on 22 July 2014; it was announced that the team observed ‘elevated methane levels, about 10 times higher than background sea water’ in the surrounding water column. A borehole through the shelf sediment produced methane.

In January 2016 a report by the Laptev Sea Programme, a Russian-German field research project which has been operating since the 1990s, revealed an extraordinary development. Since 2007 a research station mooring has been in place on the shelf, at a water depth of 40-50 metres, measuring water temperatures from the surface to the bottom and the thickness of the ice. In the remarkable summer of 2012 the instruments recorded an early retreat of the ice cover followed by a warming of the water at mid-depth, driven by heat from the Lena River outflow and penetrating solar radiation. The heat mixed downwards towards the seabed, but took time to do so, so it was winter before the seabed water warmed up, to 0.6^oC in January 2013, spending 2.5 months at that temperature. This would have a melting effect on the sediments, and links warmer water to the methane observed by ‘SWERUSC-3’. Model studies agree that the Laptev Sea could be a larger source of methane than the East Siberian Sea.

The strength of activity in this second Arctic shelf sea leads to the conclusion that methane emission from the seabed is not confined to the East Siberian Sea but is found in more, possibly all, Arctic shelf seas. So our estimates of methane emission are probably still too modest. In situ monitoring of the methane levels in the Arctic atmosphere have revealed occasional peaks lying well above background levels (called ‘dragon breath’ by Jason Box of the Geological Survey of Denmark and Greenland, since each one seems to represent some exceptional emission from a single source). They may originate in individual, unobserved mega flares. The record from a methane monitoring station at Alert, on the northern tip of Ellesmere Island, shows that methane levels, which had stabilized at about 1,852 parts per billion (ppb) in 2000, have been rising at an accelerating rate, and have now reached 1,940 ppb, most of the increase having occurred in the last three to four years.

Possibly relevant also is the appearance of three mysterious craters with smooth vertical walls in August 2014 in the north Siberian tundra, surrounded by deposited soil material. The most plausible explanation is that they arose from underground methane explosions, where the melting permafrost allowed a build-up of methane underneath a cap of sediment, and finally blew the cap out in a great explosion.

All of these events strongly suggest that increased methane emission is already under way in the Arctic coastal regions, making use of mechanisms that have never been observed before. It is important that we recognize the threat that this poses to the climate, a threat which is immediate even though it has been belittled by the IPCC in its Fifth Assessment.”