The need for geo-engineering

[content on this page dates back to 2011 and 2012 and is preserved on this page for archival purposes]

---

Abstract

This article, originally written in 2011 and updated February 2012, builds on three earlier articles to elaborate on the need for geo-engineering methods to be included in a comprehensive plan of action to reduce the risk of large methane releases from the seafloor of the Arctic Ocean.

Possible geo-engineering methods include carbon dioxide removal, reflecting more sunlight back into space, reducing pollutants other than carbon dioxide in the atmosphere and oceans, and otherwise creating conditions to improve the situation such as by targeted heat transfer and increased outward radiation.

For emergency applications, the most immediate cooling effect can be expected from methods that can help reflect more sunlight back into space and that can break down or capture methane in the Arctic. 

This article explains why both carbon dioxide removal and further geo-engineering methods must be included as indispensable parts of a comprehensive climate action plan. The article starts to set out the case for carbon dioxide removal, and concludes that the same two sets of feebates that can best bring down carbon dioxide levels can also be instrumental in saving Arctic sea ice. 

The article ends with a call for the UN to set up a committee to oversee further action. 

---

Historic records

Ice cores reveal that temperatures and levels of greenhouse gases in the atmosphere fluctuate in parallel in a sawtooth pattern, increasing sharply to decrease only gradually, and moving inside a band with a well-defined range of temperatures (+2 and -9 degrees Celsius) and levels of carbon dioxide (280 and 190 ppm) and methane (0.78 and 0.32 ppm).

The cycles are a result of the way Earth moves around the Sun. The sawtooth pattern is due to the long time it takes nature to absorb carbon dioxide through weathering and vegetation. As Professor Tim Lenton recently said: "Although plants are still cooling the Earth’s climate by reducing atmospheric carbon levels, they cannot keep up with the speed of today’s human-induced climate change. In fact, it would take millions of years for plants to remove current carbon emissions from the atmosphere.”

Increases, on the other hand, can take place much faster; forests and peatlands can burn within days, releasing not only carbon dioxide, but also aerosols that can further amplify warming and settle down on ice, resulting in accelerated melting. Feedback effects that amplify warming include:

• release of greenhouse gases and aerosols from land and oceans
• warmer oceans evaporize more water, which ends up in the atmosphere and acts as a greenhouse gas
• disappearing sea ice causes more and more sunlight to be absorbed, and less sunlight to be reflected back into space and less sunlight to produce hydroxyl (which breaks down methane)

What should we aim for? And how can we reach such a target?

In 2011, carbon dioxide reached a level of 394.97 ppm at Mauna Loa  — 41% above the 280ppm it had been for thousands of years before the Industrial Revolution started. 

Professor Hans Joachim Schellnhuber said, in a 2009 radio interview, that a carbon dioxide level of 280ppm, the pre-Industrial level, is the safe upper limit. "If you look back at paleo-climatic dynamics and also how our biosphere has evolved, it is clear that the Earth was in a sort of, self organised, dynamical equilibrium of many, many hundred thousands of years, and we shouldn't touch upon that equilibrium, which is precisely what we are doing now."

Professor Schellnhuber: "I think the best and the most simple principle is to bring the atmosphere back where it used to work and operate in a very stable equilibrium, that used to be before the Industrial revolution. We have to find ways, if you like, to come to 'negative emissions'. So, that’s the 'beyond zero' thing that means to extract CO2 again from the atmosphere somehow, and we can talk of course now about possible ways."

Question: By what sort of time frame should we get back to 280ppm of carbon dioxide in the atmosphere?

Professor Schellnhuber: "We have to first bend the global emissions curve down around (the year) 2020, 2025 at the latest, because otherwise the reduction gradient becomes simply too steep; no economy could deliver that. Then we need to reduce emissions at least, at the very least, by 50% globally by 2050, which means by the way for industrialised countries like Australia a reduction of 80-90, maybe even 95%, then as I said a complete phasing out of CO2 by the end of the century. And if you take into account the effect of the past emissions, which is currently still masked by aerosols, we will have to do even better than a 50% reduction by 2050. Starting from 2030-2040, I guess, we need to start developing large scale carbon extraction methods, i.e. geo-engineering."

Sam Carana advocates two types of feebates, i.e. Energy Feebates and Agriculture, Land use & Construction Feebates. Such feebates are typically implemented locally, working separately yet complimentary, to get emissions cut 80% by 2020 and to get carbon dioxide on the way back to 280ppm.

Many carbon dioxide removal methods are energy-intensive. As long as the energy used is expensive and polluting, not much can be achieved. A rapid shift to clean energy is necessary, which is best facilitated through energy feebates.

As the number of solar and wind facilities grows, large amounts of clean electricity will become available at off-peak hours, when there's little demand for electricity. This will make such electricity cheap, bringing down the cost of methods such as enhanced weathering, which can take place at off-peak hours. Such energy will also make carbon dioxide removal more effective, since the energy is clean to start with.

Energy Feebates

Energy feebates as pictured above can best clean up energy, while other feebates can best raise revenue for carbon dioxide removal. 

Energy feebates can phase themselves out, completing the necessary shift to clean energy within a decade. Carbon dioxide removal will need to continue for much longer, so funding will need to be raised from other sources, such as sales of livestock products, nitrogen fertilizers and Portland cement.

Agriculture, Land use & Construction Feebates

A range of methods to remove carbon dioxide would be eligible for funding under such feebates.

To be eligible for rebates, methods merely need to be safe and remove carbon dioxide.

Methods could remove carbon dioxide from the atmosphere and/or from the oceans.

Rebates favor methods that also have commercial viability. In case of enhanced    weathering, this will favor production of building materials, road pavement, etc. Such methods could include water desalination and pumping of water into deserts, in efforts to achieve more vegetation growth. Selling a forest where once was a desert could similarly attract rebates.  

Some methods will be immediately viable, such as afforestation, accelerated weathering and biochar burial. It may take some time for such methods to become more economically and politically viable, but when they do, they can take over where afforestation has exhausted its potential to get carbon dioxide back to 280ppm.

Overall picture with both types of feebates

Will aiming for 280ppm be enough?

The question is whether the above is enough. As mentioned before, feedbacks can sharply accelerate temperature rises. The image below, from the Copenhagen Report, shows temperatures over the past 2000 years.

The image below (by Kinnard et al. 2011) shows a reconstructed history of late-summer Arctic sea ice extent over the period AD 561–1995.

Ice extent has continued to decrease dramatically since 1995. Below are the figures for recent years.

Year of September Average Extent	Extent (million sq. km)
2002	5.96
2003	6.15
2004	6.04
2005	5.57
2006	5.89
2007	4.28
2008	4.67
2009	5.36
2010	4.90
2011	4.61
September Average Extents, 2002-2011, National Snow and Ice Data Center NSIDC

The rate at which Arctic sea ice is disappearing becomes even more evident when also looking at ice volume.

Above chart, by Wipneus based on PIOMAS data, shows the dramatic loss of Arctic sea ice. The feedback effects from the disappearing sea ice will make the situation a lot worse. The biggest danger is that large amounts of methane will be released rather abruptly from hydrates in the Arctic. 

The Methane threat

In 2011, methane levels globally reached 1820 ppb in Mauna Loa, Hawaii, while nitrous oxide levels (another potent greenhouse gas)  reached 325 ppb. 

The Arctic experiences even higher levels of methane, as illustrated below. 

Particularly worrying is that, in the past, methane concentrations have fluctuated up and down in line with the seasons. Over the past seven months, however, methane has shown steady growth in the Arctic. Such a long continuous period of growth is unprecedented, the more so as it takes place in winter, when vegetation growth and algae bloom is minimal. The most obvious conclusion is that the methane is venting from hydrates. 

See animation methane levels July 2011 - January 2012

How much methane is there in the Arctic?

Some 1700 Gt of organic carbon is contained in  terrestrial permafrost stores. This is projected to result in emissions of 30 - 63 Gt of Carbon by 2040, reaching 232 - 380 Gt by 2100 and 549 - 865 Gt by 2300. These figures are carbon dioxide equivalents, combining the effect of carbon released both as carbon dioxide (97.3%) and as methane (2.7%), with almost half the effect likely to be from methane.

In addition to these terrestrial stores, there is methane in the oceans and in sediments below the seafloor. More than 99.9% of the organic carbon in Earth’s crust is located in sedimentary basins, which have biological, inorganic and clastic deposits totaling 15,000,000 Gt of carbon. To put these numbers into perspective, the IPCC in TAR estimated cumulative carbon emissions associated with carbon dioxide at a level of 450ppm at 670GtC.

The IPCC further explains that, since the onset of the industrial revolution, some 300GtC stored in fossil fuels have been oxidized and released to the atmosphere. The utilization of all proven conventional oil and gas reserves would add another 200GtC, and those of coal more than 1,000 GtC. 
Methane is contained in hydrates, while there is also methane in the form of free gas. Hydrates contain primarily methane and exist within marine sediments particularly in the continental margins and within relic subsea permafrost of the Arctic margins.

Hunter and Haywood estimate that globally between 4700 and 5030 Pg (Gt) of Carbon is locked up within subsea hydrate within the continental margins. This does not include subsea permafrost-hosted hydrates and so those of the shallow Arctic margin (<~300m) were not considered. 

Dallimore and Collett (1995) found high methane concentrations in ice-bonded sediments and gas releases suggest that pore-space hydrate may be found at depths as shallow as 119 m. Recent studies indicate that hydrate formation can occur in upper gas-saturated horizons (up to 100-200 m) of permafrost. Furthermore, methane hydrates have been found in Siberia at depths as shallow as 20 m. 

Shakhova et al. estimate the accumulated methane potential for the Eastern Siberian Arctic Shelf (ESAS, rectangle on image right) alone as follows:

- organic carbon in permafrost of about 500 Gt;
- about 1000 Gt in hydrate deposits; and
- about 700 Gt in free gas beneath the gas hydrate stability zone.

The East Siberian Arctic Shelf covers about 25% of the Arctic Shelf and additional stores are present in submarine areas elsewhere at high latitudes. Importantly, the hydrate and free gas stores contain virtually 100%  methane, as opposed to the organic carbon which the above study estimates will produce emissions in the ratio of 97.3% carbon dioxide and only 2.7% methane when decomposing.

How stable is this methane?

It does take time for heat to be transferred down sediments. What can take place much more rapidly, though, is for heat to be transferred down fluids in cracks and openings in the rock and sediment, called pingos. 

The image right, from Hovland et al., shows pingo-like sediment features, formed by local accumulation of hydrate (ice) below the sediment surface, and methane migrating upwards through conduits.

A recent study by Serié describes geophysical signatures of different development stages associated with the formation and dissociation of shallow gas hydrate, as well as their link to deep-rooted plumbing systems that allow thermogenic fluid migration from several-kilometers-deep sedimentary basins. 

Paull et al. describe pingo-like-features on the Beaufort Sea Shelf, adding that a thermal pulse of more than 10 degrees Celsius is still propagating down into the submerged sediment and may be decomposing gas hydrate as well as permafrost.

The sensitivity of gas hydrate stability to changes in local pressure-temperature conditions and their existence beneath relatively shallow marine environments mean that submarine hydrates are vulnerable to changes in bottom water conditions (i.e. changes in sea level and bottom water temperatures). Following dissociation of hydrates, sediments can become unconsolidated, and structural failure of the sediment column has the potential to trigger submarine landslides and further breakdown of hydrate. The potential geohazard presented to coastal regions by tsunami is obvious.

Further shrinking of the Arctic ice-cap results in more open water, which not only absorbs more heat, but which also results in more clouds, increasing the potential for storms that can cause damage to the seafloor in coastal areas such as the East Siberian Arctic Shelf (ESAS, rectangle on image left), where the water is on average only 45 m deep.

Much of the methane released from submarine stores is still broken down by bacteria before reaching the atmosphere. Over time, however, depletion of oxygen and trace elements required for bacteria to break down methane will cause more and more methane to rise to the surface unaffected.

There are only a handful of locations in the Arctic where (flask) samples are taken to monitor the methane. Recently, two of these locations showed ominous levels of methane in the atmosphere (images below).

Such high levels of methane can get masked or even rejected in models that calculate average values over longer periods, as in the updated image below. 

Anyway, the graphs higher up with the high values at Svalbard and Barrow illustrate the height that levels can reach in the Arctic and, as such, they constitute ominous signs of what could eventuate at many locations in the Arctic. The danger is that large abrupt releases will overwhelm the system, not only causing much of the methane to reach the atmosphere unaffected, but also extending the lifetime of the methane in the atmosphere, due to hydroxyl depletion in the atmosphere.  

Shakhova et al. consider release of up to 50 Gt of predicted amount of hydrate storage as highly possible for abrupt release at any time.

What would be the impact of methane releases from hydrates in the Arctic? 

If an amount of, say, 1 Gt of methane from hydrates in the Arctic would abruptly enter the atmosphere, what would be the impact? 

Methane's global warming potential (GWP) depends on many variables, such as methane's lifetime, which changes with the size of emissions and the location of emissions (hydroxyl depletion already is a big problem in the Arctic atmosphere), the wind, the time of year (when it's winter, there can be little or no sunshine in the Arctic, so there's less greenhouse effect), etc. One of the variables is the indirect effect of large emissions and what's often overlooked is that large emissions will trigger further emissions of methane, thus further extending the lifetime of both the new and the earlier-emitted methane, which can make the methane persist locally for decades. 

The IPCC (2007) gives methane a lifetime of 12 years, and a GWP of 25 as much as carbon dioxide over 100 years and 72 as much as carbon dioxide over 20 years.  

The image below, based on Dessus (2008), illustrates how methane's GWP depends on the time horizon over which its impact is calculated.

Drew Shindell (2009) points out that the IPCC figures do not include direct+indirect radiative effects of aerosol responses to methane releases that increase methane's GWP to 105 over 20 years when included.

Over shorter periods, methane's GWP is even higher. Unlike carbon dioxide, methane's GWP does rise as more of it is released.

Using a GWP of 105 would mean that a 1 Gt abrupt release of methane would add a warming impact equivalent to 105 Gt of carbon dioxide, or about three years worth the current global carbon dioxide emissions, all at once.

And while carbon dioxide is emitted (and partly absorbed) all over the globe, the methane from such abrupt releases in the Arctic would - at least initially - be concentrated in a relatively small area, and likely cause hydroxyl depletion in the atmosphere, extending methane's lifetime, while triggering further releases from hydrates in the Arctic.

In conclusion, a release of 1 Gt of methane in the Arctic would be catastrophic and the methane wouldn't go away quickly either, since this would be likely to keep triggering further releases. While some models project rapid decay of the methane, those models often use global decay values and long periods, which is not applicable in case of such abrupt releases in the Arctic.

Instead, the methane is likely to stay active in the Arctic for decades at a very high warming potential, while the resulting summer warming (when the sun doesn't set) is likely to keep triggering further releases in the Arctic.

As said, pollutants in the atmosphere and oceans have now reached such proportions that a confluence of events is now threatening to dramatically aggravate the situation and result in catastrophe at a scale unprecedented in human history.

One event is the warming in the Arctic as a result from emissions. A second event is the disappearing sea ice. The associated albedo change means that much more sunlight will instead be absorbed in the Arctic, accelerating the warming that is already now occurring at much faster rates than elsewhere in the world. A third event is an increase in the release of methane in the Arctic. These three events are pictured below:

1. Emissions are causing climate changes around the globe, and the Arctic is particularly vulnerable. Ocean currents and rivers feed warm water into the Arctic Ocean. On land, thawing permafrost threatens to release additional greenhouse gases at rates not foreseen by the IPCC. Droughts and heatwaves threaten to trigger wildfires that can rage fiercely in peatlands and tundras at higher latitudes, with little or no prospect of recovery of such ecosystem in the near future.

2. In addition, the Arctic is particularly vulnerable to soot that is emitted globally and can settle down on Arctic ice, causing albedo changes. This contributes to the accelerated warming in the Arctic that causes sea ice loss, which creates a feedback that causes further albedo changes, in a vicious cycle further accelerating warming in the Arctic.

Without sea ice, less sunlight will be reflected back into space (albedo change). As the sea ice disappears, it no longer acts as a buffer that holds the energy that previously went into melting the ice.

As the sea ice heats up, 2.06 J/g of heat goes into every degree Celsius that the temperature of the ice rises. While the ice is melting, all energy (at 334J/g) goes into changing ice into water and the temperature remains at 0°C (273.15K, 32°F). Once all ice has turned into water, all subsequent heat goes into heating up the water, at 4.18 J/g for every degree Celsius that the temperature of water rises.

As the image shows, the amount of energy absorbed by melting ice is as much as it takes to heat an equivalent mass of water from zero to 80°C. As the ice disappears, this energy will cause temperatures to rise in the Arctic.

3. A third kind of warming comes from methane releases from melting permafrost and submarine hydrates. This adds a further feedback contributing to accelerated warming in the Arctic.

Altogether, these three kinds of warming combine to cause the Arctic to warm up much more vigorously than elsewhere on Earth.

Due to oxygen depletion in the shallow waters of the Arctic and the lack of hydroxyl in the Arctic atmosphere, this methane, with its high initial warming potential, now threatens to escalate into runaway global warming.

The above chart is based on historic NASA land-surface air temperature anomaly data (see interactive map at the bottom of this page). The projection shows that the average temperature anomalies in the Arctic (latitude 64 North and higher) look set to reach 10 degrees Celsius within decades. This is based on current data, while the albedo and methane feedbacks can make temperatures rise even faster.

Moreover, above anomalies are based on annual averages that are also averaged over a huge area. The NASA image on the left shows temperature anomalies of over 10 degrees Celsius for the month December 2011.

More detailed analysis shows that, over December 2011, an average temperature anomaly of 12.8933 degrees Celsius was recorded in an area in the Kara Sea (latitudes 79 - 81 and longitudes 73 - 89).

The animated image below, based on NOAA data for the period December 7, 2011 - January 21, 2012, shows that specific areas can already experience anomalies of over 20 degrees Celsius for specific days. 

(Note: this is a 3MB file that may take some time to fully load.)

For individual days and locations, the anomaly can be even more striking. On January 6, 2011, the minimum temperature in Coral Harbour, located at the northwest corner of Hudson Bay in the province of Nunavut, Canada, was –3.7°C (25.3°F), i.e. 30°C (54°F) above average.

The danger is that very high peak temperature anomalies can be expected in methane hotspot areas. Above temperature anomalies have yet to incorporate the full impact of the various feedback effects. As an example of such feedbacks, as the sea ice retreats, sunlight that was previously reflected back into space will get absorbed in the Arctic. Furthermore, various studies have found an accelerating frequency and intensity of storms in the Arctic, attributed to progressively warmer waters. As the sea ice decreases, the larger area of open water allows for even larger storms to develop, which could erode coastal soils, cause landslides and destabilize methane hydrates, resulting in large abrupt methane releases with a high warming potential.

Methane is predominantly broken down by hydroxyl. The IPCC gives methane a lifetime of twelve years. However, more methane causes hydroxyl depletion and methane levels in the atmosphere are already such that they have caused a 26% decrease in hydroxyl. In the Arctic, where hydroxyl levels are already very low, this situation looks set to deteriorate even further.

At the moment, the sea ice reflects light, helping formation of hydroxyl through tropospheric photolysis. Once the sea ice is gone, more light will instead get absorbed in the Arctic. One study projects that this can cause late-summer hydroxyl concentrations to decrease by up to 60% in an ice-free Arctic.

The danger is that if relatively large amounts of methane are released abruptly into the atmosphere in the Arctic, they will persist for decades, triggering yet further temperature rises and methane releases, in a vicious cycle leading to runaway global warming, even if the world did manage to take the necessary steps to dramatically reduce emissions.

Above image illustrates how much organic carbon is present in the melting permafrost. The recent firestorms in Russia provide a gloomy preview of what could happen as temperatures keep rising in the Arctic. Much of the soot from firestorms in Siberia could settle on the ice in the Himalaya Tibetan plateau, melting the glaciers there and causing short-term flooding followed by rapid decrease of the flow of ten of Asia’s largest river systems that originate there, with more than a billion people’s livelihoods depending on the continued flow of this water.

All this calls for a comprehensive plan of action that includes geo-engineering. Indeed, just like geo-engineering is an indispensable part of an action plan that aims to bring carbon dioxide levels back to safe levels, geo-engineering is also an indispensable part of efforts to avoid catastrophic methane releases in the Arctic.

Why geo-engineering is necessary to cool down the Arctic

Firstly, warming will continue because current emissions contain aerosols (in particular sulfur) that mask the warming effect of (the other) emissions. Therefore, when we dramatically reduce such emissions (as we should do), this masking effect will also stop and the Earth will suddenly feel the full impact of all the emissions of the past.

Secondly, there is considerable lag in how the climate responds to human emissions. Emissions caused by people have already committed us to an amount of warming that is still to be realized. Forests, oceans, snow and ice which currently still buffer a lot of heat, but rising temperatures could turn them from sinks into sources contributing to further warming, and this could happen rather abruptly and could be hard to reverse. 

Once ecosystems have fundamentally changed, natural recovery may take a long time, such as in the case of ocean acidification, loss of snow cover, sea ice and glaciers, and destruction of rainforest and tundra; as temperatures rise, heatwaves and wildfires could destroy large parts of the Amazon rain forest in a matter of years. In the oceans, stronger currents and more circulation and mixing of increasingly warmer water can cause it to release carbon dioxide relatively abruptly. 

In the Arctic, melting can cause more water circulation and vertical mixing, not only due to the flow of meltwater, but also since open waters are more prone to storms, which are more likely to occur with greater intensity as temperatures rise anyway. This could warm up bottom waters and sediments under the sea that contain huge amounts of methane, which could be released rather abruptly when certain temperature thresholds are crossed. 

Thirdly, warm water holds less oxygen than cold water. This means oxygen will get depleted earlier in case methane releases from hydrates. Increasingly, methane that would otherwise be oxidized in the ocean will rise through the water and enter the atmosphere unchanged.

For the above three reasons, emission reduction alone is not enough to avoid that warming will continue in the Arctic for a considerable time, at the risk of triggering runaway warming.  

While many people believed it was safe to allow a 2 degrees of Celsius warming from pre-industrial times, even a small amount of warming globally will result in a lot of warming in specific places, particularly in the Arctic, where huge amounts of methane are ready to be released abruptly, with flow-on impacts elsewhere. 

Interestingly, the same two sets of local feebates discussed above in regard to bringing down carbon dioxide, can also help avoid warming in the Arctic. This is because these feebates can - apart from bringing down carbon dioxide levels - also reduce emissions other than carbon dioxide most effectively.

Nonetheless, even with those policies firmly in place, there will remain an unacceptably large risk that warming in the Arctic will continue in a catastrophically dangerous way and trigger large abrupt releases of methane from hydrates and free gas in sediments. Such methane releases in the Arctic could trigger runaway warming that feeds on itself. 

The potential scale and intensity of such catastrophic developments and the precautionary principle mean that action must be taken immediately to reduce this risk. Reducing pollution levels is imperative, but geo-engineering methods should also be considered to be included in such action. 

Conclusions

Obviously, reducing emissions should be part of any climate action plan. Moreover, levels of greenhouse gases need to be brought back dramatically, making geo-engineering methods such as carbon dioxide removal an indispensable part of ways to achieve this. In the light of the danger of runaway warming, there is an additional need for methods that have dramatic cooling effect rapidly, in particular methods that can cause more sunlight to be reflected back into space in the Arctic and that can capture or break down methane in the Arctic. 

In conclusion, both in regard to long-term climate restoration and in regard to immediate emergency measures, responsible geo-engineering must be included as an indispensable part of a comprehensive climate action plan. Two sets of local feebates are recommended as key policy instruments to achieve the necessary results. Furthermore, global commitment is as imperative as decisive action at international levels, and it is recommended for the UN to establish a high level committee to oversee action points such as:  

• declaration of a planetary emergency situation in the Arctic
• fast-tracking R&D and testing of possible intervention methods
• deployment of possible geo-engineering methods in the Arctic
• establishment of a ban on commercial drilling in the Arctic
• establishment of a ban on direct flights over the Arctic
• establishment of a ban on agricultural waste disposal by open fires
• establishment of a ban on deforestation
• increased monitoring and reporting of conditions in the Arctic

Further reading, sources and references: 

The way back to 280ppm
The potential for methane releases in the Arctic
Warming in the Arctic