A Miss by Myles: Why Professor Allen is wrong to think carbon capture and storage will solve the climate crisis

(This post was co-written by rustneversleeps and Andy Skuce).

A recent opinion piece in the British newspaper Mail on Sunday by University of Oxford climate scientist Myles Allen argues that the best way to combat climate change is to pass laws requiring fossil fuel producers to capture and sequester a rising proportion of the carbon dioxide emissions that the fuels produce. We argue here that such a policy, with its emphasis on carbon sequestration, would not be successful in achieving the carbon emission reductions that Allen himself advocates—for a variety of political, economic, technological and logistical reasons. A more recent article by Allen in The Guardian covers the same ground.

Nevertheless, Allen’s prescription does succeed in focussing the mind on the scale of the problem that we face in mitigating climate change.


This is a very long post, so here is a clickable summary.

A good starting framework, then... Allen's diagnosis is clear and his framing of targets in terms of cumulative emissions is unabiguous. But his prescription is flawed.

Politics There is no reason to assume a fixed emissions cap schedule would be easier to sell to the public than a carbon tax. Caps would produce greater certainty of longer-term emission reductions at the cost of uncertain economic consequences.

Economics (i): Efficiency Imposing emissions caps without allowing trading through brokers would be very inefficient. It is not clear whether Allen supports or opposes trading.

Economics (ii) Innovation by fiat? Prescribing one form of technology as the principle solution is risky. Nobody can predict how technology will evolve and what problems may emerge in future.

Economics (iii): The information conveyed by prices The cost of one technology should not be used as a basis for carbon pricing. There is a wide range of mitigation options, with highly variable prices, all with variable and uncertain potential to contribute to solutions. Experience in British Columbia shows that even a modest carbon tax can reduce emissions significantly without harming the economy.

Scaling it up to climate relevance Even promoters of aggressive deployment of carbon capture and storage (CCS) do not envision it as more than a partial contribution to mitigating climate change by 2050.

Timing and feasibility The mass of the CO2 to be sequestered is about double the mass of the fossil fuels themselves. To develop a new industry, from scratch, to capture, transport and dispose of these quantities will involve vast amounts of capital and many decades, even if it were technically possible.

Hazards The magnitude of the CO2 to be sequestered in the subsurface is such that environmental risks from leakage, aquifer contamination and induced earthquakes are likely to be much larger than those from the already contentious shale gas industry. Getting  public licence for CCS projects in inhabited areas is likely to be very difficult and time consuming.

Summing up The climate crisis is so vast that we need to throw everything we have at it. Claiming that any single technology will solve the problem can lead to complacency that the fix is simple. It isn't.

 A good starting framework, then...

Allen grounds his approach—correctly, in our opinion—on the overarching goal of limiting cumulative human emissions of carbon to roughly 1 trillion tonnes (or about 3.7 trillion tonnes of carbon dioxide).

Limiting emissions to one trillion tonnes means that global warming would most likely warming be limited to 2 degrees Celsius, the generally agreed-to level above which climate change would become dangerous. According to the Trillionth Tonne calculator (a website hosted by the University of Oxford and based on an influential 2009 Nature paper by Allen et al), we are more than halfway (57%) to that trillion tonne level already and, on recent emissions trends, we will be passing the trillionth tonne in 2043. If emissions were to fall by about 2.5% per year, we would avoid crossing the trillion tonne threshold.

Having correctly identified that we need to limit our future carbon emissions, Allen then sketches out a path that he, at least, thinks could get us there:

So with a trillion tonnes to go, we need to increase the fraction we bury at an average rate of one per cent for every 10 billion tonnes of global emissions.

That’s not a policy – that’s a fact. For every 10 billion tonnes we emit without increasing this sequestered fraction by one per cent, we will just have to bury more later in order to catch up.

If this is what needs to be done, why not just make it a condition of licensing to extract or import fossil fuels? In forestry, if you fell trees, the law obliges you to replant.

We must use the same principle: a law to compel a slowly rising percentage of carbon dioxide emissions to be sequestered and stored.

Maybe that works on a spreadsheet, but it fails when exposed to reality. Let’s explore some of the problems.

Return to summary/index


Clearly frustrated with current state of international climate action, Allen writes:

Since Kyoto, world emissions haven’t fallen – they’ve risen by 40 per cent. And these vast jamborees – some involving more than 10,000 people – haven’t even started to discuss how we are going to limit the total amount of carbon we dump in the atmosphere, which is what we actually need to do to avoid dangerous climate change.

But he then contrasts that failure of the world’s politicians to implement even the mild emissions reductions under the Kyoto protocol, with an expected favourable response to his more drastic proposal:

You might argue that this would need a cumbersome agreement. Not so. All the countries who take this seriously have to do is make it clear we won’t import goods from China unless they have been made using fossil carbon treated in the same way.

One might wonder why all of these policy wonks didn’t think to ask a physicist for advice on policy and world trade earlier!

It is true that an emissions cap policy—which, in essence, is what Allen is proposing—would set a clear path to climate stability, compared to the alternative of a carbon tax. But capping emissions on a fixed schedule would produce its own uncertainties:  if alternatives to fossil fuels (e.g., renewables, efficiencies, carbon sequestration) materialized more slowly than planned, demand would not be met and price rises would ensue. Governments might well be tempted to relax the caps. As a policy choice, governments and their electorates might prefer the economically predictable effects of a rising carbon tax to the consequences of a steadily tightening emissions cap. There is, anyway, no certainty that an emissions-cap policy would have more political appeal than a carbon tax.

Return to summary/index

Economics (i): Efficiency


Of course, there will be a cost, passed on to the long-suffering consumer. But making carbon capture mandatory would trigger a headlong race to find the cheapest sources of carbon dioxide and places to bury it.

Frankly, I’d rather pay an engineer in Poland to actually dispose of carbon dioxide than some Brussels eco-yuppie to trade it around.

So, let’s cut out the corrupt middlemen and brokers and all we’ll have to do is phone up Wiktor in Kraków, whenever we need to crank up the thermostat? Or maybe when his sequestration pump needs servicing, he’ll call us and ask us to throw on another sweater on for a day or two!

More seriously, we employ brokers throughout economies because they provide an economic service by connecting buyers with sellers. We might use emissions traders for the same reason that we buy milk at the supermarket rather than the dairy farm.

Return to summary/index

Economics (ii): Innovation by fiat?

Allen also has enormous faith in the ability of markets to solve whatever carbon capture and storage (CCS) challenge is thrown at it:

Fossil fuel industrialists will need a few years to gear up, but they won’t need taxpayer-funded subsidies. 

They’ll simply need to do this to stay in business. All past evidence suggests that when industry is faced with technical challenges it needs to overcome, it’s ingenious at finding ways of doing so.

But will CCS really get continually cheaper as we move to sequester more and more of it over the next few decades? Although this makes intuitive sense—we would expect technological advances and economies of scale kick in—it is not necessarily so. For example, a study of CCS in Alberta showed that, as volumes of sequestered CO2 rise, costs also go up as the best opportunities for capture and storage are used up.

The generally impressive history of innovation in the modern world is nonetheless surprisingly mixed when you dive into the details. Many of the great technical changes were not widely anticipated: the development of oil and gas, the automobile, the computer, etc. Conversely, many confidently-made predictions have failed to materialize:

"Our children will enjoy in their homes electrical energy too cheap to meter... It is not too much to expect that our children will know of great periodic regional famines in the world only as matters of history, will travel effortlessly over the seas and under them and through the air with a minimum of danger and at great speeds, and will experience a lifespan far longer than ours, as disease yields and man comes to understand what causes him to age."

As Kenneth Boulding wrote in 1980 for the National Academy of Sciences:

… The great uncertainties here are in the area of the future of human knowledge, know-how, and skill. There is a nonexistence theorem about prediction in this area, in the sense that if we could predict what we are going to know at some time in the future, we would not have to wait, for we would know it now. [...] In preparing for the future, therefore, it is very important to have a wide range of options and to think in advance about how we are going to react to the worst cases as well as the best. [...] It is a fundamental principle that we cannot discover what is not there.

In other words, it is generally better to have a number of potential solutions competing to succeed in the marketplace, rather than going all-in on a single silver-bullet technology which might never materialize.

And the best way to ensure the market is receiving appropriate incentives for innovation? Via price signals...

Return to summary/index

Economics (iii): The information conveyed by prices

All economies face the so-called “knowledge problem.” As described by conservative economist Friedrich von Hayek, it refers to “the observation that the data required for rational economic (decision-making) are distributed among many individual actors, and thus unavoidably exist outside the knowledge of a central authority.”

The solution, in market economies, is to rely heavily on prices freely determined by the myriad interactions between buyers and sellers. The prices then help convey the best information and allow discovery of the most efficient basket of goods and services to produce.

Hayek and other economists also generally agree, however, that when certain goods and services produce negative impacts as a by-product of their production and use, then free-market established prices may not properly reflect our best “knowledge”. As Hayek notes: “In such instances we must find some substitute for the regulation by the price mechanism.

In the case of the damages from carbon emissions, most economists agree that the simplest way to correct for this deficiency is to impose an additional price on those emissions. Then, standard economic theory suggests the market will incorporate the new information from this price, and individual interactions will again determine the best “solution” to what goods and services to produce.

And Myles Allen does, in fact, recognize that his CCS recommendation ultimately translates into price signals in the broader economy:

The impact on petrol prices is even less dramatic: 50 per cent carbon capture, which we might reach by the 2040s, might add 10p to the cost of a litre of petrol. That’s well under what we already pay in fuel taxes which, we are told, are supposed to help stop climate change.

We might eventually decide to build more windfarms, or drive electric cars, or just to reduce our dependence on Russian oil and gas.

But rather than having a broad price on carbon  to help guide the most efficient way to cut emissions, Allen’s proposal seems to arbitrarily dictate that the level of pricing should primarily be determined by the marginal cost of increasing amounts of carbon capture and storage.

Several studies have been done of the costs for various carbon abatement strategies, such as this one by McKinsey&Company. A graph of their results shows that CCS-based strategies (highlighted here in red) tend to be amongst the most expensive opportunities (height on the curve below). Furthermore, they seem to have limited absolute potential for emission reductions (width). 


 Source of figures above.

Allen seems to either imply that all the other cheaper and/or competitive abatement solutions are “a waste of money”, OR he is advocating to set the eventual price signal higher than most economists think it needs be to get large reductions in emissions.

It’s odd that Allen has so much confidence in the ability of the market to deliver vast amounts of cheap CCS, once we insist that the fossil fuel producers do so, but has so little faith in other parts of the market to contribute cheap abatement solutions as well. Other than simply saying that he disagrees with his Oxford colleague, Dieter Helm, about carbon taxes, Allen doesn’t provide much backup for the economic thinking behind his proposal. Real world experience shows that carbon taxes work.

In British Columbia, a carbon tax was introduced in 2008 and has gradually been ramped up to the current level of C$30 per tonne of CO2. Since the tax was introduced, BC's consumption of petroleum fuels has fallen by 15%, while the consumption of the rest of Canada has risen by 1%. As Tom Pedersen and Lee Thiessen wrote recently in the Vancouver Sun:

Despite what you might have heard from some campaign quarters, the tax is doing its job—it is helping to reduce greenhouse gas (GHG) emissions while keeping income and corporate taxes low. Per capita consumption of all petroleum fuels has dropped an impressive 16 per cent in B.C. since the tax was introduced in 2008, and the scaremongering didn't come to pass—the provincial economy has grown faster than the Canadian average in that time.

The BC tax is a revenue-neutral tax, meaning that any carbon tax collected is returned, in full, to inviduals and businesses. BC has the lowest income taxes of any province in Canada (even lower than oil-rich Alberta) for individuals earning less than C$100,000 per year and among the lowest corporate tax rates. The BC economy has not suffered from carbon taxation, because it is designed not as a tax grab, but as a tax shift. And the tax is working, as is documented thoroughly in a report by Sustainable Prosperity. 

Carbon taxes need not swell government coffers, they do not wreck economies and, even modest taxes like BC's, produce real reductions in emissions.

Meanwhile, the impact of CCS on consumer prices may not be as benign as  Allen suggests. Coal-fired electricity generation with CCS could add 11 cents per kilowatt-hour to costs, which would double the price for some consumers, as Joe Romm wrote:

This yields a “levelised cost of electricity on a 2008 basis [that] is approximately 10 cents/kWh higher with capture than for conventional plants”. So pick your favourite price for new coal plants—Moody’s had a 2008 price of about 11 cents/kWh—and add 10 cents and you get over 20 cents/kWh.

And while it is true that Allen’s proposed cap-and-(maybe) trade system would reduce European demand on Russian fossil-fuel producers, it would be likely to increase dependence on Russian CO2 sequestration services, since it is probable that massive CO2 sequestration would be more publicly acceptable and cheaper in Siberia than, say, the Paris or Wessex Basins.

Return to summary/index

Scaling it up to climate relevance

There is no doubt that carbon capture and sequestration can work. There are working prototypes in Norway and Algeria that take CO2 from gas plants—natural gas often contains small quantities of CO2 that have to be removed from the methane before it can be transported or sold—and sequesters the waste gas in saline aquifers. There are other cases in North America where waste-product CO2 is injected into mature oil fields to enhance the oil recovery (EOR). Many more examples like these, plus the addition of carbon capture and storage at coal or gas electricity generating stations, could eventually add up to make a significant contribution to reducing emissions.

Source: IEA World Energy Outlook 2012.

For example, the figure above shows how aggressive climate-friendly policies in the IEA’s ambitious “450 Scenario” (a 2 degree-focussed scenario) could provoke increases in deployment of renewable energy and efficiencies, as well as a relatively big contribution from CCS by 2035. But note that this is only one wedge and not the largest one and that it does not really start to appear before 2025.

Let’s compare where we are today and where we will be in a few years’ time when additional CCS pilot projects are brought on stream and try to grasp the scale of the challenge we face in mitigating current emissions.


The figure shows the amount of CO2 being captured by CCS today(blue), how much is expected to be captured when CCS projects under construction are operational in 2015 (red) (source) and how much CCS capture there is projected to be in the IEA 450 scenario (lime green). The grey circle, that is only partially shown, represents the volume of emissions today and is roughly equivalent to the amount that Myles Allen considers to be captured and stored by 2050.

Even in the aggressive IEA 450 pathway, the build-out of CCS is making only a small dent, compared to the scale of the fossil fuel emissions. CCS can certainly be part of the solution, but even the boosters of the technology make no claims that CCS is a solution all by itself: the Global CCS Institute claims that by 2050, CCS could account for 31% of the power sector emissions reductions. It is important to remember that this only covers emissions from electricity generation. Emissions from non-stationary or dispersed emitters (transportation, homes and factories) will be harder or impossible to capture and store at source.

And the alternative—capturing CO2 directly from the air—faces some daunting cost challenges. The American Physical Society completed a two-year study of direct air capture (DAC) of CO2 in 2011. Using many optimistic assumptions about technology developments and energy efficiency, they estimated that the cost:

is estimated to be of the order of $600 or more per metric ton of CO2. Significant uncertainties in the process parameters result in a wide, asymmetric range associated with this estimate, with higher values being more likely than lower ones. Thus, DAC is not currently an economically viable approach to mitigating climate change.

To put that cost into perspective, a tonne of CO2 is what you get from about 2.5 barrels of oil. So, at today’s prices, the cost of removing the CO2 from the air is more than double the value of the oil that put it there.

(There are prototype projects that hope to eventually get the cost of DAC CCS down to $100/tonne CO2. See this June 2013 Physics World article for discussion.)

Now, it might be said in Myles Allen’s defence that he is aware that reductions in emissions will not just be achieved by CCS—he said as much in a quote above. However, he really does seem to believe that CCS can bear the biggest burden among all mitigation measures. For example for the scenario in slide 11 of this presentation he claims:

Meeting 2°C target requires >10GtC/year sequestered by 2050.

That’s > 36,667 Mt/yr of CO2; far beyond even the most optimistic projections of the IEA.

Return to summary/index

Timing and feasibility

One thing that we have to bear in mind is the vast quantity of CO2 involved. The mass of CO2 emitted, some 33 billion tonnes in 2011, is more than twice the mass of all of the fossil fuel produced in that year, a little under 15 billion tonnes (we have to sequester two atoms of oxygen with every atom of carbon). In terms of volume, the emitted CO2 is similarly immense. Even when converted to the supercritical fluid form used to inject it into reservoirs, the volume of CO2 we currently produce is roughly equivalent to the flow of the River Rhine, about 2000 cubic metres per second. In the form of gas at atmospheric temperature and pressure, the volume is ~240 times that.

It has taken over a century and huge capital investment to develop the infrastructure of the planet’s oil, gas and coal industries. It is highly implausible that we could construct—in just a few decades—a parallel industry dedicated solely to the disposal of a waste-product with over twice the mass of the fossil fuels themselves.

As energy economist and historian Vaclav Smil puts it in an article in the American Scientist (our emphasis):

Let us assume that we commit initially to sequestering just 20 percent of all CO2 emitted from fossil fuel combustion in 2010, or about a third of all releases from large stationary sources. After compressing the gas to a density similar to that of crude oil (800 kilograms per cubic meter) it would occupy about 8 billion cubic meters—meanwhile, global crude oil extraction in 2010 amounted to about 4 billion tonnes or (with average density of 850 kilograms per cubic meter) roughly 4.7 billion cubic meters.

This means that in order to sequester just a fifth of current CO2 emissions we would have to create an entirely new worldwide absorption-gathering-compression-transportation-storage industry whose annual throughput would have to be about 70 percent larger than the annual volume now handled by the global crude oil industry whose immense infrastructure of wells, pipelines, compressor stations and storages took generations to build. Technically possible—but not within a timeframe that would prevent CO2 from rising above 450 ppm. And remember not only that this would contain just 20 percent of today’s CO2 emissions but also this crucial difference: The oil industry has invested in its enormous infrastructure in order to make a profit, to sell its product on an energy-hungry market (at around $100 per barrel and 7.2 barrels per tonne that comes to about $700 per tonne)—but (one way or another) the taxpayers of rich countries would have to pay for huge capital costs and significant operating burdens of any massive CCS.

Perhaps Allen would again counter:

Fossil fuel industrialists will [...] simply need to do this to stay in business. All past evidence suggests that when industry is faced with technical challenges it needs to overcome, it’s ingenious at finding ways of doing so.

Nobody doubts the technical ingenuity of the fossil fuel industry nor their ability to turn a societal need into a handsome return to shareholders. But consumers do not like handing over their hard-earned cash to oil companies any more than they do to governments as tax. 

Return to summary/index


Injecting fluids into the Earth at high pressure can have consequences: causing earthquakes; contamination of surface aquifers by the injected fluid; and by connecting together of shallow geological layers (like drinking water aquifers and shallow coals seams) by the numerous boreholes required, some of which will have faulting casing and cement jobs.

All of this has been brought to the public’s attention by the recent increase in hydrofracking by the oil and gas industry. The volumes involved in CO2 sequestration will be much higher than quantities of water used in fracking. For example, the Marcellus Shale fracking is said to have used some 2.5 million cubic metres of water in a year in hundreds of wells over a large area encompassing several US states. These regional volumes are smaller the quantities injected in a single CO2 sequestration project. A study from Duke University has pointed to the potential of contamination of drinking water by large scale sequestration.

Induced earthquakes are an additional potential hazard arising from storage of carbon dioxide, according to Stanford geophysicist Mark Zoback. He said:

It is not the shaking an earthquake causes at the surface that creates the hazard in this instance, it is what it does at depth," Zoback said. "It may not take a very big earthquake to damage the seal of an underground reservoir that has been pumped full of carbon dioxide.

In other words, not only might certain storage sites fail at keeping the CO2 effectively sequestered for the required centuries-plus time scales, but leaks could potentially threaten the health of those living nearby.

Given the controversy and delays associated with fracking in North America and Europe, it might be expected that CO2 sequestration will meet with similar or greater levels of public resistance than fracking already has. Rapid deployment would be far from guaranteed, even if the technical, logistical and economic  obstacles had been overcome.

Return to summary/index

Summing up

Myles Allen’s technical diagnosis of the climate challenge is clear and spot-on. We have to limit cumulative emissions and aiming for a trillion-tonne carbon target provides us with a technically sound and clearly-defined goal that is measurable and achievable. We would join in him lamenting the failure of governments over the past few decades to negotiate workable policies to limit climate change.

However, his prescribed focus on a single technology—carbon capture and storage—is naïve and dangerous, in our view. Naïve, because this technology cannot be scaled up as a stand-alone solution, for all of the reasons we have detailed above (see also the excellent take-down by Joe Romm). Dangerous, because claiming that any one technology can solve the climate crisis, particularly when that technology has not been proven at the required scale, can breed false complacency. In the case of CCS, this provides cover for the notion that we can can continue to exploit fossil fuels in a business-as-usual manner, because we will be one day bailed out by technology. As if liposuction was a cure for overeating.

It is no surprise that CCS is a technology favoured by fossil fuel companies; it extends the economic life of their assets in the ground, while providing them with a potential future source of revenue as they leverage their subsurface expertise. That enthusiasm may provide sufficient reason for many climate contrarians to embrace the technology, even if they dismiss or downplay the clear climate-science diagnosis delivered by Myles Allen and the overwhelming majority of his climate scientist colleagues.

On the other hand, it is tempting, for the same reason, for climate activists to dismiss CCS altogether. This would be a mistake. Some form of CCS may be necessary one day to reverse or partially remediate the damage caused by emissions. CCS is already making a small contribution to mitigation and it is possible, though far from certain, that it can be scaled up over decades to come to provide a wedge-sized contribution (i.e., 1 Gt C/yr or 3,700 Mt CO2/yr in 50 years’ time) to the emissions challenge. However, 19 and perhaps as many as 31 such wedges are required to reduce emissions to zero by 2060, and 9 wedges just to stabIlize emissions at current levels (Davis et al., 2013). In this perspective, we can’t afford to discard any of our options. But, it would be folly to pick any single one of the options we have and promote it as the stand-alone solution.

It is possible that Myles Allen was just being provocative, using phrases like “Brussels eco-yuppie” to get climate contrarians to pay attention to an article that actually contains an uncompromising prognosis if we stick to business-as-usual:

Do I think we’re doomed to disastrous warming? Absolutely not. But do I think we are doomed if we persist in our current approach to climate policy?

I am afraid the answer is yes.

By playing up CCS as a silver-bullet solution, he may have also been deliberately provoking the rest of us to start paying more attention to solutions.

That may be the case. But, even so, the debate over what to do about climate change is not being conducted in an academic common room but in the public sphere where people have short attention spans and limited understanding of the complex issues involved. Anyone, who read Myles Allen’s article could be forgiven for concluding that the climate crisis is urgent but can be solved relatively quickly and painlessly with simple policies that will force oil companies to rebury the carbon. That is not “refreshing heresy” as some have claimed, but dangerously misleading. With this article, Allen could well have reinforced the sense of complacency that has so far stymied action on climate, which we are sure was not his intention.

Posted by Andy Skuce on Tuesday, 11 June, 2013

Creative Commons License The Skeptical Science website by Skeptical Science is licensed under a Creative Commons Attribution 3.0 Unported License.