Current CO2 concentration in the atmosphere

The most terrifying climate finding you’ve seen (so far)

If you think the title of this post is a reference to the ongoing discussion about the Whiteman, Hope, and Wadhams piece in Nature about the economic impacts of a 50 billion ton methane release from the Arctic (see Climate science: Vast costs of Arctic change), a.k.a. the “$60 trillion” debate, here’s the good news: It isn’t.[1]

The bad news is that I’m going to try to jump start a discussion about another finding, one that is literally the most terrifying thing I’ve read in the more than 10 years I’ve been studying climate and energy issues. It’s a paper that should be causing considerable turmoil, yet has appeared and all but disappeared without so much as an online peep. Before I dive into this, I would beg you not to succumb to the temptation of “single study syndrome”, i.e. placing undue emphasis on a single research finding. It could well be that the paper mentioned below will be rejected as a flawed outlier and not accepted as part of the scientific consensus in time. It’s also possible that it will go down in history as a landmark finding that helped to reveal the true urgency of this mess we’ve created. If there are follow-on publications that support this finding or include other factors (explained below) that show it was optimistic(!), then those would certainly earn the title of most terrifying climate finding and explain the “so far” part of this post’s title.

Enough foreplay. The paper is If anthropogenic CO2 emissions cease, will atmospheric CO2 concentration continue to increase?. The abstract (emphasis added):

If anthropogenic CO2 emissions were to suddenly cease, the evolution of the atmospheric CO2 concentration would depend on the magnitude and sign of natural carbon sources and sinks. Experiments using Earth system models indicate that overall carbon sinks dominate, such that upon cessation of anthropogenic emissions, atmospheric CO2 levels decrease over time. However, these models have typically neglected the permafrost carbon pool, which has the potential to introduce an additional terrestrial source of carbon to the atmosphere. Here we use the University of Victoria Earth System Climate Model (UVic ESCM), which has recently been expanded to include permafrost carbon stocks and exchanges with the atmosphere. In a scenario of zeroed CO2 and sulphate aerosol emissions, we assess whether the warming induced by specified constant concentrations of non-CO2 greenhouse gases could slow the CO2 decline following zero emissions, or even reverse this trend and cause CO2 to increase over time. We find that a radiative forcing from non-CO2 gases of approximately 0.6 W m−2 results in a near balance of CO2 emissions from the terrestrial biosphere and uptake of CO2 by the oceans, resulting in near-constant atmospheric CO2 concentrations for at least a century after emissions are eliminated. At higher values of non-CO2 radiative forcing, CO2 concentrations increase over time, regardless of when emissions cease during the 21st century. Given that the present-day radiative forcing from non-CO2 greenhouse gases is about 0.95 W m−2, our results suggest that if we were to eliminate all CO2 and aerosols emissions without also decreasing non-CO2 greenhouse gas emissions, CO2 levels would increase over time, resulting in a small increase in climate warming associated with this positive permafrost carbon feedback.

If you’re seeing this for the first time and aren’t a climate scientist, I suspect it will take a few minutes for the implications of this study to crystallize in your mind. While that’s happening, let me offer some thoughts while you ponder this in background mode, and we can do a cognitive docking maneuver in a few paragraphs.

The scenario modeled in this paper is essentially the most utopian dream of environmentalists: The immediate and complete cessation of fossil fuel burning, worldwide. The overwhelming assumption by every environmentalist I’ve ever communicated with about emissions and public policy is that such dramatic action would be ideal from a climate standpoint — “there’s no emissions like no emissions” — but has zero chance of happening for all the obvious reasons of short-term human and economic impact, infrastructure lock-in, political corruption, etc. Further, people assume (again, in my experience) that such a fantasy scenario would be more than enough to avoid climate armageddon, so we still have a place to stand somewhere between our disastrous business-as-usual path and utopia where we’ll be “safe” if we come up with a schedule of carbon emissions reductions, such as the endlessly mentioned 80% reduction from 1990 levels by 2050, or some other formulation that keeps us from exceeding the roughly 485 billion tons of CO2 remaining on our “carbon budget”.[2]

The problem is that almost no one talks about the really big picture, taking into account Arctic permafrost, methane hydrates, albedo flip[3], and aerosol whiplash.[4][5] Even this study doesn’t hit all those points; it seems, based on the abstract, to include the permafrost carbon and aerosol whiplash but not methane hydrates. I don’t know what, if anything, it says about albedo flip.

Even without including those Brobdingnagian methane hydrate deposits (the subject you’re relieved I’m not talking about, I would remind you) the result is that if we zero out CO2 and aerosol emissions and don’t make major cuts to non-CO2 greenhouse gases then not only will we see continued warming due our current thermal disequilibrium, a.k.a. the warming “in the pipeline” scientists have been talking about for decades, but feedbacks will ensure that the level of CO2 in the atmosphere will continue to rise. In other words, this paper is saying that we’ve already passed the mother of all tipping points.

Does anyone here think that we can approach a complete fossil fuel cessation and a simultaneous major cut to non-CO2 greenhouse gas emissions?

Does anyone here think that falling short of that extreme goal — especially at a time when worldwide CO2 emissions are still rising — will do anything other than make our situation considerably worse?

Does anyone here think that this paper is describing anything short of a runaway effect that could only be stopped by geoengineering?

See also:

Offshore permafrost decay and massive seabed methane escape in water depths >20 m at the South Kara Sea shelf

Sea Level Rise ‘Locking In’ Quickly, Cities Threatened

UT grad student images water system beneath Antartic Glacier (Hint: It’s as big as the US state of Florida.)

[1] Please don’t leap to the conclusion that I reject or disagree with anything the authors of that piece said.

[2] The 485 number is based on the 565 billion tons figure mentioned in the Do the Math video, which I’ve updated to account for emissions since the end of 2010. And yes, you should watch Do the Math, even if you’ve seen it already:

[3] More formally “Arctic amplification”, albedo flip refers to the increased warming from the loss of ice and snow in the Arctic region. Open ocean and bare land absorb much more sunlight than do snow and ice, which means considerable additional warming during the Northern Hemisphere summer.

[4] Aerosol whiplash refers to perhaps the nastiest part of our climate mess, what James Hansen has called our “Faustian bargain with coal”. Aerosol emissions, largely from power plants, have a fairly strong cooling effect because they reflect sunlight. The problem is that as reduce our aerosol emissions the atmospheric level of aerosols will decrease in weeks to months, vastly quicker than would the atmospheric level of CO2 if/when we cut those emissions. We’re already cutting aerosols, as the EU has been aggressively installing scrubbers on power plants, with the US playing catch up. China is now getting very serious about curtailing their aerosol emissions, in an effort to improve local air quality. For the purposes of this posting, the thing to keep in mind is that if we end coal burning we also end aerosol emissions, which means the CO2 already in the atmosphere stays around essentially forever in human planning terms while we lose the benefit of the aerosols very quickly and suffer a jump in warming. Hence the whiplash.

[5] I’m sure experts could/would add other items to this list of factors for a “big umbrella” assessment of our climate situation, such as changes in ocean and atmosphere currents. I’ve only included those that are major factors according to the current scientific consensus, as best I understand it.

9 comments to The most terrifying climate finding you’ve seen (so far)

  • Thanks for the pointer to the paper – good to see we’re starting to get studies that take multiple feedback processes into account. However, I think your interpretation of the findings is wrong. The paper is basically another update on the work that Weaver and Matthews have been doing over the last few years. In their earlier studies, without the permafrost feedback, they found that the “committed warming” was just about balanced out by the thermal inertia, resulting in approximately steady temperatures in the zero emissions scenario (e.g. see this paper:
    This new paper shows that if you put in the permafrost feedback, you nudge the temperature rise back to slightly positive, which is where most climate scientists thought they were anyway, because they hadn’t thought much about how carbon inertia and thermal inertia interact. If I understand the paper properly (and so far I’ve only skimmed it), there’s no evidence here of a runaway feedback loop. For that to happen, the feedback effect from the permafrost would have to dominate the trend towards a new thermal equilibrium for the planet as a whole.
    So the bottom line is that this is more evidence that projections from global circulation models are likely to under-estimate the warming (because they omit some of these feedbacks), but we already knew that anyway.

    • Lou

      Check the last line of the abstract (which is all I have access to). It says CO2 levels will continue to increase with zero anthro. emissions. How is that not a runaway condition?

      • Marcel

        Rates of permafrost thaw are presumably something of an unknown. If permafrost takes longer to thaw out the deeper down it is, it could mean that even if for a number of years the CO2 levels continue to increase with zero anthro. emissions, eventually (after 100′s of years?) levels could drop to zero and start reducing.

        IANACS but to my layman’s understanding it’s still bad, but not necessarily a runaway.

        This page talks about how little is known about the thaw rates:

        Someone needs to set up an experiment where they take a huge column of permafrost, bring it back to the lab frozen, then do an accelerated test to see how surface temperatures affect rates of thaw!

      • rustneversleeps

        Positive feedback ¬= runaway

        • Lou

          No kidding. Of course positive feedbacks do not denote a runaway condition. We have positive feedbacks happening now. The paper is talking about CO2 levels rising beyond our control.

          So, for the sake of argument, let us forget the word “runaway”.

          How about: If we cut 100% of CO2 emissions and 100% of aerosol emissions, then the atmospheric level of CO2 will actually increase thanks to feedbacks, and that’s an unimaginably bad position to be in, because it means we won’t be able to avoid hideous climate impacts without additionally resorting to geoengineering.

          Does that make people feel better?

          If it does, then I would suggest those people are too busy splitting hairs and not spending enough time looking at the big picture.

  • Mike

    Which non-co2 gases are they writing about? Methane mostly? If we are to constrain co2 emissions to zero, you’d have to bet that anthropogenic emissions of other gases will be pretty minimal, although biogenic sources will remain.

    Lou, the “do the math” video won’t play. I looked around for a new link, but unsuccessfully so far.



  • Lewis Cleverdon

    Lou – thanks for this. It’s good to see even provisional partial assessments of the null efficacy of instant Emissions Control alone with a sole feedback partially incorporated.

    A more rational and informative scenario, which we can only wait for our scientific superiors to be so kind as to spare time for, would be less fanciful and more inclusive.

    Less fanciful – in taking a best case of Emissions Control – say near-zero anthro-CO2 by 2050 – which would itself require unprecedented co-operation to avoid both resource bottlenecks and economic turmoil.

    More inclusive – in accounting the net pipeline warming + the phase-out emissions warming + the fossil sulphate parasol warming + the seven interactive mega-feedbacks now accelerating. If any aren’t aware of those seven, in order of seniority they are:
    Water Vapour, Albedo Loss, Microbial Peat Decay, Permafrost Melt, Forest Combustion, Soil Desiccation, and Methyl Clathrates. Two may warrant further detailing:
    Microbial Peat Decay is the response of peat bogs’ microbial ecology worldwide to rising airborne CO2. Outflow streams were first observed worldwide in ’62 to have a 6%/yr rise in DOC (dissolved organic carbon) which rapidly outgasses as CO2. The rate of increase has held at 6% and, if airborne CO2 levels maintained their rising trend, then Microbial Peat Decay would give an annual CO2 output by ~2070 equal to the present anthro-CO2 output.

    Apart from the ESAS CH4, Permafrost Melt is perhaps the most obviously underestimated feedback given not only the wide range of drivers in addition to AGW
    - such as warmer winds coming off the de-iced Arctic Ocean, and the decline of snow cover, and the forest combustion blackening the surface, and the projected massive increase in rainfall (see Algui Dai) due to the poleward migration of rainfall already under way
    - but also due to additional drivers favouring the fraction of carbon emitted as methane due to warmer saturated conditions, including both the rainfall increase, and the heatwaves due to Jetstream destabilization, and the spreading melt pools and landslip dam lakes collecting both solar heat and leaves from the advancing deciduous tree species.

    Conventional optimism bias on Permafrost projects not more than 2.7% of carbon being emitted as methane, but this finding (Archer et al) restricts drivers solely to AGW. To put this in context, 3% as CH4 doubles the combined output’s CO2e value, 6% triples, and 9% quadruples, give or take a smidgen.

    With the most advanced mega-feedback, Albedo Loss, having been cited in GRL in 2010 as already imposing a warming equal to that from about 30% of annual anthro-CO2, in combination the feedbacks’ advance under just 0.8C of AGW makes an utter nonsense of the assumption that Emissions Control alone will resolve the problem.

    Until more scientists find the balls to provide the comprehensive information on which a commensurate policy that includes GEO-E can be built, it appears that we’ll continue sliding towards the edge of terminal decline.



  • Joan Savage

    The abstract doesn’t specify which non-greenhouse gases they considered for the 0.6 W m−2 radiative forcing. As they didn’t specify, presumably all of them.

    If it was only anthropogenic gases, that would be the trio of methane, nitrous oxide, and fluorinated gases, over which humans are supposed to have some control, at present.

    The growth of fluorinated gases should be of exceptional concern, as those gases have very long half-lives, in the tens of thousands of years, as compared to a few hundred years for CO2.

    However, water vapor and ozone are also greenhouse gases, and we’ve already heated up the atmosphere enough to increase the water vapor. I didn’t pay for a full copy of the article, but it would be useful to resolve ambiguity about what they modeled, particularly what percentages they were giving to each of the non-CO2 GHGs.

    Their recommendation is to get a handle on the non-CO2 GHGs, soon, before the permafrost really kicks in. That’s a solid basic take-away message, none-the-less.

  • TheTracker

    This is precisely the same conclusion I came to looking at the permafrost carbon feedback:

    Without some geoengineering, the 2C target is out of reach. I am cautiously optimistic, though, that that geoengineering need not be an impractical sci-if climate hack nor need it persist for thousands of years:

    Love the blog!