Putting an Upper Bound on Climate Change: Permafrost

Disclaimer:  I am now retired, and am therefore no longer an expert on anything.  This blog post presents only my opinions, and anything in it should not be relied on.

When trying to figure out the most likely rate of global warming in our “business as usual” society, I always go back to that amusing quote from the movie version of Tolkien’s Fellowship of the Ring:  “Are you scared?  Not scared enough!”  The reason I remember this is that up to now, all of our scientific models of how climate change will play out over the next 100 years or so have consistently underestimated the rate of warming, not to mention the rate at which weather extremes have become the “new abnormal”.  And the reasons for this underestimation in the scientific community, from what I can tell, are (a) conservatism (even though the modeler may know that an effect will add to global warming, until the effect can be bounded with 95% likelihood in a positive direction it is assumed to have neither positive nor negative effect) and (b) lack of knowledge (for example, for a long time, the effect of increased cloudiness on global warming was not clearly understood).

Over time, these certainties and uncertainties in models have tended to be sorted out into the categories of Donald Rumsfeld’s pernicious classification:  knowns, known unknowns, and unknown unknowns:

1.       The knowns are now, effectively, all those direct effects (e.g., the direct effect of CO2 and methane on global warming) and feedback or “knock-on” effects (e.g., the effect of release of black carbon by wildfires on decreasing Arctic albedo, where global warming from CO2 increase causes increased wildfires, which causes black carbon to fall on snow and ice, decreasing Arctic albedo and hence causing more Arctic warming).

2.       The known unknowns are those effects that we know will play a role, but which are not incorporated in models because we don’t know the extent of the role they will play.  Of these, perhaps the largest ones are (i) melting permafrost and (ii) methane (CH4) releases other than from permafrost melting, including methane clathrates, release of methane locked in land and ocean repositories, and release of methane from human action, including cow emissions and natural-gas processing emissions.

3.       The unknown unknowns are, obviously, effects that we probably can’t anticipate until they happen.  We can, however, take a guess at them by looking at the relationship between CO2 atmospheric emissions and global temperature during the last two major rises in CO2 – 250 million years ago (mya) and 55 mya (the Paleocene-Eocene Thermal Maximum, or PETM, also known as “hell and high water”).  If known unknowns turn into knowns, then the difference between what our models tell us will happen and what past experience suggests is probably due to unknown unknowns.
  

If we can turn the known unknowns into knowns, there appears to be a reasonable prospect that the revised models will track with past PETM (and other) experience, which would mean that the unknown unknowns probably don’t have a significant impact on global warming one way or the other.  In fact, the only remaining known unknown upside to be accounted for might be the fact that we are accelerating carbon emissions at a far greater pace than they have ever increased before.  The usual thinking is that this may speed up the process of actual global temperature warming, (e.g., in our case, a doubling of atmospheric carbon since 1850 should lead to a 2-2.8 degree C temperature increase in the next 100 years, and a total 4 degrees C temperature increase over the next 1000 years, rather than a 1200-year gradual doubling leading to an increase of 1.3 degrees C by 100 years from now, but with the same total 4 degrees C increase at the end), although it is possible that our rush to emit carbon will trigger unknown unknowns that will lead to an even greater rise in temperature at the end of each doubling.  In a nutshell:  with permafrost and methane accounted for, we would begin to see a plausible upper bound for the amount of long-term global warming from increased atmospheric CO2 and related greenhouse-gas emissions.

In long-ago posts, I said that research seemed to suggest that increased methane emissions would not deliver a major boost to global warming.  Without going into horrendous detail, I saw the research as suggesting that if even if methane clathrates, sea-floor methane, and permafrost methane suddenly started rising into the atmosphere a very high rates over the next 100 years, it was still unlikely that methane would achieve “saturation” in the atmosphere, which would cause methane to stay in the atmosphere much longer and therefore have a much greater warming effect – perhaps as much as 1 degree F while it lasted.  Nothing in the second edition of Joe Romm’s “Climate Change:  What Everyone Needs to Know” (summarizing findings as of early 2018) suggests that this scientific consensus has changed.

And so, I would argue, getting at the likely effect of melting all the permafrost gives us a reasonable shot at an upper bound for the most likely effects from “business as usual” increases in carbon emissions.  Or, as I like to say, we may finally know how scared is scared enough.

New Findings on Permafrost

Only nine years ago, it wasn’t clear that permafrost melting was happening yet.  Now, it’s not only clear that it’s happening, it seems to be happening faster than anticipated.   So a plausible upper bound for global warming is increasingly seeming more like a “middle bound” – the most likely case – for business-as-usual global warming.

Here’s a back-of-the-envelope SWAG at the effects of full permafrost melt.  According to “Climate Change”, there are 1.5 trillion tons of carbon locked in the world’s permafrost.  Studies cited in CC also suggest that for each 90 tons of permafrost carbon released, atmospheric carbon goes up about 60-80 ppm.  Thus, total melt means about 1000-1333 ppm of added atmospheric carbon.
 
The effect on global warming then depends on how fast it happens and what else is going on.  More specifically, if melt is fast, then most of its effect falls within the first two doublings of atmospheric carbon.  Thus, very loosely, with a 2 degree C direct effect of non-permafrost added atmospheric carbon on temperatures, we are talking about somewhere around a 4 degree C direct effect of total permafrost melt.

But direct effects of carbon are not the only effects.  A significant proportion of the carbon in permafrost will almost certainly be emitted as methane (CH4), not carbon dioxide.  While much of this will eventually turn into CO2 in the atmosphere, the rise in per-decade atmospheric methane during quick permafrost melt can add up to another degree C to warming over centuries-to-millenia time periods, because methane can be up to 86 times as powerful a greenhouse gas as carbon dioxide.  And the lack of snow cover and increased vegetation associated with the peat bogs likely to be a result of permafrost melt mean a shift in albedo that, to my mind, should significantly up global warming as well.  In sum, aggressive total permafrost melt seems to me at first glance to result in an average 4-6 degrees C of global warming for at least the next 1000 years – and, considering the unknown unknowns, that may be conservative.

It should be pointed out that if we are effective in ending business-as-usual global warming in the near future, much less of the permafrost will ultimately melt.   At another SWAG, we may have locked in melt of 5% of permafrost already (the result of non-permafrost feedbacks adding 0.5 degrees C to the temperature henceforward plus the effects of additional heating from initial permafrost melt).  The rest is still in play.

Glimpses of an Upper Bound

What strikes me about this estimate (4-6 degrees C) is that it seems to fit in more nicely both with the experience of past catastrophic global warming and with recent Hansen papers trying to set an upper bound.  In the PETM, carbon ppm started at about 1000 and apparently went up to about 2000, but the temps seem to have gone up 6 degrees C in the process, 4 degrees more than direct atmospheric carbon effects can explain.  One recently raised suggestion is that elimination of stratocumulus clouds, which should thin out as air temps get warmer, can add 8 degrees C to global warming.   But this is not likely to take effect until well beyond the 1000 ppm level, below which aggressive permafrost melt plays its major role.

In trying to square the PETM with the likely direct effects of atmospheric carbon, Hansen conjectured, iirc, that each doubling of atmospheric carbon would lead, without human interference, to a little more than 2 degrees C of direct effects plus a little less than 2 degrees C of feedback effects, for a total of 4 degrees C.  But the feedback effects he identified only seemed applicable to the first doubling (to 550-odd ppm), and the temps in the PETM seemed too high, imho, for a 2000-ppm atmospheric carbon level.  Some combination of permafrost melt and cloud thinning, with permafrost melt playing its major role before 1000 ppm (at the time of the late Cretaceous global warming 11 million years before, with loss of sea ice and therefore probably loss of permafrost) and cloud thinning after, may therefore explain some of the PETM warming (whether it played a role in the global warming/mass extinction 250 mya, which involved a super-continent with no extreme north or south land, is not clear).

This also squares with another Hansen paper (iirc). In it, he argues that using up most of our present fossil fuel reserves will lead ultimately to 16 degrees C/30 degrees F average warming, or 30 degrees C/54 degrees F average warming in the Arctic and Antarctic.

All of the above statements contain large fudge factors, of course, and can be and are argued over.  However, I do now believe it is much more probable now that something like this upper bound is in store for us, if business as usual continues indefinitely. 

Implications for Armageddon

With all of the above in mind, I like to think of the worst as four stages – horrible, catastrophic, apocalyptic, and decimating, the first three corresponding to doublings of atmospheric carbon, and the last comprising only feedback effects.  Aggressive permafrost melt gives the second and third stages, which have yet to arrive, a similar warming effect over time to our first stage.  And that arbitrary classification leads to a couple of assertions:

1.       Each stage is much worse in its effects than the previous one; and

2.       Each stage becomes harder to prevent or ameliorate than the previous one.  In a sense, each stage adds momentum and size to the downward rolling snowball, because (a) new feedback effects are triggered, like permafrost melt and Arctic/Antarctic ice melt, (b) saturation effects start occurring, like inability of the ocean to store oxygen or saturation of atmospheric methane leading to longer stays in the air, and (c) our ability to mitigate is increasingly hindered by having to undo greater  amounts of now-inappropriate fossil-fuel infrastructure and optimized-for-the-wrong-temperature energy-inefficient dwellings.

      And that is why I disagree with both the optimists and pessimists.  Yes, failure to undo business as usual is rife in the air.  Yes, we may well have effectively crossed the boundary into Stage 2, with horrible effects of climate change already locked in.  Yes, the ultimate end of Stage 4 may well be the decimation of the human race and the rest of the natural world.  But until 100 or 200 years from now, we still have the ability to break utterly the cycle of business as usual, harder though it gets as time goes by.  And therefore every setback in our quest demands not despair but a fiercer effort, more confrontation of the guilty, and greater demands for even more to be done.

It’s like a crooked poker game.  Yes, right now it’s the only game in town, and yes, right now you can’t win or fold, just keep on losing.  But the game will end, and your job is to lose as little as possible, as the stakes go up and up, so that when the game is over you may have lost your heart’s desire but you will not be an indebted slave for life, or killed because you couldn’t pay your debt.  Or most of the effects may fall on your great-great-grandchildren; but they’re still the same shattering effects.  On everybody.

Oh, and one extremely small hopeful note:  it seems to me that to avoid business as usual resurfacing at any time over the next 1000 years, we will have to change fundamentally.  To sustainability.  To a method of living that almost certainly will not destroy the remainder of humanity.  Stage 4 will be as bad as it gets.

And so, hopefully, understanding the effects of permafrost and having a likely upper bound means you are scared enough.  Absolutely, totally scared stiff.  With decimation staring you in the face.  And so, said the psychiatrist in Portnoy’s Complaint, now ve may perhaps to begin, yes?