Geoengineering as Critical Care

A smoking-cessation skeptic

Suppose your grandmother comes into the emergency room with a severe pneumonia. Probably she should have gone to her doctor last week when she started with a wet cough and a low-grade fever, maybe a little short of breath, but she decided to tough it out. Now her respirations are shallow and fast, she is pale and sweaty, and the toxic byproducts of the bacteria in her bloodstream have stunned her heart, dropped her blood pressure and shut down her kidneys.

She's dying. What do you want the ER doctor to do?

The first option is to do what should have been done last week; prescribe some oral antibiotics, bedrest, lots of fluids. But while that's was right thing to do last week, today that therapy will accomplish exactly nothing, unless you put the nurse in your personal time machine and send him back to last week.

So we are going to come at this a little harder; we are going to treat it like the emergency it is. IV fluids to correct dehydration, IV antibiotics to tackle the infection, supplemental oxygen to give a head start to her struggling lungs.

But sometimes that doesn't work either. The blood pressure doesn't correct with IV fluids; other organs begin to fail; her lungs cannot maintain her body's oxygenation requirements, even with the supplemental O2. She is still breathing fast and shallow and now starting to have heart and liver dysfunction to go with the kidney failure.

At this point your only option(*) is critical care. A breathing tube will do what her lung muscles no longer can. Vasopressive drugs will support her blood pressure. She may need supplemental electrolytes; she may need insulin to control an elevated glucose.

An important fact to realize about critical care is that all of these interventions -- all of them -- are terrible for the body and fraught with life-threatening side effects. None of them are remotely as safe and effective as going in to see your family doctor when you(**) are coughing with a fever and shortness of breath. But, again, no time machine.

Those vasopressors will clamp down your peripheral circulation and can cause skin ulcers, gut ischemia, maybe further cardiac damage. Intubation can lead to long-term respiratory failure, barotrauma (you put too much air in the lungs!), or oxygen toxicity. The IV fluids will leak out of the vessels and cause edema, and so on.

Critical care -- all medicine, really, but especially critical care -- is a matter of trade-offs. We support your critical needs -- especially adequate and well-oxygenated blood flow to your heart and your brain -- at the expense of the normal, orderly functioning of your body. That makes them temporizing measures. Only an idiot would do these things and not also treat the underlying infection with powerful IV antibiotics. Without the antibiotics and functioning immune system, none of the other measures are likely to accomplish anything except to briefly prolong a painful death.

Geoengineering is similar to critical care. It is absolutely inferior to timely mitigation. However we have not carried out timely mitigation, and are now sitting on a massive stockpile of melting permafrost and an inefficient economic system generating huge volumes of CO2 and other GHGs with a large amount of inertia. Even if we were to embark on an ideal program of mitigation today, we would likely end up over the 2C threshold.

There is no point in geoengineering if we do not also intensively mitigate. It is bound to be at best a partial solution, with many side effects, and much more expensive than it looks on paper. Mitigation is like the antibiotics; the critical care in essence buys time for the real solution to work.



-----------------------------------------------
* Other than hospice, which doesn't really work with this metaphor.
** Grandmother/otherwise elderly you. If you're under fifty, it's probably just a cold, you big baby.

Adapt, Geoengineer, Mitigate (AGM)

Source: National Research Council. 2011. Climate Stabilization Targets: Emissions, Concentrations, and Impacts over Decades to Millennia. p.101. Washington, DC: The National Academies Press

So it's time to talk about geoengineering. Like it or not.

The reality that a significant amount of carbon dioxide and methane are going to emerge from melting permafrost seems at last to be making an impression on the popular press (thanks latterly to a recent UN report). The implication, that it is extraordinarily improbable that mitigation alone will be able to limit warming to < 2C above preindustrial, does not seem to have sunk in yet, but the logic is fairly inescapable.

Investigations of climate sensitivity continue to come back with values clustering around 3C/doubling. Based on those values, the total amount of CO2 that can be added to the atmosphere and still leave us with the hope of keeping warming below the 2C target is about 1,000 gigatons or a trillion tons. Total CO2e emissions to date are between 500 and 600 gigatons. Permafrost emissions by 2200 are estimated to be between 246 to 415 gigatonnes. Take the midpoints of both ranges and add them together and you get 880 gigatons. Right now the world is adding to that figure at a rate of 30 gigatons of carbon dioxide per year, plus sundry other GHGs.

Even if you could instantly cut GHG production by 90%, you'd still cross the threshold within 40 years or so.  Any sort of a realistic program -- and in that I include a WWII-style crash program to cut emissions, consuming a significant chunk of the planet's GDP over the next few decades -- would come nowhere close to meeting the trillion ton target.

Taking the most likely case -- that scientists' warnings continue to fall on deaf ears for at least a half a decade -- we will commit ourselves absolutely in four to five years -- perhaps less, if the world economy grows at a brisk pace.

It's possible, of course, to stick with the party line despite the inevitability of crossing the trillion-ton mark. Emission cuts as fast as possible; adaptation; and hold on to something, because the 21st century is looking like a bumpy ride. But it is more in the spirit of climate realism to face the facts honestly and, where necessary, change our strategy.

What are those facts? Fact number one: carbon-cycle feedbacks will put the 2C target out of reach through mitigation alone. Fact two: the severity of the climate impacts we are seeing at 0.8C above preindustrial suggests the 2C is a hard target. Two and a half times the warming we have seen to date is already, probably, outside any reasonable boundary of "safe" temperatures. Fact three: the Arctic permafrost is not the only game in town. There is also carbon under Antarctica. There are methyl hydrates. There is carbon locked up in the Amazon and other forests vulnerable to die-back.

If you take the 2C limit seriously, you have to consider that the time may be approaching where we will need geoengineering as a bridge to lower levels of GHGs.

Geoengineering has a bad reputation. People fear it as a quick fix, a barrier to the changes we need, and a long walk off a short pier into the Bay of Unintended Consequences. It has the potential to be all of those things. But it seems increasingly unlikely that we will get through the next two centuries in one piece without it.

The AGM strategy has three elements:

Adapt: Prepare our defenses and infrastructure for multi-meter sea level rise and the storms of the 21st century. Prepare our water resources for droughts, salinization, and flooding. Prepare our emergencies services, diversify our food crops, improve the robustness (and efficiency) of our infrastructure.

Geoengineer: Start planning with small-scale tests now; larger-scale tests as soon as feasible; infrastructure for large-scale deployment as soon as we have a workable technology or set of technologies. Then set a hard upper limit well back from the 2C boundary -- like 1.5C, or at the first sign of a catastrophe like massive methyl hydrate degassing. At 1.5C over preindustrial, geoengineering kicks in.

Mitigate: Agreement to severe and ongoing cuts in GHG emissions between a few large powers, with serious diplomatic and economic arm-twisting as necessary to enlist the rest of the world. Our goal should be to get back to 350ppm CO2e

Could wildly successful geoengineering decrease the pressure for an agreement on serious mitigation? Sure it could. But you have to ask yourself if you believe the science.

If you do believe the science, and understand that as we approach 2C our civilization and most of the species we share the earth with are in mortal danger, then while that perverse incentive matters, it can't be paramount, any more than the fear that people will eat too much and not exercise is a reason to not put a heart attack survivor on blood pressure and cholesterol-lowering medications. Yes, they have side effects. Yes, they are in some respects an artificial compensation for a failure in self-control. Nevertheless, letting the patient drop dead is a bad option. Better to use the artificial support, and continue to campaign for the lifestyle changes.

The permafrost carbon feedback, ctd

Field information links permafrost carbon to physical vulnerabilities of thawing

Deep soil profiles containing permafrost (Gelisols) were characterized for organic carbon (C) and total nitrogen (N) stocks to 3 m depths. Using the Community Climate System Model (CCSM4) we calculate cumulative distributions of active layer thickness (ALT) under current and future climates. The difference in cumulative ALT distributions over time was multiplied by C and N contents of soil horizons in Gelisol suborders to calculate newly thawed C and N. Thawing ranged from 147 PgC with 10 PgN by 2050 (representative concentration pathway RCP scenario 4.5) to 436 PgC with 29 PgN by 2100 (RCP 8.5). Organic horizons that thaw are vulnerable to combustion, and all horizon types are vulnerable to shifts in hydrology and decomposition. The rates and extent of such losses are unknown and can be further constrained by linking field and modelling approaches. These changes have the potential for strong additional loading to our atmosphere, water resources, and ecosystems. 

Similar general estimates as compared with this paper: 68 billion tons to 508 billion tons in 2100 versus 436 billion tons. This is estimated thawed carbon, though; it's not clear to me from the abstract if they even attempt to estimate how much of that ends up in the atmosphere. This is particularly important in the case of nitrogen, given that NO2 is a powerful greenhouse gas (with 310 times the warming potential of CO2) as well as an ozone-eating chemical.

Committed warming

Andrew Weaver, the faculty author on Nature Geosciences' new paper on the Arctic permafrost carbon feedback (not to be confused with the new paper implying the risk of a catastrophic Antarctic carbon feedback), is talking about the paper from his perch at the Huffington Post. He offers his take on the committed warming in the pipeline:

Instrumental records have clearly revealed that the world is about 0.8°C warmer than it was during pre-industrial times. Numerous studies have also indicated that as a consequence of existing levels of greenhouse gases, we have a commitment to an additional future global warming of between 0.6 and 0.7°C. Our analysis points out that the permafrost carbon feedback adds to this another 0.4 to 0.8°C warming. Taken together, the planet is committed to between 1.8 and 2.3°C of future global warming -- even if emissions reductions programs start to get implemented.

So <+2C is off the table, unless we geoengineer (Planet 3.0 is revisiting the subject now). So there's that.

The last time the Earth was two degrees warming was the mid-Pliocene. That was about three million years ago. So common sense would suggest we prepare for a world that looks similar to the mid-Pliocene. Well . . .

* Sea levels were between 15m and 25m higher.
* The WAIS repeatedly melted back (repeatedly), potentially uncovering large carbon reserves.
* The Arctic was 10C-20C hotter than the present day (suggesting that while our climate models can't fully reproduce the degree of Arctic amplification we observe today, it's likely to be a real and persistent feature of the climate system.)

It will be interesting to see how cost/benefit estimates change once this permafrost carbon feedback is "priced in" to economic models of climate change. On the one hand, damages will increase. On the other hand, the differences between BAU and intensive mitigation scenarios will decrease, because the permafrost feedback will cause less warming in a hotter world, and more warming in a cooler world. Just to give a simple illustration, consider a carbon feedback that adds 100ppm of CO2 to the atmosphere. Then overlay that atop two scenarios:

* Aggressive mitigation -- 450ppm CO2
* BAU -- 800ppm CO2

Those numbers go into this equation:

 Where C(0) is preindustrial CO2 (280ppm), and C is 450ppm and 800ppm, respectively. Do that and we get a forcing of 2.5W/m^2 vs 5.6W/m^2. But look what happens when we add the permafrost carbon:

* Aggressive mitigation -- (450 + 100=) 550ppm CO2 = 3.6W/m^2
* BAU -- (800 + 100=) 900ppm CO2 = 6.2W/m^2

Both forcings have increased, but the mitigation scenario has increased far more, making the difference between mitigation and carbocide somewhat smaller: 2.6W/m^2 compared to 3.1W/m^2.

It's not a huge difference, but it narrows the difference between action and inaction, and least in terms of forcing. At the same time all scenarios get more expensive and destructive. It'll be interesting to see which effect is stronger.