Fan Fiction
I see a few kinds of wrongheadedness in how a lot of ‘climate technology’ gets developed:
- Tangled-up thinking about why we might want to address climate change in the first place.
- A misunderstanding of the nature of progress in science and technology.
- Signaling value overriding unit economics in funding decisions.
The technology that really brings all three of these together like no other is air capture of carbon dioxide.
-----------------------------------------
Direct air capture of carbon dioxide, “DAC”. Air separation technology that cleans CO2 from the air. Some call it a necessary piece of our net-zero-by-whenever economy. But can it really help?
Just think a minute: we’d need to fully remove the CO2 from ~2km3 of air per second to keep pace with current emissions[1]. As Neal Stephenson says “that’s gonna take some big fans”.
With about 10 billion people on earth, this is conservatively ~100 liters of air per person per second, assuming 100% removal efficiency. That’s more than the airflow of a big engine running full speed, one engine per person on planet earth, somehow sucking 400ppm CO2 all the way down to zero.
Global numbers are often astounding, since multiplying anything by 1010 makes for big numbers. But when the numbers are huge even on a per capita basis, it might be cause for further reflection.
And that’s just the airflow piece. You still have to separate the CO2 from air, which might take a third or so of world energy[2]. Then you need to store it. The storage proposals that don’t involve injecting CO2 into oil wells to get yet more oil often turn the CO2 back into solid carbon, which is of course tantamount to ‘unburning’ the fuel and would require nearly all of the energy that society currently uses[2]. Don’t forget that the whole gambit has to run on carbon-free energy or there’s no point—oh, and unfortunately the decarbonized share of US primary energy has only ticked up from 9% to 13% over the last 35 years[3].
We still pay coal miners just dollars per ton to pull carbon out of the ground. You really think the same economy is going to pay Silicon Valley hundreds of dollars per ton to put it back? Huh. No one is going to pay for that except to varnish their reputations, and varnishes get applied in thin layers….
“But wait”, they say, “these $400/ton DAC plants are just learning investments to get lower cost DAC in the future”. Do you really think that? Every step in DAC is a process that the chemical industry has been refining for a hundred years—moving air, separating gasses, injecting CO2 into the ground or reducing it to solid carbon. The remaining ground to cover on better fans, a better adsorbent, a better injection technique are 1.1x improvements. Maybe 2x. Not 100x.[1]
We’re operating against a backdrop of obvious stagnation in all of the physics-driven fields[4]. Despite claims to the contrary, there’s no sequel to Moore’s law, the Manhattan Project, or the Moonshot waiting behind the century-old technology that underlies DAC[5].
So big fans? More like fan fiction. Like in our movies these days, Jurassic Park 4 or something, nice effects can’t save plotless riffs on the stories of the past.
What you think about DAC shouldn’t depend on whether you see climate change as chaos and inundation, or as an annoyance that might cost a few years global GDP. I happen to think climate change could be serious[6]. But these cargo-cult technologies are not helping[7]. That goes for DAC, but also to varying extents for the other climate technologies that clearly can’t compete on the market. Fantastic projections for 100% grid penetration of solar and wind[8] without storage, ill-advised ventures filling planes or ships with smartphone batteries, or easily falsifiable claims of economical electrolytic production of fertilizer, cement or steel.
Climate change presents a tragedy of the commons dynamic between nations but also between generations; the time constants of the earth system are so long that most of our investment in climate won’t pay off (in terms of climate damage avoided) until after the careers of all decision makers involved[9]. That’s why we’re misallocating our resources so badly. The strongest incentives now are virtue signaling incentives, and those incentives are best followed by playing out dubious but good-feeling charades like DAC, or building other technologies (electrified x, bio-derived y) that conform to a popular conception of what decarbonization should look like, whether or not they can work at scale.
It’s also interesting to ponder what’s spinning up these DAC fans in the first place. “Climate change” these days collectivizes many peoples’ fear and loathing over all the externalities of hosting ~10 billion people on planet earth. If you cared about human suffering, you’d worry more about air pollution, which will certainly kill more people this century than climate change. If you cared most about the biosphere, you’d focus on agriculture and habitat loss, which has caused 99% of species loss so far and will still cause most even in a radically warming world. Commodity fetishism unites these concerns, perhaps counterproductively, around CO2 (when a metric becomes a target…). So of course the most literal thinkers want to remove the CO2 directly. But even if that were feasible, it wouldn’t solve the problems they say they care about most.
It's usually not good practice to disparage particular technical approaches. But DAC seems to earn an exception. Partly because there are real solutions out there—some of them technological. Electrification and renewables yes, but also things getting far less support: different ways of producing food, reducing waste, closing the biggest free energy leaks in the commodity economy. Climate technologies that don’t just react to a fears about the future, but that are visions of a better future in their own right.
In the end, all the DAC fans are doing is sucking the air out of the debate on tractable solutions. New energy technologies that could make money and spread naturally in a capitalist system. A real conversation about acceptable tradeoffs in our relationship with the earth system. And yes, even promising strategies for dealing directly with the planet’s excess CO2[10].
Footnotes
[1] It's true that there are other DAC proposals that might make more sense than brute force mechanical air separation. Crushing up rocks in the desert, or harvesting biomass and either burying it in plastic bags or dropping it into anoxic seas. And many other with varying probabilities of success. But DAC via direct air separation is the best known, so it’s the punching bag here.
[2] Worldwide power consumption, all forms = 6e20 J/year = 19 TW. Worldwide electrical power = 24 PWh/year = 2.7 TW. CO2 --> C + O2 @ 30 Gt/year >= 8.4 TW. Climeworks 2MWh/t DAC @ 30 Gt/year = 6.8 TW.Total DAC + solidification @ 30 Gt/year >= 15 TW
[3] Vaclav Smil via Michael Cembalest, 2021
Also Vaclav: Kindergarten algebra is all you need: in order to remove just 15% of today's annual emissions we would need a 6 Gt/year industry, larger in mass terms than the global oil industry that took over a century to develop and that earns (today's price) about $600 per ton lifting that liquid, while shoving the gas (or liquefied gas) underground with no guarantee of it staying there would be pure cost.
[4] There’s a much-discussed tension between the popular narratives of dizzying growth (AI, synbio, Kurzweil etc.) and relative stagnation (Smil “Invention and Innovation”, Cowen “The Great Stagnation”).
But I don’t think it’s too mysterious, or requires too much navel gazing about human capital or institutional malaise. The great leap in physics from Maxwell to Dirac paid 50 years of unbelievable dividends in the physics-driven fields. Now they’re drying up. Later fundamental breakthroughs (Watson & Crick, Turing’s computer) are paying dividends in computer science and biology still, but it’s not clear how relevant they’ll be to the giant commodity industries with the biggest climate impacts. To expect meteoric returns without further foundational breakthroughs is to misunderstand the nature of progress.
[5] In the 40’s, they’d just discovered a new fundamental force and had the wartime budget of a nation behind them. These days, our loudest DAC evangelicals made their money delivering groceries, running credit cards or making zombies out of people with poor impulse control.
[6] How do we respond to the tipping pointers, the doomers, the zero-times-infinity existential risk crowd? (i.e., maybe we study geoengineering for a century only to find all known routes to be impractical, and then the permafrost melts and we're headed for borderline uninhabitable Earth. We are a century behind in developing machines to put carbon away.) I think there is a risk, probably small, and to some extent I’m glad that Carbon Engineering et al. have been tossed >$3B over the last decade to calculate the upper bound on what it will cost to solve this problem. But don’t kid yourself that we could actually implement it even then without a radical (and probably radically anticapitalist) change in global governance.
[7] Feynman, Cargo Cult Science
[8] Rebuttal of Jacobsen’s “100% renewables with no new technology” that got Chris Clack sued.
[9] Greed motivates people. Ken Caldeira’s “Breakeven year” for when climate mitigation pays the mitigators is 2080, which is robust to different mitigation costs and climate damage scenarios. The world will really mobilize when people feel the heat, not sooner. Until then, we have to work hard to find technologies that are simply better, and that requires thinking much more expansively than we are now.
[10] It’s worth studying geoengineering. If you honestly count up the ways we already geoengineer the planet, cover 15% the habitable surface of the earth in crop monocultures, graze purpose-bred animals on 30%, artificially seed half of the clouds in the northern hemisphere with soot aerosols, redirect half the freshwater flows, routinely vandalize the sky with contrail-seeded cirrus clouds, fill the ocean with more plastic than fish — if you countenance that wild geoengineering we do, it might take relatively minor changes to the chemistry of the ocean or beaches to pull CO2 from the sky, tiny changes to the stratosphere or marine clouds to bounce sunlight back to space. Certainly less meddling with the earth system than the Gobi’s worth of PV you’d need to power your DAC fantasy.