> Lot's of commenters on HN do not seem able to (or willing to) exercise critical thinking. Some might not have the requisite scientific background or real-world experience to understand what are otherwise simple arguments rooted in physics.
Sure, that's true, but are those the people whose opinion you care about? If you tailor your argument for them and leave out all the reasoning, you'll look like one of them, and people who are able and willing to exercise critical thinking on the topic will write you off. Don't tempt people to confuse you with dumbfucks like Ted Turner and Noel J. Brown.
> You cannot filter air at a planetary scale, ... These systems are far more likely to "filter" the same air over and over again than to actually "remove all of it".
You might be right that you need more than a single giant inverse volcano sucking carbon dioxide out of all of the air.
The diffusivity of CO₂ in air is 16mm²/s, or 1.6 × 10⁻⁵ m/s in SI units. Suppose, conservatively, that there are no jet streams or any other wind, so you have to rely entirely on diffusion, and that you've reduced the CO₂ around your scrubber to the pre-industrial 280ppm (0.012mol/m³), while the opposite side of the planet, 20000km away, is still at 400ppm (0.017mol/m³). If the concentration gradient were constant, it would be 0.25 nmol/m³/m over that distance. Using the given value of the diffusivity, this results in 4 × 10⁻¹⁵ moles per second per square meter flowing over that gradient. The atmosphere is in effect about 8 km tall, so at the halfway point of this flow, the great circle halfway along the path, you have 320000 km² of cross-sectional area, through which your flow is about 0.000128 mol/s. If you're trying to clean up the air over, say, 20 years, that works out to about 800 kilomoles, only about 36 tonnes of CO₂.
But we had to remove 2.3 gigatonnes of the atmosphere's 7.8 gigatonnes of CO₂, not 36 tonnes. We need a system that's 64 million times more powerful, which means that instead of being 20000 km away from the farthest points on the planet, it needs to be about 20000/8000 = 2.5 km. (Also, the above is assuming that the gradient remains constant, which of course it can't; you'll have a much stronger cross-sectional gradient around anything like a point sink or source of carbon, because the same mass flow is spread over a much smaller area. I don't know the exact solution for diffusion on the surface of a globe, but I'm confident the above is in the ballpark.)
But we do have wind; to take the most extreme example, the jet streams travel about 50m/s, so any parcel of air in them circles the globe about once a week. So if you set up your antivolcano near a jet stream and create a locally very low CO₂ concentration, CO₂ will diffuse rapidly out of the jet stream as it travels through your area and diffuse rapidly back into the jet stream on the other side of the planet three days later.
So if you have winds that bring all of the atmosphere within 2.5km of your antivolcano at some point within those 20 years, you can scrub it all. All of it.
> Notable eruptions in recent years appear to have affected climate. One example is the 1991 eruption of Mount Pinatubo in the Philippines, which injected nearly 20 million tons of SO2 into the stratosphere that became dispersed around the globe in about 3 weeks. The recorded effect was a 0.5 degrees C (0.9 degrees F) drop in temperature for the following two years.
So in practice the timescale on which an antivolcano's CO₂-scrubbing effect would reach the air on the other side of the planet is only a few weeks, not the hundreds of millions of years that the naïve diffusion calculation suggests.
You say:
> In the end, you cannot create a clean room by filtering the enclosed volume.
I haven't tried, but I'd venture to guess that that's because dust, like flour, doesn't diffuse through air (at human temperatures on human timescales). It's a thousand times denser than air, so it can settle on surfaces and get trapped in certain kinds of airflow patterns. Carbon dioxide doesn't behave like that. It behaves like water vapor.
> Scrubber systems in the corner-case applications where this isn't an option are incredibly expensive and complex. They do not scale well at all.
Unlike cleanroom filters, scrubbers, which are gas-exchange devices used for removing unwanted gases from air, are actually very cheap and simple, and they scale extremely well; they are already being operated at many-megawatt industrial scale, which is how we ended the acid rain problem caused by coal power plants. A CO₂ scrubber can be as simple and low-tech as a wall painted with whitewash, though soda-lime scrubbers (a few percent of lye added to the whitewash, which is balled up in little beads to let the air pass through) are much more common because they are so much more compact. They are in wide use for general anesthesia, scuba diving, and decompression chambers. On the order of a million recreational scuba divers own their own scrubbers. You can buy a bottle of 5 liters of soda lime for €32 at https://www.diveavenue.com/en/high-pressure/1860-chaux-spher... if you stick a porous breathing tube into the bottle, it becomes a scrubber, although you'd probably die if you tried diving with it.
Other kinds of gas scrubbers are available as disposable cartridges from 3M for their respirator masks, so it's absolutely not true that they're "incredibly expensive and complex", but let's stay focused on CO₂ scrubbers here, since that's what's relevant to climate change.
The disadvantage of lime CO₂ scrubbers is that, although they are very simple, regenerating the lime requires heating it up to 900°. People have been doing this for 12000 years (it's probably literally the oldest human chemical process) but it takes a lot of energy compared to other sorbents like triethanolamine, which is why amine scrubbing is currently the mainstay of both submarine CO₂ scrubbers and point-source CO₂ capture pilot projects. There's a nice process flow diagram of a submarine CO₂ scrubber at https://sites.psu.edu/mooneypassionblog2/2022/03/22/how-subm....
As for scaling, lime burning, in the slightly altered form of portland cement making, is one of the world's largest-scale industrial processes, producing 4 billion tonnes of cement per year and emitting 1 billion tonnes of CO₂, which you'll notice is about 2½ years for the mass of the carbon dioxide we have to remove. (Portland cement, unlike Neolithic-style lime cement, only reabsorbs a fraction of that CO₂ when it sets.) So we already know we can scale lime kilns up to the requisite levels; we just have to capture their carbon dioxide output. There are also a bunch of startups trying to scale up amine scrubbing and other processes to direct air capture, but I'm skeptical that their supply chains will be able to compete for scale with the world limestone-and-other-calcareous-stone industry.
> My point is that the idea of filtering our entire atmosphere (or a substantial enough portion of it) only works on paper, research grant applications and proposals for funding by politicians eager to continue the narrative that drives unthinking masses to vote for them.
Every engineering idea only works on paper until you build it. That's how engineering works: you make things work on paper, you validate your assumptions with prototypes, you learn from your mistakes, and finally you create something that never existed before. Your criticism here is a fully general criticism of every technical innovation in history; it doesn't distinguish between perpetual motion machines (which work on paper if you screw up your calculations badly enough) and any routine civil engineering project such as a bridge embankment.
As shown in https://news.ycombinator.com/item?id=42439752 my numbers above for the total atmospheric carbon dioxide load and the amount to remove are wrong by orders of magnitude. The total is 3.3 teratonnes, of which we will need to remove 1.1 teratonnes to get back to preindustrial levels, not just 2.3 gigatonnes. I believe this reduces the 2.5km number above (for the range over which diffusion alone would transport the necessary CO2 flux) to some 400m. Assuming I got the rest of the calculation right, which I'm uncertain of.
This also means that the world cement industry is not currently large enough to carry out the necessary direct air capture over a timespan of decades. You need a carbon dioxide capture industry 10 or 20 times larger; at current energy and material prices, this would cost on the order of 6 trillion dollars per year, about 6% of the world GDP of US$105 trillion per year, nominal, World Bank estimate: https://en.wikipedia.org/wiki/List_of_countries_by_GDP_(nomi.... World GDP has been growing about 3.9% per year, so rather than setting us back to the Stone Age, or even to the 18th century, this staggering expense would set us back to about the time Silicon Valley Bank collapsed, Hamas invaded Israel, OpenAI released GPT-4, and Microsoft bought Activision Blizzard.
This clearly demonstrates that it is technically feasible already, just not economically/politically. Rapidly falling energy prices thanks to the transition to super-cheap renewable energy will ease the economic difficulties, though the project may still require international diplomacy.
Also, https://earthscience.stackexchange.com/questions/994/how-lon... says, "The time scale of interhemispheric tropospheric transport is in the order of one year," not a few weeks. I assume that the difference from the Mount Pinatubo number is because the troposphere mixes more slowly than the stratosphere because in general the winds in the troposphere are slower. (However, the jet stream in particular is at the tropopause, where the two meet.)
Sure, that's true, but are those the people whose opinion you care about? If you tailor your argument for them and leave out all the reasoning, you'll look like one of them, and people who are able and willing to exercise critical thinking on the topic will write you off. Don't tempt people to confuse you with dumbfucks like Ted Turner and Noel J. Brown.
> You cannot filter air at a planetary scale, ... These systems are far more likely to "filter" the same air over and over again than to actually "remove all of it".
You might be right that you need more than a single giant inverse volcano sucking carbon dioxide out of all of the air.
The diffusivity of CO₂ in air is 16mm²/s, or 1.6 × 10⁻⁵ m/s in SI units. Suppose, conservatively, that there are no jet streams or any other wind, so you have to rely entirely on diffusion, and that you've reduced the CO₂ around your scrubber to the pre-industrial 280ppm (0.012mol/m³), while the opposite side of the planet, 20000km away, is still at 400ppm (0.017mol/m³). If the concentration gradient were constant, it would be 0.25 nmol/m³/m over that distance. Using the given value of the diffusivity, this results in 4 × 10⁻¹⁵ moles per second per square meter flowing over that gradient. The atmosphere is in effect about 8 km tall, so at the halfway point of this flow, the great circle halfway along the path, you have 320000 km² of cross-sectional area, through which your flow is about 0.000128 mol/s. If you're trying to clean up the air over, say, 20 years, that works out to about 800 kilomoles, only about 36 tonnes of CO₂.
But we had to remove 2.3 gigatonnes of the atmosphere's 7.8 gigatonnes of CO₂, not 36 tonnes. We need a system that's 64 million times more powerful, which means that instead of being 20000 km away from the farthest points on the planet, it needs to be about 20000/8000 = 2.5 km. (Also, the above is assuming that the gradient remains constant, which of course it can't; you'll have a much stronger cross-sectional gradient around anything like a point sink or source of carbon, because the same mass flow is spread over a much smaller area. I don't know the exact solution for diffusion on the surface of a globe, but I'm confident the above is in the ballpark.)
But we do have wind; to take the most extreme example, the jet streams travel about 50m/s, so any parcel of air in them circles the globe about once a week. So if you set up your antivolcano near a jet stream and create a locally very low CO₂ concentration, CO₂ will diffuse rapidly out of the jet stream as it travels through your area and diffuse rapidly back into the jet stream on the other side of the planet three days later.
So if you have winds that bring all of the atmosphere within 2.5km of your antivolcano at some point within those 20 years, you can scrub it all. All of it.
https://www.usgs.gov/news/volcano-watch-global-reach-volcani... says of Mount Pinatubo:
> Notable eruptions in recent years appear to have affected climate. One example is the 1991 eruption of Mount Pinatubo in the Philippines, which injected nearly 20 million tons of SO2 into the stratosphere that became dispersed around the globe in about 3 weeks. The recorded effect was a 0.5 degrees C (0.9 degrees F) drop in temperature for the following two years.
So in practice the timescale on which an antivolcano's CO₂-scrubbing effect would reach the air on the other side of the planet is only a few weeks, not the hundreds of millions of years that the naïve diffusion calculation suggests.
You say:
> In the end, you cannot create a clean room by filtering the enclosed volume.
I haven't tried, but I'd venture to guess that that's because dust, like flour, doesn't diffuse through air (at human temperatures on human timescales). It's a thousand times denser than air, so it can settle on surfaces and get trapped in certain kinds of airflow patterns. Carbon dioxide doesn't behave like that. It behaves like water vapor.
> Scrubber systems in the corner-case applications where this isn't an option are incredibly expensive and complex. They do not scale well at all.
Unlike cleanroom filters, scrubbers, which are gas-exchange devices used for removing unwanted gases from air, are actually very cheap and simple, and they scale extremely well; they are already being operated at many-megawatt industrial scale, which is how we ended the acid rain problem caused by coal power plants. A CO₂ scrubber can be as simple and low-tech as a wall painted with whitewash, though soda-lime scrubbers (a few percent of lye added to the whitewash, which is balled up in little beads to let the air pass through) are much more common because they are so much more compact. They are in wide use for general anesthesia, scuba diving, and decompression chambers. On the order of a million recreational scuba divers own their own scrubbers. You can buy a bottle of 5 liters of soda lime for €32 at https://www.diveavenue.com/en/high-pressure/1860-chaux-spher... if you stick a porous breathing tube into the bottle, it becomes a scrubber, although you'd probably die if you tried diving with it.
Other kinds of gas scrubbers are available as disposable cartridges from 3M for their respirator masks, so it's absolutely not true that they're "incredibly expensive and complex", but let's stay focused on CO₂ scrubbers here, since that's what's relevant to climate change.
The disadvantage of lime CO₂ scrubbers is that, although they are very simple, regenerating the lime requires heating it up to 900°. People have been doing this for 12000 years (it's probably literally the oldest human chemical process) but it takes a lot of energy compared to other sorbents like triethanolamine, which is why amine scrubbing is currently the mainstay of both submarine CO₂ scrubbers and point-source CO₂ capture pilot projects. There's a nice process flow diagram of a submarine CO₂ scrubber at https://sites.psu.edu/mooneypassionblog2/2022/03/22/how-subm....
As for scaling, lime burning, in the slightly altered form of portland cement making, is one of the world's largest-scale industrial processes, producing 4 billion tonnes of cement per year and emitting 1 billion tonnes of CO₂, which you'll notice is about 2½ years for the mass of the carbon dioxide we have to remove. (Portland cement, unlike Neolithic-style lime cement, only reabsorbs a fraction of that CO₂ when it sets.) So we already know we can scale lime kilns up to the requisite levels; we just have to capture their carbon dioxide output. There are also a bunch of startups trying to scale up amine scrubbing and other processes to direct air capture, but I'm skeptical that their supply chains will be able to compete for scale with the world limestone-and-other-calcareous-stone industry.
> My point is that the idea of filtering our entire atmosphere (or a substantial enough portion of it) only works on paper, research grant applications and proposals for funding by politicians eager to continue the narrative that drives unthinking masses to vote for them.
Every engineering idea only works on paper until you build it. That's how engineering works: you make things work on paper, you validate your assumptions with prototypes, you learn from your mistakes, and finally you create something that never existed before. Your criticism here is a fully general criticism of every technical innovation in history; it doesn't distinguish between perpetual motion machines (which work on paper if you screw up your calculations badly enough) and any routine civil engineering project such as a bridge embankment.