Hacker News Reader

> The 5 largest subtropical deserts in the world have enough room to hold 30 HEUs worth of solar farms. If we built all those and turned them all on at once

Funny enough, this was done in the SF novel “The Hail Mary Project”, and derided by the scientists in the book for furthering climate problems dramatically by turning a normally heat reflective surface into a heat absorbent surface.

Worth investigating before committing our deserts to power collection.

It's not a real concern. Might make for a good science Fiction story though. https://yaleclimateconnections.org/2009/11/superfreakonomics...

This is a fantastic analysis. I think a lot of sibling comments are missing an important point: Say we build enough solar to produce enough energy for peak demand. That’s going to end up wasting multiple HEU’s of electricity during the day (unless we build an insane number of batteries).

For reference, cities currently cover 3 500 000 km^2 of the earth. Also, solar density is still improving, and other sources of energy exist.

I find it funny that we're carbon neutral if we build two HEUs. Surely, we can just build the one, shut down all other energy production, and call it a day?

No, because people in Bangladeshi villages don’t have induction stoves.

Wouldn't it cost less to replace their stoves than to double energy costs?

It would.

By simple thermodynamics we can be 100%, not 99%, 100% sure, that the energetic cost of removing CO2 from the freaking atmosphere is higher than the benefit of putting it there in the first place.

Maybe there is a way to be sure of that but "benefit" isn't even a concept in "simple thermodynamics". That's not an argument for anything.

If you want the process to be sustainable then it has to be reversible. Burning 1000W worth of fuel only to spend >1000W of electricity to remove it from the atmosphere and turn it back into fuel is clearly a net cost compared to just using 1000W of electricity to cook your food. Or less than that with non-inductive methods, since electric heat pumps can produce more than 1000W of heat from 1000W of electricity by removing some of it from the environment.

A non-sustainable process that converts fuel to CO2 and then into a solid or liquid that isn't fuel could theoretically have positive efficiency, but it is by no means guaranteed to (especially when the competition is heat pumps), even if it did by a small margin it could still cost significantly more, and it implies that you're eventually going to both run out of fuel and convert it all into an enormous amount of industrial waste.

By benefit I mean the energy effectively extracted from the oxidation of carbon.

That's very different from benefit. Also I don't see how that's related to what your original comment was arguing for (cost of replacing stoves).

Maybe I should add that I do agree with your conclusion (that replacing the stoves makes sense). I just disagree with the argument from "basic thermodynamics" which gives you "100% confidence".

Oxidizing carbon fuels is a irreversible process, meaning an increase in entropy. Dispersion of CO2 in the atmosphere also increases entropy.

We will always, 100% of the times, expend more energy to reverse a irreversible process than we could possible extract from this irreversible process.

This statement is valid regardless of the path chosen.

Which naturally spawns the argument 'let's not reverse it, let's do CH to CO to Cx, all we have to do is find x'.

The problem is this is not valid because we burn a lot of fuel.

Even if we found x, it would not be enough to sustain our rate of carbon emissions. In 2022 the world consumed 5.8*10^12 liters of crude oil.

So it would be necessary to include y. And z. Etc until n.

Inevitably including a regeneration step Cn + H to CH + n.

Now the path looks like:

CH -> CO -> Cx -> Cy -> (...) -> Cn -> CH

Which is a closed loop, meaning net power loss. With 100% confidence.

Maybe the one who oxidizes CH pays the bill to remove the CO2 emitted. Is it even possible to put such measure worldwide? Logistics would suddenly be prohibitively expensive for all but the most valuable products per weight/volume. Globalization is addicted to fossil fuels.

Well, that is literally all we did and build and invest in the last century or so. The very power grid of the world is based on carbon fuels. We can't run our carbon removal machinery on dirty power or we would be emitting more than we could possible remove.

It becomes 100% clear the winning strategy is not removing CO2 from the atmosphere, it is replacing all machines and appliances that burn carbon. Worldwide.

Not all are possible, ofc. But a stove ayy.

Again, I agree with your claims about entropy. I also agree with the claim that replacing stoves is generally good.

One still does not imply the other. The question was about what makes sense. That's economically, socially, politically, regionally, etc. Amount of energy extracted via some processes is not the same as benefit. However, benefit is what matters. I guess this discussion isn't moving forward from here.

Yeah, the missing key is seeing power and energy as money, because that's essentially how our economy works.

We can convert one in the other, at different conversion rates granted, but still.

Energy input cost is a line in every company spreadsheet, it can block or allow companies to succeed. Today this line is kept artificially low because we base our economies on a very exothermic open loop CH to CO. Closing this loop is not an option because basic thermodynamics, which is the very logic behind carbon removal.

We have to stop entering this path as much as possible.

Cuz some people in developing countries burning wood is the true cause for global warming… (/s)

While this is a tiny contribution to global warming, funny enough, it turns out that reducing this kind of emissions is also very cheap. To the point where you get very solid emissions reduction per dollar spent.

You're ignoring a huge part that makes this (as well as virtually all other carbon capture plans) useless: where do you store all the carbon?

We are currently emitting ~37B tons of CO2 per year. That's ~0.375 C, so assuming we only store the carbon, that's ~10B tons of C per year.

For reference, the Empire State Building is ~365K tons - so you'd need to build ~30 thousand empire state buildings per year with the captured carbon.

In fairness, the Empire State Building is mostly empty space. If what you have is elemental carbon, the same mass is a cube with 6m sides.

Which, yes, you then need 30,000 of so it doesn't make the whole thing that much easier.

10B tons of C is a mere 10 cubic kilometers. For context, we have individual open pit mines of over 2 cubic kilometers.

If we had emitted 37 billion metric tons of CO2 a year each year for the last 400 years that would be a total of about 1.48 x 10^19 grams of CO2. That's more than we now have in the atmosphere so this gives us a convenient upper limit.

12/44th of the mass of CO2 is from carbon, so that is 4 x 10^18 g of carbon to deal with if we wanted to store all the carbon current in atmospheric CO2.

The density of carbon depends on what form it is in--diamond is a lot denser than graphite, which is a lot denser than carbon powder. Let's store it as graphite, which has a density of 2.2 g/cm^3.

We'd need 1.8 x 10^18 cm^3 for that much powdered carbon, or 1.8 x 10^12 m^3, or 1800 km^3.

But remember that we are getting all this carbon we need to deal with using energy from massive solar farms. Put the solar panels a few meters off the ground, and dump the graphite under the solar panels. If the graphite was dumped into a pile 1 m tall it would need 1800 km^2 of area.

That would be just a tiny fraction of the area available under the amount of panels we'd have to be using to capture that much carbon. Spread evenly under all the panels it would be less than a millimeter.

I'm curious, did you try to calculate the mass of ores that would need to be mined?

No, I just wanted to know if we covered some large deserts with solar panels could that power enough DAC to make a difference.

A quick Googling to find current worldwide solar panel production suggests that it would take about 90 years of current production for the panels for a 500 000 km^2 solar farm, so to build enough to just cancel out current emissions would need about 180 times the resources that currently go annually into solar panel production.

So we'd probably need to increase mining (and several other things) by about 20x if we wanted to use this approach to get enough DAC to bring emissions to net zero.

I think that's about as far as I can go. I don't know enough about the availability of various resources to figure out if the various things needed are abundant enough and accessible enough that we could increase production by 20x.

Prices of panels keep halving, so that suggests potential production keeps doubling.

i would be beyond surprised if covering 10% of the planet's land in solar didn't require any new science or engineering. i doubt we even have the labor force or materials to pull it off, timeline notwithstanding

It’s only 10% if we want something that can lower CO2 to preindustrial levels in under a decade. It’s under 0.7% if we just want to stop the increase.

Plants are basically self-replicating solar panels with efficient CO2 removal & storage.

Plant matter is carbon neutral unless you cut it down and find a way to stop it from burning down or decomposing for thousands of years.

It's carbon neutral on the order of magnitude of a century. If we were to plant a forest in the Sahara, it'll largely be net positive capture for at least 3 decades, and then we'll be able to replace coal with that wood, making a global net zero (but we'd already have reduced co2 in the atmosphere). Replacing oil with wood would probably require quite a lot of work though (as making the desalination plants to make a forest)

Or you can permanently increase the area given over to plant matter

That's a one-time thing though and plants have already naturally spread to basically everywhere that they can spread.

So, for example, we have cleared forests to make way for farmland. Reversing that would increase the total amount of carbon held.

Perhaps I should have said increasing the amount of equilibrium biomass or something to more clearly include rewilding or ecosystem changes. It’s not just plants that are a carbon store, after all.

I don’t think this is likely to represent a full solution, if that is the subtext to your comment.

Everything is carbon neutral on a long enough scale. Unless you send that carbon out to space.

> the output of a 500 000 km^2 solar farm per year is about equal to the total yearly energy use of humanity

And discovering this factoid didn't lead you to realize how insane the whole idea is?

> Just a lot of money, politics, manufacturing, and construction.

Oh yes. Surely just a matter of that. I understand you're more interested in getting ballpark numbers than actually solving this problem though, so thanks for sharing.

500 000 km^2 = 708 km x 708 km

A single 708 km x 708 km would have these corners (Paris, Berlin, Vienna, Milan):https://goo.gl/maps/DTxdVCcK3aYUEirw8 (Rough approximation!)

30 of these: 21,240 km x 21,240 km

500000 km^2 is a little less than the entire area of France, for anyone wondering.

Irrigate the desert and plant 500 000 km^2 of trees seems a lot more practical. And we get lots of building materials as a side effect.

We need to store about 10B tons of C per year if we were to be carbon neutral. We're going to need to build a lot of megastructures out of wood to even keep pace.

Plus wood for building is NOT a permanent carbon storage mechanism. You would need to entomb that wood deep underground where it cannot decay.

We still need building materials, and trees are cut down all the time for building materials and paper. Besides, wood decays very, very slowly. There's a couple trees nearby that were felled 25 years ago, and they are just starting to rot.

Besides, they don't have to be buried that deep. People are still finding intact bodies in bogs in England from thousands of years ago. They're just a few feet down.

In the desert things decay pretty slowly, too. See Sylvester, the cowboy mummy in a museum in Seattle that was found in a sand dune in Arizona with a bullet hole in him.

you think 25 years is enough to store CO2 in? even 1000 years is not enough since we will need to entomb all available wood. But that is all stupid anyway because of this:

To start reversing the CO2 emission we would need as many trees it takes to take up the current CO2 emissions and then a bit more.

From what I found you can estimate every m³ (a bit less than 1 American fridge) of Wood stores 1 ton of CO2. Say each hectare of Forest with fast growing trees produces ~ 10m³ per year (more when they are cut down but they need to grow over a decade+. So if they take 10 years that is 100m³ wood yield at the end).

Humanity produced by some estimates 37 billion tons of CO2 in 2020. That means you need 3.7 billion hectare of forest.

The USA has a land area of 157.7 million hectare. So you need an area of 23 times the LAND area of the United States of America just for trees just to HALT CO² at today's level.

How well are fast growing trees growing in the Arizona desert btw?

Things grow very well in the Arizona desert when they're watered. Lots of sunlight!

You're right that trees cannot be 100% of the solution. But they can be a big part of it.

> you think 25 years is enough to store CO2 in?

It's starting to rot at 25 years, lying in the rain and mud and beetles. There are logs much much older than that lying around the local woods. There are stumps over a century old. My house is 25 years old, and the wood structure that has been kept dry has no rot at all. None. Zero. Simply by keeping it dry.

My point is that 25, 100, or 5000 years is not a long time in this case. The also issue with wood is not that it "cannot be part of the solution". The issue is that it is part of green-washing "solutions" that always calm people down for a bit because something is being done. then nothing else happens for 10 years.

Also I live in a house made from wood, stone, straw and dirt. Depending on how you count its about 400 years old. before that it burned to the ground once. And the Wooden parts aren't actually that old they have been replaced as recently as 100 years ago. the inside has been redone in the 80ies tho and a lot of the wood got taken out.

Its silly to argue for wood as a storage medium even on the basis of that.

It doesn't need to last 5000 years. It just has to last until we transition to other power sources, of which there are many.

> It's starting to rot at 25 years, lying in the rain and mud and beetles. There are logs much much older than that lying around the local woods. There are stumps over a century old.

Limited to rot resistant trees, such as cedars. Even then, not shielded from the rain for 25 years and not seeing much rot would be amazing.

I can send you a photo if you like.

Bogs are a "special" type of environment that are good at preserving or mummifying tissue, and the majority of bog bodies are not preserved, they're in various states of decay. It's a big deal when an intact and/or well-preserved bog body is found because of it.

It can't be too hard to recreate the preservative conditions of a bog. If I recall correctly, it's simply a very low oxygen environment.

The ancient Egyptian tombs preserved things very well as long as the tomb stayed dry.

You'd need literal oceans of preservatives for this scheme to work.

> Sylvester, the cowboy mummy

This story sounded so cool, I just looked it up. Unfortunately, the bullet hole is apparently fake and the reason he didn't decay is because an embalmer deliberately mummified him immediately after his death by injecting him with an arsenic-based fluid, which killed off all the bacteria and insects invading the body.

Hmm, this wasn't known when I last saw Sylvester. I guess there have been some investigations since.

Sounds good, but... Where's the desalination tech which allows to use water in the desert without diverting the existing limited resources from surrounding areas? How do you transport and establish the usable quantities of soil where you have a desert now? That's not a "just plant some trees" solution.

You can gradually move from an existing forest line towards the desert. Forests do alter their micro-climate as well.

Given enough time, yes. We need solutions within decades though - not hundreds of years needed for something like this.

> use water in the desert

Cheaper than building a zillion solar panels.

Have you got the numbers to support that?

For the record, back of a napkin calculation says: 500000km2 of medium density forest of medium water requirement trees needs ~100MW of energy just for desalination. Then the construction of the irrigation grid, maintaining it, pumping the water, staffing the initial works, machinery, etc. goes on top of that.

How much CO2 do you expect that 500,000 km^2 to capture?

Looks like about 250 million tons a year. Less than a hundredth of the solar farm and carbon capture machines.

What happens to the CO2 in the extraction model you used? Collected into tanks that we then have to deal with?

So it’s completely unreasonable. We’re not going to be building 100km2 solar plants let alone a 1000km2 one. Even if we do, something like that is enough for the energy requirement of a few countries so the CO2 problem would probably fix itself

The largest solar plant seems to be ~100 square km: https://en.m.wikipedia.org/wiki/Bhadla_Solar_Park

It says it covers an area of 56 square kilometers. But, it's not laid out as a square, large amounts of that area are empty.

Also: "The Bhadla Solar Park has faced some challenges due to its location and scale. One of the main challenges has been dust accumulation on the solar panels, which reduces their efficiency and output. The park is also located in an arid region that experiences frequent dust storms and sandstorms."

Whether it's unreasonable or not, I don't think we have much of a choice.

Obviously the goal is to move as much stuff off of GHG emissions, but we've done a lot of damage to the atmosphere. If you look at a CO2 graph in the atmosphere over 40,000 years our GHG emissions are basically a vertical line:

https://en.wikipedia.org/wiki/File:CO2_40k.png

I would think that if we stop putting it out there and let the trees take care of the rest we might make it. But this is an uninformed opinion.

Nature would eventually take care of it, but that would take centuries. On top of that, it's unlikely that we will be able to cut all GHG emissions. Having some amount of carbon capture seems important.

Fortuitous timing from my perspective. I was just thinking in the past day or two about fiddling with numbers in the more mundane 'replace current energy use' w/ 'renewables' realm. The impetus being simply having an extra moment for my mind to regale me w/ impending potential horrors in light of Exxon's release only days ago of their "ha ha, we're definitely sailing right past CO2 etc. required to stay under 2C of global average temp increase by 2050, M'Fers!" report ...

I thought, of course, of the always attractive "breakthrough" possibility, but ... in particular, fusion isn't it.* So, deployment of renewables is the obvious solution ... and, solar, in particular, has made massive strides in the past decade+.

What you're talking about is actually much more attractive, though. Sounds like a potentially great implementation / execution. And, all of this - much more rapid deployment for (electric) power generation, or, even better, as you propose, I think - seems like exactly the context in which a new "TVA" (Tennessee Valley Authority) or similar (maybe even "Apollo Project") would be the way to actually get it done with the urgency that seems warranted.

Of course, the US coffers have been so raided, it may take the kinds of "national pain" experienced around a century ago to generate the political "housecleaning" required to, well really, "right the "USS USA". As usual, it's not clear the will is there in the electorate. But, I'd say this much for sure: as old as I am now, the behavior of people older than me has disgusted me in many ways for years, and anyone younger than me should be thoroughly pissed off, I think. I don't want to veer into more of a rant - suffice it to say ... if you look at the benefits a certain generation enjoyed, then consider their voting for policies / tax breaks likely to deny such benefits to future generations (already happening, of course - take a look at college tuitions in the US, for just one example), it's a travesty.

In any case, I think another commenter raised a good point regarding the materials requirement. I'd imagine that would be "paid back" many times over very easily by the scheme overall, but don't know for sure ... I know there have been issues with having enough of the right type of sand to make high (enough) purity silicon wafers for chip manufacture ... but, my knowledge is very limited in these areas in general. If anyone has a real handle on what the materials side of this type of scheme looks like, I'd be interested...

* It was always such an attractive idea ... sounds so perfect ... until you know about neutron generation / flux and realize that something like a "tokamok", in particular, is just not likely to be economical AT ALL (the problem being that without some method to contain neutrons or method for preventing generation, you bombard your materials and end up having to replace / "decontaminate" / etc. far too quickly ... at least, based on my sense of best methods available around 10 years ago ... last time I was involved in any work even marginally related to anything in that field). In any case, even if we have or can determine ways of further reducing neutron generation or ... doing something w/ Higgs or something to deal better w/ the neutron problem, seems like it'd take more like a "miracle" than a breakthrough at this point for it to have any bearing on the climate problem on the timescale relevant.

Edit: realized couple oversights / clarifications (not that it's likely to matter, but, prefer to put in anyway): I have nothing against fission apart from its merits. It's been expensive and track record hasn't been good - plus waste, proliferation, etc. concerns. Practically, it's the cost issue (generally related to the other issues I mentioned, of course). I keep waiting for the advanced / modular / "standard design" reactors we've been promised for years... the track record vis-a-vis AP-1000 (already rather old, but, at least deploying) is not confidence inspiring.

Regarding "certain generation", I've known too many from that generation who sat in taxpayer funded jobs essentially not fulfilling THEIR responsibilities as they RAILED against taxes and the POOR. I have a dim view based on personal experience - too many I knew were spoiled and hypocritical to an unbelievable degree. Of course, that doesn't mean there aren't plenty who were decent etc. So, obviously, I don't mean to blanket condemn any generation ... usual qualifications etc. apply.