[PeterisP]

Land isn't that cheap compared to the (not that large) effect, especially if we want that effect to be meaningful in the short term.

Buying land and planting trees there will cost something like $2000 per acre and retain something like one ton of CO2 per acre per year (an order of magnitude estimate - depending on details both the cost and CO2 effect can be very different).

Industrial carbon capture at power plants can do that for something like 70$ per ton. That's much cheaper than forestry, but that's still not good enough. ycombinator is obviously looking for technologies that scale better than these existing approaches, something that might achieve large scale carbon removal at maybe $10/ton or less, at which stage the option "just pay a lot of money to reverse the effect of our emissions" might be plausibly considered affordable to our society.

[strainer]

>Industrial carbon capture at power plants can do that for something like 70$ per ton.

And then you have a lot of captured carbon dioxide on your hands - next big cost is the storage/conversion.

IPCC summary on cost of forest sequestration :

> Estimates of the private costs of sequestration range from about US$0.10-US$100/tC, which are modest compared with many of the energy alternatives (see Table 3.9 and Figure 4.9). Additionally, it should be noted that most forest projects have positive non-market benefits, thus increasing their social worth

http://www.ipcc.ch/ipccreports/tar/wg3/index.php?idp=171#fig...

[ed_balls]

Why do you want to buy the land? Raise the money for seedling, get volunteers/robots to plant them. Pay people to maintain trees on land (which should require 0 effort). You can come up with some clever designs to make it a tourist attraction and make some extra cash. Plenty of room for improvements.

Let say you can have 100 000 trees per square km. If 40 trees gives you 1 ton of carbon per year then spending $25 000 gives you $10/ton.

[PeterisP]

Well, because to the first approximation pretty much all land is used - any land thats suitable to be a forest but is not already a forest is only that way because it's used for grazing or farming, otherwise it would overgrow naturally (though not as efficiently as with planting). If you want to increase the amount of woodland, you have to decrease the amount of pastures or farmland, so you have to either buy that land from the previous owner (because they won't be able to graze or plant there anymore) or take it from them by force.

[ed_balls]

I pretty sure that are that 10 mln km^2 around the word that is not use (Canada, Alaska, Syberia and patches of land around the world)

[PeterisP]

As I said, I'm talking about "any land thats suitable to be a forest but is not already a forest". These large uninhabited areas in Canada, Alaska and Siberia don't have such land - if any spot there is suitable for trees, then trees already have filled that area for hundreds of years (I mean, these areas already have massive forests), and in the areas where trees aren't growing naturally, it's for a reason, planting won't make a difference.

If you want to convert not-woodland into woodland, then that limits you to farmland or pastures - because there's no such thing as "unused natural potential woodland", any potential woodland that's not used and left alone becomes actual woodland; any potential woodland that's not woodland only became that way when we cut down the trees and cleared the land because we wanted to use it otherwise.

[ed_balls]

I agree that my estimate could be too big, but I do think there is a potential to increase the forestation in the north. Because of the warmer climate, Iceland can grow aspen. The government is planning to grow tree in a large part of the island, but the have a problem with free roaming sheeps. The land the can only sustain shrubs and moss could grow trees now.

Some countries are increasing the forestation as well https://www.citylab.com/environment/2018/01/northern-forest-...