3 ms·
>there’s no significant externality to tossing a few squishy potatoes in the compost bin (or a landfill for that matter) If in the process you used practically
by modo_mario 4y ago
>there’s no significant externality to tossing a few squishy potatoes in the compost bin (or a landfill for that matter)
If in the process you used practically finite fertilisers in the form of phosphorus from Morocco and nitrogen from fossil fuels, etc then you may wish to rethink that.
- pfdietz 4y agoUltimately phosphate is going to have to be recycled or extracted from much less concentrated sources. The average abundance in the Earth's crust is 0.1%, so supplying current world phosphorus demand (about 22 million tonnes/year; equivalent to 50 million tonnes/year of P2O5) would require mining about 9 cubic kilometers of rock per year. Difficult but not impossible.
- RugnirViking 4y agoaren't fertilizers mainly produced using the haber-bosch process which takes nitrogen, the most abundant element in the atmosphere, directly from air and combines it with hydrogen which is split from water, producing oxygen as a waste product which can itself be sold for oxygen cutting, scuba, and more? of all the things that are limited in this world, this seems like it ranks among the least concerning. Sure, it takes electricity, but that itself can be much more efficiently collected going into the future
- pfdietz 4y agoThat's nitrogen fertilizer. Phosphate and potassium (the other two major plant nutrients) are currently obtained by mining. Potassium will ultimately be available from seawater (although the estimated land resource would supply current global demand for thousands of years), so is not as much a problem as phosphate. But you are right that making nitrogen fertilizer doesn't require fossil fuels. It would be a major market for "green hydrogen", which would be a great dispatchable demand to help smooth variability in renewable power output and in other demand.