I know its not that sexy, but soil is a hugely diverse ecosystem that is barely understood. There is lots of science to be done trying to classify and work out the mechanics of how nutrient is filters transmuted and transported
It we want to feed the world, when that world is throwing more extreme weather at us, we need to work out how to do companion planting at scale. (think how east coast indians did farming) IF we can make practical farm robots, we can not only remove the need for herbiscides (direct manual intervention, ie physically weeding buy pulling out the seedlings) but also keep ground cover even after cropping, meaning much less water loss.
Soil degradation is a real threat. the way we farm now means we have massive monocultures, large tracks of land that are bare for weeks on end. All of this requires lots of inputs to be productive. The promise of non-pesticide farming is that you get much richer soil, because you're not killing off the stuff that lives there.
But we need to understand what makes a soil productive, however that changes based on location and crops.
There has been an explosion in no-till and low till farming in the UK as energy prices (no till uses less diesel) and fertilizer prices spiked. Cover crops to avoid bare soil became big news.
But how do you kill the cover crop so you can grow wheat again? How do you kill the weeds? The answer for hundreds of years has been ploughing, but that is exactly what we are trying to avoid. The only viable answer today is...Roundup (glyphosate).
And there is the rub. To farm with better soil health, and less ploughing today requires a chemical that we are not happy with using.
A robot to pull the wild oats out of a wheat field sounds practical. A robot to pull 100 acres of white mustard and weeds is what?
There is some work with special rollers that can kill leafy cover crops, and there are tractor pulled mowers, of course, but it is a partial solution. Afterwards you still have a field of dandelions and black grass. So they use roundup.
Then there is the break crop issue. After wheat you would plant rape or beans, perhaps, but only rape will make you a profit, but this is a tremendous risk. A flea beetle outbreak will kill the entire crop. The solution until recently was neonicitinoid coated seed, but that is now banned. So what do you grow?
Part of the solution for me is mixed farming. Wheat followed by fodder beet, graze it off by sheep. Also the drought tolerant lucerne (Alfalfa to the rest of the world). Then seed grass and put cows or sheep (and hat and silage) followed by poultry (bird flu dependent) in a paddock grazing system. Then plough it and back to wheat.
Smaller automated machines could allow smaller fields and a more diverse patchwork I suppose. Cooperation needs to increase massively between farms, so a dairy farm partners with a arable farmer on one side and a sheep farmer on the other.
All depending on your soil type and topography of course. Lots of ground is grazing only.
Then you need to make the economics work. Small farms don't pay. Of course 1000 acres of mountain ground is totally different to 1000 acres of flat arable ground.
We definitely need some innovative in the economics. The current model of subsidy is laughable. Farmers being incentivised to grow no crops at all can't be the answer to food security!
I would love to work in this sector. I feel with better automation and better economics we could make smaller farms that are more like a market garden with many different crops could work. A practical (and cheap) way to harvest grain on a small field would be the biggest breakthrough for me.
> A robot to pull 100 acres of white mustard and weeds is what?
Lets step back and say: could a bunch of humans with hand tools do this?
Yes, they could. (As opposed to, say, eradicating horsetail: if we can't even do it ourselves, then no, we definitely aren't getting a robot to do it).
One way could be the lasers-on-trailer approach, burn the undesirable plants so they're sufficiently uncompetitive. Another could have arms and cutters to reach down and sever the plant below the surface.
Either of the above could instead be a smaller autonomous robot working as part of a swarm day and night.
> But how do you kill the cover crop so you can grow wheat again? How do you kill the weeds? ... The only viable answer today is...Roundup (glyphosate).
I don't agree, and I note that you also answered your question differently later in your post with the note about 'mixed farming' (grazing it off).
There are, of course, other answers than herbicides. Seasonal crops, harvesting and then seed-sowing amongst the stubble (provides some mulch & eroson protection), intensive strip-grazing (bovine, ovine, caprine, or fowl, all effective options), or even a cycle or two of fallow.
Article is a bit short. Here's a few more to flesh out the topic and the plant, tho honestly only by a smidge (none of these links require javascript enabled)
Maybe it is best to think of many plants as photochemical food factory extensions for mycorrhizal fungi. Some plants can do without but many will suffer without a specific mycorrhizae.
I am more and more convinced that the separation into different living beings is somewhat artificial. If there are multicellular organisms, then an ecosystem as a whole can also be considered a life form.
Maybe a key distinction is collaboration vs. competition. The more collaboration between individual "units" (e.g. cells in a multicellular organism or organisms in an ecosystem) the more they behave like a single thing. Ant colonies are also a strong example.
To an extent, yes. I'd say the difference is government systems. A single organism, or something like the human body, has more evolved, more sophisticated government mechanisms. The body is mostly a cooperative civilisation of cells. Of course, there's still natural competition among them, in many shapes and forms. The cells are held together in a coherent, agile, resilient organism by governance systems strong enough to keep internal Darwinism from becoming civil war.
This isn't really a new thought. It's exactly what's meant by terms such as "circle of life" or "ecosystem". The separation of individual beings is entirely artificial, or if your being more charitable and technical, analytical and descriptive.
Science is not reality. We abstract reality to make nice and useful models. Reality violates our models constantly.
Didn’t claim it was. It’s just something difficult to really accept, at least to me, inhabiting a body that definitely feels very distinct from my environment.
I was afraid that would happen. My comment was really more aimed at being a comment to yours, than a reply. The fact that you're starting to "feel" this as being more true is not negated or impacted by it being an existing thought. Thoughts like this take time to settle into experienced truth, and i appreciate that. Had we been conversing that would have been a non-sequitur, and i would not have made it.
One of the problems with comment systems though is that we are at once conversing and broadcasting. The comment was more intended on being a broadcast than a direct reply to you, as a breadcrumb for anyone interested in the path you were taking to maybe seek it out in existing literature.
16 comments:
I know its not that sexy, but soil is a hugely diverse ecosystem that is barely understood. There is lots of science to be done trying to classify and work out the mechanics of how nutrient is filters transmuted and transported
It we want to feed the world, when that world is throwing more extreme weather at us, we need to work out how to do companion planting at scale. (think how east coast indians did farming) IF we can make practical farm robots, we can not only remove the need for herbiscides (direct manual intervention, ie physically weeding buy pulling out the seedlings) but also keep ground cover even after cropping, meaning much less water loss.
Soil degradation is a real threat. the way we farm now means we have massive monocultures, large tracks of land that are bare for weeks on end. All of this requires lots of inputs to be productive. The promise of non-pesticide farming is that you get much richer soil, because you're not killing off the stuff that lives there.
But we need to understand what makes a soil productive, however that changes based on location and crops.
There has been an explosion in no-till and low till farming in the UK as energy prices (no till uses less diesel) and fertilizer prices spiked. Cover crops to avoid bare soil became big news.
But how do you kill the cover crop so you can grow wheat again? How do you kill the weeds? The answer for hundreds of years has been ploughing, but that is exactly what we are trying to avoid. The only viable answer today is...Roundup (glyphosate).
And there is the rub. To farm with better soil health, and less ploughing today requires a chemical that we are not happy with using.
A robot to pull the wild oats out of a wheat field sounds practical. A robot to pull 100 acres of white mustard and weeds is what?
There is some work with special rollers that can kill leafy cover crops, and there are tractor pulled mowers, of course, but it is a partial solution. Afterwards you still have a field of dandelions and black grass. So they use roundup.
Then there is the break crop issue. After wheat you would plant rape or beans, perhaps, but only rape will make you a profit, but this is a tremendous risk. A flea beetle outbreak will kill the entire crop. The solution until recently was neonicitinoid coated seed, but that is now banned. So what do you grow?
Part of the solution for me is mixed farming. Wheat followed by fodder beet, graze it off by sheep. Also the drought tolerant lucerne (Alfalfa to the rest of the world). Then seed grass and put cows or sheep (and hat and silage) followed by poultry (bird flu dependent) in a paddock grazing system. Then plough it and back to wheat.
Smaller automated machines could allow smaller fields and a more diverse patchwork I suppose. Cooperation needs to increase massively between farms, so a dairy farm partners with a arable farmer on one side and a sheep farmer on the other.
All depending on your soil type and topography of course. Lots of ground is grazing only.
Then you need to make the economics work. Small farms don't pay. Of course 1000 acres of mountain ground is totally different to 1000 acres of flat arable ground.
We definitely need some innovative in the economics. The current model of subsidy is laughable. Farmers being incentivised to grow no crops at all can't be the answer to food security!
I would love to work in this sector. I feel with better automation and better economics we could make smaller farms that are more like a market garden with many different crops could work. A practical (and cheap) way to harvest grain on a small field would be the biggest breakthrough for me.
> A robot to pull 100 acres of white mustard and weeds is what?
Lets step back and say: could a bunch of humans with hand tools do this?
Yes, they could. (As opposed to, say, eradicating horsetail: if we can't even do it ourselves, then no, we definitely aren't getting a robot to do it).
One way could be the lasers-on-trailer approach, burn the undesirable plants so they're sufficiently uncompetitive. Another could have arms and cutters to reach down and sever the plant below the surface.
Either of the above could instead be a smaller autonomous robot working as part of a swarm day and night.
> But how do you kill the cover crop so you can grow wheat again? How do you kill the weeds? ... The only viable answer today is...Roundup (glyphosate).
I don't agree, and I note that you also answered your question differently later in your post with the note about 'mixed farming' (grazing it off).
There are, of course, other answers than herbicides. Seasonal crops, harvesting and then seed-sowing amongst the stubble (provides some mulch & eroson protection), intensive strip-grazing (bovine, ovine, caprine, or fowl, all effective options), or even a cycle or two of fallow.
Note: coauthor Toby Kiers received the Tyler Prize for Environmental Achievement of 2026. She also created SPUN: https://en.wikipedia.org/wiki/Society_for_the_Protection_of_...
Article is a bit short. Here's a few more to flesh out the topic and the plant, tho honestly only by a smidge (none of these links require javascript enabled)
https://naturalhistory.si.edu/research/botany/news/plant-pre...
https://www.earth.com/news/native-fungi-native-trees-plants-...
https://matjournals.net/pharmacy/index.php/IJPPR/article/vie...
https://ijpsr.com/?action=download_pdf&postid=97498
Maybe it is best to think of many plants as photochemical food factory extensions for mycorrhizal fungi. Some plants can do without but many will suffer without a specific mycorrhizae.
I am more and more convinced that the separation into different living beings is somewhat artificial. If there are multicellular organisms, then an ecosystem as a whole can also be considered a life form.
Maybe a key distinction is collaboration vs. competition. The more collaboration between individual "units" (e.g. cells in a multicellular organism or organisms in an ecosystem) the more they behave like a single thing. Ant colonies are also a strong example.
To an extent, yes. I'd say the difference is government systems. A single organism, or something like the human body, has more evolved, more sophisticated government mechanisms. The body is mostly a cooperative civilisation of cells. Of course, there's still natural competition among them, in many shapes and forms. The cells are held together in a coherent, agile, resilient organism by governance systems strong enough to keep internal Darwinism from becoming civil war.
Collaboration and competition are simultaneous. Just look at humans working in an office!
This isn't really a new thought. It's exactly what's meant by terms such as "circle of life" or "ecosystem". The separation of individual beings is entirely artificial, or if your being more charitable and technical, analytical and descriptive.
Science is not reality. We abstract reality to make nice and useful models. Reality violates our models constantly.
Didn’t claim it was. It’s just something difficult to really accept, at least to me, inhabiting a body that definitely feels very distinct from my environment.
Gaia theory - James Lovelock.
You find it difficult to accept, or is it just your brain that finds it difficult to accept?
> Didn’t claim it was
I was afraid that would happen. My comment was really more aimed at being a comment to yours, than a reply. The fact that you're starting to "feel" this as being more true is not negated or impacted by it being an existing thought. Thoughts like this take time to settle into experienced truth, and i appreciate that. Had we been conversing that would have been a non-sequitur, and i would not have made it.
One of the problems with comment systems though is that we are at once conversing and broadcasting. The comment was more intended on being a broadcast than a direct reply to you, as a breadcrumb for anyone interested in the path you were taking to maybe seek it out in existing literature.
Nice looking fern at left of the scientist in the image appears to be Microsorum pustulatum, aka "Kangaroo fern", a climber/spreading rhizome.