Imagine if someone invented machines to suck carbon out of the atmosphere — machines that were absurdly cheap, autonomous, and solar powered, too. Wouldn’t that be great? But we already have these gadgets! They’re called plants.
The problem is, plants die. So there’s one hurdle remaining: We have to figure out how to lock away the carbon in dead plants so that it doesn’t just return to the atmosphere. The obvious place to put that carbon is into the ground. And so, for years, scientists and governments have been urging farmers to leave their crop residue — the stalks and leaves — on the ground, so it would be incorporated into the soil. The trouble is, sometimes this doesn’t work: Farmers will leave residues on a field and they won’t turn into carbon-rich soil — they’ll just sit there. Sometimes, the whole process ends up releasing more greenhouse gasses than it locks away.
This has left people scratching their heads. But now a simple idea is spreading that could allow farmers to begin reliably pulling carbon out of the atmosphere and into their soil.
Clive Kirkby was one of those government agents urging farmers to leave dead plant residues in their fields. He was working in New South Wales, Australia, where farmers traditionally have burnt off their wheat stubble after harvest. Kirkby implored farmers to stop. Instead of torching all that plant residue and releasing the carbon into the air, he told them, let it stay on the ground. It seemed like a win-win: The carbon was harmful in the air, where it contributed to the greenhouse effect, and beneficial in the ground, where it made the soil rich.
As he was proselytizing, Kirkby began to bump heads with an agronomist namedJohn Kirkegaard. “Look, Clive,” Kirkegaard would say, “the best treatment here is burn and cultivate — that’s the one that’s growing the best crops.”
This made Kirkby crazy. Burning was bad enough, and cultivation, which essentially means plowing, was also exactly the opposite of what he wanted. When farmers break up the soil with cultivation it releases some of the carbon stored there, according to conventional wisdom. But Kirkby had to admit that Kirkegaard had data on his side. The agronomist would show him the numbers, and it was clear that the soil organic matter (which holds the carbon) wasn’t increasing. In some cases, it was decreasing.
“I’ve been returning the stubble to the ground now for six years, and it’s just not going into the soil,” Kirkegaard told him.
The way that soil locks up greenhouse gas has been frustratingly mysterious, but the basics are clear: After plants suck up the carbon, the critters (microbes and fungi and insects) swarming in the topsoil chew up plant molecules, subjecting them to one chemical reaction after another as they pass through a fantastically complex food web. If everything goes right, the end result is microscopic bricks of stable carbon, which form the foundation of rich black soil.
Kirkby knew that there must be some mysterious quirk of this topsoil ecosystem that was thwarting him. But how do you investigate a complex, microscopic community that lives underground? There are just so many different organisms eating each other, and cooperating, and parasitizing one another, that we have no clue what’s going on there. People are studying it — but mostly they are reporting that the soil microbiome, as it’s called, is far more confusing than anyone suspected.
Kirkby, however, came up with an idea, that in theory, might allow farmers to manipulate the soil microbiome without having to understand everything that was going on in that black box. He pursued this idea for years, and though he was already nearing retirement age, went back to school and earned a PhD as he assembled evidence. If he’d simply tried to win his original confrontation with Kirkegaard, they’d have remained locked in a stalemate. Instead, because they allowed their minds to be shifted by the evidence, that adversarial relationship was tremendously productive. Kirkby came full circle when Kirkegaard took him on as a post-doctoral fellow (at the age of 66, Kirkby had to be one of the oldest postdocs ever).
The idea that drove Kirkby was elegant in its simplicity. “The way you get carbon into the ground,” he said, “is to take plant residue and turn it into microorganisms.” To grow microorganisms you have to feed them.
They will eat corn stalks and wheat straw, but that, alone, is not a balanced diet. That’s like giving people nothing to eat but a mountain of sugar. There are certain elements that all creatures on earth need to build the bodies of the next generation: carbon, nitrogen, phosphorus, sulfur, oxygen, and hydrogen. These six elements are the basic ingredients of living organisms. By leaving stalks and stems on the fields they were providing a lot of carbon, and oxygen and hydrogen comes easily from the air, but the bugs were lacking in nitrogen, sulfur, and phosphorus. Provide enough of these missing building blocks, Kirkby figured, and the soil microbes would finally be able to consume the plant residue. He tried it. It worked.
One lab test provides a dramatic visual of how this works. The scientists added wheat straw to two pans of sandy soil, and fertilized one with nutrients. That pan looks like rich compost. The untreated control looks as lifeless as the surface of Mars.

Dirt
Courtesy of CSIRO Plant Industry / CSIRO Agriculture (CA Kirkby, JA Kirkegaard, AE Richardson)
I saw this picture recently when I met, via Skype, with Kirkby, Kirkegaard, and another collaborator named Alan Richardson. All work at the Australian government’s Commonwealth Scientific and Industrial Research Organisation. They crowded together in front of the computer in Kirkegaard’s Canberra office.
“That’s moist soil with chopped up wheat straw on the left,” Kirkegaard said. “There’s no reason why that shouldn’t have decomposed, except for the fact that nutrients are missing. When you give them the nutrients, all the wheat straw is gone, and you get the results of the microbial activity and their bodies and it creates a whole lot of…”
“Humus!” cried Kirkby. He spoke with enthusiastic, rapid-fire intensity, his accent pinching the vowels through the nose: “With the right balance of nutrients you get a population explosion. And that’s what you want. The carbon is in the soil’s organic matter, and that’s essentially dead bug bits. And live bugs. Humus!”
Richardson, who stood leaning against the far wall, chimed in, gruff and sedate compared to Kirkby. “Historically we’ve fertilized the crop,” he said. “We’ve been interested in the crop. The paradigm shift is in thinking that you have to fertilize the system, the microbes and all that. And through that you support the crop.”
Instead of simply trying to optimize for the plants, they’ve realized, you can optimize soil along with the plant — you can optimize the whole system.
The three men explained that, when they looked at soil organic matter from around the world, the proportions of nutrients — the ratio of carbon atoms to nitrogen, for instance — are stunningly consistent. The organic matter is microbes. And if you want to build more of it, you have to give the microbes the right ratios of nutrients to build more tiny, cellular bodies.
Instead of trying to identify every soil microbe and understand what it’s doing, they have hit upon a way of treating the whole mess like a super-organism that responds in predictable ways.
The scientist Richard Jefferson, who introduced me to this work, calls it breeding by feeding: We don’t actually know what these microbes are that we’re breeding; we only know that when we set out the right proportions of food, they click into high gear.
All this helps explain why organic farms often capture more carbon. In adding compost to amend the soil, organic farmers are adding the same ratios of nutrients. The organic claim that fertilizing with synthetic nitrogen kills off soil life actually makes sense, Kirkby said; it’s just that the problem has nothing to do with the nitrogen’s artificiality. The trouble is that farmers are applying the nitrogen without the other nutrients necessary to nurture the microbiome.
“As agronomists, we talk about nutrient-use efficiency,” Kirkegaard said. “Now, the best way to have high nutrient-use efficiency is to mine the organic matter, because that comes to you for free. You’re wanting to put on juuuust enough nutrients to feed the crop and not have any left over. And that just means the other crop, under the soil, the microbial crop, misses out. As a result, we’ve lost about half the organic matter in land we’ve been using for agriculture.”
I wanted to get a reality check from another scientist, because this all sounded almost too good to be true. So I got in touch with a true authority in the field,Rattan Lal, president elect of the International Union of Soil Sciences. Lal took a look at look some of the work and pronounced his judgment: “I agree,” he said. “This phenomenon is well understood.” A colleague of Lal’s was teaching students to apply exactly the same ratios of nutrients 50 years ago, he said.
This stopped me. If this is old news, why haven’t we been putting it to work? Why the confusion when no-till fails to capture carbon? Why the mystery surrounding the ability of organic farming to do so?
Sometimes good information simply doesn’t spread everywhere it should go, Lal said, with a note of weariness. This isn’t a exactly breakthrough, he said, but he welcomed the work and said he hoped people would pay attention this time. When he followed up with an email, he wrote: “The theme addressed is very important and it must be brought to the attention of general public and policy makers.”
When I initially spoke with Kirkby, Kirkegaard, and Richardson, they had been forthright in telling me that we’ve known about this golden ratio of nutrients for a long time. They also noted that there were other scientists like Sébastien Fontainepublishing similar papers. In a follow-up email, Richardson wrote, “What we think is new is the direct connection between the soil microbiome and the [soil organic matter], which is mediated by [the ratio of nutrients].  We think that our set of recent papers provides some of the first real evidence that underpins this connection and shows evidence that the dynamics can in fact be changed.”
Jefferson says the Australians are being modest, and conservative with their claims. Connecting the well-known nutrient ratios with the microbiome truly is a breakthrough, he said.
“Now they have a mechanism to explain how this works, which allows you to make predictions, so you can imagine experiments driving this forward. One of the things that’s exciting for me is that this really bridges empiricism and scholarly science nicely. There have been tens of thousands of anecdotes noted about the performance of small scale, traditional agriculture — empirical studies or stories of small farmers who do exciting things in terms of performance and resilience. It has been largely dismissed by the hard-core science community because it has not been scalable and replicable. We can’t take one farmer’s success and move it to the next farmer or the next ecosystem because we have no understanding of how it works — complex systems don’t extrapolate well, they don’t work out of context.”
In other words, when we see an organic farmer building up the soil and achieving amazing results, it’s hard to copy it because we don’t know what to imitate. What is it that makes this work? The type of fertilizer? The local microclimate? The prayer the farmer says before breakfast? The work coming out of Australia provides the traction to separate superstition from the stuff that gets results.
Both Lal and the Australian scientists agree that there’s still one more major hurdle, which may have kept this information from spreading: These nutrients cost money. If farmers were paid for locking up carbon, they would gladly buy the fertilizers, Lal said, but right now the reimbursements are far too low. Even at the high point of the carbon markets, when people were paying $30 per ton, it would not be enough to reimburse farmers. “It costs $800 a ton of CO2 to do geological sequestration, you know, pumping carbon underground,” he said. “If farmers could get even a tenth of that, $80 a ton, I know many soil-poor farmers would participate.”
Kirkby thinks that, by tinkering with the soil microbiome, farmers might see enough gains to pay for the extra inputs. There’s already evidence that the soil microbes can help suppress plant disease and improve dirt quality. Extending this concept of growing a healthy system, not just a healthy crop, could yield profits.
“We’re probably not going to increase yields incredibly, but we might be able to improve incrementally,” Kirkby said. “In a sandy soil we might improve water-holding capacity. In a heavy clay soil we might reduce diseases a little bit — added together it might pay for the nutrients at the end of the day.”
One thing is certain: If agriculture were able to switch from an emitter of carbon to an absorber of carbon, the effect would be huge. Plants, those cheap carbon-removal machines that nature has given us, work well. If we can get them to make our dinner while they are also sucking up greenhouse gas, what a coup that would be.
But it would be an even greater coup if we could begin, as these scientists have done, to understand how to manipulate whole ecological systems — rather than just systems that have been simplified and stripped down to easily controllable parts.