What is regenerative agriculture? Why is it different from sustainable agriculture? And how do I reconcile what practitioners of this system are claiming with the scientific evidence? These were all going through my mind when, a couple weeks ago at an advisory committee meeting of the WSU Center for Sustaining Agriculture and Natural Resources, we watched a YouTube video of Gabe Brown’s TEDx talk in Grand Forks, North Dakota. Brown farms near Bismarck, ND, and has become the American face of regenerative agriculture in the past decade. Here is what I learned.
Brown is a good speaker, in high demand for conferences and events. Our meeting organizers invited him to speak, but he is booked a year out, so we watched the video. Brown spoke of his 20+ year journey from a conventional row crop farmer to a regenerative farmer. In the video, he answered my first question, “what is it?” by giving five principles of regenerative agriculture. However, Brown’s version of regenerative agriculture is not the only one.
In my past explorations of regenerative ag, I had found that there are multiple versions of these principles, each with a different flavor. Rodale and partners offer their strictly organic version with a new certification program attached. Project Drawdown includes regenerative ag in its plan to reverse global warming, and California State University at Chico has their regenerative ag initiative. Table 1 shows Brown’s principles/practices compared to these other versions, and to conservation agriculture.
Principle #1 Minimize or eliminate tillage
Building or rebuilding soil is the primary focus of “regenerative” practices, all the versions agree on this (see Table 1). While some versions extend this to restoring animal health, human health, and communities, it all starts with soil health. To achieve this, they all agree that farming should minimize or eliminate tillage. Brown’s #1 principle is “least amount of mechanical disturbance possible.”
This is also the main focus of conservation agriculture. This set of principles grew out of the development of no-till in the 1970s. I see it as one of the predecessors of regenerative agriculture. Not much new here, but not anything to disagree with. So far, so good.
Principle #2 (and #3) Protect the soil
Brown says the second principle of regenerative ag is “armor on the soil surface.” Keeping the soil covered to eliminate erosion is important because you can’t build soil while it is blowing or washing away. Related to this principle is Brown’s #3, “living plant roots in soil as long as possible.” I think the idea is that the soil will always be covered if there is no tillage and there always a living plant growing.
Like principle #1, this is one that few will disagree with, but which is hard to implement with crops like potatoes or carrots, because they grow underground and require tillage to harvest them, and with small-seeded vegetables, because they require precise shallow planting which is difficult to achieve with crop residues on the soil surface. It is interesting that the climate-change and organic versions do not include a principle related to protecting the soil, at least explicitly (see Table 1). They are also the two versions that explicitly ban synthetic pesticides and fertilizers. If you can’t use herbicides, it is very difficult to always keep the soil covered with either dead crop residues or living plants.
Principle #4, Biodiversity
The next principle is to increase biodiversity. It is shared by all the regenerative ag versions, and conservation ag, although the latter does not often refer to it as biodiversity, per se. Brown implements his “diversity of plants” through intercropped cash crops and high-diversity cover crops that total 70 species. Impressive. Crop rotations and cover crops are, like the earlier principles, hard to disagree with. These are basics of sustainable agriculture and when markets and cropping seasons allow, they should be used.
Principle #5, Integrate livestock
In the video, Brown’s last principle is “animal impact.” In all versions of regenerative agriculture, this is crucial to making regenerative agriculture work, and the main way to get the animal impact is through grazing. The particular type of grazing promoted by regenerative agriculture is management-intensive grazing, the holistic management of Allan Savory (click for a discussion of Savory and his practices). No argument here. Grazing livestock adds diversity to the products produced on the farm, adds value to cover crops (really annual forage crops), and recycles nutrients through manure. Nothing new here either – I helped organize management-intensive grazing workshops in NE Nebraska 25 years ago. So what is new about regenerative agriculture?
Well, first, regenerative agriculture seems to be a mashup of several systems of principles. It can be viewed like this:
From what I have seen, one of the actual new things about regenerative agriculture is the intense focus on multi-species cover crops. The cover crop mixes in regenerative agriculture are not just 2- or 3- or even 5-way mixes, they range from 10 to 60 or more species. Brown and other proponents of regenerative agriculture claim that these cover crop mixes stimulate the soil microbial population to supply plants with the nutrients they need, greatly reducing or eliminating the need for synthetic fertilizers.
I have written previously about the lack of evidence supporting the use of cover crop mixtures over monoculture cover crops. Gabe Brown even commented on my essay, as did other fans of regenerative agriculture. I may have a “monoculture mindset” as Brown wrote, and be an “externalist” as another commenter suggested, but for all the comments, I did not receive any published evidence that cover crop mixtures are consistently better than monocultures. If cover crop mixtures are so beneficial, those benefits are sure hard to detect. Nor have I found evidence showing that intercropping is better than a diverse rotation of monocultures. If you know of evidence contrary to my conclusions, please let me know.
The other thing that characterizes regenerative agriculture are claims by practitioners and scientist proponents that go against all published soil science evidence, indeed they seem truly miraculous by the standards of what we think we know about the soil. There are many examples, but let’s look at just one, given in Brown’s TED talk.
An Improbable Increase in Soil Organic Matter
During his talk, Brown offers the following slide showing the increase of his topsoil depth and soil organic matter over his 20-year transition from conventional farming to regenerative practices.
Topsoil depth increases from 3” to 14” while soil organic matter (SOM) increases from 1.7% to 11.1%. Increasing soil organic matter by a few percentage points is normally thought of as a long, difficult process, unless you use a lot of imported manure or compost. Here, however, Brown claims to have increased SOM by over 9 percentage points. How? According to the slide, by cover crops, multi-species cover crops, and livestock integration. Let’s do the numbers according to what current soil science tells us this would require.
First, some assumptions. My calculations are for the top 6” of soil for all 20 years. This ignores the increased topsoil depth shown on the slide and is therefore conservative. I am assuming that what Brown is showing is real organic matter, and not just undecomposed plant roots or shoots. Soil organic matter is not all organic material in the soil, it is the result of a complex biological process, with the resulting organic matter having very different properties from plant roots or shoots.
In the process from plant (or microbe) biomass to SOM, losses of mass (CO2 released to the atmosphere) range from 80-90%. I assumed a loss of 85%, equivalent to a plant/microbe mass to SOM mass conversion rate of 15%. I took the nutrient contents of SOM from this NRCS publication.
For ease of calculations, I assumed a constant rate of SOM increase. In reality, it is generally easier to increase SOM when levels are lower and more difficult as they get higher. Now we are ready for the calculations.
First, the amount of plant biomass required to obtain Brown’s increase in SOM. Given the 15% conversion rate, he would have had to add 31 tons (dry), per acre, of plant or other biomass to the soil, every year, for 20 years (see figure 1). If 31 tons does not mean much to you, it is more than the entire aboveground biomass of a fully fertilized, irrigated corn crop. It is more than a full season, four cuttings, of irrigated alfalfa hay production. It’s a lot of biomass. And this amount of biomass was added to the soil; what was harvested as a crop, or as meat through livestock grazing, is in addition to this 31 tons per acre per year.
Building soil organic matter requires more than biomass; nutrients are also needed, either in the added biomass or from the soil. SOM averages 5% nitrogen and 0.5% phosphorus. So then, Brown’s SOM increase requires 470 lb. of nitrogen and 47 lb. of phosphorus per acre, each year, for 20 years. This is more nitrogen than is applied to a high yielding irrigated potato crop, and as much as is harvested in a 9 ton per acre alfalfa crop. And this 470 lb of nitrogen per acre is in addition to what is needed to produce a crop or to produce meat.
To top this all, Brown states (after mentioning his land with 11.1% SOM), “We’ve done this without the use of any synthetic fertilizers, pesticides, or fungicides.”
We are to believe that biodiversity-powered microbes free up large amounts of phosphorus, fix large amounts of nitrogen from the air, while plants produce 31 tons of biomass in a short North Dakota season, while also producing harvested crops and livestock?
I cannot say that this scenario is impossible, but I find it highly improbable, because if this is true, then it means that science has missed an astounding, extraordinary process. And it has been missed by not just agricultural soil scientists, but also those who work in prairies and forests, because, according to regenerative agriculture, this is how it works in nature. And we have been studying nature for a long time. And this is not just about a claim made by Gabe Brown; similar claims are commonplace in regenerative ag circles. If this and similar claims are true, then we are talking about a revolution in agriculture, which is what regenerative farmers and their supporters say it is.
However, there is another principle here: extraordinary claims require extraordinary evidence. What counts as evidence are peer-reviewed publications in scientific journals – I have looked for the evidence to support the claims of regenerative agriculture. What I have found are lots of YouTube videos, testimonials, articles and interviews. None of these sources are extraordinary evidence.
Extraordinary claims also require scrutiny, which is why I wrote this piece. I cannot disprove with words and calculations what Brown says he has observed in the field, but words and calculations can show that this is extraordinary, and so demand more evidence. I also wrote it to show the regenerative agriculture community the reasons why people like me, scientists and researchers, and those who believe in the scientific process, are skeptical of their claims.
If the claims of regenerative agriculture are real and repeatable, then they are of such magnitude (i.e. 1.7 to 11.1% SOM) that they should be easy to measure. So here is a challenge to regenerative agriculture. Provide the extraordinary evidence. If it exists, let me know and I will post it here. If the research still needs to be done, connect with researchers to start the process. Don’t let regenerative ag become the cold fusion of agriculture. Pursue rigorous science to demonstrate its value.
Here is a 2018 paper that offers itself as an evaluation of the “relative effects of regenerative and conventional corn production systems on pest management services, soil conservation, and farmer profitability and productivity throughout the Northern Plains of the United States.”
The “most” regenerative farms were defined as using multi-species cover crops, “never-till”, used no insecticides, and grazed livestock on their cropland. None of the conventional farms used cover crops, almost all of them used tillage and none of them grazed their cropland (Table 1 in the paper). What did they find?
Pest management services
The abstract states:
Pests were 10-fold more abundant in insecticide-treated corn fields than on insecticide-free regenerative farms, indicating that farmers who proactively design pest-resilient food systems outperform farmers that react to pests chemically.
However, the paper states that “none of these pests [in either system] were at economically damaging levels.” Both types of farms managed their pests, so this 10-fold difference does not really matter. The paper also tells us that the treatment in “insecticide-treated fields” consisted of “genetically modified insect resistant varieties and neonicotinoid seed treatments.” Not really a high concern scenario in terms of insecticides.
Although the paper measured soil organic matter levels on all the farms (Table 2 in the paper), and particulate organic matter, a more biologically active part of the organic matter, it does not directly compare these values for conventional and regenerative farms. I would guess that this is because such a comparison would inappropriate given that they did not control for region, or soil texture, manure application, etc. Although the abstract promises an evaluation of soil conservation, the paper does not deliver any such evaluation.
The authors found that corn yields on regenerative farms, despite having 10x fewer insect pests, were 29% less than those on conventional farms. This reinforces the point that the insect pest differences did not matter. The yield difference is explained:
Yield reductions are commonly reported in more ecologically based food production systems relative to conventional systems.
Given the lower yields, it might be a surprise that regenerative farms were found to be nearly twice as profitable as conventional farms. But the profits included in the calculations were not just from corn yields; the regenerative farm’s profits include meat production from grazing, organic premiums, and direct marketing. It is not possible to tell from the paper how much these influenced the net profits, but given that the regenerative farms started with 29% less yield, I assume grazing and marketing made up a large part of the difference. On the cost side, the regenerative farms had lower fertilizer costs due to the use of legume cover crops, and lower seed costs because they, I assume, did not plant GM corn.
Curiously, profits were plotted against soil organic matter and soil bulk density (Figure 3 in the paper) and a positive correlation was found for both. One of the beliefs of regenerative agriculture is that profits are directly related to soil health/organic matter/carbon. This was implied in the paper’s conclusions, except the language switches from correlation, which they rightly showed, to causation, which they did not show: “Soil organic matter was a more important driver of proximate farm profitability than yields were…”
What did we learn?
We know that tillage degrades soil, cover crops improve soil, and organic premiums and direct marketing can improve profits; nothing new there. And I see nothing that supports the extraordinary claims of regenerative farming.
Also, don’t rely on just the abstract; read the paper if it is available.
I don’t have to review this because Alan Franzluebber, a respected soil scientist, has already done it. See his long comment at the very bottom of the webpage. Franzluebbers concludes, “Although the enormous rate of soil organic C accumulation reported from this study is an outcome of contention, the real concern is the lack of robust experimental procedures used to obtain the estimate. High quality and replicated data are needed to make such bold proclamations.”
Robust research in agriculture is not easy to do, and not every peer-reviewed paper is of equal worth.
van Groenigen etl al. 2017
van Groenigen, J.W., C. van Kessel, B.A. Hungate, O. Oenema, D.S. Powlson, and K.J. van Groenigen. 2017. Sequestering Soil Organic Carbon: A Nitrogen Dilemma. Environmental Science & Technology 51(9): 4738–4739. doi: 10.1021/acs.est.7b01427.
I am not the only one who wonders where the nitrogen comes from. Here is a paper that calculates the nitrogen needed to increase organic matter % in soils globally by 0.4% per year. They conclude that the studies done to assess this goal “overlooked limitations imposed by nutrient availability.”
Sullivan et al. 2014
Sullivan, Benjamin W., W. Kolby Smith, Alan R. Townsend, Megan K. Nasto, Sasha C. Reed, Robin L. Chazdon, and Cory C. Cleveland. 2014. “Spatially Robust Estimates of Biological Nitrogen (N) Fixation Imply Substantial Human Alteration of the Tropical N Cycle.” Proceedings of the National Academy of Sciences, May, 201320646. https://doi.org/10.1073/pnas.1320646111.
This study made robust measurements of biological nitrogen fixation in a very diverse environment of tropical forests, both primary and secondary. The highest rate they measured was about 20 lb N/ac per year in contrast to earlier estimates by Cleveland et al. 1999, which I mention in a comment below. Diversity does not guarantee high or even moderate rates of nitrogen fixation in natural ecosystems, although the nitrogen fixation here could be limited by available P.
Reed et al. 2011
Reed SC, Cleveland CC, Townsend AR. 2011. Functional ecology of free‐living nitrogen fixation: a contemporary perspective. Annual Review of Ecology, Evolution, and Systematics 42: 489–512.
Free-living N fixing organisms may contribute up to 10–15 lb/ac per yr in some ecosystems. These ecologists seemed to be impressed with this rate, but this agronomist is not.
Kallenbach et al. 2016
Kallenbach, C.M., S.D. Frey, and A.S. Grandy. 2016. Direct evidence for microbial-derived soil organic matter formation and its ecophysiological controls. Nature Communications 7: 13630. doi: 10.1038/ncomms13630.
Here is a recent paper that shows that microbes are the source of much of the soil’s organic matter. However, they find that 75% of added C from plants through microbes to SOM is lost. This is lower than my estimate, but not too far off. “Using SOC stocks as an integrator of mass C balance, the majority (>75%) of total substrate-C added was lost via respiration across all treatments by 18 months (Table 1).” https://www.nature.com/articles/ncomms13630
Castellano et al. 2015
Castellano, M.J., K.E. Mueller, D.C. Olk, J.E. Sawyer, and J. Six. 2015. Integrating plant litter quality, soil organic matter stabilization, and the carbon saturation concept. Glob Change Biol 21(9): 3200–3209. doi: 10.1111/gcb.12982.
This paper has a table listing papers that measured the amount of C addition converted to soil organic carbon. The field measurements in the table from nine studies range from 3% to 33%, or 67-97% loss.