Where Agriculture Fits in the Carbon Markets Story

Written by Dan Kane, Director of Science

tl;dr: Soil carbon removal is a legit natural climate solution that we can enact now that has real co-benefits. But soil carbon by its nature is non-permanent, meaning there’s always a chance carbon in soils can be re-emitted to the atmosphere. By and large, the Voluntary Carbon Market (VCM) conceptually treats biological and geological removals as the same. That mismatch is arguably what underpins the uncertainty of the agricultural VCM (and other parts of the VCM). Uncertainty can be managed, but it’s never fully going away.

New climate finance and policy tools that reward temporary storage and better align expectations with reality could help. But even if the carbon story is uncertain, our collective commitment to and investment in regenerative ag is absolutely critical due to the multitude of co-benefits.

In the past several years, the voluntary carbon market (VCM) has exploded in size. The 2025 State of the Voluntary Carbon Market report shows that since 2019 both the cumulative value and volume of credits has nearly doubled relative to the prior 15+ years [1]. Agricultural carbon projects focused on reducing GHG emissions and removing carbon from the atmosphere into soil organic carbon stocks in particular are having a moment. Agricultural practices such as cover cropping, no tillage, and managed grazing, appear to have real, albeit variable, potential to increase soil organic carbon stocks [2]. These practices also have real potential to improve soil health, structure, and function. By all appearances, soil carbon sequestration in agricultural lands is a win-win. I’ve worked in and around the agricultural VCM for several years now and focused on soil carbon in my academic career before that, so I’ve been along for the ride.

Despite the good vibes, the agricultural VCM has been harried by uncertainty and has been slow to scale. It remains the smallest sector of the VCM, and only a limited number of projects have made it to the point of credit issuance [1]. From one perspective, the uncertainty could be chalked up to market growing pains. The VCM is working its way through the hype curve. Fast, sloppy work is getting filtered out. Good projects and good actors who take their time will succeed when the noise dies down. Buyers are holding while they wait for clarity and quality to emerge. Trust the process.

From another, the problems of the market are more fundamental issues of certainty and measurement. Several studies have called into question the veracity of prior evidence and synthesis about the removal rates of key practices, such as cover crops, no tillage, and managed grazing, suggesting some practices have much lower effects than previously thought or no effect at all [3], [4], [5]. Plus, while it’s not impossible to measure changes in soil carbon stocks at scale, it’s still really hard. Soil carbon is highly variable over space and time, and each context is different. All that means there is no one size fits all solution you can pull off the shelf and you have to collect A LOT of samples to distinguish signal from noise[6], and those samples are pricey.

I have opinions on the ag VCM hype curve and the extent to which uncertainty in evidence of effect and methods should matter. But I’m not going to talk about those topics here…at least not directly or in detail. Instead I’m more interested in exploring fundamental questions of “product-market fit,” for lack of a better term. Likewise, I’m going to focus this conversation on soil organic carbon stocks and not other GHG emissions from agriculture. Not because they don’t matter, but because I want to focus on just wrapping my head around the CO2 story for now and what it means for carbon markets.

Get in, loser, we’re going to the Ordovician Period

To start, let’s go back 450 million years to the Ordovician Period. STAY WITH ME. I PROMISE THIS IS GOING SOMEWHERE. At its start, atmospheric CO2 levels were as high as 5000 ppm at that time, that’s over 10x current levels (~427 ppm) and 15x the approximate level (300 ppm) under which human life evolved. During the Ordovician, the first land plants evolved, leading to massive swamp forests and setting off a sustained decrease in atmospheric CO2 that lasted 150 million years through the Carboniferous period. These forests led to the formation of peat that eventually was buried and became coal deposits. At the same time, the rise of terrestrial plants also accelerated mineral weathering resulting in atmospheric carbon removal into carbonates [7], [8], [9], [10].

After a brief uptick in atmospheric carbon during the Permian and Triassic periods, carbon again started to sharply decrease in the Jurassic and Cretaceous periods with the rise of marine phytoplankton, which removed carbon from the atmosphere through both carbonate formation and photosynthesis, leading to the formation of oil and natural gas deposits [11]. This period also saw massive continental uplift events, such as the collision of the Indian and Eurasian tectonic plate that led to formation of the Himalayas, exposing massive amounts of rock leading to even more silicate weathering [12].

Why the 500 million year detour?

All this activity is what led to atmospheric carbon levels of 200-300 ppm that gave us the temperate climate under which mammals and human life have flourished, and it’s a great demonstration of the major carbon pools. In biogeochemistry, we tend to think of global carbon pools as being geological vs. biological or inactive vs. active. Geological storage generally refers to carbon that’s in some type of carbonate mineral. These minerals cycle on really long timescales and only under specific conditions. They are effectively inactive. Sedimentary rock and buried marine sediments are the main forms of geological storage. Geological stores are primarily responsible for creating the modern climate.

Biological or active pools include mostly organic carbon compounds that are actively cycled in biological processes (e.g. photosynthesis, cellular respiration, etc.) on shorter timescales. This includes plants, soils, the atmosphere, peatlands, permafrost, and surface oceans—anything on the earth’s surface really. In some cases, biological pools can remain inactive for centuries or longer, but usually because for one reason or another they’re inaccessible to living organisms. Think permafrost—it’s frozen, so microbes can’t eat up and respire all that carbon at an appreciable rate.

Geological stores account for hundreds of millions of tons of gigatons of CO2e and are more durable than biological stores, which collectively account for less than 100,000 gigatons of CO2e [13], [14]. We’ve included a full accounting of the relative size of different pools in the table at the end for the curious reader.

Fossil fuels are a curious case, as fossil fuel formation is essentially a process by which biological carbon is converted to inactive or geological carbon stores over (very) extended periods of time. For the most part, fossil fuels are not used in biological processes and are so buried as to be effectively inert until extracted and burned. Fossil fuel extraction and use has released about 2500 Gt CO2 from that pool, effectively reallocating that carbon from inactive/geological pools into active/biological pools. Given current policy and energy trends globally we’re arguably on track over the next century for a middle of the road scenario where we’ll see warming of 2.7 C and burn about another 2500+ GtCO2 in fossil fuels [15] [16].

To date, surface oceans and terrestrial carbon sinks have acted as a buffer, absorbing a little over half of fossil fuel emissions to date, leaving about 1100 Gt of excess CO2 in the atmosphere relative to pre-industrial levels [13]. But that won’t continue forever. In 2023, the terrestrial carbon sink (soils, forests, etc.) just….didn’t work. Between massive forest fires and droughts that limited plant growth, net uptake of CO2 by terrestrial carbon sinks was close to zero[17].

Surface oceans continue to chug along, absorbing a good portion of CO2 emissions at a relatively consistent rate. But as oceans absorb CO2, their acidity increases, reducing their capacity to absorb more CO2. By the end of the century oceans may reach a point where they can no longer absorb CO2[18], and if the Atlantic Meridional Overturning Circulation (AMOC), which brings warm water from near the equator to northern latitudes collapses (yes, that’s a very real possibility now due to warming), that would also reduce the ocean’s capacity to absorb CO2[19].

What about soils?

Soils are the largest terrestrial store of carbon at ~1650 Gt C (6056 Gt CO2e). But just because they’re the largest store doesn’t mean it has the highest potential as a climate solution. Soils would need to be able to accrue more carbon above and beyond current levels to make a meaningful impact on carbon removal.

The most comprehensive estimate of soil carbon lost to human activity, which is a reasonable proxy for the maximum total potential for soil carbon sequestration, is 133 Pg C (487.7 Gt CO2) [20]. So if we maxed out soil carbon sequestration globally we’d account for a little under half of legacy fossil fuel emissions, or less once you start to include future emissions. But that’s the maximum total potential and would require returning to pre-agricultural levels of soil carbon through new practices or converting ag land back to natural habitat. Considering the necessity to continue using agricultural lands to produce food, adoption/learning curves for key practices (e.g. cover crops, reduced tillage, etc.), current trends in agricultural expansion/contraction, and the complexities of land tenure, actual achievable removals are likely much lower[21], [22].

What’s more, as our understanding of soil carbon dynamics has evolved, we’ve learned that no part of the soil carbon pool is completely protected from decomposition. Nothing is truly sequestered, microbes can eat anything if they put their minds to it. Soil carbon levels are much more like a bank account—some amount is coming in all the time from plants, and some amount is always being lost as microbes in the soil eat carbon for energy and emit CO2. So it’s all about balancing inputs and outputs to achieve any kind of increase, and you have to maintain that balance for a long time to achieve appreciable increases. Stopping a practice or having a particularly bad growing year could upset the balance, resulting in delayed progress or even losses.

Now, it’s true that we could be underestimating the total potential of soil carbon removals, and there is healthy debate around the carbon removal rates of different practices and how well we understand how carbon is stabilized in soils. But regardless, all current evidence suggests the total contribution of soil carbon removal to climate change mitigation is small, will continue to be uncertain, and will always be subject to real world constraints. It’s a non-permanent store of carbon, and the human element makes it even more so.

I thought this was about the VCM?

For all intents and purposes, the VCM treats geological and biological pools as functionally equivalent and all uncertainties around permanence as manageable. Emitters buy credits that correspond to one ton of carbon removed to a biological pool (typically), or one ton of land use emissions avoided, to offset a ton of “unavoidable” fossil fuel emissions. Advocates of this approach argue “a ton is a ton is a ton.” The planet does not know the difference, and bickering about the difference undermines action.

Since it’s virtually impossible to guarantee that on a ton-by-ton basis carbon removed by a project will not be re-emitted (i.e. we guarantee this tree will never burn down, and/or decompose), VCM projects manage non-permanence at the project level. That is, the project as a whole has to result in net carbon removal, and total removals need to be maintained for a period of 100 years after the initial project period ends. Projects generally also have to contribute to a “buffer pool” which serves as an aggregate pool of insurance credits that are canceled if the soil or forests in a project experience reversals in the future (e.g. a forest fire happens).

The VCM is clever about aggregating and managing risk, but those tools can only go so far and require us to trust that they’ll work long into the future. For my money, uncertainty around permanence is a huge part of what’s underneath the uncertainty in the agricultural VCM and why it’s been slow to scale. That uncertainty can be managed, but it’s a fundamental constraint that will never go away, and savvy buyers and investors are aware of that. The same is true for other project types, such as reforestation efforts—a single forest fire could wipe out years of removals within days. But the faster ROI in terms of both dollars and carbon removal for those is just better, so buyers are more apt to take the risk.

So what now?

Beyond the next few decades, biological carbon sinks are not likely to be an adequate substitute for much more permanent, geological sinks. Recently, the researcher that originally coined the term “net-zero” published a paper arguing for a new concept “geological net zero.”[23] This concept suggests that since fossil fuels are a form of geological carbon storage, they must be offset 1-to-1 with geological carbon removals, which could include Direct Air Capture (DAC), Carbon Capture and Storage (CCS), or Enhanced Rock Weathering (ERW). Such solutions are still really expensive and would require low-carbon energy to be net-positive over deployment cycles, but we’ll need them (or something similar) eventually.

Until then, climate change still demands an “all-hands-on-deck” approach and absolutely nothing should be left off the table. Agricultural carbon sequestration can be seen as a climate stability wedge[24], and studies suggest that even temporary storage could help us stay on track for less damaging future climate scenarios[25]. It’s harm reduction, and that’s a valid goal.

But that goal may need a different finance mechanism that’s based on rewarding action towards temporary carbon storage as opposed to permanent removal, or at least a reframing of buyer’s expectations or goals. To be fair, this analysis is not a Dan Kane original! Many other VCM participants and observers have raised the same issues and looked for solutions on both the supply and demand side. BUT, did they spend several hundred words in the Paleozoic and Mesozoic Eras? I bet not. Next time, we’ll explore what people are thinking about the relative value of temporary storage and how to develop financing solutions to match.

One parting thought. Even if the carbon story is uncertain, our collective commitment to and investment in regenerative ag is absolutely critical due to the multitude of co-benefits. Cover crops, crop rotation, managed grazing, perennializing field margins, among many other practices all have benefits that are important in their own right. They can improve soil health, foster resilience to extreme weather, provide habitat for animals, and reduce agricultural pollution. We already demand so much of the land, we can’t expect it to also save us from ourselves. To live in reciprocity with the land also means we need to be honest and to not focus our attention on just one goal.

References

[1] Procton, Alex, “State of the Voluntary Carbon Market 2025.” Ecosystem Marketplace, 2025. Available: https://www.ecosystemmarketplace.com/publications/2025-state-of-the-voluntary-carbon-market-sovcm/

[2] D. A. Bossio et al., “The role of soil carbon in natural climate solutions,” Nat. Sustain., pp. 1–8, Mar. 2020, doi: 10.1038/s41893-020-0491-z. Available: https://www.nature.com/articles/s41893-020-0491-z

[3] M. Abdalla et al., “Critical review of the impacts of grazing intensity on soil organic carbon storage and other soil quality indicators in extensively managed grasslands,” Agric. Ecosyst. Environ., vol. 253, pp. 62–81, Feb. 2018, doi: 10.1016/j.agee.2017.10.023. Available: https://www.sciencedirect.com/science/article/pii/S0167880917304693

[4] L. Xiao, S. Zhou, R. Zhao, P. Greenwood, and N. J. Kuhn, “Evaluating soil organic carbon stock changes induced by no-tillage based on fixed depth and equivalent soil mass approaches,” Agric. Ecosyst. Environ., vol. 300, p. 106982, Sep. 2020, doi: 10.1016/j.agee.2020.106982. Available: https://www.sciencedirect.com/science/article/pii/S0167880920301675

[5] V. Chaplot and P. Smith, “Cover crops do not increase soil organic carbon stocks as much as has been claimed: What is the way forward?,” Glob. Change Biol., vol. 29, no. 22, pp. 6163–6169, 2023, doi: 10.1111/gcb.16917. Available: https://onlinelibrary.wiley.com/doi/abs/10.1111/gcb.16917

[6] P. Stanley, J. Spertus, J. Chiartas, P. B. Stark, and T. Bowles, “Valid inferences about soil carbon in heterogeneous landscapes,” Geoderma, vol. 430, p. 116323, Feb. 2023, doi: 10.1016/j.geoderma.2022.116323. Available: https://linkinghub.elsevier.com/retrieve/pii/S0016706122006309

[7] R. A. Berner, “The Rise of Plants and Their Effect on Weathering and Atmospheric CO2,” Science, vol. 276, no. 5312, pp. 544–546, Apr. 1997, doi: 10.1126/science.276.5312.544. Available: https://science.sciencemag.org/content/276/5312/544

[8] R. A. Berner, “The carbon cycle and CO2 over Phanerozoic time: the role of land plants,” 1998.

[9] D. L. Royer, R. A. Berner, I. P. Montañez, N. J. Tabor, and D. J. Beerling, “CO2 as a primary driver of Phanerozoic climate,” GSA Today, vol. 14, no. 3, p. 4, 2004, doi: 10.1130/1052-5173(2004)014<4:CAAPDO>2.0.CO;2. Available: http://www.geosociety.org/gsatoday/archive/14/3/pdf/i1052-5173-14-3-4.pdf

[10] P. Kenrick and P. R. Crane, “The origin and early evolution of plants on land,” Nature, vol. 389, no. 6646, pp. 33–39, Sep. 1997, doi: 10.1038/37918. Available: https://www.nature.com/articles/37918

[11] P. G. Falkowski et al., “The Evolution of Modern Eukaryotic Phytoplankton,” Science, vol. 305, no. 5682, pp. 354–360, Jul. 2004, doi: 10.1126/science.1095964. Available: https://www.science.org/doi/10.1126/science.1095964

[12] M. E. Raymo and W. F. Ruddiman, “Tectonic forcing of late Cenozoic climate,” Nature, vol. 359, no. 6391, pp. 117–122, Sep. 1992, doi: 10.1038/359117a0. Available: https://www.nature.com/articles/359117a0

[13] P. Friedlingstein et al., “Global Carbon Budget 2024,” Earth Syst. Sci. Data, vol. 17, no. 3, pp. 965–1039, Mar. 2025, doi: 10.5194/essd-17-965-2025. Available: https://essd.copernicus.org/articles/17/965/2025/

[14] “Chapter 5: Global Carbon and other Biogeochemical Cycles and Feedbacks.” Available: https://www.ipcc.ch/report/ar6/wg1/chapter/chapter-5/

[15] thomas harrisson, “Explainer: How ‘Shared Socioeconomic Pathways’ explore future climate change,” Carbon Brief, Apr. 19, 2018. Available: https://www.carbonbrief.org/explainer-how-shared-socioeconomic-pathways-explore-future-climate-change/

[16] Z. Hausfather and G. P. Peters, “Emissions – the ‘business as usual’ story is misleading,” Nature, vol. 577, no. 7792, pp. 618–620, Jan. 2020, doi: 10.1038/d41586-020-00177-3. Available: https://www.nature.com/articles/d41586-020-00177-3

[17] P. Ke et al., “Low latency carbon budget analysis reveals a large decline of the land carbon sink in 2023,” Natl. Sci. Rev., vol. 11, no. 12, p. nwae367, Dec. 2024, doi: 10.1093/nsr/nwae367. Available: https://doi.org/10.1093/nsr/nwae367

[18] M. O. Chikamoto, P. DiNezio, and N. Lovenduski, “Long-Term Slowdown of Ocean Carbon Uptake by Alkalinity Dynamics,” Geophys. Res. Lett., vol. 50, no. 4, p. e2022GL101954, 2023, doi: 10.1029/2022GL101954. Available: https://onlinelibrary.wiley.com/doi/abs/10.1029/2022GL101954

[19] Q. Zhang, T. Ito, and A. Bracco, “Modulation of regional carbon uptake by AMOC and alkalinity changes in the subpolar North Atlantic under a warming climate,” Front. Mar. Sci., vol. 11, Apr. 2024, doi: 10.3389/fmars.2024.1304193. Available: https://www.frontiersin.org/journals/marine-science/articles/10.3389/fmars.2024.1304193/full

[20] J. Sanderman, T. Hengl, and G. J. Fiske, “Soil carbon debt of 12,000 years of human land use,” Proc. Natl. Acad. Sci., vol. 114, no. 36, pp. 9575–9580, Sep. 2017, doi: 10.1073/pnas.1706103114. Available: http://www.pnas.org/content/114/36/9575

[21] E. E. Oldfield et al., “Greenhouse gas mitigation on croplands: clarifying the debate on knowns, unknowns and risks to move forward with effective management interventions,” Carbon Manag., vol. 15, no. 1, p. 2365896, Dec. 2024, doi: 10.1080/17583004.2024.2365896. Available: https://doi.org/10.1080/17583004.2024.2365896

[22] S. C. McClelland et al., “Managing for climate and production goals on crop-lands,” Nat. Clim. Change, pp. 1–8, May 2025, doi: 10.1038/s41558-025-02337-7. Available: https://www.nature.com/articles/s41558-025-02337-7

[23] M. R. Allen et al., “Geological Net Zero and the need for disaggregated accounting for carbon sinks,” Nature, vol. 638, no. 8050, pp. 343–350, Feb. 2025, doi: 10.1038/s41586-024-08326-8. Available: https://www.nature.com/articles/s41586-024-08326-8

[24] S. Pacala and R. Socolow, “Stabilization wedges: solving the climate problem for the next 50 years with current technologies,” science, vol. 305, no. 5686, pp. 968–972, 2004, Available: http://www.sciencemag.org/content/305/5686/968.short

[25] H. D. Matthews et al., “Temporary nature-based carbon removal can lower peak warming in a well-below 2 °C scenario,” Commun. Earth Environ., vol. 3, no. 1, pp. 1–8, Mar. 2022, doi: 10.1038/s43247-022-00391-z. Available: https://www.nature.com/articles/s43247-022-00391-z