Where soil carbon credit revenue actually comes from

Regenerative practices are spreading faster than the ability to prove what they deliver. Reported sequestration rates for cover cropping, reduced tillage and residue retention carry uncertainty bands approaching half their central values, which is why a site-specific measured figure is worth paying for. Soil carbon is also reversible: a season of intensive tillage can undo years of accumulation. Permanence provisions and continuous monitoring therefore matter more than any single point-in-time measurement.

The practices, and what they actually deliver

Regenerative agriculture is a family of practices rather than a protocol. Five appear consistently across European and North American systems, and they differ substantially in both carbon effect and how confidently that effect can be stated.

Two things stand out from the ranges rather than the central values.

First, the uncertainty is large relative to the effect. A practice sequestering 0.32 tonnes of carbon per hectare per year with a stated range of plus or minus 0.08 has an uncertainty band of roughly half its own value. For a single field over a single year, that is not a measurement, it is a probability distribution. Credible claims require either long time series or large areas, usually both.

Second, these are averages across soils, climates and starting conditions that vary enormously. A degraded soil low in organic matter has more headroom and will typically sequester faster than a soil already well managed. The farms with the most to gain are frequently the ones least equipped to prove it.

Adoption: real, and harder to quantify than it looks

Practice adoption across Europe and the United States is genuinely substantial, crop rotation is near-universal, reduced tillage covers a significant minority of arable land, cover cropping is expanding quickly.

Treat these figures as indicative. Adoption statistics for regenerative practices vary widely between sources, and the reason is definitional rather than statistical. Does reduced tillage mean no-till, minimum till, or one fewer pass? Does a cover crop count if it is grazed or terminated early? Does crop rotation count as regenerative when it has been standard practice for centuries?

This is not pedantry. It is the same problem that makes the carbon claims hard to verify, appearing earlier in the chain. If a practice cannot be defined precisely enough to count consistently, it cannot be verified consistently either.

“Half the disagreement about regenerative agriculture is people using the same word for different things and then comparing the numbers.” — Spacenus team

Why the proof is the bottleneck

The economic logic of carbon farming is straightforward: a farmer changes practice, soil carbon increases, someone pays for the increase. Every element works except the middle one, which has to be demonstrated to a standard that survives audit.

The signal is small and slow

Annual soil carbon change is a small fraction of total soil carbon stock. Detecting it against natural spatial variability requires either dense sampling, which is expensive, or modelling validated by targeted sampling, which requires the model to be right.

It is reversible

Carbon accumulated over five years can be released by one season of intensive tillage. This is why credit frameworks require permanence provisions, buffer pools and continuous monitoring rather than a single point-in-time measurement. A soil carbon certificate without a monitoring obligation attached is a photograph presented as a guarantee.

Verification cost has been the binding constraint

At 25–50% of project budget, conventional MRV puts a floor on viable programme size and excludes most smallholders entirely. This is the constraint that has actually held the sector back, not farmer willingness, not agronomic uncertainty, not buyer demand.

It is also the one that has moved. In the ESA SatMRV programme, verification across 8,845 hectares of European arable farmland came to 4.1% of projected credit revenue, using satellite-guided stratified sampling. That is a change in what is economically possible rather than an incremental improvement.

What changed in 2026

Two developments matter for anyone deciding whether regenerative claims are now provable.

The EU Carbon Removal Certification Framework, in force since December 2024, had its first carbon farming methodologies adopted in July 2026. European carbon farming now has an operative certification pathway with a defined verification standard, and that standard requires verification by a party independent of whoever operates the programme.

Separately, the GHG Protocol Land Sector and Removals Standard takes effect on 1 January 2027 and requires primary, field-level data for companies with significant land-sector activity. That obligation applies whether or not a single credit is issued, which means food companies now need field-level evidence from regenerative supply chains regardless of their position on carbon markets.

What a provable claim looks like

• A defined practice. Not "regenerative" but a specific, observable change: cover crop established by a date, tillage passes reduced from three to one.

• Evidence the practice occurred. Satellite time series showing the spectral signature at the right time, corroborated by geotagged field evidence and cross-checked against comparable neighbouring land.

• A baseline. What the soil carbon was before, measured or modelled with stated uncertainty, and what it would have done without the change.

• Continuous monitoring. Evidence that the practice persisted and the carbon was not released again.

• Independent verification. Checked by a party with no financial interest in the answer.

• The last one is the shortest and does the most work. An operator that designs the programme, recruits the farmers, issues the credits and verifies the outcome has a structural conflict, which is precisely why the CRCF addresses it directly.

Where this leaves the sector

Regenerative agriculture is not short of ambition, funding or agronomic evidence. What it has been short of is affordable proof, and that has changed enough in the last two years to alter what is buildable.

The organisations that will do well from here are the ones that treat verification as infrastructure to be built early rather than a compliance cost to be minimised late. The 2027 deadline arrives regardless, and the evidence base cannot be assembled retrospectively, you cannot go back and observe a field you were not watching.

Spacenus verifies regenerative practice claims for programmes operated by others, through SatMRV and RegenScope3. We hold no credits and run no farmer programmes, which is what allows the same evidence to serve the operator, the buyer and the auditor without anyone taking a supplier’s word for it.

Common questions

How much carbon can regenerative practices sequester?

Reported rates for individual practices generally fall below one tonne of carbon per hectare per year, with uncertainty ranges often approaching half the central value. Rates depend heavily on soil type, climate and how degraded the starting condition was.

Is soil carbon sequestration permanent?

No. Accumulated soil carbon can be released by a return to intensive tillage or land use change, which is why credit frameworks require permanence buffers, defined monitoring periods and continuous verification.

Why do regenerative adoption statistics vary so much between sources?

Mostly definitional. Reduced tillage, cover cropping and rotation are all counted differently depending on the source, so headline percentages are not directly comparable.

Do I need carbon credits to benefit from regenerative practice data?

No, and increasingly the reporting obligation matters more. GHG Protocol LSR from January 2027 requires primary field-level data for significant land-sector activity regardless of whether any credit is issued.