Somewhere in the paperwork of a mid-sized grain farm in Kansas, a number disappears. The farmer enrolled two years ago, paid for soil sampling at multiple depths, changed his tillage practice, and waited. The verifier's report comes back with a sequestration figure roughly forty percent lower than the project developer's estimate. No explanation travels back to the farmer in plain language. The registry's methodology committee had, some months earlier, quietly revised its additionality baseline upward, and the verifier, working from an internal protocol update that hadn't been publicly announced, applied the new threshold retrospectively. The credit issuance was reduced. The farmer's carbon payment shrank accordingly.
This is how governance actually shapes outcomes in soil carbon markets. Not through scandal or fraud, but through the slow, structural machinery of who sets the rules, who interprets them, and what the verifier is actually checking when they sign off on a claim.
The body that writes the rules nobody reads
Every major soil carbon registry, whether Verra's Verified Carbon Standard, Gold Standard, the American Carbon Registry, or a newer entrant like Soil Carbon Initiative, operates through a layered governance structure. At the top sits a board, typically composed of scientists, NGO representatives, corporate buyers, and occasionally farmer advocates. Below that, methodology committees draft and revise the technical protocols: how soil samples must be collected, at what depth increments, using which statistical models for uncertainty discounting, and what reference scenarios count as a valid baseline.
The critical word in that last sentence is "baseline." A baseline is the counterfactual, what would have happened to soil carbon on this land if the farmer had done nothing differently. Set it low, and most farms can show additionality, meaning genuine new sequestration above and beyond business-as-usual. Set it high, and vast swathes of agricultural land simply can't generate credits, because the modeled business-as-usual trajectory already predicts carbon gains from regional trends in rainfall, temperature, or already-shifting practices.
Methodology committees revise baselines periodically, often in response to new science. Legitimate. Necessary, even. The governance question is whether those revisions apply retroactively to projects already in the pipeline, and whether the rationale is publicly legible before it affects a farmer's payout. On both counts, the record across registries is uneven, and "uneven" is the polite word for it.
What a verifier is actually checking
Think of the verifier less like an auditor and more like a referee who is also reading the rulebook for the first time. Verifiers are accredited third parties, often consultancies with soil science capacity, contracted either by the project developer or by the registry itself. The distinction matters enormously: a verifier hired by the developer has an implicit interest in finding a number that makes the project viable. A verifier contracted directly by the registry, paid from a fee structure that doesn't depend on credit issuance, has somewhat better insulation from that pressure.
In practice, verification involves reviewing the sampling methodology, checking the chain of custody for soil cores, running the reported figures through the registry's approved quantification model, and applying whatever uncertainty buffers the current protocol mandates. Uncertainty buffers are the quiet workhorses of credit reduction. If a protocol requires a 20% buffer deduction to account for measurement uncertainty, a project claiming 10,000 tonnes of CO2 equivalent will issue only 8,000 credits. Raise that buffer requirement to 30%, and the same physical soil produces 7,000 credits. The science behind buffer sizing is real. The specific number chosen, though, is a governance decision, made by a committee, and it compounds across thousands of projects like interest on a loan nobody agreed to.
Verifiers also apply what the protocols call "permanence discounts." Soil carbon can be lost: a drought, a change in ownership, a return to conventional tillage. Registries address this by withholding a percentage of credits into a pooled buffer account. The size of that pool, and the formula for calculating each project's contribution to it, is again a governance variable. A project in a region the methodology committee has classified as high drought-risk might contribute 25% of its credits to the buffer rather than the 10% assigned to a lower-risk project. That classification is made at the registry level, not by the farmer, and it is not always transparent which climate-risk tiers cover which geographies.
The moment the claim gets quietly trimmed
Consider two farmers, Maria and David, who both enrolled in the same registry program in the same year, farming comparable acreage in the same state. Maria used a project developer who submitted her sampling data promptly and whose methodology aligned closely with the registry's approved quantification tool at the time of submission. David's developer submitted six months later, by which point the methodology committee had issued an addendum tightening the minimum sampling depth from 30 centimeters to 50 centimeters for certain soil types. David's samples hadn't reached 50 centimeters. His verifier flagged the gap. The registry's internal review panel determined his project was non-compliant and issued zero credits for that monitoring period.
Zero.
David wasn't committing fraud. His developer wasn't either, necessarily. A protocol update issued with no grace period for projects already mid-monitoring turned a technical revision into a financial wipeout. Maria, whose paperwork cleared the gate before the addendum, received her credits. Same soil type, same practice change, same regional carbon dynamics. Different governance moment.
This is not an invented edge case. It reflects a structural tension that soil scientists and carbon market researchers have documented repeatedly: the pace of methodological refinement in soil carbon science genuinely is fast, because the field is still maturing. Registries that freeze their protocols to protect enrolled farmers risk issuing credits that later science would not support. Registries that update aggressively protect scientific integrity but introduce a kind of regulatory whiplash that falls hardest on smaller operators without legal counsel to track protocol amendments. Both failure modes are real. Pretending there is a clean path between them is the kind of reassurance that belongs in a sales brochure, not a governance document.
The permanence problem nobody has solved cleanly
Soil carbon is not like a wind turbine, which either generates electricity or it doesn't. It is a biological process operating across decades, subject to reversal by events the farmer cannot control. A registry's governance must therefore answer a genuinely hard question: if carbon is lost after credits have been sold, who bears the liability?
The pooled buffer approach is the dominant answer, but its governance has a structural flaw worth naming plainly. The buffer pool is managed by the registry itself. The registry decides when a reversal event is large enough to trigger a drawdown from the pool. The registry also decides how to replenish the pool if drawdowns deplete it. There is no external trustee, no independent actuarial review of buffer adequacy, and in most cases no public disclosure of the pool's current balance relative to its theoretical liability. Buyers of soil carbon credits are, in effect, trusting the registry's governance to hold a promise that extends thirty to a hundred years into the future, administered by an organization that may not exist in its current form for anywhere near that long.
Ask yourself: would you buy a fire insurance policy from a company that also decided, internally and without external audit, whether your house had actually burned down?
This is the crust that builds up inside the system, invisible until something breaks. Not dishonesty, but accumulated governance decisions, each defensible in isolation, that collectively produce a structure where the entity most exposed to long-term liability is also the entity least accountable to external oversight.
The one honest caveat
It would be wrong to read all of this as a condemnation of soil carbon markets or the registries that run them. The governance problems described here are real, but they are also known and actively contested. Several registries have moved to publish methodology committee minutes, establish farmer advisory panels, and introduce formal comment periods before protocol changes take effect. Verra's VCS has faced significant public criticism over its forestry credits, which has created pressure for more rigorous soil protocols. The American Carbon Registry has published comparatively detailed documentation of its uncertainty quantification approach. Progress is uneven and slow, but it is not absent.
The honest caveat is this: most critiques of soil carbon markets focus on measurement accuracy, which is a science problem. The governance problem is distinct and harder to fix, because it requires registries to accept constraints on their own discretion. Science can tell you how accurately we can measure a tonne of carbon in topsoil. Governance determines whether the farmer who sequestered it gets paid, and whether the buyer who purchased the credit can trust it in twenty years. Those are political and institutional questions, not technical ones, and they do not resolve themselves just because the sampling methodology improves.
If you are a landowner considering enrollment in a soil carbon program, the question worth asking is not only how the registry measures carbon, but who sits on the methodology committee, how often protocols change, whether those changes apply retroactively, and what the buffer pool's disclosed balance is. Most program brochures won't answer any of those questions. The ones that do are telling you something important about how seriously they take the word "permanent," and that gap, between the brochure and the methodology committee minutes, is where a farmer's income goes to get quietly revised downward.