Carbon markets run on trust, and trust requires proof. The mechanism that generates that proof — Measurement, Reporting, and Verification — sounds procedural until you realize it's the entire reason a soil carbon credit is worth anything at all. Without rigorous MRV, a carbon credit is just a promise with a price tag attached.
Breaking Down the Three Letters
Measurement is the most technically demanding component. In the context of soil organic carbon (SOC), it means physically quantifying how much carbon is stored in a given volume of soil — expressed as a percentage of dry soil mass, typically across defined depth intervals (0–30 cm, 30–60 cm, 60–90 cm). The measurement challenge isn't conceptual; it's statistical. Soil is heterogeneous. SOC concentration can vary by a factor of two within a single field, driven by micro-topography, organic matter inputs, drainage patterns, and prior tillage history. A single soil core is almost meaningless. A statistically defensible measurement requires a sampling design — grid density, stratification, minimum sample count — that captures that spatial variability.
Reporting is the documentation layer. Every data point — sample location coordinates, collection date, depth intervals, lab analysis method, chain-of-custody records, bulk density measurements — must be documented in a format that a third-party verifier can follow from field to spreadsheet without ambiguity. Reporting requirements are specified by the protocol under which credits are issued. Verra's Verified Carbon Standard (VCS) methodology VM0042, for instance, requires specific output data fields and statistical uncertainty characterizations. Reporting is where many first-time projects fail: the measurement data exists, but the documentation trail is incomplete enough to block registry issuance.
Verification is third-party confirmation that measurement and reporting were conducted correctly. A Verra-approved Validation/Verification Body (VVB) reviews the project documentation, may conduct field audits, and issues a verification statement that allows credits to be issued to the registry. For corporate buyers, this is the legal and reputational backstop — the document they point to when their ESG disclosures are audited.
Why Precision Is Not Optional
There's a reason the carbon market has had credibility problems, and it's not mostly fraud — it's imprecision masquerading as measurement. Early voluntary carbon methodologies, particularly for soil carbon, accepted modeled SOC estimates derived from IPCC Tier 1 default emissions factors. These are national-level averages applied to individual farm fields. The uncertainty range on a Tier 1 estimate can exceed ±50% of the estimated value.
Consider what that means practically: a farm generating a nominal 200 tonnes of CO₂e per year under a Tier 1 estimate might actually be sequestering anywhere from 100 to 300 tonnes. The credit buyer is paying for the midpoint of a range wide enough to drive a combine through. For buyers with net-zero commitments that face external audit, that uncertainty is not acceptable — it migrates into their Scope 3 accounting and creates material restatement risk.
We're not saying model-based estimation is always wrong. We're saying it shouldn't be the final word when the asset being sold is a specific quantity of carbon from a specific field.
The IPCC Tier 3 methodology standard requires project-specific, direct measurement combined with local calibration data. That's a meaningful precision step up — but it's also more expensive and operationally complex, which is why many early soil carbon programs avoided it. The market is now correcting toward Tier 3 as buyer scrutiny has intensified.
Soil Carbon's Specific MRV Challenges
MRV for soil carbon is technically harder than for most other carbon project types, and it's worth being direct about why.
First, the baseline problem. Unlike avoided deforestation, where you can observe whether trees were cut, SOC has no visible proxy. You cannot determine a farm's pre-project carbon stock without physically sampling it. This means baseline establishment — the reference measurement against which all future sequestration is calculated — must be done carefully at project initiation, because there's no way to retroactively recover a baseline from historical satellite data alone.
Second, the temporal dynamics. SOC is not static. It responds to management practices, weather events, crop rotations, and disturbance. A drought year can suppress the biological processes that build organic matter. An unexpected tillage event can release years of accumulated carbon within a single field operation. MRV frameworks must account for this by requiring ongoing monitoring at defined intervals — typically annual or biennial — rather than a one-time measurement.
Third, the depth distribution matters. Regenerative practices that increase SOC tend to do so preferentially in the topsoil (0–30 cm). But deep-rooted prairie species and cover crops with extensive root systems can contribute to subsoil carbon accumulation at 30–90 cm depth. Sampling only to 30 cm misses a real portion of the carbon stock, while also being more susceptible to disturbance from surface management changes. Protocols differ on required sampling depths, and that choice materially affects the reported credit quantity.
The Dual-Validation Approach
Ground-based soil core sampling provides accurate, direct measurements of SOC at sampled points. But sampling every acre of a large farm is economically prohibitive — at typical lab analysis costs, whole-farm dense sampling would cost more than the credits are worth. The practical solution is to use ground-truth samples to calibrate a satellite-based model that can then extrapolate SOC concentration estimates across unsampled areas.
This is where shortwave infrared (SWIR) spectral reflectance from platforms like Sentinel-2 and Landsat 9 becomes relevant. Bare soil reflectance in the SWIR bands (approximately 1,550–1,750 nm and 2,090–2,350 nm) correlates inversely with SOC concentration — soils with more organic matter absorb more SWIR radiation and appear darker. That correlation is real and well-documented in the soil science literature, though it's modulated by soil moisture, clay content, iron oxide concentration, and surface roughness.
Terrabit's methodology uses physical soil cores to anchor the spectral model on each enrolled field. The core samples serve two functions: they provide direct SOC measurements for the sampled points, and they calibrate the farm-specific spectral reflectance relationship so that satellite imagery can be used for continuous monitoring between sampling campaigns. Neither source alone is sufficient; together, they bound the uncertainty to a level that supports credit issuance at ±0.3% SOC precision.
What MRV Quality Means for Credit Value
The market is increasingly pricing MRV quality into credit transactions. Buyers' due diligence processes — particularly for large volume purchases for Scope 1/2 offsetting — now routinely include methodology review. A credit package with full chain-of-custody documentation, direct SOC measurements at appropriate sampling density, and a third-party verification statement commands a premium over modeled estimates. The spread between high-quality and low-quality soil carbon credits has widened as the market has matured.
For growers considering enrollment in a carbon program, MRV quality directly affects whether credits can be sold at all in premium voluntary markets, and what price range they'll command. A program that uses opaque or model-only methodology may generate more credits on paper — but those credits will face growing resistance from buyers with sophisticated ESG programs, and increasingly from the registries themselves as standards tighten.
The question isn't whether MRV is worth doing well. It's whether the cost of doing it well — the sampling campaigns, the lab analysis, the satellite calibration work, the documentation — can be efficiently enough to leave growers with meaningful net revenue. That's the engineering problem Terrabit was built to solve.
Amara Diallo is the founder and CEO of Terrabit. She holds a background in soil biogeochemistry and remote sensing from the University of Iowa. Questions about MRV methodology can be directed to [email protected].