Soil carbon is not a container being filled. It is a balance between what enters, such as root exudates, litter, residues, and manures, and what leaves, almost entirely as CO₂ from microbial respiration. Once you see it as a flow, the frustrating shape of real SOC monitoring data stops being a surprise.
The Balance
Stock rises when inputs exceed losses. But losses are not fixed: microbial decomposition scales with the amount of decomposable carbon present. Add more carbon and you also feed the organisms that consume it.
| Phase after a management change | Input vs output | Observed stock change |
|---|---|---|
| Early | Inputs raised, outputs still low | Fastest gain |
| Middle | Outputs rising toward new inputs | Gain decelerating |
| New equilibrium | Outputs match inputs | No further gain |
This is why the honest expectation for a soil carbon intervention is a decelerating curve approaching a new plateau, not a straight line. Projects that model SOC gain as linear over thirty years are modelling something that does not occur.
Why the Ceiling Exists
The plateau is not arbitrary. It is set by how much carbon the soil can physically and chemically protect from decomposition, with clay content and mineralogy being the dominant controls, alongside climate setting the decomposition rate.
A sandy soil in a hot, wet climate has a low ceiling and reaches it quickly. A clay-rich soil in a cooler climate has a higher ceiling and takes longer. The same management change produces quite different outcomes on the two, which is why SOC results are so site-specific and why borrowing a sequestration rate from another region is unreliable.
The Implication for Crediting
Two consequences follow, and both are uncomfortable for project economics.
First, the creditable period is finite. Gains taper. A project claiming steady SOC sequestration in year twenty-five should expect to be asked why, and the honest answer is usually that it should not be.
Second, the gain is reversible. Because the stock sits at an equilibrium maintained by continuing inputs, reverting the management reverts the carbon and losses occur faster than gains, because decomposition is not rate-limited the way accumulation is. This is the basis for the strict permanence requirements attached to soil carbon crediting.
What This Means for Monitoring Design
If the expected change is largest early and diminishing later, monitoring effort should not be uniform.
Dense sampling in the first monitoring periods captures the signal when it is strongest relative to the noise. Later periods, where the annual change may be smaller than the measurement uncertainty, produce intervals that cannot distinguish the change from zero no matter how many cores are taken.
The uncomfortable conclusion some projects reach: after a certain point, continuing to measure SOC costs money and generates no creditable result. Planning for that in advance is better than discovering it in year twelve.
Turn climate goals into a verified portfolio
TREEO connects the full carbon cycle, encompassing eligibility, simulation, real time monitoring, and registry ready reporting, thereby combining expert consulting with dMRV technology so the evidence exists before anyone asks for it.



