5 August 2026 ยท Carbon accounting
Measuring what matters: why bulk soil carbon is not the whole story
Grasslands hold most of their carbon belowground, not in the grass and animals that are visible, but in the soil itself (Poeplau, 2021), a single fact that goes a long way toward explaining why regenerative grazing has become such an attractive climate story: get the soil right, the argument goes, and a carbon sink follows almost for free. I do not disagree with the potential, but I would push back on how confidently we are currently measuring whether it is actually happening.
Most carbon claims attached to regenerative grazing schemes and soil carbon credit markets still rest on bulk soil organic carbon measurements, fed into simplified turnover models, a reasonable starting point, but a blunt one. A tonne of carbon measured this year tells us almost nothing about whether that carbon is stable or about to be lost again through microbial respiration the moment conditions change, and two fields with identical bulk carbon numbers can be on completely different trajectories.
Splitting carbon into pools that actually mean something
Part of my PhD work is built around carbon fractionation, separating soil carbon into pools that behave differently rather than treating it as one undifferentiated number. Permanganate oxidisable carbon (POXC) captures the labile, biologically active fraction, the carbon currently driving microbial activity and nutrient turnover, and the part most likely to be respired straight back out, whereas mineral-associated organic matter (MAOM) represents carbon that is physically and chemically protected, stabilised through interaction with soil minerals, and much more likely to still be present in a decade (Six et al., 2002; Cotrufo et al., 2019).
A field with high bulk carbon dominated by the labile fraction is not the same asset as a field with the same bulk number dominated by mineral-associated carbon: one is a claim waiting to be reversed by next year's drought or a change in management, while the other is closer to genuine long-term storage, and at present almost none of that distinction makes it into farm-level or scheme-level carbon reporting.
Why this is not merely an academic distinction
As investment in regenerative grazing and soil carbon credit projects accelerates, claims are increasingly made without mechanistic biological validation behind them, a scientific problem, but also a financial one. A microbially and biologically informed carbon model can flag where projected carbon gains are likely overstated, a transition risk for anyone relying on those credits, and where microbial activity is vulnerable to drought or warming, a physical climate risk, which is directly relevant to nature-related financial disclosure frameworks such as TNFD, against which companies are increasingly expected to report.
None of this is an argument against regenerative grazing, but an argument for being honest about what our current measurements can, and cannot, tell us. The rest of this project, and the posts that follow, examine the biology that is mostly missing from that picture: how grazing intensity actually changes carbon inputs on the ground, and how the organisms living in the soil determine whether that carbon remains there.