The soils under the world’s croplands and pastures contain, in aggregate, more carbon than the atmosphere and the global biomass combined. They contain less, however, than they did before they were brought into agricultural production. The historical decline in soil organic carbon under conventional agriculture is, depending on region, between thirty and sixty percent of the pre-cultivation baseline. Recovering some fraction of that carbon, while continuing to produce food on the same land, is one of the few proposed climate interventions that does not require taking land out of production.
What soil organic carbon actually is
Soil organic carbon is not a single substance. It is a family of carbon-containing compounds — plant residues at various stages of decomposition, microbial biomass, and stable mineral-associated organic matter — that accumulate when biomass enters the soil faster than soil microbes respire it as carbon dioxide. The relevant practices are those that increase biomass inputs (cover crops, perennials, manure additions) and those that reduce decomposition rates (reduced tillage, perennial root systems that protect aggregates).
The practical agronomic effects of higher soil carbon are well documented: better water-holding capacity, improved soil structure, more resilient performance in drought years, lower fertilizer requirements over time, reduced erosion. These are first-order benefits that justify the practices on conventional production economics, independent of any carbon claim.
The carbon side, honestly
Soil-carbon sequestration claims have at times been overstated by their advocates. A few constraints matter. First, sequestration rates plateau over decades as the soil approaches a new equilibrium; it is not an indefinite carbon sink at the same per-hectare rate. Second, sequestration is reversible — a return to conventional tillage can release the gained carbon over a few seasons. Third, accurately measuring soil-carbon change at field scale is expensive and noisy, which has made carbon-credit markets in this category contentious.
The honest ranges suggest that aggressively pursuing soil-carbon-friendly practices on global cropland and grazing land could plausibly remove several tens of gigatons of CO2 from the atmosphere over the coming decades — meaningful but not transformative against annual emissions in the high-30s of gigatons.
The uncomfortable bit for the climate framing
The strongest case for regenerative practices is not actually their carbon contribution. It is that they appear to make the underlying productive asset — the soil — more resilient to the climatic shocks already in the pipeline. Whether they sequester an additional gigaton or not, agricultural systems that retain water through droughts and structure through floods are the ones likely to keep functioning. The carbon argument is partly a way to attract subsidy and credit-market financing to changes that have first-order agronomic merit.