What “soil carbon” means in a CDR context
Soil carbon as a removal pathway is the deliberate addition — or protection — of organic carbon in agricultural and grassland soils so that atmospheric CO₂ ends up stored as soil organic matter rather than circulating in the air. In practice, almost every commercial project in the directory today is doing this through biochar: pyrolysing biomass into a stable, carbon-rich solid and burying it in farmland. A smaller share of projects pursue management-based sequestration (cover crops, no-till, compost, agroforestry) where the carbon gain comes from shifting the balance between plant input and microbial decomposition. The two approaches share a destination — carbon in soil — but the durability profiles and the science behind them are very different, which is the single most important thing for a buyer or journalist to internalise before going further.
How it works
Soil organic carbon (SOC) accumulates when carbon inputs (root exudates, residues, amendments) exceed microbial mineralisation back to CO₂. Recent work has overturned the older view that SOC is mostly plant-derived humus. Liang et al. (2021) showed that microbial necromass — dead bacterial and fungal cell walls — is the dominant precursor to stable soil carbon across global ecosystems. Georgiou et al. (2022) estimated mineral-associated organic matter (MAOM), the most persistent SOC pool, at roughly 899 Gt C globally with substantial unfilled capacity in temperate cropland. The implication: durability depends on whether carbon ends up bound to mineral surfaces (decadal-to-centennial) or sitting as particulate matter (years-to-decades, and climate-sensitive). Practices that boost root exudation and mycorrhizal turnover — reviewed by Panchal et al. (2022) — feed the MAOM pathway more efficiently than surface residue dumping.
Biochar is a different mechanism. Biomass is heated to 450–750 °C in low-oxygen conditions, driving off volatiles and leaving a recalcitrant aromatic carbon skeleton. Yaashikaa et al. (2022) reviews the feedstock-and-temperature design space; Lehmann et al.-style reviews of agronomic effects summarise the co-benefits (water retention, cation exchange, reduced nutrient leaching). The carbon in biochar resists mineralisation because its fused-ring structure is not a useful substrate for most soil microbes — half-lives are typically modelled in centuries, though the tail of labile carbon decays within years.
Who’s doing it
The directory is biochar-heavy, which mirrors the actual commercial market — management-based SOC has struggled to issue durable credits.
- NetZero operates mid-size >600 °C pyrolysis plants on tropical crop residues (coffee husks, sugarcane bagasse) in Brazil and Cameroon, and is one of the larger Puro.earth-certified suppliers by volume.
- Pyreg is a German equipment manufacturer — its PX500/PX1500 reactors are the workhorse hardware behind many European biochar projects, particularly those processing sewage sludge.
- Carbo Culture (Finland) runs its own “Carbolysis” high-temperature reactors and has sold forward volumes to Microsoft and others.
- MASH Makes operates pyrolysis plants in Karnataka, India, fed largely by cashew-industry waste, with bio-oil as a co-product.
- Bio-Logical runs what it describes as Africa’s largest biochar plant near Mt. Kenya, blending biochar into a fertiliser product distributed to smallholders.
- Syncraft sells wood-gasification plants whose primary product is electricity and heat, with biochar as a byproduct — a useful contrast to dedicated CDR-first operators.
- Aymium targets industrial off-take (coal/coke replacement in metals), which routes the carbon outside soil entirely and raises a different durability accounting question.
- AquaGreen focuses on sewage-sludge pyrolysis, a feedstock with disposal-cost economics that don’t depend on the credit price.
These illustrate the main axes of variation: feedstock (woody vs. crop residue vs. sludge), end-use (soil vs. industrial), business model (project developer vs. equipment OEM), and geography.
Durability and MRV
For biochar, the Schmidt et al. EBC framework and Puro.earth methodology converged on H/C ratios (typically <0.4 or <0.7 depending on the standard) as a proxy for centennial stability, with permanence discounts applied accordingly. This is workable but coarse — field decay data beyond 10–20 years is sparse, and the IPCC’s 100-year-permanence convention is a policy choice, not a measurement.
Management-based soil carbon is harder. Bai & Cotrufo (2022) catalogue the problems in grassland systems: high spatial heterogeneity, slow signal-to-noise, and saturation behaviour in MAOM. Reversal risk under tillage or drought is real and not hypothetical. Most management-credit registries rely on modelled outcomes with sample sparse soil sampling, which is why buyers seeking durable tonnes have largely avoided them.
The honest summary: biochar has a defensible (if imperfect) MRV stack and is what the durable CDR market is actually buying. Soil-management credits are a different product — co-benefit-heavy, durability-light.
Open questions
A reader returning in a year should look for:
- Field decay data. Long-term in-situ measurements of aged biochar that either confirm or revise the centennial half-life assumption, particularly in tropical soils where the bulk of new capacity sits.
- Feedstock competition. As pyrolysis hardware scales, whether agricultural residue supply tightens — and whether projects shift to purpose-grown biomass, which changes the LCA materially.
- MAOM-targeted agronomy. Whether anyone successfully commercialises a management-based credit that ties payment to measured mineral-associated carbon rather than modelled bulk SOC.
- Industrial-use accounting. How standards treat biochar used as a metallurgical reductant (Aymium, CHAR), where the carbon is oxidised within years but displaces fossil coke.
- Methane and N₂O. Net-GHG accounting in paddy and fertilised systems, where pathway differences can erode or amplify the headline CO₂ removal.