Durability & Permanence
Durability (often used interchangeably with permanence) is the expected time that a tonne of carbon dioxide, once removed from the atmosphere, stays out of it. It is the single variable that most cleanly separates a nature-based credit priced in the low tens of dollars from an engineered credit priced in the high hundreds — and it is the variable that determines whether a corporate buyer can defensibly count a purchase against a net-zero claim. As Fankhauser et al. (2021) argue, “net zero” is only coherent if residual emissions are matched with removals of comparable longevity; a fossil CO₂ molecule released today and a soil-carbon credit that reverses in 30 years do not cancel out in any physically meaningful sense.
The mechanics and the tensions
Durability is usually thought of as a spectrum. At the short end sit soil organic carbon and standing forest projects, where storage half-lives are measured in decades and are exposed to management change, fire, drought, and pest outbreaks. In the middle sit biochar (centuries, with a mean residence time that depends on pyrolysis temperature and soil conditions) and enhanced rock weathering (millennia, once bicarbonate reaches the ocean). At the far end sit geologic pathways — mineralized CO₂ in basalt, saline aquifer injection under monitored Class VI wells — where storage is typically claimed at 10,000+ years.
Reversal risk is what durability is really about. Drivers include physical loss (wildfire, tillage, pyrolyzed-carbon oxidation in high-temperature soils), economic reversal (a landowner clears a stand when timber prices rise), regulatory reversal (a policy incentive lapses), and measurement error dressed up as reversal. Registries manage this in two ways. The first is buffer pools: a fraction of issued credits is withheld into a shared insurance reserve that is drawn on when a project reverses. The second is ton-year accounting, which discounts a short-lived tonne against a long-lived one — an approach that is intellectually clean but disliked by many buyers because it converts a physical claim into an accounting equivalence with contested parameters.
The life-cycle literature complicates the picture further. Terlouw et al. (2021) show in their review of CDR life-cycle assessments that upstream emissions, energy source, and end-of-life fate can materially shrink the net durable tonne, especially for pathways where the “removal” and the “storage” are decoupled in time and space. For biochar specifically, Lehmann et al. (2021) note that mean residence times vary by more than an order of magnitude depending on feedstock and production conditions — a headline “hundreds to thousands of years” figure hides a distribution that matters for pricing.
How it plays out across pathways
The concept bites differently depending on where the carbon ends up.
For solid-sorbent Direct Air Capture with basalt mineralization, operators like Climeworks and Heirloom sell against a >10,000-year claim, which is why offtake prices sit in the high hundreds of dollars per tonne. The durability argument depends on the storage partner (mineralization vs. saline aquifer) more than on the capture technology; a CarbonCapture Inc. or Skytree tonne is only as durable as the geology it lands in.
For bioenergy with carbon capture and storage (BECCS), the durability claim is again geologic — Carbon America is building around Class VI wells in Colorado, and Svante’s acquired Carbon Alpha assets target the same storage class. Reverion, by contrast, captures biogenic CO₂ but the storage question is answered downstream.
For biochar, the durability claim is inherently probabilistic and site-specific. NetZero, Bio-Logical, Dark Earth Carbon, Alcom Carbon Markets, Aymium, Green Carbon, and Carbon Remove all pyrolyze different feedstocks under different conditions, and registries apply a decay factor (Puro.earth’s methodology, for example, discounts the tonne based on H:C ratio). The agronomic meta-analysis by Schmidt et al. (2021) is often used to defend century-plus persistence, but field measurement over that horizon does not exist.
For afforestation and soil carbon, the durability claim is the weakest and buffer pools are largest. Klim and Grow Indigo operate in regulatory and biophysical environments where a 30- to 100-year commitment is the ceiling. Buyers who need “like-for-like” durability against fossil emissions increasingly treat these as avoidance-adjacent rather than as removals for net-zero claims.
The hard parts
Three questions remain genuinely open.
First, there is no agreed exchange rate between a 100-year tonne and a 10,000-year tonne. Ton-year accounting, buffer-pool discounts, and outright exclusion of short-duration credits from net-zero use are all live positions among Science Based Targets initiative, ICVCM, and individual buyer frameworks.
Second, monitoring, reporting, and verification (MRV) horizons do not match durability claims. A biochar project sells a 1,000-year tonne after a few years of field sampling; a mineralization project sells a 10,000-year tonne after a monitoring window measured in decades at best. The gap is bridged by models, and the models are not independently validated at horizon.
Third, buffer pools are actuarially untested. If correlated reversals — a bad fire decade across a portfolio-heavy region — hit before pools are capitalized to steady state, insolvency is possible, and the loss falls on buyers who already retired the credits. Durability, in other words, is not just a physical property of the stored carbon; it is a property of the institution standing behind the claim.