How long does biochar store carbon? Ask a soil scientist, a carbon registry and a coal petrologist and you get three different answers: centuries, at least 200 years, and more than 1,000 years. All three are describing the same black solid. They disagree because each one measures something different and counts over a different horizon.
This article lays out the three horizons, the 2025 dispute between them, and what actually sets the number for a specific batch of char. We design gasification and waste-to-energy plants that produce biochar alongside power and heat, so for us the question is never abstract. It is: which fraction of our char will a registry credit, and for how many years.
The Short Answer: Centuries, for the Stable Fraction
Biochar is biomass carbon that has been heated without oxygen until most of it is rearranged into condensed aromatic rings. A 2025 review in Communications Earth and Environment puts the result plainly: biochar has a mean residence time in soil of hundreds to thousands of years, and the factors that control it are feedstock type, the highest pyrolysis temperature and how long the material is held there.
For scale, compare the material it came from. Bulk soil organic matter has a mean residence time that averages 50 years across studies, and even lignin is 90 percent mineralised within a year of being left on a field. Charring moves the carbon from the "years to decades" bin into the "centuries" bin.
The most authoritative single statement on the question came in November 2025, when fourteen of the field's leading researchers, including Lehmann, Woolf, Cowie and Schmidt, published a policy commentary in GCB Bioenergy. It states that soil-applied biochar has mean carbon residence times exceeding 1,000 years, that no carbonaceous material is completely inert, and that policy should estimate persistence on the order of centuries rather than millennia. Hold both halves of that sentence. The material lasts a very long time, and the honest claim is still centuries.
Why There Are Three Numbers
100 years: the incubation decay models
The 100-year figure comes from the way persistence was first measured. Researchers mix biochar into soil, incubate it for a year or more, track the CO2 that comes off, and fit a decay curve. The reference model is Woolf and colleagues, 2021. Across their dataset, 63 to 82 percent of biochar carbon remains unmineralised after 100 years at the global mean cropland temperature of 14.9 C. They fitted at least a two-pool model, a fast fraction and a slow one, and recommended a 100-year horizon for the same reason global warming potentials use one: short enough to measure progress, long enough to matter.
That model became the market's arithmetic. Puro.earth's earlier biochar methodology required a molar H/Corg ratio below 0.7, treated material below 0.2 as hardly degradable, and set the 100-year permanence factor at 1.04 minus 0.64 times H/Corg for a 14.9 C soil. Plug in a typical industrial char at H/Corg 0.4 and the formula returns 78 percent of the carbon still in place at year 100. At H/Corg 0.2 it returns 91 percent. The ratio of hydrogen to organic carbon is a proxy for how condensed the aromatic structure is: fewer hydrogen atoms per carbon, more ring-to-ring bonds, less for microbes to attack.
200 years and more: what registries now guarantee
In 2025 the registries moved. Puro's Edition 2025 methodology, approved on 12 June 2025, raised the durability claim to several centuries, well beyond 200 years, issued as CORC200+, and notes that at least 200 years is considered permanent removal under EU legislation. Its own account of the decision is careful: the registry went to several centuries but not yet to millennia, and random reflectance measurements are encouraged in reporting but do not yet affect quantification. The same post records that IPCC working groups discussed a 1,000-year basis and concluded the methods were not suited to extrapolate that far.
Isometric's protocol lands in the same place by a different route. It cites Amazonian Terra Preta soils radiocarbon-dated to up to 7,000 years, requires storage of more than 200 years, and credits only the highly durable fraction of the organic carbon. Verra's VM0044 v1.2 has been active since 27 June 2025 for soil and non-soil uses including biochar in concrete. In the EU, the Carbon Removals and Carbon Farming Regulation's biochar methodology credits a quantifiable fraction of stable biochar expected to store carbon for at least several centuries. We unpack what those rules demand of a plant in our guide to CRCF permanent carbon removal credits.
Notice the pattern. Every registry credits a fraction, not the whole char, and every registry now writes "centuries" rather than "100 years". For how that fraction becomes a tradable unit, see how biochar carbon credits work.
1,000 years and more: the inertinite benchmark
The third number comes from coal petrology rather than soil science. Sanei and colleagues at Aarhus proposed measuring char under a microscope the way geologists rank coal. Their 2024 paper sets a random reflectance of 2 percent as the inertinite benchmark, finds that 76 percent of the commercial biochar samples studied sit entirely above it, and models roughly 100 million years for half the carbon in such a char to degrade even in an oxidising environment at 30 C.
The Ithaka Institute's position is similar in spirit. For char made above 550 C with an H:C ratio below 0.4, it holds that 75 percent of the carbon persists for more than 1,000 years regardless of soil and climate, while the remaining 25 percent is semi-persistent with a residence time of 50 to 100 years.
The argument against the 100-year models is methodological. A 2025 evaluation in the journal Biochar notes that total carbon loss in incubations runs from 15 percent for char made at 350 C down to 1 percent at 800 C, which means the experiments have mostly measured the labile fraction and tell us little about how the condensed fraction behaves over centuries.
Here is where the November 2025 commentary pushes back. Its authors write that biochar and inertinite-ranked fossil coals cannot be equated, because inertinite spent millions of years buried under pressure and sealed in minerals, while char made in a modern reactor is a different material in a living soil. Declaring any soil-applied char 100 percent persistent over millennia, they argue, is inconsistent with current evidence. The reflectance method measures relative stability well. Whether it measures absolute permanence is the open question, and that is exactly why Puro reports it but does not yet pay for it.
What Sets the Number for Your Batch
The three horizons are the market's problem. For a plant, the number is set by physics you control.
Temperature and residence time, read through H/Corg
Higher peak temperature and longer residence time push more carbon into the condensed pool. Isometric's protocol puts the threshold where stability climbs at pyrolysis above 500 C. Woolf's dataset groups chars into low, medium and high classes at 350 to 450 C, 450 to 600 C and above 600 C, and finds H/Corg a better predictor than temperature alone, because carbonisation also depends on how long the material is held.
Temperature is not free. A 2025 analysis in Environmental Research Letters estimates that raising pyrolysis temperature by 100 C increases the energy requirement of production by about 20 percent. The same paper quantifies how blunt the standard proxy is: the most common single-parameter model explains only 32 percent of the variance in durability outcomes, the H/C 0.7 threshold corresponds to a predicted 80 percent retained after 100 years, and the authors' own re-analysis exceeded the Woolf estimate in 76 of 83 observations. Their conclusion is that today's standards are conservative and probably undervalue biochar, which is a defensible place for a credit to sit.
Feedstock and ash: the gasification caveat
This is the part the SERP never mentions, and it matters most to anyone considering a gasification route. Woolf's compilation shows that the organic carbon content of char varies with feedstock and process from as low as 7 percent for gasification of biosolids to 79 percent for wood pyrolysed above 600 C. Gasification chars are made at higher temperatures than pyrolysis chars, so they carry low H/Corg ratios that correlate with high persistence, and the paper notes that H/Corg is the right proxy precisely for ash-rich material of this kind.
Read those two facts together. A sewage-sludge gasification char can be extremely durable per unit of carbon and still contain little carbon per tonne, because most of its mass is ash. Durability and quantity are separate questions, and a credit needs both. Our public five-step process treats biochar plus verified CO2 offsets as one output stream of a plant whose main job is dispatchable power and heat, which is why the carbon fraction of the char has to be measured and modelled before anyone promises offset volumes. The carbon balance upstream of the reactor is covered in how biochar removes CO2 from the air, and the sludge-specific case in biochar from sewage sludge.
Feedstock consistency decides how repeatable the H/Corg result is. We have argued that thermal pre-treatment is non-negotiable for moisture reduction, energy density and homogenisation. A wet, uneven feed produces uneven carbonisation, and uneven carbonisation forces a lab to sample more often and a registry to assume the worst batch. The thresholds a batch is tested against are collected in biochar quality standards.
Soil temperature, end use and the 25 percent that is not permanent
Where the char ends up changes the credited number. Every decay model is temperature adjusted. In Puro's 100-year regression the coefficients shift from 1.13 and minus 0.46 at a 5 C soil to 0.98 and minus 0.66 at 25 C. For a char at H/Corg 0.4 that is 95 percent retained in a cold northern soil and 72 percent in a tropical one. The same reactor output earns a different factor in Saxony than in Lagos, and a developer in a warm climate should budget for that before signing an offtake.
The 2025 commentary adds that persistence depends on soil mineralogy and biological activity as well, and that aging is not only decay: interaction with minerals and aggregation with soil particles also stabilise the char over time. Non-soil uses sidestep the soil question. Concrete and other building materials are eligible under VM0044 and under Isometric's materials module, and they remove most of the climate and biology variables from the model.
Finally, be honest about the labile pool. Whichever camp is right about the condensed fraction, every source above agrees that a share of fresh char, roughly the low tens of percent for mid-temperature material, is gone within decades. Methodologies already exclude it. Marketing copy often does not.
What This Means if You Build or Buy
Three rules follow from the evidence.
- Test H/Corg on every batch, and record the temperature history that produced it. The ratio is what every registry quantifies on, and the 0.7 cut-off that Puro, VM0044 and Isometric all reference is a hard eligibility line, not a target.
- Pick the horizon your buyer needs, not the biggest one on offer. A corporate buyer matching fossil emissions wants the 200-year-plus claim with a conservative decay deduction. A research-grade claim of millennia, backed by reflectance data, is worth collecting now because registries have said they may upgrade later, but it is not yet a credit.
- Separate durability from volume. Ask how much organic carbon is in a tonne of the char, not only how stable that carbon is. For gasification char from ash-rich feedstock this is the number that decides project economics.
If you are on the buying side, the same logic becomes a due-diligence list, which we set out in whether biochar carbon credits are reliable.
So how long does biochar store carbon? For the stable fraction of a well-made char: centuries, with credible evidence pointing to far longer. That is the claim the science, the registries and the EU all now share, and it is the one a plant can prove, batch by batch.
FAQ
Does biochar break down in soil?
Yes, slowly and unevenly. A small labile pool is lost within years to decades. The condensed aromatic pool persists for centuries or longer. Field and laboratory studies show measurable but small mineralisation for every char type studied, which is why no methodology credits 100 percent of the carbon.
Does biochar decompose faster in hot climates?
Yes. Decay models are adjusted for mean annual soil temperature, and the permanence coefficients fall between a 5 C and a 25 C soil. The same char therefore earns a lower 100-year permanence factor in a tropical field than in a northern one, and projects in warm regions should model the local value rather than the global 14.9 C default.
Is biochar carbon storage counted as permanent in the EU?
Yes, for the stable fraction. The CRCF biochar methodology treats it as permanent carbon removal on the basis that the stable fraction is expected to store carbon for at least several centuries, and Puro notes that a durability of at least 200 years is considered permanent under EU legislation.
How is biochar durability measured in a lab?
The standard method is elemental analysis of hydrogen and organic carbon to compute the molar H/Corg ratio, which feeds a temperature-adjusted decay regression. The emerging method is random reflectance microscopy, which compares the char to the inertinite found in coal. Puro's Edition 2025 encourages reporting reflectance but does not yet use it for quantification.