Agroforestry carbon credits: what farmers earn, and how it is verified

A farmer plants trees on land that is already growing a crop. Years later, someone pays for the carbon those trees have pulled out of the air. Between those two moments sits a measurement problem, and it is the reason agroforestry carbon credits are harder to issue than almost any other nature-based credit.

How agroforestry carbon credits are generated

Agroforestry sits in the same family as afforestation, reforestation and revegetation — the removal class usually written as ARR — and it integrates trees into farming systems rather than replacing them. The trees build carbon stock above and below ground while the land keeps producing, and alongside the carbon they improve soil health, support biodiversity and add climate resilience. The credit is the tradable form of that stored carbon: a buyer offsets emissions, and the payment gives the farmer a direct reason to plant trees and, more importantly, to keep them.

An agroforestry system sequesters in the range of 5–15 tonnes of CO₂e per hectare per year. That is a wide band because it is doing a lot of work: species, planting density, tree age and site conditions all move a project within it, and they move it season by season rather than once at the start.

What actually sets the size of the payment

Two things decide what a project earns from agroforestry carbon credits: how many it can issue, and what each one sells for. We do not publish a price, and neither should anyone else in the abstract — it moves with the standard, the buyer, the vintage and the market. Any figure promising a farmer a fixed income per hectare before a single tree has been measured is a guess wearing a number.

Volume is the half a project can control, and it is not the same as carbon in the ground. It is carbon that can be proved to be in the ground. Sequestration a verifier will not accept earns nothing, which makes the field record as much a financial instrument as the trees are.

Several deductions sit between the two. Additionality has to be argued before anything is issued: the trees must not be ones that would have gone in anyway. Leakage subtracts the emissions a project pushes elsewhere, such as grazing or fuelwood collection displaced onto neighbouring land. Permanence is handled through a buffer pool — a share of every issuance withheld against the risk of the trees not surviving the crediting period, released only if they do. And the same tonne earns differently depending on its vintage and the standard that issued it.

None of those are field measurements, but every one of them is settled with field evidence. A buffer contribution is sized from a non-permanence risk assessment that a credible survival record can lower.

Why it is counted tree by tree, not per hectare

This is now written into the methodology itself. Verra’s ARR methodology VM0047, which replaced the older CDM methodologies AR-ACM0003 and AR-AMS0007, routes dispersed plantings that keep their existing land use — agroforestry, in other words — through a census-based approach requiring a complete census of planting units.

Most carbon methodologies work on plots. Agroforestry does not, because biomass — and therefore sequestered carbon — is estimated from the dimensions of individual trees through species-wise allometric equations. Species, age, height and girth (DBH) all feed that calculation, and all of them change every season.

The consequence is arithmetic. A project with a hundred thousand trees needs a hundred thousand accurate, located, photographed, re-visitable records, gathered by field teams usually working well beyond mobile coverage. Our companion piece on measuring agroforestry carbon tree by tree covers why that reshapes the whole survey.

What the Agroforestry dMRV app captures in the field

The Agroforestry dMRV app exists to make a hundred thousand of those records survivable. What it takes at each plant:

• Per-tree census — species, age, height and girth (DBH) logged against each individual plant and GPS-geotagged, not averaged across a block.
• Survival monitoring — each tree re-measured in later monitoring periods, so mortality and replanting are recorded rather than assumed.
• Location per plant — every tree carries its own coordinates rather than sitting somewhere inside a plot boundary, so a re-visit two seasons later finds the same tree.
• Automated and manual measurement — height and girth (DBH) can be captured either way, so a canopy or a light condition that defeats the automated route does not stop the survey.
• Photographic evidence — a photo and a location per plant, which is what gives a verifier something to check rather than something to trust.
• Full offline operation — no connection required, because no connection is the normal state of a plantation landscape.

It belongs to the same family as our other field applications, set out on the Apps page.

The dashboard, and why two-step verification protects the farmer

Collection is half the job. The monitoring dashboard is where field data becomes evidence:

• Live survey monitoring — progress and incoming records visible from the desktop while teams are still out.
• Validation of everything collected — every record from the mobile app is checked, not a sample of them.
• Two-step verification — a second review pass before a record goes forward, aimed at complete accuracy rather than at an acceptable error rate.
• Two-way communication — validators and ground teams resolve queries directly, so a correction happens while the team is still on site.
• Integrated analytics — survey effort is measured and optimised, which saves field time and cost.

Why this matters to a farmer and not only to a project manager is timing. A discrepancy caught during the survey costs a message. The same discrepancy caught at verification can cost a season — and the season it costs is the one the farmer was going to be paid for.

What a verifier will ask for

A verifier — formally the validation and verification body, the VVB — reviewing agroforestry carbon credits is not assessing whether agroforestry sequesters carbon; that is settled. They are assessing whether these trees exist, stand where the project says they stand, and measure what the project says they measure. That is a documentary question, and it is answered with per-tree records carrying coordinates and photographs — exportable as shapefile, KML or GeoJSON, and cross-checked against satellite canopy and NDVI change over the same parcels — not with summary tables.

Projects that lose credits usually lose them here rather than in the field biology. We have written separately about the data failures that cost carbon projects their credits. Programme data is handled under our ISO/IEC 27001:2022 certification, which starts to matter once a project is carrying the personal details of the farmers being paid.

What the farmer gets besides the credit

The carbon payment is additional income, not replacement income — the land is still farmed. Around it sit the effects that made agroforestry worth doing before carbon markets existed: better soil, more biodiversity, more resilience to a bad season, and tree cover that keeps growing in value while it keeps sequestering. The credit is the part that can be sold. It is rarely the largest part of the case.

Related reading

The Agroforestry dMRV app page covers the platform in detail. For how agroforestry sits alongside the other methodologies we measure, see DMRV methods.

• Measuring agroforestry carbon tree by tree
• The 5M approach to running a digital MRV programme
• The field and verification failures that cost credits