What Route 6 is
R6 issues verified ex-post avoided-emission tonnes from two activity classes. REDD+ reduces emissions by stopping at-risk deforestation or degradation in a defined area: instead of cutting trees that would otherwise have been cut, the project keeps them standing. IFM (Improved Forest Management) reduces emissions or increases stocks by changing harvest practice: longer rotations, reduced-impact logging, selective harvest, or conservation set-asides on a working forest.
Both activity classes are reductions (avoided emissions), not removals (active sequestration). The tonne is the difference between what would have been emitted under the baseline and what is emitted under the project.
R6 operates only on the CCP-eligible methodology family. Project-scale REDD+ runs on Verra VM0048 (CCP-approved). Jurisdictional REDD+ runs on Verra JNR v4.1 (CCP-approved) or ART TREES v2.0 (CCP-approved). IFM runs on Verra VM0045 with the most recent methodology corrections applied. The locked accounting plan at S02 records the methodology source and version on-chain; any subsequent restatement follows the documented re-issuance SOP, never ad-hoc.
The 2024 market filter
REDD+ went through a hard 18 months across 2023-24. Investigative journalism in early 2023 raised serious questions about the integrity of legacy project baselines, particularly the use of dynamic baselines that allowed projects to claim more avoided deforestation than the historical evidence supported. Buyers paused. Volumes fell. The Ecosystem Marketplace tracking shows REDD+ volume down approximately 52% in 2024 with average prices at approximately $6.37 per tonne (down 5.5% year-on-year); transaction counts compressed; weak projects exited or restated.
The market response was not to abandon REDD+ but to filter it. ICVCM approved three REDD+ or jurisdictional methodologies for its CCP label across 2024: ART TREES v2.0, Verra VM0048, and Verra JNR v4.1. The assessed pipeline of credits under these new-generation rules is on the order of 400 Mt. The new-gen methodologies replace dynamic baselines with allocated deforestation risk maps that constrain the baseline geographically and require updates on a defined schedule. Activity-shifting leakage and market leakage are explicitly budgeted.
IFM moved in the opposite direction across the same period: transaction volume tripled while pricing firmed to approximately $15-16 per tonne across sources. The reasons are partly structural (IFM has a clearer practice-vs-baseline story than REDD+, particularly for North American legacy projects on CAR and ACR ecosystems) and partly methodology (VM0045 corrections addressed several of the methodology-level concerns that had previously haircut prices).
R6 is built for the filtered market. The protocol does not operate on legacy methodologies; the gate at S01 routes pre-CCP projects to other lanes or back to their source registries. The CCP-aligned slice is what 2025 buyers pay for.
How R6 works
After the additionality gate passes (tenure and FPIC checked, intervention enforceability confirmed, risk and leakage feasibility scored, 'protected on paper' areas explicitly screened out), R6 locks the accounting plan at S02. The plan specifies: the methodology source and version (one of Verra VM0048 for project-scale REDD+, Verra JNR v4.1 for jurisdictional, ART TREES v2.0 for jurisdictional, or Verra VM0045 for IFM); the baseline framework, including allocated deforestation risk maps where the methodology requires them; the leakage budget covering activity-shifting and market leakage; the uncertainty deductions; the permanence buffer (typically 10-20% for forest projects).
Measurement at S03 runs through the GRO evidence chain. Remote sensing (optical plus radar deforestation detection at registered cadence) covers area-level change. Field plot measurements verify carbon stock per hectare. For IFM, harvest ledger access is registered so the practice-vs-baseline calculation can be verified. GRO tokens mint per stratum and interval to the project wallet, non-transferable, consumed at issuance.
At S04, any overlapping nested REDD+ credits, biodiversity offsets, conservation easements, or jurisdictional payments for the same area are retired or immobilized in their source registry. The Issue and Retire IDs are anchored on-chain. The protocol does not mint R6 tonnes until this stage closes.
At S05, the vintage batch dossier is assembled (deforestation detection records, plot measurements, harvest ledgers for IFM, accounting plan, EAC retirements). An independent accredited verifier from the panel reviews and signs on-chain. On a passing opinion, S06 mints the vintage batch; the backing GRO is consumed-to-carbon. Sale and stablecoin settlement at S07 follow the same pattern as other routes.
The seven-stage diagram below shows the full path.
What qualifies (and what doesn't)
Typical passes: project-scale REDD+ with real deforestation pressure and enforceable interventions (community patrols, access control, alternative livelihoods, demonstrated tenure); jurisdictional or nested programmes aligning with ART TREES v2.0 or Verra JNR v4.1, where the host-country baseline is set by national-inventory methods and updates on a defined schedule; IFM projects in North America on legacy CAR or ACR ecosystems with documented harvest-rotation changes; emerging IFM projects under VM0045 with national-inventory baselines.
Typical fails: 'protected on paper' areas where no real deforestation pressure exists; projects on legacy dynamic-baseline methodologies that have not been updated to the CCP-aligned versions; projects with contested tenure or insufficient FPIC documentation; projects that cannot demonstrate the intervention enforceability (paper park designations without on-the-ground enforcement).
The gate is binary. Projects that fail are routed to Route 5 (ARR removals, where the carbon claim is sequestration rather than avoided loss), Route 8 (soils or biochar), or energy Routes 1 or 2 so the developer still earns.
Baseline integrity: the change after 2024
The most visible weakness in legacy REDD+ was baseline construction. Older methodologies allowed developers to project deforestation forward from historical trends in a way that, in retrospect, over-estimated the counter-factual. When investigative journalists, academic researchers, and integrity bodies compared issued credits to satellite-observed deforestation, the discrepancies were significant for some legacy projects.
The new-generation methodologies (VM0048, JNR v4.1, ART TREES v2.0) address this in three ways. First, baselines use allocated deforestation risk maps that constrain the geographic distribution of expected loss based on observable predictors (proximity to roads, slope, population pressure) rather than free-form developer projections. Second, baseline updates run on a defined schedule (typically every 6 years for jurisdictional approaches) so the counter-factual stays anchored to recent observation. Third, leakage budgets are explicit: activity-shifting (deforestation pressure moves to a neighbouring area when this area is protected) and market leakage (the broader market for the displaced wood adjusts) are quantified and deducted.
R6's gate at S01 requires the project to operate on a CCP-aligned methodology. The locked plan at S02 records the methodology version and the specific baseline framework. The retirement record at S07 carries that lineage, so the buyer's audit can trace the integrity of the baseline back to its source. The 2023-era criticisms apply to projects R6 does not accept.
Stacking with other routes
R6 stacks cleanly with other carbon routes when they cover different areas or different units of benefit. R5 (ARR removals on different planted areas), R7 (blue carbon on coastal land), R8 (soils or biochar on separate land) all coexist with R6 on the same project portfolio. Energy routes (R1 or R2 on separate energy assets) and Route 3 (community pool participation, which uses pool assets) are also compatible.
What R6 cannot stack with on the same area is another instrument claiming the same avoided-deforestation tonnes. Any overlapping nested REDD+ credit, biodiversity offset, or jurisdictional payment must be retired or immobilized at S04 before R6 will issue tonnes from that land. The retirement is anchored on-chain; the One-Claim Ledger reverts if anyone tries to bypass it.
Where it stands
R6 is built on the same protocol-level architecture as R5 and the other carbon routes. The R6-specific dependencies are:
Methodology version tracking. Verra VM0048 for REDD+ project-scale, Verra JNR v4.1 for jurisdictional, ART TREES v2.0 for jurisdictional, Verra VM0045 for IFM. The locked plan at S02 records the specific version; updates are governed by the re-issuance SOP.
Jurisdictional integration. For nested or jurisdictional approaches, the host-country reporting infrastructure (national inventories, ART TREES enrolment, Verra JNR programmes) needs to be integrated as a data source. The first-mover jurisdictional programmes (Guyana under ART TREES is the most-cited example) demonstrate the issuance pathway is operational.
Allocated deforestation risk map data. Where the methodology requires an ADRM, the protocol integrates the relevant geospatial data source as part of the locked accounting plan at S02.
Registry bridges. Verra VCS, ART, and the related voluntary registries for retirement anchoring at S04 and for export at S06 where buyers require registration in those registries.
R6 readiness tracks the broader market's settling on the new-generation methodology family as the procurement baseline. For the protocol-level architecture R6 depends on, see Proof-of-Verification, One-Claim Ledger, Attestor Registry, and the sibling carbon route pages: R5 Reforestation, R7 Blue Carbon, and R8 Soils & Biochar.
