What Route 4 is
R4 is the carbon lane for energy projects. The output is a verified, ex-post, vintage-batched avoided-emission tonne issued in the voluntary carbon market. The route is open only to projects that pass a strict additionality test, that operate in or shift into high-marginal-emission grids, and that have not already monetised the same kilowatt-hour as an energy attribute.
R4 is not a Verra, Gold Standard, or other registry-native programme. R4 is the protocol-level lane on EDMA that builds the tonne with audit-grade evidence and either bridges it to a recognised voluntary registry or settles directly with the buyer. R4 is also not an EU ETS or other compliance-market instrument; allowances under closed compliance schemes are a separate financial instrument that R4 does not produce.
R4 is for buyers who want measurable impact: corporates with beyond-value-chain commitments, net-zero pathways that require high-quality voluntary tonnes, or insetting programmes that need provenance from generation to retirement. The discipline that makes the tonne defensible is what limits R4's scope: the additionality gate filters out most clean-grid rooftop generation, which is correctly priced as an R1 or R2 attribute rather than a carbon offset.
The three things carbon projects fail on
Carbon projects in the voluntary market typically fail on one or more of three things, and the buyer's auditor sees all of them:
Additionality. Would the project have happened anyway? A rooftop solar system in a clean-grid OECD country with strong subsidies almost certainly would have. A new gas-displacement project in a coal-heavy grid that became viable specifically because of the EDMA-financed carbon revenue almost certainly would not. R4 tests this binary: if the answer is 'would have happened anyway', the producer is rerouted to R1 or R2 so they still earn from their generation, but no carbon tonne is issued.
Math. Are the avoided emissions computed correctly, with location-specific and time-specific factors? A project that uses annual average grid emissions to compute its avoided tonnes is mathematically weaker than a project that uses hourly marginal factors, because the actual emissions displaced depend on which generator was on the margin at the hour the clean kWh was injected. R4 enforces a hierarchy: hourly marginal where available, hourly residual-mix as fallback, monthly or annual residual-mix with added conservativeness as last resort.
Double counting. Did the same kWh also fund an energy attribute? A REC, a GO, an I-REC, or a granular certificate already represents the renewable claim for that kWh; issuing a carbon tonne from the same kWh while the attribute is still live is double counting. R4 mandates retirement or immobilization of the overlapping attribute before the carbon tonne can be queued.
R4 fixes all three by design. The gate, the EF hierarchy, and the mandatory EAC retirement are not optional features; they are blocking stages in the protocol flow.
The additionality gate
The gate tests three things and is binary in its verdict.
Causality. Did EDMA finance or incentive enable the project, or materially enlarge it? Examples that pass: new renewable capacity built with EDMA-arranged financing in markets where the carbon revenue is the difference between a viable and a non-viable project; capacity upgrades to existing installations triggered specifically by R4 enrolment; private-wire or captive-use renewable installations that displace fossil self-generation or grid imports. Examples that fail: legacy residential rooftop PV in clean-grid OECD markets with strong subsidies; new utility-scale solar in markets where wholesale renewable economics are positive without carbon revenue.
Emissionality. Is the grid emission factor high enough for avoided emissions to be material? High-marginal-EF grids (coal-dominant, gas-peaking-heavy) qualify; very low marginal-EF grids (deep nuclear or deep hydro) typically do not. Storage projects qualify if the storage dispatch demonstrably shifts clean energy into high-marginal-EF hours, with the no-storage baseline and charge-source proof documented.
Rights. Does the project owner have the legal rights to monetise the avoided emissions? Some jurisdictions reserve avoided-emission claims to the host country (for example, under Paris Agreement Article 6 with Corresponding Adjustments). R4 does not issue tonnes where the rights are unclear or contested.
The gate produces a fast verdict (1 to 7 days in most cases) with published criteria. Producers that fail are routed to R1 or R2 with explicit guidance on the alternative monetisation path. R4 is never the default; it is the strictest of the four energy routes by design.
How R4 works
After the additionality gate passes, R4 locks the emission factor hierarchy for the project's node and region. The hierarchy preference order is hourly marginal EF where available; hourly residual-mix as fallback; monthly or annual residual-mix with conservativeness as last resort. T&D losses are attached. A conservative buffer (typically 5%) is applied. The EF source and version are recorded on-chain. If the provider later restates factors, a documented re-issuance SOP applies; settled batches are never repriced.
The producer's hourly generation (or storage dispatch) is metered per the R2 base flow: meter signs at source, edge software canonicalises, METER_OP + GRID + AUDITOR quorum signs the evidence, the PoV Gate evaluates, ETTs mint with a 'carbon-eligible' flag carrying the locked EF version. For storage, additional evidence is required: charge-source proof showing the energy was charged from the qualifying renewable source, round-trip efficiency, no-storage baseline.
Before any carbon can be queued, the protocol requires retirement or immobilization of any energy attributes (REC, GO, I-REC, granular certificate) that exist for the same kilowatt-hours. The external registry Issue and Retire IDs are anchored on-chain at this stage. The corresponding ETTs are flagged 'carbon-bound' and locked from any other retirement.
Carbon issuance is ex-post. At the end of each issuance period (typically quarterly), the vintage batch is assembled: the metering records, the EF snapshots, the EAC retirement IDs, and any storage dispatch proofs. An independent accredited verifier reviews the dossier. On a signed verification opinion, the CARBON_TONNE contract mints the vintage batch and the backing ETTs are marked consumed-to-carbon and rendered permanently non-transferable.
Sale and settlement run on either a forward or spot model. The seller receives stablecoin to their Earnings Vault. A 4% total fee applies (2% seller in $EDM at claim + 2% buyer with an $EDM-discount option), of which 50% burns in $EDM until the 100 M circulating supply target.
The seven-stage diagram below shows the full path.
What qualifies (and what doesn't)
Typical passes: new PV or storage in coal-heavy grids that EDMA financing made viable; capacity upgrades triggered by R4 incentives; private-wire or captive-use projects displacing fossil self-generation; storage-enabled dispatch into peak-emission hours with audited dispatch evidence.
Typical fails: legacy residential PV in clean grids with strong subsidies (would have happened anyway); any kilowatt-hour still monetised as a REC, GO, or I-REC unless those attributes are retired or immobilized first; projects using annual average grid factors when hourly marginal factors are available for that region.
Most rooftop solar in OECD clean grids will not qualify for R4 and will be routed to R1 or R2. This is not a failure of the route; it is the additionality discipline doing what it is supposed to do. R4 is the strictest route by design because the tonne it produces has to survive a voluntary-market audit on additionality, emissionality, and no-double-counting all three.
The protocol docs include directional ranges (illustrative, not guarantees) for what producers earn under R4 in different grid types: coal-heavy grids with new PV produce in the order of 10 to 12 avoided tonnes per year per 10 kWp installed at carbon prices of $30 to $60 per tonne; mixed grids with storage shift produce in the order of 2 tonnes per 5 MWh shifted at the same price range; clean-grid OECD installations typically do not qualify at all. Each project sees a specific min-likely-max range based on its locked EF, eligible MWh, losses, buffer, and prevailing carbon price.
Stacking with other routes
R4 alongside R1 or R2 is allowed only by splitting ETTs across the routes. Some ETTs from a producer's generation can fund R1 or R2 attributes; other ETTs from the same generator can fund R4 carbon tonnes. The ETTs cannot fund both. If a producer wants to move an ETT that previously funded a R1 attribute to R4, the R1 attribute must first be retired in its home registry and the IDs anchored on-chain; only then can R4 issue a tonne from that ETT.
R4 alongside R3 is straightforward because R3 operates on pool assets that the participant does not host. The participant's own R4 issuance is independent of any R3 slot they subscribe to.
Flex revenue (R2's demand response side) does not consume ETTs and is independent of R4. A producer that earns DR payments for verified load reduction can also earn R4 tonnes from their own generation, with the relevant ETTs split between the routes.
Regional and regulatory context
Paris Agreement Article 6. For government-authorised credits with Corresponding Adjustments (ITMOs), R4 can operate a separate lane. The process is slower and paperwork-heavy; pricing can be premium. ITMOs are not R4's default product.
EU ETS and other compliance allowances. Not in scope for R4. Compliance allowances are financial instruments in closed compliance registries; R4 issues voluntary-market tonnes with no double counting against energy attributes.
Voluntary carbon market. Corporate demand concentrates where the additionality story is strongest: new builds in dirty grids, storage that visibly shifts into dirty hours with auditable provenance. R4 is positioned for that demand.
Jurisdiction by jurisdiction. The host-country rights confirmation in the additionality gate ensures R4 does not issue where the rights are reserved to a national authority or contested. Geofencing applies until the relevant rights are clear.
Where it stands
R4 is built on the R2 base flow (hourly metering with the METER_OP + GRID + AUDITOR three-role quorum) plus four R4-specific layers:
Additionality gate contract. Smart contract evaluating causality, emissionality, and rights with published criteria. Fast verdict with reroute to R1 or R2 on fail.
Accounting contract. Locks the EF hierarchy, T&D losses, and conservative buffer at issuance for each vintage batch. Re-issuance SOP applies on EF provider restatements.
EAC retirement bridge. Mandatory retirement or immobilization of overlapping energy attributes before any carbon can be queued. External registry Issue and Retire IDs are anchored on-chain.
Ex-post verifier integration. A panel of accredited verifiers (the recognised voluntary-market verification firms) reviews the vintage batch dossier off-chain and signs on-chain with their attestor identity. Batches are held until findings are resolved.
R4 readiness depends on three signals: the panel of accredited verifiers being on-boarded as AUDITOR-class entities; the EAC retirement bridges being live for the major energy attribute registries (I-REC, GO via AIB, M-RETS, NEPOOL-GIS, PJM-GATS); and the voluntary carbon market settling on its post-VCMI methodologies for hourly-marginal-EF accounting and storage-shift additionality.
For the protocol-level architecture R4 depends on, see Proof-of-Verification, One-Claim Ledger, Attestor Registry, and the sibling route pages: R1 Compliance credits, R2 Granular & Flex, and R3 Community-Share.
