Quality Scoring
Every asset submitted to NCRB receives a programmatic quality score computed off-chain by the Oracle API and then stored on-chain via QualityAssessment. The score drives the asset's rating band, which in turn informs pricing on the marketplace.
Overview
The scoring system is designed to be:
- Consistent — same framework across all 10 asset types
- Aligned with industry standards — benchmarked against Sylvera, BeZero, ICVCM, and equivalent methodologies per asset class
- Transparent — all dimension weights and band thresholds are readable on-chain via
GET /api/quality-profiles/:assetType - Institutional-grade — mapping directly to rating conventions (AAA → Not Eligible) familiar to ESG procurement teams
Scoring Dimensions
Every asset type uses the same six dimensions, though the specific criteria within each dimension vary by asset class:
| Dimension | Weight | What It Measures |
|---|---|---|
| Technical Quality | 25% | Asset-specific quality bands (Sylvera ratings, USDA land classes, ore grades, FSC certification, distinctiveness level) |
| Additionality | 20% | Evidence that the environmental benefit would not have occurred without the project (financial additionality, regulatory surplus) |
| Permanence | 20% | Long-term durability of the environmental benefit — reversibility risk, guarantee mechanisms, covenant length (100+ years preferred) |
| Certification Level | 15% | Rigour of the issuing registry and whether the asset has CCP-Approved or equivalent status |
| Social Impact | 12% | Co-benefit certifications: CCB Gold, SDG alignment, Indigenous engagement, community employment |
| Vintage / Condition | 8% | Asset age and data freshness — older vintages and stale monitoring data attract discounts |
Total: 100%
Composite Score Calculation
Score = (TechnicalQuality × 0.25)
+ (Additionality × 0.20)
+ (Permanence × 0.20)
+ (CertificationLvl × 0.15)
+ (SocialImpact × 0.12)
+ (VintageCondition × 0.08)
Each dimension is scored 0–100 by the Oracle API based on the parameters submitted with the certificate. The composite score is also 0–100.
Rating Bands
The composite score maps to one of five rating bands:
| Band | Score Range | Description |
|---|---|---|
| AAA | 85–100 | Highest integrity — premium pricing |
| AA | 75–84 | High quality |
| A | 65–74 | Good quality |
| BBB | 50–64 | Minimum eligible — compliance-grade pricing |
| Not Eligible | < 50 | Below threshold — cannot be tokenized on NCRB |
The minimum score for tokenization is 50 (BBB).
Sub-Type Scoring Bonuses
When a certificate's subType field (stored in assetMetadata) maps to a known sub-type with affectsScoring: true, bonus points are added to the raw dimension scores before the composite is calculated. Available sub-types are served by GET /api/registry-standards/sub-types?asset_type_code={id}.
Carbon credits (assetType 0):
| Sub-type | Additionality bonus | Co-benefits bonus |
|---|---|---|
compliance | +3 | — |
removal_dac | +5 | +3 |
removal_beccs | +4 | +2 |
removal_enhanced_weathering | +3 | — |
Other asset types:
| Asset | Sub-type | Dimension | Bonus |
|---|---|---|---|
| Plastic | ocean_bound | Co-benefits | +4 |
| Nitrogen | nitrous_oxide | Additionality | +3 |
| Agricultural | regenerative | Additionality | +3 |
| Phosphorus | watershed | Co-benefits | +3 |
| Water | environmental_flow | Co-benefits | +4 |
| Mining | reclaimed | Co-benefits | +4 |
| Forestry | natural_forest, agroforestry | Co-benefits | +3 |
| Biodiversity | wetland | Co-benefits | +4 |
| Biodiversity | coastal_marine | Co-benefits | +3 |
Paris Agreement Article 6.4 Bonuses
For carbon credits (assetType 0), three fields in assetMetadata trigger additional additionality bonuses. These are set during asset registration and stored as part of the on-chain assetMetadata JSON.
| Field | Value | Additionality bonus |
|---|---|---|
creditMechanism | article_6_4 | +5 |
creditMechanism | corsia | +3 |
correspondingAdjustment | applied | +2 |
correspondingAdjustment + loaStatus | applied + provided | +1 (stacked on top of the CA bonus) |
creditMechanism values:
| Value | Description |
|---|---|
vcm | Voluntary Carbon Market (no bonus) |
article_6_4 | UN Paris Agreement Article 6.4 mechanism |
corsia | ICAO CORSIA aviation offsetting |
cdm_transition | CDM credits transitioning to Article 6.4 |
correspondingAdjustment — applied means the host country has formally transferred the mitigation outcome to the buyer's NDC, preventing double-counting. This is the key integrity requirement for Article 6.4 compliance.
loaStatus — provided means a Letter of Authorisation from the host country government is on file.
SDG Co-Benefits Bonus
For all asset types, each UN SDG goal listed in sdgGoals adds +1 to the co-benefits dimension score, capped at +4. This bonus stacks on top of any sub-type co-benefits bonus.
| SDG goals aligned | Co-benefits bonus |
|---|---|
| 0 | +0 |
| 1 | +1 |
| 2 | +2 |
| 3 | +3 |
| 4 or more | +4 (cap) |
SDG goals are stored as an integer array in assetMetadata.sdgGoals (e.g. [13, 14, 15]).
Social Impact Premium
The Social Impact dimension (12%) reflects co-benefits that command price premiums from institutional buyers. Corporate buyers including Microsoft, Google, and Meta explicitly require social impact verification in procurement criteria.
| Co-benefit | Typical Price Premium |
|---|---|
| CCB Gold certification | +30–50% |
| SDG alignment (3+ goals) | +20–40% |
| Indigenous community engagement | +40–100% |
| Community employment > 50 FTE | +20–30% |
These premiums are applied on top of the base band price range.
Asset-Specific Technical Quality Criteria
The Technical Quality dimension (25%) is the most asset-specific. Here is how it is applied per asset class:
| Asset Class | Technical Quality Criteria |
|---|---|
| Carbon Credits | Sylvera / BeZero rating (minimum BBB); vintage within 5 years; third-party verification |
| Plastic Credits | Collection type (ocean-bound > landfill > recycling); geographic provenance tracking |
| Nitrogen Credits | Delivery ratio (0–1.0) — fraction of upstream reduction reaching the regulated water body |
| Phosphorus Credits | Delivery ratio (0–1.0) — phosphorus binds to sediment, transport dynamics differ from nitrogen |
| Agricultural Land | USDA Land Capability Classification: Class I (prime) → Class IV (limited); LESA score |
| Mining Rights | Reserve classification (NI 43-101 / JORC / S-K 1300); ore grade (g/t or equivalent) |
| Water Rights | Priority date seniority (senior > intermediate > junior); use type (municipal > industrial > agricultural) |
| Renewable Energy Credits | Source type; registry tier (APX, WREGIS, M-RETS); state compliance class (Class I > voluntary) |
| Forestry Rights | NHLA timber grade (FAS > No. 1 Common); FSC/SFI/ATFS certification level |
| Biodiversity Credits | UK Biodiversity Metric 4.0 distinctiveness (D6 Very High → D3 Medium minimum); NatureQuant index |
On-Chain Storage
After the Oracle API computes the score, it writes it to the QualityAssessment contract:
Oracle API computes score
│
▼
Oracle API calls QualityAssessment.recordAssessment(serialNumber, score, dimensions)
│
▼
Score stored on-chain — immutable and auditable
│
▼
ncrb-indexer reads QualityAssessment events
│
▼
Score available via GET /api/quality-profiles/score/:serialNumber
Reading the Methodology
The current dimension weights and band thresholds for any asset type can be queried at runtime:
# Quality profile (dimensions + weights)
GET /api/quality-profiles/0 # assetType 0 = Carbon Credits
# Rating bands (score ranges + price multipliers)
GET /api/quality-profiles/0/bands
# Full methodology rules
GET /api/quality/methodology
The weights stored on-chain in QualityAssessment are the authoritative source. The database-side registry standards (/api/registry-standards/quality-parameters/:registryId/:assetTypeId) provide the registry-specific parameter definitions used as inputs.
Example: Assessing a Carbon Credit
Quality assessment is triggered automatically when the Oracle processes a CertificateSubmitted event. The certificate's assetMetadata JSON (stored on-chain) provides the inputs.
Example assetMetadata for a high-scoring Article 6.4 carbon credit:
{
"methodology": "VM0015",
"vintage": 2023,
"subType": "removal_dac",
"creditMechanism": "article_6_4",
"correspondingAdjustment": "applied",
"loaStatus": "provided",
"sdgGoals": [7, 13, 14, 15],
"projectName": "Bangladesh Mangrove Reforestation",
"country": "BD"
}
Scoring result for this certificate:
| Dimension | Base score | Bonus | Final |
|---|---|---|---|
| Technical Quality | — | — | ~80 |
| Additionality | ~60 | +5 (Art. 6.4) +2 (CA) +1 (LoA) = +8 | ~68 |
| Co-benefits | ~60 | +3 (removal_dac sub-type) +4 (4 SDGs) = +7 | ~67 |
| Certification Level | — | — | ~75 |
| Social Impact | — | — | ~80 |
| Vintage/Condition | — | — | ~72 |
Bonus points are added to the raw dimension score (capped at 100 per dimension) before the weighted composite is calculated.