Back to Home
METHODOLOGY

VERIFICATION
STANDARDS

Our rigorous AI-powered verification methodology ensures every carbon credit on CarboCredit represents real, measurable, and additional environmental impact.

Version 2.1VCS CompatibleGold Standard AlignedICROA Endorsed

CORE VERIFICATION PRINCIPLES

Every project on CarboCredit must meet these five fundamental criteria to qualify for credit issuance.

01

Real

Emissions reductions or removals must have actually occurred

02

Measurable

Quantifiable using scientific methods and data

03

Additional

Would not have occurred without carbon finance

04

Permanent

Long-term storage with reversal mitigation

05

Verified

Independently validated by our AI systems

AI VERIFICATION METHODOLOGY

STEP 01

Multi-Spectral Satellite Analysis

Our AI models continuously analyze satellite imagery from multiple sources to monitor project areas.

Data Sources

  • Sentinel-2 (10m resolution, 5-day revisit)
  • Landsat 8/9 (30m resolution, 16-day revisit)
  • Planet Labs (3m daily imagery)
  • MODIS (250m daily thermal data)

Analysis Capabilities

  • NDVI vegetation health monitoring
  • Biomass estimation via SAR
  • Deforestation detection
  • Land use change tracking
STEP 02

IoT Sensor Integration

For renewable energy and industrial projects, we ingest real-time data directly from connected sensors.

Smart Meters

Real-time energy production and consumption

Weather Stations

Climate conditions affecting project performance

Soil Sensors

Carbon content and soil health monitoring

STEP 03

Machine Learning Processing

Our proprietary ML models analyze collected data to calculate carbon sequestration with high precision.

Model Architecture

Forest Carbon ModelResNet-152 + LSTM
Energy VerificationTransformer-XL
Anomaly DetectionIsolation Forest + AE
Ensemble Voting5-Model Consensus

Training Data

Historical Projects12,500+
Verified Data Points850M+
Ground Truth Samples45,000+
Continuous LearningActive
STEP 04

Multi-Stage Validation

Every verification passes through multiple validation stages before credit issuance.

Stage 1
Automated AI Check
Confidence: 95%
Stage 2
Cross-Reference Validation
Confidence: 97%
Stage 3
Anomaly Detection
Confidence: 99%
Stage 4
Consensus Verification
Confidence: 99.7%
STEP 05

Blockchain Recording

Verified credits are minted on-chain with full provenance data and IPFS metadata storage.

// On-chain verification record
{
  "credit_id": "cc_1a2b3c4d5e",
  "project_id": "proj_amazon_2026",
  "verification": {
    "timestamp": "2026-01-12T14:30:00Z",
    "ai_confidence": 0.987,
    "model_version": "v2.1.0",
    "data_hash": "ipfs://QmX7d9f...",
    "satellite_refs": ["S2_20260110", "L8_20260108"],
    "validator_nodes": 5,
    "consensus": "unanimous"
  },
  "carbon": {
    "tonnage": 12.5,
    "methodology": "VCS VM0007",
    "vintage": 2026
  }
}

PROJECT TYPE STANDARDS

🌲

Reforestation & Afforestation

VCS VM0007, VM0010
  • Minimum project area: 10 hectares
  • Native species requirement: 70%+
  • 35-year permanence commitment
  • Community benefit documentation
  • Quarterly satellite monitoring
  • Annual ground-truth verification

Renewable Energy

Gold Standard, VCS ACM0002
  • Grid-connected with smart metering
  • Real-time production monitoring
  • Baseline emissions documentation
  • Additionality demonstration
  • Monthly production verification
  • Annual performance review
🌊

Blue Carbon (Mangroves)

VCS VM0033
  • Coastal ecosystem mapping
  • Tidal zone classification
  • Soil carbon sampling
  • Biodiversity assessment
  • Monthly remote sensing
  • Bi-annual field verification
🛡️

Avoided Deforestation (REDD+)

VCS VM0015, VM0009
  • Historical deforestation analysis
  • Reference region establishment
  • Leakage zone monitoring
  • Community safeguards
  • Weekly deforestation alerts
  • Quarterly full verification

QUALITY ASSURANCE

Continuous Monitoring

Unlike traditional annual audits, our AI monitors all projects continuously, detecting issues in real-time.

Monitoring FrequencyReal-time
Alert Response< 24 hours
Full Re-verificationWeekly

Buffer Pool

A percentage of all credits are held in a buffer pool to cover any reversals or verification discrepancies.

Forest Projects20% buffer
Energy Projects5% buffer
High-Risk Zones30% buffer

Third-Party Audits

Our methodology and AI models are independently audited by leading verification bodies.

Methodology AuditAnnual
AI Model AuditQuarterly
Sample VerificationMonthly

CONFIDENCE SCORING

Every verification includes an AI confidence score that reflects the certainty of our assessment.

97-100%
Excellent
Highest certainty, multiple data sources confirm
93-96%
Good
High confidence, minor data limitations
90-92%
Acceptable
Meets threshold, additional monitoring applied
<90%
Pending
Requires additional verification or data

Note: Credits are only issued when verification confidence exceeds 90%. Projects scoring below this threshold enter a pending state and require additional data collection or manual review before credits can be issued.

TRUST IN EVERY CREDIT

Ready to explore our verified project registry?