Policy-gated federated learning for global climate intelligence.
Sovereign Mohawk Climate Global enables nation-sovereign and sensor-sovereign AI for carbon tracking, extreme weather prediction, biodiversity monitoring, sea-level risk analytics, and net-zero compliance without centralizing raw environmental data.
Sovereign Mohawk Climate Global: every sensor, satellite, and nation keeps full data sovereignty while training planetary-scale climate AI without ever sharing raw environmental data.
Core repositories:
- Sovereign-Mohawk-Proto
- Sovereign_Map_Federated_Learning
- Sovereign_Mohawk_Oncology_Global
- Sovereign_Mohawk_Agriculture_Global
- Sovereign_Mohawk_Climate_Global
- Sovereign_Mohawk_SupplyChain_Global
Canonical architecture and shared execution plans:
- docs/ECOSYSTEM_ARCHITECTURE.md
- docs/UNIFIED_PHASE2_MILESTONES.md
- docs/GOVERNANCE_PR_ONLY.md
- docs/ACTION_PINNING_POLICY.md
- docs/PRODUCTION_DATA_PLANE_PLAN.md
- Proto: runtime, zk-SNARK verification, PQC, and BFT trust controls.
- Map: streaming aggregation profile, high-scale orchestration, tokenomics telemetry.
- Oncology: policy-gated wrapper and compliance workflow pattern.
- Agriculture: geospatial dashboard + operational map interaction baseline.
Progression:
Proto -> Map -> Oncology -> Agriculture -> Climate (specialized global climate sibling).
- Signed model updates only.
- Regional policy checks and poisoned-update rejection.
- Differential privacy and secure aggregation in the climate wrapper path.
- Region drill-down by IPCC zones, ocean basins, and biome clusters.
- Climate layers: carbon flux, extreme-weather convergence, biodiversity indices, sea-level pressure, and adaptation risk.
- FL pipeline, node health, compliance scorecards, and tokenomics telemetry.
- Mapping support for Paris Agreement, NDC tracking, EU ETS, and CBAM-adjacent reporting controls.
- DPIA-style templates for environmental and satellite-linked data flows.
- Regional policy examples (for example, PQC-only weight acceptance).
- Safe prompt workflows for climate scenario planning and adaptation policy analysis.
- STRIDE baseline for sensor spoofing, greenwashing manipulation, and data integrity attacks.
index.html: climate dashboard entry.app.js: climate map, pipeline, and risk visualization logic.styles.css: UI style layer.manifesto.html: climate edition manifesto.flower_security_wrapper/: climate policy wrapper templates.mock_api/: local climate endpoint mock server.docs/: roadmap, integration, threat model, compliance evidence, beta release.scripts/: validation and artifact tooling.examples/: satellite + buoy workflow examples..github/workflows/: CI, security, and pages deploy workflows.
cd Sovereign_Mohawk_Climate_Global
python3 -m http.server 8080Open: http://localhost:8080/
docker compose -f docker-compose.mock.yml up- Dashboard: http://localhost:8080/
- Mock API: http://localhost:8088/healthz
Default endpoint targets:
/api/climate/stats/api/climate/regions/api/climate/pipeline/api/climate/risk/api/climate/health
You can override endpoints by setting window.CLIMATE_DASHBOARD_CONFIG before loading app.js.
<script>
window.CLIMATE_DASHBOARD_CONFIG = {
statsUrl: "https://your-gateway.example.com/climate/stats",
regionsUrl: "https://your-gateway.example.com/climate/regions",
pipelineUrl: "https://your-gateway.example.com/climate/pipeline",
riskUrl: "https://your-gateway.example.com/climate/risk",
healthUrl: "https://your-gateway.example.com/climate/health",
timeoutMs: 5000
};
</script>See docs/PHASE2_INTEGRATION.md for payload contracts.
python3 scripts/validate_dashboard_contract.py
python3 -m unittest -q tests/test_contract_validator.pyClimate repository scaffold complete with CI/security baseline and local mock integration.