CCUS Policy Hub

Global intelligence infrastructure for carbon capture, utilization, and storage

[English] | 简体中文 | Live Demo

---

Overview

CCUS Policy Hub is an open-source research and data platform for comparing CCUS policy frameworks, facility deployment, project-record capacity, and governance maturity across countries and regions.

It connects three layers that are usually studied separately:

• Policy and regulation: incentives, legal frameworks, permitting, liability, MRV, and cross-border rules.
• Facility and project records: status, type, sector, region, capacity, location, hub, partners, and key dates.
• Governance and deployment analytics: country benchmarking, regional project pipelines, committed capacity, and data-quality evidence.

The platform is designed for researchers, policymakers, market analysts, investors, and AI agents that need a structured and auditable grounding source rather than a collection of disconnected webpages.

Dataset Snapshot

Dataset	Current snapshot
Policies	130
Facility records	1,110
Country governance profiles	66
Facility dataset version	2026-Q2
Policy review status	Active
Last checked	2026-06-24

Counts describe records in the published dataset. A facility record is not always equivalent to one unique real-world project, especially where capture, transport, storage, and full-chain components are represented separately.

What the Platform Helps Answer

• How do CCUS incentives, permitting systems, storage rights, liability rules, and MRV arrangements differ across jurisdictions?
• Where are operational, under-construction, and planned facility records concentrated?
• How much publicly recorded capacity is still planned, and how much has reached construction or operation?
• Which regions combine stronger governance capability with greater deployment scale?

---

Core Capabilities

1. Global Policy Access Console

Navigate an interactive world map to review country and regional policy frameworks, market-access conditions, incentive mechanisms, legal weight, and policy status.

• Bilingual policy content in English and Chinese.
• Structured policy categories and analysis dimensions.
• Build-time policy counts and regional coverage indicators.
• Direct access to policy details, sources, dates, and related facilities.

2. Global Facility Intelligence

Explore the published facility dataset through maps, filters, searchable cards, and project-level pages.

• Filter by country, region, sector, status, and facility type.
• Review project-record capacity, coordinates, precision, hub, operator, partners, and key dates.
• Distinguish operational, under-construction, planned, cancelled, suspended, and decommissioned records.
• Inspect the data-quality and provenance fields available for each record.

3. CCUS Project-Record Capacity Growth

The homepage time-series workspace shows how recorded CCUS capacity accumulates across the project pipeline.

• Adjustable time window from historical records to announced future projects.
• Type filters for Capture, Transport, T&S, Storage, Full chain, and CCU.
• Two consistent series:
  • Pipeline = Planned + Under construction + Operational.
  • Committed = Under construction + Operational.
• The active rightmost year drives the headline project-record capacity figure.
• Record year priority follows operation → FID → announcement.

Project-record capacity is an aggregation of published records. It should not be interpreted as net deliverable capture capacity or net geological storage capacity.

4. 2026 Global Project Landscape

A current dataset snapshot complements the time series with project maturity and regional concentration.

• Projects by Status: operational, under-construction, and planned records within the active portfolio.
• Pipeline vs Committed by Region: nested regional bars make the subset relationship explicit.
• Toggle between recorded capacity and project count.
• Automatically generated insights for:
  • Global committed share.
  • Largest pipeline region.
  • Largest committed region.
  • Capacity-data coverage.

Cancelled, suspended, and decommissioned records are disclosed separately rather than mixed into the active portfolio.

5. Governance Benchmarking & Deployment Comparison

The comparison workspace goes beyond a single readiness score and exposes the structure behind national governance capability.

• Side-by-side comparison of regulatory pillars, including pore-space rights, liability transfer, financial assurance, permitting, CO₂ legal status, and cross-border rules.
• Governance benchmarking across selected countries.
• Governance–Deployment Matrix connecting institutional capability with project-record deployment scale.
• Global maturity views for identifying governance leaders, deployment leaders, and implementation gaps.
• Method notes and evidence boundaries presented alongside the visual analysis.

6. Trust, Quality, and AI-Readable Data

The platform treats data governance as a product capability, not a back-office process.

• SQLite single source of truth.
• Bilingual export and parity checks.
• Blocking deep-audit checks in CI.
• Public quality dashboard and dataset-version metadata.
• Static JSON endpoints and JSON Schemas.
• llms.txt and llms-full.txt documentation for AI agents.

---

Methodology and Limitations

Project-record methodology

• Capacity is derived from the maintained estimated value when available, otherwise from the announced range according to the repository's capacity rules.
• Records without a usable capacity remain in project-count views but contribute 0 Mtpa to capacity aggregations.
• Summing across capture, transport, storage, and full-chain records may include different components of the same wider value chain.
• Project-record totals are therefore analytical indicators, not net physical system capacity.

Snapshot methodology

• The 2026 Global Project Landscape describes the current state of the 2026 dataset snapshot.
• It does not mean that all displayed projects were announced, built, or commissioned in 2026.
• Current status fields should not be used as a historical reconstruction of what a project's status was in an earlier year.

Facility-policy relationship confidence

Facility records are linked to relevant jurisdiction-level policy frameworks to provide governance context. These links should not be interpreted as verified project-specific legal applicability unless direct evidence is available.

Relationship level	Indicative confidence	Meaning
Country	Low	Facility and policy share the same jurisdiction
Sector	Medium	Facility sector matches the documented policy scope
Direct evidence	High	A source explicitly connects the policy and facility

See the Facility-Policy Relationship Model for the full design.

Geographic precision

Coordinates may represent exact, state-level, or country-level locations. Country-level coordinates are visualization anchors and must not be interpreted as verified project sites.

---

Data Architecture

The platform uses a data-governed static-site architecture: the website is static, while its source data is maintained through an auditable database workflow.

graph LR
    A[(SQLite SSOT)] --> B[Deep audit and governance checks]
    B --> C[Bilingual Markdown exports]
    B --> D[Public JSON and quality metrics]
    C --> E{Astro build}
    D --> E
    E --> F[Static HTML and interactive analytics]
    E --> G[Pagefind search index]
    F --> H[GitHub Pages]
    G --> H

Loading

• SSOT: agent/ccus-ai-agent/db/ccus_master.sqlite.
• Published content layer: src/content/policies/{en,zh} and src/content/facilities/{en,zh}.
• Public data layer: public/data/*.json and public/data/schemas/.
• Frontend: Astro 5, Tailwind CSS, Chart.js, Leaflet, and Pagefind.
• Deployment: static output published through GitHub Pages.

---

AI-Readable Data Interface

Static endpoints are available for AI agents, researchers, and scripts:

Endpoint	Description
/data/manifest.json	Dataset, schema, and documentation index
/data/policies.json	130 policy records with analysis data
/data/facilities.json	1,110 facility records with capacity and location data
/data/countries.json	66 governance profiles with regulatory pillars
/data/quality.json	Quality metrics and audit status
/data/dataset-versions.json	Dataset version and last-checked metadata
/data/schemas/	JSON Schemas for validation
/llms.txt	Concise AI-agent documentation
/llms-full.txt	Extended data and field documentation

Recommended usage:

1. Fetch /data/manifest.json to discover the available resources.
2. Validate records against the published schemas.
3. Check /data/quality.json and version metadata before drawing conclusions.
4. Preserve methodology notes when using capacity or relationship fields.

---

Development and Data Governance

Local setup

git clone https://github.com/liuh886/ccus-policy-hub.git
cd ccus-policy-hub
pnpm install
pnpm dev

Validation and production build

pnpm lint
pnpm test
pnpm astro check
pnpm manage:db:audit:deep
pnpm manage:db:quality:export
pnpm manage:db:data:export
pnpm build

Governed content workflow

SQLite edit
  → deep audit
  → bilingual Markdown export
  → quality and public-data generation
  → Astro production build
  → human review and deployment

Governance rules:

• Make authoritative content changes in SQLite, not directly in generated Markdown.
• Treat src/content/**/*.md as the publication layer.
• Reverse synchronization from Markdown to SQLite is migration-only and requires explicit acknowledgement.
• Preserve provenance, audit dates, bilingual parity, and source URLs when updating records.

---

AI-Assisted Development

The project uses AI as a research and engineering copilot within a human-reviewed workflow.

GPT-5.6

• Helps define analytical methodology, metric boundaries, and acceptance criteria.
• Reviews governance and deployment logic for conceptual consistency.
• Designs bilingual information architecture, product copy, and visual QA requirements.
• Translates research judgments into implementation tasks, tests, and PR review criteria.

Codex and repository agents

• Inspect the current GitHub repository and implement focused changes.
• Create and update pull requests, tests, and documentation.
• Diagnose Astro, TypeScript, data-pipeline, and CI failures.
• Hand off environment-specific validation to local Codex workflows when needed.

graph LR
    A[Research question] --> B[GPT-5.6 methodology and acceptance criteria]
    B --> C[Codex or GitHub implementation]
    C --> D[Automated tests and CI]
    D --> E[Human review]
    E --> F[Merge and release]

Loading

AI-generated changes are not accepted solely because they compile. Data definitions, evidence, and release decisions remain subject to maintainer review.

---

Citation

If you use CCUS Policy Hub in research, policy analysis, or derivative datasets, please cite the archived Zenodo record:

Liu, Z. CCUS Policy Hub: Global Intelligence Infrastructure for Carbon Capture, Utilization, and Storage. Zenodo. https://doi.org/10.5281/zenodo.21110615

@software{liu_ccus_policy_hub,
  author = {Liu, Zhihao},
  title = {CCUS Policy Hub: Global Intelligence Infrastructure for Carbon Capture, Utilization, and Storage},
  publisher = {Zenodo},
  doi = {10.5281/zenodo.21110615},
  url = {https://doi.org/10.5281/zenodo.21110615}
}

License

Code is released under the MIT License. Dataset records retain their original source attribution and should be reused with the methodology and provenance limitations described above.

---

☕ Support CCUS Policy Hub on Ko-fi

_{Built for the global climate and CCUS research community by liuh886}