Skip to content
r4

Methodology & integrity

Show your working.

For a policy audience, provenance is the argument. Here is exactly how the numbers were derived, how the corpus was processed, and what we are still unsure about.

How the corpus becomes claims

The site is built on a small, deterministic data pipeline — not hand-typed numbers:

  1. Corpus. A curated bibliography of 24 sources, each with type, primary/secondary flag, pillars, status and key claims, lives in one structured file.
  2. Ingest. A script fetches each open source, extracts readable text, and chunks it (~280-word windows). Paywalled or JavaScript-gated pages fall back to the curated abstract, so every source still has at least one resolvable excerpt — nothing is silently dropped.
  3. Insights. Each source’s key claims are flattened into the evidence dataset and matched back to the chunk they came from, so the explorer can show provenance.
  4. Gate. A check step fails the build if any claim does not resolve to a real source. A claim with no source does not ship.

Where each headline figure comes from

FigureHow it was derived
Methane is ~80× more potent than CO₂ (20-yr) IPCC AR6 WG1 Table 7.15: fossil-CH₄ GWP20 ≈ 82.5 with carbon-climate feedback. We round to “~80×”.
52.4% sustainable energy mix Cambridge CCAF 2025 study: 9.8% nuclear + 42.6% renewables, sample covering ~48% of mining, public methodology.
76 GW flexible load at 0.25% curtailment Duke Nicholas Institute “Rethinking Load Growth” (2025) — institutional working paper, not yet peer-reviewed; labelled as such.
~2,187 t CH₄ mitigated Rudd, Batten et al. (2024), J. Cleaner Production — Monte-Carlo model of a representative 1.14 MW LFGTE+mining facility. Paywalled; figure drawn from abstract/accepted text.
99.9% combustion efficiency Crusoe operator field data; paired with peer-reviewed flare-gas economics (Risks 2022). Operator claim, flagged as secondary.
1,312 MW / ~59% AU curtailment AEMO Quarterly Energy Dynamics, Q4 2025 — official NEM dataset.
~0.4% of global electricity IEA “Energy and AI” (2025) framing of crypto’s share of global electricity.

What we’re open about

These flags are carried straight from the corpus and surfaced rather than buried:

  • Working papers (SSRN, institutional) are labelled 'working-paper', not peer-reviewed.
  • Paywalled publishers (Elsevier/Nature/ACS) are flagged 'paywalled'; only abstract / accepted-manuscript text is quoted.
  • The Batten landfill paper is titled '…model' on SSRN but published as '…framework' in J. Cleaner Production — same study; the JCP DOI is canonical.
  • No 2023 Bratspies-authored Greenpeace 'Change the Code' report exists; that attribution is excluded. Her relevant work is the 2018 Michigan Tech Law Review article.
  • No AU-specific peer-reviewed landfill/coal-seam + mining study was found; the transferable model is Rudd et al. 2024 plus Clean Energy Regulator / DCCEEW inventory data.

Working papers (2)

Labelled as working papers, not peer-reviewed journal articles: Norris 2025, Carter 2023.

Paywalled, abstract-only (6)

We could not ingest full text and quote only the abstract / accepted manuscript: Rudd 2024, Lal 2023, Vries 2021, Masanet 2019, Stoll 2019, Mora 2018.

Known gaps

There is no AU-specific peer-reviewed landfill or coal-seam + mining study in this corpus; we use the transferable Rudd et al. (2024) model and flag that Australian abatement potential should be quantified against Clean Energy Regulator / DCCEEW inventory data before being asserted as a domestic figure.

Corrections

If a figure here is wrong or a stronger source exists, we want to know. The corpus is designed to be edited: change the source, and the claim, the evidence explorer and the figures update together.