Executive summary — ESG Map is an interactive, dark-themed world-map atlas that recolours by national sustainability indicators and opens a decision-grade detail panel for any country: its energy mix, disclosure posture, environmental indicators, and a 25-year historical trend. It is wired to real, dated open data from Our World in Data and the World Bank, augmented with an hourly near-real-time grid-operator layer, and ships as a fully static site for analysts, researchers, and educators who need provenance-first sustainability intelligence they can cite.
- Real data, never invented. Headline metrics are ingested from Our World in Data and the World Bank by a reproducible build script; missing values render grey / "no data" and are emitted as explicit
NA— never fabricated, never zero. - Seven map layers — renewable power share, grid carbon intensity, CO₂ per capita, air quality (PM2.5), forest cover, climate-risk exposure, and a documented composite sustainability score.
- D3 world atlas —
geoMercator, full-bleed, latitude-clipped to remove polar smear, with zoom/pan, hover tooltips, and click-to-select fly-to that centres on a country's largest landmass. - Near-real-time grid layer — hourly live renewable-share and grid-carbon readings from national grid operators (UK, US, ~25 European countries), labelled distinctly from annual values and never blended into them.
- Built to be cited — per-country and dataset-level BibTeX / RIS / APA export, a
CITATION.cff, deep-link permalinks that encode full app state, a static per-resource JSON "API", and a JSON Schema codebook. - Reproducible & verifiable — content hash + version + git SHA +
build-manifest.jsonstamp every edition; the whole dataset can be regenerated with one command and the site deploys with no network access. - Accessible & offline — screen-reader data-table fallback for the map, keyboard focus rings, colourblind-safe (cividis) and greyscale palettes, a print stylesheet, and an installable PWA with a service worker.
ESG Map turns scattered open sustainability datasets into a single, navigable atlas. Pick one of seven map layers and the world recolours by that metric; click any country to open a detail panel with its energy mix, disclosure status, environmental indicators, and a 2000→latest historical trend. A time slider scrubs the renewable and carbon layers across history.
Supporting views extend the map into an analytical workbench:
- Rankings — a sortable table across any metric, with explicit accounting of how many indicators each country was scored on.
- Regional trends — an overlay tracing metric trajectories by world region.
- Compare — a two-country side-by-side view.
- Methodology — every source documented with its retrieval date, the composite-score formula, and one-click citation export.
- Score Lab — re-weight the composite with live sliders and watch the ranking move, without ever overwriting the published score.
- Explore — scatter any two indicators, coloured by region, with explicit missing-data accounting; plus a Spearman correlation matrix among sub-indicators and convergent-validity checks against the Yale EPI.
The full app state — layer, year, selected country, pins, view, and palette — lives in the URL hash, so any view is a shareable, citable, embeddable link.
Per Monarch Castle doctrine — evidence before assertion. Every value in ESG Map is traceable to a named source and stamped with the date it was retrieved, and that provenance is rendered live in the in-app Methodology view. Countries with no upstream value for a metric are emitted as null and shown as "no data" — figures are never invented.
Ingested annual metrics (scripts/build-data.mjs):
| Metric(s) | Source | License |
|---|---|---|
| Renewable share, grid carbon intensity, electricity use per capita, electricity mix, 2000→latest history | Our World in Data — Energy (Ember + Energy Institute Statistical Review) | CC BY 4.0 |
| CO₂ per capita | Our World in Data — CO₂ & GHG (Global Carbon Project) | CC BY 4.0 |
| Forest cover (% land) | World Bank AG.LND.FRST.ZS (FAO FRA) |
CC BY 4.0 |
| Air quality (PM2.5, µg/m³) | World Bank EN.ATM.PM25.MC.M3 (WHO / IHME GBD) |
CC BY 4.0 |
Near-real-time overlay (scripts/build-live.mjs, refreshed hourly via cron) — instantaneous generation-mix snapshots from national grid operators, labelled distinctly from annual figures:
| Source | Coverage | Token |
|---|---|---|
| National Energy System Operator (UK) | United Kingdom | none |
| U.S. EIA Grid Monitor | United States | EIA_KEY (free, instant) |
| ENTSO-E Transparency Platform | ~25 European countries | ENTSOE_TOKEN (free, email request) |
Curated, dated layer — slow-moving categorical fields (Paris status, 2030 NDC target, net-zero year, IFRS S1/S2 adoption, EV sales share, ND-GAIN climate-risk exposure) maintained as versioned JSON with retrieval stamps, compiled from UNFCCC, Climate Watch (WRI), the Net Zero Tracker, the IFRS Foundation, and the IEA.
Derived metric — the 0–100 sustainability score is a transparent, weighted composite (clean-power 30%, grid carbon 25%, CO₂ per capita 20%, disclosure readiness 15%, climate risk 10%), renormalised over whichever sub-scores are available per country. It is documented, not borrowed from a third-party index; the exact weights live in scripts/build-data.mjs and are surfaced in the Methodology view.
The committed dataset lets the app build and deploy without network access; CI attempts a fresh ingest on each deploy and falls back to the committed copy if a feed is down. Full provenance, licensing, renewable-classification rules, and emission factors are documented in DATA_SOURCES.md; the maintenance model and responsible-use posture are in GOVERNANCE.md.
OWID Energy CSV ─┐
OWID CO₂ CSV ─┤
World Bank API ─┼─► build-data.mjs ─► src/data/countries.json ─► app
curated layer ─┘ (ISO-3 join, derive score, stamp dates)
| Concern | Choice |
|---|---|
| Framework | React 18 + TypeScript 5 |
| Build | Vite 5 (static output, relative base for Pages) |
| Map / scales | d3-geo, d3-zoom, d3-scale, d3-selection, d3-transition, d3-interpolate, d3-ease, topojson-client |
| Geometry | Natural Earth 110m (world-atlas), self-hosted under public/geo/ |
| Fonts | IBM Plex Sans + IBM Plex Mono |
| Data | Our World in Data, World Bank, national grid operators, curated policy layer |
| Automation | GitHub Actions — build + hourly live-data refresh (cron) |
| Hosting | GitHub Pages (fully static, no backend) |
npm install
npm run dev # http://localhost:5173Live site: https://monarchcastletech.github.io/esgmap/
| Command | What it does |
|---|---|
npm run dev |
Vite dev server |
npm run build |
Type-check (tsc -b) + production build to dist/ |
npm run preview |
Serve the production build locally |
npm run typecheck |
Type-check only |
npm run build:data |
Re-ingest the annual open-data feeds → src/data/countries.json |
npm run build:live |
Refresh the near-real-time overlay → src/data/live.json (UK works with no token) |
npm run build:meta |
Regenerate the curated policy layer from the design handoff |
Add repository secrets under Settings → Secrets and variables → Actions:
EIA_KEY— register instantly at https://www.eia.gov/opendata/register.phpENTSOE_TOKEN— emailtransparency@entsoe.eu(subject "RESTful API access"), then generate the token in your account settings.
npm run build:live # UK only (no token)
EIA_KEY=... ENTSOE_TOKEN=... npm run build:live # full coverageThe repo ships .github/workflows/deploy.yml. Push to master, set Settings → Pages → Source = GitHub Actions, and the workflow installs, optionally refreshes data, builds, and publishes dist/. vite.config.ts uses base: "./" and public/.nojekyll is included, so the same build works at both a user page and a project page with no config change.
A product of Energy Intelligence · Monarch Castle Technologies — an operating company of Monarch Castle Holdings. Sister companies: Monarch Castle Technologies · Strategic Data Company of Ankara
Application code: MIT. Upstream datasets retain their own licenses — see DATA_SOURCES.md and LICENSE. © 2026 Monarch Castle Holdings · Ankara, Türkiye.