🌍 Carbon Intensity API

Real-time and historical electricity carbon intensity (gCO₂eq/kWh) plus prices, load, and cross-border flows for 45+ countries and grid regions.

🔓 Zero API keys. Zero registration. Zero tokens. All data sources are public:

• 🇪🇺 Fraunhofer Energy-Charts (Europe — 30 countries via the German research institute Fraunhofer ISE)
• 🇺🇸 EIA Hourly Electric Grid Monitor (US — bulk CSVs, public domain)
• 🇬🇧 UK NESO Carbon Intensity API (UK — National Energy System Operator)
• 🇧🇷 ONS Dados Abertos (Brazil — Operador Nacional do Sistema Elétrico)

Built with both production-based and consumption-based (flow-traced) calculations following IPCC AR6 emission factors and the ECON-PowerCI methodology (Nature 2025).

---

🎯 What this delivers

For each hour and country/region:

Field	Description
carbon_intensity_gco2_kwh	gCO₂eq per kWh of electricity
renewable_pct	% from wind, solar, hydro, biomass, geothermal
fossil_free_pct	renewable + nuclear
total_generation_mwh	total electricity generated
fossil_only_intensity_gco2_kwh	CI of fossil portion only
generation_mix_mw	breakdown by source: wind, solar, gas, coal, etc
load_mw	total system demand (consumption)
net_load_mw	load minus variable renewables (= load to be served by dispatchable sources)
day_ahead_price	wholesale spot market price
day_ahead_price_currency	EUR (Europe), USD (US), GBP (UK), BRL (BR)
day_ahead_price_zone	bidding zone code (e.g., "DE-LU", "FR", "IT-NORTH")
cross_border_flows_mw	list of imports/exports with neighbouring zones

Available in 4 methods (all calculated, choose which to return):

• production_lifecycle (default) — IPCC AR6 lifecycle factors, only domestic generation
• production_operational — combustion-only factors
• consumption_lifecycle — flow-traced through cross-border imports, lifecycle factors
• consumption_operational — flow-traced, operational factors

---

🌐 Coverage

Europe (30 countries via Fraunhofer Energy-Charts): AT, BE, BG, CH, CZ, DE, DK, EE, ES, FI, FR, GR, HR, HU, IE, IT, LT, LU, LV, ME, MK, NL, NO, PL, PT, RO, RS, SE, SI, SK

United Kingdom (via NESO): GB

United States (10 Balancing Authorities via EIA bulk CSV): US-CISO (California), US-ERCO (Texas), US-MISO, US-PJM, US-NYIS, US-ISNE, US-BPAT, US-SWPP, US-SOCO, US-FPL

Brazil (4 SIN submarkets via ONS): BR-SE (Sudeste/CO), BR-S (Sul), BR-NE (Nordeste), BR-N (Norte)

---

⚙️ Input

{
  "mode": "latest",
  "hours": 48,
  "method": "production_lifecycle",
  "countries": ["DE", "FR", "GB", "US-CISO", "BR-SE"]
}

That's it. Just run it.

Modes

• latest — last N hours (most recent data, default 48h)
• today — current UTC day so far
• yesterday — full previous UTC day
• full_refresh — entire history since 2015 (long run)

Optional skip flags

• skip_europe, skip_us, skip_uk, skip_brazil — disable specific data sources

---

📦 Output

Each hour × country yields one record:

{
  "country_code": "DE",
  "country_name": "Germany",
  "region": "europe",
  "source": "fraunhofer",
  "timestamp_utc": "2026-05-05T14:00:00+00:00",
  "method": "production_lifecycle",
  "carbon_intensity_gco2_kwh": 312.4,
  "renewable_pct": 47.2,
  "fossil_free_pct": 52.8,
  "total_generation_mwh": 58420.5,
  "fossil_only_intensity_gco2_kwh": 658.3,
  "generation_mix_mw": {
    "wind": 16850.2,
    "solar": 7106.4,
    "nuclear": 3268.5,
    "gas": 13088.4,
    "coal": 8358.2,
    "hydro": 3855.7,
    "biomass": 2400.0,
    "oil": 290.1
  },
  "load_mw": 60140.0,
  "net_load_mw": 29928.5,
  "day_ahead_price": 86.43,
  "day_ahead_price_currency": "EUR",
  "day_ahead_price_zone": "DE-LU",
  "cross_border_flows_mw": [
    { "counterparty": "FR", "flow_mw": 5200.0, "direction": "import" },
    { "counterparty": "PL", "flow_mw": 1100.0, "direction": "import" },
    { "counterparty": "AT", "flow_mw":  800.0, "direction": "export" }
  ]
}

---

🧮 Methodology

Production-based

CI_production = Σ(generation_i × emission_factor_i) / Σ(generation_i)

Per hour, sum each fuel's MW × its IPCC AR6 lifecycle factor (gCO₂eq/kWh), divide by total. Doesn't account for imports.

Consumption-based (flow-traced)

For interconnected grids, what matters is what you consume, not what you produce. We solve a linear system per hour:

(G_i + IM_i) · x_i = own_emissions_i + Σⱼ(flow_jᵢ × x_j)

where x_i is the consumption-based CI of country i, G_i is domestic generation, IM_i is total imports, and flow_jᵢ is the physical power flow from j into i.

Solved with numpy.linalg.solve per hour. Method follows ECON-PowerCI (Nature Scientific Data, 2025).

Emission factors (IPCC AR6 WG3, Annex III median values)

Source	Lifecycle (gCO₂eq/kWh)	Operational
Wind	11	0
Solar	48	0
Nuclear	12	0
Hydro	24	0
Biomass	230	90
Gas	490	370
Coal	820	760
Oil	650	580

---

🚀 Use cases

• ESG / Scope 2 carbon accounting — accurate hourly CI for corporate reporting
• Climate tech & green software — schedule loads when grid is cleanest
• Energy trading — backtesting and signal generation
• Academic research — replicable, IPCC-aligned dataset
• Grid-aware applications — compute, EV charging, smart home

---

🔄 Schedule

Apify scheduler can run this actor every hour to keep your dataset fresh. Output is overwritten/upserted by (country, timestamp, method) so reruns are idempotent.

---

📄 Data sources

All data sources are public and free, no authentication required:

• Fraunhofer Energy-Charts — operated by Fraunhofer ISE (German federal research institute), data aggregated from ENTSO-E. License: CC BY 4.0
• EIA Hourly Electric Grid Monitor — US Energy Information Administration. Public domain (US gov)
• NESO Carbon Intensity API — UK National Energy System Operator. License: CC BY 4.0
• ONS Dados Abertos — Operador Nacional do Sistema Elétrico (Brazil). License: CC BY

Methodology citations: IPCC AR6 (Working Group III, Annex III) + ECON-PowerCI (Nature Scientific Data 2025, doi:10.1038/s41597-025-05978-7).

Built by Jenko Systems.