Carbon Monitor Scraper

Extract real daily CO2-emissions estimates from Carbon Monitor's public global dataset. Filter the data by country or region, sector, and inclusive date range, then receive normalized JSON records ready for analysis, dashboards, research pipelines, spreadsheets, or data warehouses.

Carbon Monitor is an international scientific initiative that publishes regularly updated daily estimates of fossil-fuel and cement-production CO2 emissions. Its website states that the data are freely available under a fair-use open-data policy and encourages users to cite the Carbon Monitor research. This Actor is an independent community integration and is not affiliated with or endorsed by Carbon Monitor or its supporting institutions.

What it extracts

Every result contains:

• stable recordId built from country, ISO date, and sector;
• country or region exactly as published;
• normalized date in YYYY-MM-DD form;
• Carbon Monitor sector;
• numeric emissionsMtCO2PerDay;
• explicit unit of MtCO2/day;
• source release date and source filename when provided by the download response;
• direct dataset URL, source-page URL, and scrape timestamp.

The supported sectors are Domestic Aviation, Ground Transport, Industry, International Aviation, Power, and Residential. Leave the sector as all to return every sector matching the other filters.

Input example

{
  "country": "Austria",
  "sector": "Power",
  "dateFrom": "2025-01-01",
  "dateTo": "2025-01-31",
  "maxResults": 100
}

country is an exact case-insensitive filter. Leave it empty to include all published countries and regions. dateFrom and dateTo are optional inclusive ISO dates. maxResults caps unique emitted rows; maxRowsScanned, maxDownloadMbytes, maxRunMillis, and maxRetries provide operational safety limits. An optional Apify Proxy configuration is accepted, though direct access normally works.

Data quality and responsible operation

The Actor downloads the official CSV once per run, validates the HTTP status and CSV content type, enforces a response-size guard, and parses rows as a stream. It validates dates and numeric emissions, removes duplicate country-date-sector identities, and omits unavailable optional metadata. It never generates placeholder records, mock values, or hardcoded success.

Downloads use a consistent browser identity, bounded exponential retries, and one request at a time. Runs have an Actor-side deadline below five minutes.

Important limitations

Carbon Monitor is a living scientific dataset. Values can change as new activity data arrive, errors are corrected, or methods are revised. These are estimates, not direct measurements at individual facilities and not official national greenhouse-gas inventories. Sector and geographic coverage follow the source dataset. Results should be interpreted with Carbon Monitor's methods and cited research, and users remain responsible for analytical conclusions based on the data.

The source CSV may grow over time. If it exceeds the configured size or scan guard, the Actor fails clearly so callers can raise the limit deliberately rather than receiving a silently truncated dataset.