Proposed dataset: FIFA World Cup carbon footprint per edition (2010–2034)

[amautadev]

Description

A compact, maintained reference table of the greenhouse-gas footprint of each FIFA men's World Cup, one row per tournament, with the air-travel share broken out. Covers the four most recent past editions (2010–2022) plus the three projected expanded tournaments (2026, 2030, 2034). The 2026 edition — the first 48-team, 104-match, three-nation finals — is projected to be the most polluting ever (~9.0 Mt CO₂e), ~86% of it from air travel.

Why it belongs in core data

• Topical & high-demand: sustainability of mega-events is a live debate; the 2026 World Cup kicks off June 2026 and "most polluting ever" is already circulating in coverage.
• Nothing open & maintained exists: the figures are scattered across an advocacy report (PDF) and FIFA's per-tournament sustainability audits — there is no single, tidy, openly-licensed CSV.
• Small but reusable: a clean per-edition series others can chart, extend, or join to host-country / attendance data.

Potential data source

• Primary projection: Scientists for Global Responsibility / New Weather Institute, FIFA's Climate Blind Spot: The Men's World Cup in a Warming World (2025) — https://www.sgr.org.uk/publications/fifa-s-climate-blind-spot-men-s-world-cup-warming-world (PDF Table ES-1). Gives 2026 = 9.02 Mt, air travel 7.72 Mt, and a recalculated 2010–22 average (4.71 Mt) on a consistent fuller scope; 2030 = 6.1 Mt, 2034 = 8.6 Mt.
• Per-edition history: FIFA host sustainability audits, via Play the Game — https://www.playthegame.org/themes/sport-and-climate-change/understanding-sports-carbon-emissions/ (South Africa 2010 2.75, Brazil 2014 2.27, Russia 2018 2.16, Qatar 2022 3.63 Mt — FIFA self-reported scope).

Format (proposed)

emissions_per_edition.csv — one row per tournament:

column	type	notes
year	int	2010–2034
host	string	host nation(s)
co2e_mt	float	total emissions, million tonnes CO₂e
basis	string	FIFA self-report | SGR projection (flags the methodology per row)
source	string	citation
source_url	string	link

Plus emissions_reference.csv (SGR recalculated 2010–22 average; 2026 air-travel vs other split).

License

SGR/New Weather: not explicitly CC — figures cited with attribution (factual data, standard journalistic/academic reuse). FIFA audits: public reporting, attribution. Underlying numbers are factual, not copyrightable.

⚠️ Methodology note (important)

There are two non-comparable scopes: FIFA's self-reported per-edition audits (2.2–3.6 Mt) use a narrower boundary than SGR's fuller-scope projection. The honest comparison is 2026 (9.0, SGR scope) vs SGR's recalculated 2010–22 average (4.71, same scope) = +92% — not 2026 vs the raw FIFA self-reports. Any packaged dataset must carry this flag, not blend the two.

Open questions for discussion

1. Scope: should a core dataset present only the SGR consistent series (2010–2034, like-for-like), or carry both scopes side by side with the flag? I lean toward both-with-flag.
2. Projections: include 2030/2034 (projections, not actuals) in the same table, or split actual vs projected into separate resources?
3. Maintenance: what's the best primary to watch for the actual 2026 post-tournament footprint (vs the current projection), and who could own updates?
4. Granularity: worth adding a category split (air travel / stadiums / accommodation) where the source provides it, even if uneven across editions?