Methane Outliers

Operators burn methane for the same reason people throw away packing peanuts: they wanted the thing in the box, not the filler. In West Texas the prize is oil, the filler is associated gas. Gas needs a pipeline while oil just needs a truck. When drilling outruns midstream, or Waha prices go negative and buyers effectively charge you to take it, the arithmetic is brutal and boring: burn the cheap byproduct, keep selling the expensive one. Add completions flowback, plant turnarounds and the occasional Texas freeze, and a flare is really just the relief valve for a midstream system’s bad day.

Alberta solved the nowhere-to-put-it problem decades ago and got stuck with the subtler one: gas that leaks by design. Pneumatic controllers running on produced gas, tank vapors, dehydrator stills, heavy-oil casing gas too thin and remote to be worth a compressor: all quietly exhaled rather than dramatically burned. That distinction matters, because venting is 100% methane while a lit flare destroys about 98% of it. But aerial surveys keep finding flares that are unlit or burning poorly, dragging real-world destruction down to roughly 91% and tripling the methane that flaring was assumed to emit. It is also why the maps below count venting and flaring together. The last decade’s real progress came less from regulators than from investors, LNG buyers and satellites that could finally count the flames from orbit, and the rules, in both jurisdictions, have spent 2025 and 2026 quietly loosening again.

For scale: 1 MMcf, the unit in the tables below, is about a year’s natural gas for a dozen homes, or the energy in ~9,000 gallons of gasoline.

12 months ending …

show outliers only (cloud shows the latest window)

Alberta facilities

Facility	flared+vented MMcf	% of own production

Texas leases

Lease	flared+vented MMcf	% of own production

Same lens, different regimes

The measure. Flared+vented gas divided by gas-equivalent production (gas + liquids at energy equivalence), summed over a 12-month window: a unitless ratio computed identically on both sides (Alberta volumes converted at 1 e3m³ = 35.3147 MCF). A unit needs ≥2,000 MCF flared+vented and real production in the window to be ranked; dot size is flared+vented volume.
The units. Alberta reports against facilities (batteries, plants); Texas against leases. Both are the operator’s reporting unit above the well. Neither jurisdiction publishes complete well-level attribution, so this is as close as public data gets.
The windows align exactly. The slider steps through every 12-month window both public archives share. Alberta’s archive is a rolling ~5 years; Texas publishes back to 1993 but its two newest months are still filling in.
Transparency differs. Alberta publishes vented and flared volumes separately for every facility. Texas operators have reported the split since 2021, but every bulk product the RRC publishes folds it back into one “vented or flared” number, so the combined series is what both maps show.
Not a scoreboard. Different rules, different measurement, and both jurisdictions under-report: satellite and aerial campaigns routinely observe more flaring and venting than official filings admit; see Plant et al. 2022 (Science) on unlit and inefficient flares, and the IEA Global Methane Tracker on the gap between reported and observed emissions. Absolute levels aren’t comparable; the ratio, the outlier structure, and the trends are.