Woodfibre flare emissions: evidence and data acquisition roadmap

Original research cut-off: 5 September 2026; bounded flare-record update: 11 September 2026. This document identifies the next records and measurements needed for defensible predictions. It does not establish current operational exposure, and it does not claim every public record has been exhausted. The accompanying JSON register specifies fields, units, timing, owners, fallbacks and acceptance checks.

The highest-value next step is to obtain the current accepted Flaring Management Report, its engineering attachments, and the operational air authorization application. A more elaborate transport model cannot compensate for an unknown time-varying source. Before operation, a defensible result would be a prediction conditional on documented engineering cases, with uncertainty. Actual event predictions and independent validation will additionally require operational records and measurements.

What the current public record adds

A dated engineering plan now improves the flare-location evidence. The District-linked overall plot plan, revision 6 issued for construction on 16 August 2024, depicts the HP/LP flare. A scale-qualified read-off places its base approximately at E482010/N5501780 in UTM Zone 10 NAD83(CSRS), about 490 m from the existing map proxy under an explicitly approximate datum comparison. This is a dated location candidate, not surveyed current tip geometry. Its 205 m radiation-evaluation annotation is not an established stack height. The next record target is the current source schedule and accepted Flaring Management Report linked to this drawing and any successor. Drawing evidence and exact source.

Finding	Primary source and locator	What it resolves—and what it does not
The April 2026 facility permit identifies HP and LP flares and a 2016 Flare Basis of Design. Its conditions require a Flaring Management Report before installation, to the regulator’s satisfaction, and normally at least 18 months before commissioning, subject to written exemptions.	BCER facility permit, latest amendment 8 April 2026, original conditions 14 and 21–23; PDF pp. 15–16.	Identifies specific existing-record targets. It does not supply a complete current source model or prove installation occurred or a condition was discharged. Request the accepted revision, attachments, acceptance correspondence and exemptions.
A Squamish Nation update describes an approximately 134 m flare structure planned for construction in May 2026; it says natural gas was not then in the system and startup flaring was expected in 2027 after necessary permits.	Squamish Nation newsletter, 30 April 2026, PDF p. 15, printed p. 29.	A useful recent height reference, not surveyed HP/LP tip elevations, final coordinates, or proof of the facility’s September status. Older 100 m/140 m modelling references cannot silently substitute for as-built geometry.
Council approved a request for stack parameters from both BCER and Woodfibre LNG on 2 June 2026, following a May committee recommendation.	12 May committee minutes, PDF pp. 5–6; 2 June Council minutes, PDF pp. 3–4.	The committee listed five fields; the adopted minutes list four (count, height, exit velocity, internal diameter), omitting exit temperature. Verify the outgoing letter and any reply. No technical answer was found in the text of the May correspondence or June correspondence reviewed.
The 2025 federal annual report describes a construction reporting year; its operational flaring condition 4.2 is marked not applicable for that period.	2025 Federal Decision Statement annual report, 31 March 2026, Appendix A, PDF p. 35.	This report is not an operational flare emissions time series. Future events and unlocated existing design records are different kinds of gap.
Public onsite ambient reporting extends to June 2026 in the filings reviewed. June’s report uses FAQMMP Rev. 8, dated 2 June 2026, and labels its data Level 0 validation.	June 2026 AQMS report, dated 10 August 2026, §1/PDF p. 6; Table B.1/PDF p. 32; filings index.	PM₂.₅, PM₁₀, TSP and NO₂ are continuous measurements; SO₂ and VOC as hexane are passive measurements. Daily summaries and hourly charts are useful context, but not a complete validated event dataset or flare attribution. Preserve missing values and detection limits.
A new air discharge application gives source-level rates and exhaust parameters for the separate Fortis Squamish compressor station.	BCER notice and PA112277 notice/application summary, 26 August 2026, Tables 1–2, p. 3.	This materially improves the cumulative-source inventory. It is neither the Woodfibre flare authorization nor a reason to attribute all ambient pollution to flaring. The full Part 3 discharge form and assessment remain acquisition targets.

The public BCER Woodfibre page also links PE111578, which concerns construction effluent. It should not be described as an operational air permit. A current Woodfibre operational air source table was not retrieved in this search; its existence, filing status and any subsequent authorization should be checked explicitly against facility 00026917 / AD100105360. BCER Woodfibre project page

Acquisition order

“P0” means the missing evidence blocks a current facility-specific concentration prediction. “P1” is needed for event validation or interpreting community concentrations. “P2” improves historical traceability. Proposed sampling resolutions below are scientific recommendations, not claims about legal requirements.

P0—Get the current design and authorization package first. Request the accepted Flaring Management Report and revisions, final flare basis, vendor performance package, operational air application/technical assessment/source tables, regulator review and any commissioning approvals. Follow the District’s existing stack-data request before creating a duplicate request. Establish an explicit design revision and source-ID crosswalk.
P0—Replace the source-location proxy with surveyed geometry. Use the recovered Rev 6 plan as a record lead, then obtain each active tip’s current coordinates and datum, ground and tip elevations, effective exit area, tip type, orientation, staging, pilot locations and relevant nearby structures. “134 m structure” alone cannot determine heat release height or building effects.
P0—Obtain event histories of gas flow and composition. Separate flare headers, pilots, purge and assist gas. Request native historian records, preferably 1–10 seconds during fast changes, with one-minute aggregates and integrated event totals. Include standard temperature/pressure, wet/dry basis, meter calibration, low-flow substitution, detection limits and overload flags. Pair them with gas chromatography or representative samples before and during composition changes. Pre-operation, use engineering time histories for explicitly named cases.
P0—Resolve species emissions and buoyancy together. For each case, obtain g/s for CO, NOₓ, PM₂.₅/PM₁₀, methane, named VOCs and sulfur species, and the method behind every number: measurement, vendor guarantee, calculation or generic factor. Request net heating value, heat release, radiative fraction, exit conditions and flame/plume treatment. A flare is not simply a conventional stack with a hot exhaust: the selected model’s equivalent-source method must avoid counting plume rise twice.
P1—Acquire existing monitoring files before purchasing instruments. Obtain raw and final QA ambient records, calibration/audit files, flags, clocks, instrument/inlet metadata and regional reference data. The June report identifies the onsite AQMS at NAD83 UTM10 E481569/N5501374; this is a monitor location, not the flare. June Table B.1 identifies filter-tape outages and an unexplained NO₂ gap. Neither missing values nor below-detection passive samples should become zero. June AQMS report, §§1, 4 and Appendix B.
P1—Plan synchronized validation around actual events. Match operator event records to near-source plume observations, meteorology and upwind/downwind ambient measurements. Choose operating loads and wind regimes from the uncertainty analysis. Include separate NO/NO₂ and ozone information where NO₂ conversion matters, plus targeted VOC speciation where health interpretation is intended. A short campaign can test a case; it cannot establish annual event frequency.
P1—Complete the cumulative inventory. Include the Fortis station, Woodfibre’s other combustion sources, vessels and construction equipment with source-specific schedules; treat regional smoke and urban background separately. Keep total particulate distinct from PM₂.₅. The Fortis notice describes two boilers but presents a single boiler row; reconcile the full application before multiplying counts or rates.
P2—Retrieve the original EA model archive. It can reproduce the historical assessment and resolve weighting conventions, but cannot make historical WWGF/CDGF rates current. Preserve those rates only as explicitly historical conditional scenarios.

For every acquisition, retain the native file, revision/date, provenance, units, missing-value conventions and uncertainty. A usable source dataset must pass mass/energy and timestamp checks, reconcile to reported event totals, and distinguish an actual measurement from an estimated pollutant mass. Unknown SO₂ or VOC emissions remain unknown; the inlet H₂S limit is not a measured flare-stream sulfur profile.

What comparable flares can teach us

Primary studies support measurement methods and uncertainty tests, not substitution of another facility’s emissions or efficiency for Woodfibre’s.

Evidence	Transferable approach	Boundary on its use
Evans et al., 7 March 2024, with 87-test dataset	Test actual tip designs across flow, heating value, dilution and wind; measure combustion products and propagate uncertainty.	Controlled upstream-flare experiments do not certify Woodfibre’s larger final flare system. Neither a universal 98% efficiency nor a universal heating-value threshold follows.
Shaw et al., 26 January 2023	Aircraft measurements of 58 North Sea plumes combine background subtraction, multiple carbon species and NOₓ observations.	The reported median CH₄ destruction efficiency, 98.5%, is specific to that sample. Nearby diesel sources can confound NOₓ attribution; this is not a Woodfibre rate.
Plant et al., 30 September 2022	Record lit/unlit status and distinguish combustion performance from the frequency of unlit operation.	The reported 91.1% effective methane destruction combines inefficient and unlit flares in three US production basins. It is not an LNG-flare in-flame efficiency.
Explicit/Equinor FlareE final report, public 2025, §§5.2–5.3	Match measured feed composition with repeated CH₄/CO₂ plume flux surveys; use about one-minute operational history to select representative conditions.	This is an industry project report with a third-party uncertainty assessment. CH₄/CO₂ alone does not establish NO₂, CO, PM or named VOC emissions. Any field campaign requires qualified, site-approved operators; this is not a proposal for community flights near the flare.

BCER’s released LNG Canada records are especially useful templates. Its Flaring Management Report describes a design DRE of at least 99.53%; this is a design claim in that report, not a Woodfibre measurement. More importantly, its September 2024 commissioning report separates metered flare gas volumes from factor-calculated pollutant masses. It uses upstream meter sums below flare-meter detection, AP-42 factors for several species, and a sulfur mass balance for SOₓ. CO and VOC factors are rated “E—Poor” in its table. Request that same provenance distinction for Woodfibre; a reported pollutant mass is not automatically a measured plume emission. BCER2026-007 release, PDF p. 33; BCER2025-023 release, PDF pp. 188–189.

Combustion efficiency and destruction efficiency are also different quantities. Destroying a feed chemical does not prove complete conversion to CO₂, nor establish the other combustion products. Generic AP-42 factors are an identified fallback with uncertainty, not local validation. EPA AP-42 §13.5, February 2018, pp. 3–4.

Draft records scope—not submitted

For Woodfibre LNG facility 00026917 / AD100105360, seek the latest accepted Flaring Management Report and its appendices, superseded revisions relevant to source design, acceptance correspondence and written exemptions; the current flare basis of design and final source parameters; the operational air authorization application, technical assessment, source inventory and review correspondence; and any commissioning flaring plan, event-reporting specification and monitoring-data release arrangements. Request existing records in their native electronic formats, with a document index and an indication of any records withheld or not yet created. Separately, ask the District for the outgoing letter and replies following its 2 June 2026 adopted request, and reconcile the exit-temperature omission relative to the 12 May recommendation. Use BCER’s public records/FOI route only after checking public and informally available records; no request has been sent.

Once these source gates are closed, combine them with the separate weather and transport-method work. Community maps should report conditional concentrations and uncertainty for stated averaging periods. They should not label a coloured overlap a health-risk zone, treat a 20 km map edge as a plume boundary, or infer regulatory attainment from one simulated event.