Methane Emissions & LDAR · Fundamentals

Specific Methane Sources: Tanks, Pneumatics, Compressors, Flares

Engineering reference for the four largest discrete methane sources in upstream and midstream oil & gas operations. Covers tank flashing emissions (Vasquez-Beggs flash GOR), pneumatic device venting (high/low/zero-bleed conversion), reciprocating compressor rod-packing and centrifugal seal vents (EPA Subpart W §98.233), and flare combustion efficiency (Allen-Torres 2017 / 40 CFR 60.18).

High-bleed pneumatic

28.6 scf/h

EPA §98.233 emission factor for high-bleed continuous device. Zero-bleed (electric / IA) eliminates this entirely.

Recip rod packing

11.5 scf/min/cyl

Aged packing baseline per §98.233(p). Replace at 26,000 hr per OOOOa.

Flare DRE

98% min

40 CFR 60.18 default assumption. Allen 2017 field measurements often well below this in adverse weather.

Run the calculations

Source-specific calcs

Tank flashing, pneumatic conversion, compressor seal venting, and flare combustion efficiency.

Open methane source calcs →

1. Overview

Methane emissions from oil & gas operations are dominated by a small number of source types. EPA's Greenhouse Gas Inventory and EDF field measurements consistently rank the top contributors:

Source categoryApprox. share of US O&G CH₄Mitigation difficulty
Pneumatic devices (high-bleed)15–20%Easy: convert to low-bleed/zero-bleed
Tank flashing & vents10–15%Moderate: VRU installation
Compressor rod packing & seals10–15%Easy: scheduled replacement
Flares (incomplete combustion + venting)10–20%Hard: combustion physics + operator behavior
Component fugitives (LDAR-addressable)10–15%Moderate: quarterly OGI
Liquids unloading & well workovers5–10%Hard: process changes
Super-emitter releases (episodic)20–30%Aerial / satellite detection

This page focuses on the four discrete-source categories handled by EPA Subpart W §98.233 / §98.234. Each has standard emission factors, established mitigation paths, and predictable economics — the "easy wins" of methane emission reduction.

The 80/20 rule: For most upstream / midstream facilities, addressing pneumatic conversions + rod-packing replacement + tank VRU installation + flare optimization captures 50–70% of the methane reduction available — at near-zero net cost when gas-recovery value is included. The hard cases (super-emitters, fugitive components) come after these baseline programs are in place.

2. Tank Flashing Emissions

Crude oil, condensate, and produced-water storage tanks at oil & gas production sites operate at atmospheric pressure. Liquids leaving the separator at 30–200 psig flash when they enter the tank — the pressure drop releases dissolved gas to vent stack.

The Vasquez-Beggs flash GOR equation

Vasquez & Beggs (1980) provide the industry-standard correlation for solution gas-oil ratio (Rs) at any P, T:

Rs = C₁ · γ_g · P^C₂ · exp(C₃ · γapi / (T_F + 460)) Rs = solution GOR at conditions (scf/STB) γ_g = gas specific gravity (vs air) γapi = API gravity of crude T_F = temperature (°F) P = pressure (psia) For γapi ≤ 30°: C₁ = 0.0362, C₂ = 1.0937, C₃ = 25.724 For γapi > 30°: C₁ = 0.0178, C₂ = 1.187, C₃ = 23.931 Flash GOR = Rs(separator) − Rs(stock tank, 14.7 psia)

Composition of flash gas

Flash gas composition depends on the crude/condensate and separator conditions. Typical splits for medium crude (38° API) at 50 psig separator:

ComponentApproximate fraction
CH₄ (methane)60–75%
VOC (C₃+)20–35%
Other (N₂, CO₂)2–8%

For richer condensate or higher separator P, VOC fraction rises and CH₄ fraction falls. Light crude with H₂S produces additional emission category requiring sour-service equipment.

Vapor Recovery Units (VRU)

Tank vapor capture via VRU is the standard mitigation. Typical VRU is a small reciprocating compressor that pulls a slight vacuum on the tank vent and delivers vapor to the gas sales line. Capture efficiency: 95–99% with properly sized and maintained VRU.

NSPS OOOOa thresholdRequirement
VOC potential ≥ 6 tpyVRU capture or 95% destruction (flare)
VOC potential 6–10 tpyQuarterly OGI inspection
VOC potential < 6 tpyAnnual / semi-annual inspection sufficient

3. Pneumatic Devices

Pneumatic instruments (level controllers, valve positioners, valve actuators) traditionally use available pressurized natural gas as the working fluid. Each actuation vents some gas to atmosphere, and continuous-bleed designs constantly leak gas through pilot orifices.

EPA Subpart W §98.233 emission factors

TypeEF (scf CH₄/h)Annual (lb/yr at 95% NG)Typical use
High-bleed continuous28.6~ 9,500Older controllers; pre-OOOO design
Intermittent-bleed13.5~ 4,500Snap-action level controllers
Low-bleed continuous1.39~ 460Modern OOOOa-compliant pneumatic
Zero-bleed00Electric or instrument-air actuated

EPA OOOOa / OOOOb requirements

  • OOOOa (2016): All new sources must use low-bleed (≤ 6 scf/h) for continuous-bleed devices
  • OOOOb (2024): Zero-bleed required for all new sources where electric or instrument air available
  • OOOOc (state implementation): Existing-source rules — varies by state implementation plan

Conversion economics

Conversion CAPEX per device:

ConversionCAPEX per deviceAnnual gas savings (at $3.50/Mcf)Simple payback
High-bleed → low-bleed$2,000–4,000~ $80/yr per device25–50 years
High-bleed → zero-bleed (electric)$3,000–8,000~ $90/yr per device33–90 years
Intermittent → low-bleed$1,500–3,000~ $36/yr per device40–80 years

Pure economics rarely justifies pneumatic conversion — the case is regulatory compliance (OOOOa/b) plus avoided social cost of carbon. With $51/tCO₂e SCC, the abatement cost per tCO₂e for high-bleed → zero-bleed conversion is ~$50–100 — competitive with other climate actions.

4. Compressor Vents

Reciprocating and centrifugal compressors used for gas gathering, transmission, and processing are major methane sources. Two distinct loss mechanisms:

Reciprocating compressor rod packing

Reciprocating compressors use packing rings around the piston rod to seal between the high-pressure cylinder and atmosphere. Packing wear over time degrades the seal and increases methane leakage. EPA Subpart W §98.233(p) emission factor:

EF = 11.5 scf CH₄/min/cylinder (aged, > 26,000 hr since replacement) EF = 0 scf/min/cylinder (newly replaced, < 26,000 hr) Linear ramp between these limits with operating hours. For a 4-cylinder recip operating 8000 hr/yr at full age: 4 × 11.5 × 60 × 8000 = 22,080,000 scf CH₄/yr ≈ 22 MMscf/yr × 19.2 kg/Mcf = 423,000 kg CH₄/yr per recip = 423 t/yr = ~ 12,000 tCO₂e/yr

EPA OOOOa requires rod-packing replacement at 26,000 hours or 36 months whichever first. Typical replacement cost: $5,000–10,000 per cylinder including consumables and labor; payback in 1–2 years from gas-recovery value alone.

Centrifugal compressor seals

Centrifugal compressors use either wet seals (oil film) or dry gas seals (gas-cushion non-contacting):

Seal typeEF (scf CH₄/min/compressor)Notes
Wet seal (oil film)24.7Older design; degassed seal oil vents to atmosphere
Dry gas seal1.6Modern industry standard; near-zero seal leakage

Wet-to-dry seal conversion typical CAPEX: $300,000–500,000 per centrifugal compressor. Payback varies widely with operating hours; for a continuously-running pipeline compressor (8000 hr/yr), payback can be 5–10 years from gas-recovery value alone, faster with SCC.

Standby and venting modes

EPA Subpart W also accounts for compressors in standby (pressurized but not running) and not-operating-depressurized states:

ModeRecip EF (scf/min)Centrifugal EF (scf/min)
Operating11.5/cyl (aged)24.7 (wet) / 1.6 (dry)
Standby pressurized4.4/unit0.95 (dry seal)
Not operating, depressurized0 (vented during depressurization)0

5. Flare Combustion Efficiency

Industrial flares are designed to combust waste gas, converting hydrocarbons (mostly CH₄ and VOC) to CO₂ and H₂O. The destruction efficiency (DE) determines the methane release: any uncombusted hydrocarbon vents as fugitive emissions through the flare tip.

40 CFR 60.18 standard

The EPA flare design rule specifies:

  • Minimum 98% destruction efficiency (DRE) — assumed by default for compliant flares
  • Minimum heating value 200 BTU/scf (without supplementary fuel)
  • Maximum exit velocity per Equation (40 CFR 60.18(c)(4)):
    v_max [ft/s] = exp((H_v + 1212) / 850) for HHV in BTU/scf, capped at 60 ft/s for HHV ≥ 1000
  • Pilot flame must be present and continuously monitored

Allen-Torres 2017 field measurements

Allen et al. (Univ. of Texas) and Torres et al. measured actual flare CE in production-field service using sampling drone hovering above the flare tip. Findings:

  • Average CE: 91% (vs 98% assumed by EPA)
  • 5% of measurements: CE < 50% (mostly during high wind / low gas heating value events)
  • 30% of total measurements: CE < 95%

The implication: actual methane release from flares is typically 2–5× higher than EPA inventory assumes, because the 98% assumption rarely holds in field conditions.

Driving factors for CE

FactorCE impact
Wind/exit velocity ratioAbove 3, CE drops sharply
Heating valueBelow 200 BTU/scf, flame instability; below 100, no sustained combustion
Assist (steam vs air vs unassisted)Steam-assisted: 99%+; air-assisted: 98%+; unassisted: 95% baseline
Pilot maintenanceFailed pilot = 0% CE (full venting); requires real-time monitoring
Liquid carryoverPulls flame into yellow-tip region; reduces CE significantly

Modeling CE for engineering screening

The Allen-Torres-inspired correlation used in calc C19:

CE = base_CE × HHV_factor × wind_factor base_CE: 0.97 unassisted, 0.985 air-assisted, 0.99 steam-assisted HHV_factor: HHV ≥ 300: 1.0 HHV 200–300: 0.95 + (HHV−200)/100 × 0.05 HHV 50–200: 0.85 + (HHV−50)/150 × 0.10 wind_factor (wind/v_exit ratio r): r < 1: 1.0 − 0.05·r 1 ≤ r < 3: 0.95 − 0.10·(r−1)/2 3 ≤ r < 10: 0.85 − 0.30·(r−3)/7 r ≥ 10: 0.55

6. Standards & References

  • EPA NSPS Subpart OOOOa (2016) and OOOOb/c (2024)
  • EPA Greenhouse Gas Reporting Rule, Subpart W §98.232–.236 (40 CFR Part 98)
  • EPA AP-42, Compilation of Air Pollutant Emission Factors, Chapter 5.3
  • EPA TANKS 4.09 (legacy) — superseded by Subpart W §98.234 for tank emissions
  • 40 CFR 60.18 — General control device requirements (flare design)
  • API Manual of Petroleum Measurement Standards 19.4 (evaporative loss from oil storage)
  • Vasquez, M.E., Beggs, H.D. (1980). "Correlations for fluid physical property prediction," JPT 32(6), 968–970.
  • Allen, D.T. et al. (2017). "Methane emissions from process equipment at natural gas production sites in the United States: Liquid unloadings," Environ. Sci. Technol.
  • Torres, V.M. et al. (2012). "Industrial Flare Performance at Low Flow Conditions," Ind. Eng. Chem. Res.
  • McDaniel, M., Pohl, J.H. (1986). EPA-600/2-86-080, Flare Efficiency Study
  • API Compendium of GHG Emissions Methodologies (2021)
  • OGMP 2.0 Reporting Framework (2020)

Frequently Asked Questions

What is tank flashing and why does it matter?

Tank flashing occurs when liquid hydrocarbons leaving a separator (typically at 30–200 psig) enter an atmospheric storage tank — the pressure drop releases dissolved gas (mostly methane and light HCs) that vents to atmosphere. EPA Subpart W §98.234 covers tank flashing emissions; for a typical 500 bbl/d oil well the flash gas is ~5–50 t CH₄/yr. NSPS OOOOa requires VRU (vapor recovery unit) capture if VOC > 6 tpy.

What are high/low/zero-bleed pneumatic devices?

Pneumatic devices (level controllers, valve actuators, position controllers) traditionally use natural gas as their working fluid, venting some to atmosphere with each actuation. EPA Subpart W §98.233 emission factors: high-bleed continuous 28.6 scf CH₄/h, intermittent-bleed 13.5 scf/h, low-bleed 1.39 scf/h, zero-bleed (electric or instrument air) 0. EPA OOOOa requires zero-bleed for all new sources; existing high-bleed devices typically converted to low-bleed first ($2–5k CAPEX), then to zero-bleed if utilities support it.

What is reciprocating compressor rod-packing leakage?

Reciprocating compressors use packing rings around the piston rod to seal between the high-pressure cylinder and atmosphere. As packing wears (typically over 26,000 operating hours), the seal degrades and methane vents around the rod. EPA Subpart W §98.233(p) emission factor: 11.5 scf CH₄/min/cylinder for aged packing (= ~ 60 t CH₄/yr per cylinder at full operation). EPA OOOOa requires packing replacement at 26,000 hours or 36 months whichever first; typical replacement cost is $5–10k per cylinder.

What is flare combustion efficiency and how is it measured?

Flare combustion efficiency (CE) is the fraction of hydrocarbon mass entering the flare that is fully oxidized to CO₂ and H₂O — the remainder vents as unburned hydrocarbon. 40 CFR 60.18 specifies 98% minimum destruction efficiency assumed by default. Allen et al. 2017 measured field CE often well below 98% under high-wind, low-exit-velocity, or low-HHV conditions — sometimes as low as 70% for unassisted flares in adverse weather. The actual CE depends on exit velocity, wind speed, gas heating value, and assist type (steam, air, or unassisted).

How are these methane sources quantified for inventory reporting?

EPA Subpart W §98.232–.236 provides default emission factors that operators can apply to component counts for annual reporting. Best practice (OGMP 2.0 Level 3+) replaces defaults with measured rates: tank flashing via Vasquez-Beggs flash calculation or process simulator; pneumatic vents by counting devices and applying type-specific factors; rod-packing via age-weighted factor or direct measurement; flare CE via mass balance + measured discharge composition. Measured rates typically diverge from defaults by 30–300% — usually higher, because heavy-tailed distributions of broken/leaking equipment are systematically under-counted in default factors.