By ShipCalculators.com · Published 8 May 2026

Per-fuel well-to-wake intensity: LNG (Otto vs Diesel cycle)

Liquefied natural gas (LNG) is the most widely deployed alternative marine fuel in 2026, but its well-to-wake (WtW) intensity is not a single number. The two thermodynamic combustion paths used in marine dual-fuel engines, the low-pressure Otto cycle (Wartsila DF, MAN ME-GA at part load, Caterpillar M46DF) and the high-pressure Diesel cycle (MAN ME-GI), produce very different methane-slip profiles, and that single difference propagates into a WtW gap of roughly 9 gCO2eq/MJ between the two technologies. Under MEPC.391(82) Annex 1, the typical default for LNG-Otto is approximately 85 gCO2eq/MJ at AR5 GWP100, while LNG-Diesel sits near 76 gCO2eq/MJ. The same hierarchy is repeated in FuelEU Maritime Annex II, with engine-technology-specific factors that operators must declare and verify. The upstream well-to-tank (WtT) component, covering extraction, liquefaction, shipping, and regasification, contributes roughly 18 to 22 gCO2eq/MJ on the IMO defaults but varies materially by sourcing basin (US shale, Qatar North Field, Australia North West Shelf, Russia Yamal). Methane is 28 times more potent than CO2 over a 100-year horizon under AR5 GWP100, the metric the IMO LCA Guidelines and FuelEU mandate, and 84 times more potent over 20 years under AR5 GWP20, a sensitivity that materially recasts the LNG-versus-VLSFO comparison covered in /wiki/per-fuel-wtw-vlsfo-mgo. Operators size LNG WtW exposure with /calculators/fuel-wtw-lng, the bio-LNG variant with /calculators/fuel-wtw-bio-lng, and dual-fuel blend cases with /calculators/fuel-wtw-blend. Operational compliance still references MARPOL Annex VI Regulation 14 sulphur and Regulation 13 NOx separately from the lifecycle GHG path.

Contents

Background: LNG as a transition fuel

LNG entered deep-sea shipping in volume during the 2018 to 2024 wave of dual-fuel newbuildings, driven by the IMO 2020 sulphur cap, competitive bunker economics in the post-pandemic recovery, and the maturity of large-bore two-stroke gas engine technology. By 2026, LNG-fuelled vessels exceed 600 in service with another 800 on order, spread across container ships, car carriers, dry bulk, tankers, and LNG carriers themselves running on their own boil-off gas (BOG).

The fuel offers immediate and uncontroversial reductions in SOx (essentially zero), NOx (Tier III compliance achievable without selective catalytic reduction in lean-burn Otto-cycle engines), and particulate matter (close to zero). The carbon dioxide reduction at the ship’s funnel is approximately 23 to 27 percent per MJ of fuel burned compared to VLSFO, because methane (CH4) carries four hydrogen atoms per carbon atom and therefore yields more energy per unit of CO2 produced than the long-chain hydrocarbons in residual fuel oil.

The complication is methane slip: a small fraction of unburned methane that passes through the engine to the exhaust. Because methane is itself a strong greenhouse gas, even a small slip percentage erodes the funnel-stack CO2 advantage when measured on a CO2-equivalent basis. The magnitude of slip is determined by combustion physics, and combustion physics is determined by which thermodynamic cycle the engine runs.

Under the IMO Net-Zero Framework adopted at MEPC 83, every fuel is benchmarked on its WtW gCO2eq/MJ intensity, not just funnel CO2. That construction is what forces the Otto-versus-Diesel distinction onto the regulatory landscape: two ships burning the same LNG molecules can attract materially different GHG Fuel Standard (GFS) outcomes simply because their engine technology differs.

This page walks through the cycle physics, the per-engine slip behaviour, the MEPC.391(82) and FuelEU defaults, the upstream WtT pathway with sourcing variability, and the GWP sensitivity that determines how the LNG story reads in different policy frames.

Otto cycle vs Diesel cycle in marine dual-fuel engines

The thermodynamic distinction is foundational and dictates everything downstream.

The Otto cycle, named for Nikolaus Otto’s 1876 engine, is a constant-volume combustion cycle in which a premixed air-fuel mixture is compressed and then ignited at or near top dead centre by an external ignition source (a spark plug in automotive practice; a small pilot fuel injection in marine dual-fuel engines). Because gas and air are premixed during the compression stroke, combustion proceeds as a deflagration through a homogeneous charge. Compression ratios are limited by knock resistance, typically 11 to 13 in marine dual-fuel applications, and the gas is admitted at relatively low pressure (typically 5 to 10 bar) into the inlet duct or directly into the cylinder before compression begins.

The Diesel cycle, named for Rudolf Diesel’s 1893 patent, is a constant-pressure combustion cycle in which only air is compressed to high pressure and high temperature, and the fuel is injected directly into the hot compressed air at or near top dead centre. Combustion proceeds as a non-premixed diffusion flame; the fuel auto-ignites without a separate ignition source. Compression ratios reach 16 to 22, and in marine dual-fuel Diesel-cycle gas engines, the gas itself is injected at very high pressure (typically 300 bar) into the cylinder after compression.

Three consequences flow from this distinction.

Combustion efficiency and unburned fuel. In an Otto-cycle premixed charge, gas can lodge in cylinder crevices (top-land clearance, ring pack, head-gasket recess) and escape combustion. That gas is then expelled with the exhaust, producing methane slip. In a Diesel-cycle non-premixed charge, gas is injected as a high-pressure jet into hot air and burns essentially as it leaves the injector tip; little or no gas reaches the crevices because the cylinder is filled with hot air during compression, not premixed mixture. Diesel-cycle slip is consequently an order of magnitude lower than Otto-cycle slip.

Knock and load envelope. Otto-cycle gas engines must run lean (excess air ratio lambda typically 2.0 to 2.2) to avoid knock and to control NOx, which constrains the achievable specific power and forces operation in a narrow window around design load. Off-design operation, particularly part-load and transient loading, pushes the engine into regimes where slip rises sharply. Diesel-cycle gas engines, by contrast, are not knock-limited and can run the same load envelope as a heavy-fuel diesel engine.

Pilot fuel and dual-fuel operation. Both cycles use a small pilot diesel injection (typically 1 to 5 percent of full load energy) to ignite the gas charge. In the Otto cycle the pilot triggers premixed flame propagation; in the Diesel cycle the pilot establishes a high-temperature ignition kernel into which the high-pressure gas is then injected. The fuel-mode switching logic and the pilot injector hardware differ between manufacturers but all marine dual-fuel engines share this architecture.

The cycle choice is therefore not a marketing label but a thermodynamic commitment with measurable WtW consequences. Sections 3 and 4 describe the engine families that embody each path.

Low-pressure DF engines: Wartsila DF, MAN ME-GA part-load, Caterpillar M46DF

Low-pressure (LP) dual-fuel engines run the Otto cycle and admit gas at fuel-supply pressures of 5 to 10 bar. They dominate the four-stroke segment (auxiliary engines, smaller propulsion engines on ferries and offshore vessels) and have a substantial two-stroke presence through the MAN ME-GA family.

Wartsila DF four-stroke engines. The Wartsila 31DF, 34DF, 46DF, and 50DF series are the reference low-pressure dual-fuel engines for medium-speed marine and stationary power applications. They use port-fuel admission of natural gas during the inlet stroke, premixed charge compression, and a micro-pilot diesel injection (less than 1 percent at full load) to ignite the lean mixture. Tier III NOx compliance is achieved without SCR by running lean. Methane slip in gas mode is reported by Wartsila at full load near 2.5 to 3.5 g/kWh for the older 50DF and 1.5 to 2.5 g/kWh for the latest 31DF generation, with significant rise at part load below 50 percent MCR. Expressed as a percentage of fuel mass, slip ranges from about 1.5 to 3.0 percent at full load.

MAN ME-GA two-stroke engines. The ME-GA is MAN Energy Solutions’ low-pressure two-stroke gas engine, introduced as a counterpoint to the high-pressure ME-GI for owners willing to trade slip performance for fuel-supply system simplicity and cost. The engine uses gas admission valves on the cylinder liner during the upward scavenge stroke, with compression of the premixed charge and pilot ignition. Slip at full load is reported around 1.0 to 1.5 percent of fuel mass, rising to 2.0 to 2.5 percent at part load. The ME-GA is specified primarily for container ship and tanker projects where fuel-system capex savings outweigh the WtW penalty, although the 2024 to 2026 wave of orders has shown a swing back to ME-GI on stricter FuelEU compliance arithmetic.

Caterpillar M46DF four-stroke engine. The M46DF is Caterpillar’s medium-speed dual-fuel platform, sold by the MaK brand for ferry, ro-pax, offshore, and inland-waterway applications. It uses the same low-pressure Otto-cycle architecture as the Wartsila DF family, with port-fuel admission, lean premixed combustion, and micro-pilot ignition. Slip behaviour is comparable to Wartsila’s, in the 1.5 to 3.0 percent range at full load, and the engine is widely deployed on ro-pax services in northern Europe and the US west coast.

The shared characteristic across the LP Otto-cycle family is that slip is intrinsic to the combustion architecture. Manufacturer R&D programmes have made meaningful progress (the latest Wartsila 31DF and MAN ME-GA generation have slip around 30 to 50 percent below 2010-era engines), but no Otto-cycle engine reaches the slip floor of a Diesel-cycle engine. The gap is structural.

High-pressure DF engines: MAN ME-GI

The MAN ME-GI (Man Engine, Gas Injection) is the flagship high-pressure dual-fuel two-stroke engine and the only large-bore Diesel-cycle gas engine in series production for deep-sea propulsion as of 2026. It is offered in the same bore sizes as MAN’s heavy-fuel ME-C platform (50 cm, 60 cm, 70 cm, 80 cm, 90 cm) and is mechanically a derivative of the standard ME-C electronically controlled two-stroke with the addition of a high-pressure gas injection system.

The fuel supply system delivers natural gas at 300 bar to the cylinder through a high-pressure double-walled pipe network, an inert-gas-purged annulus, and a Cryostar or Burckhardt high-pressure gas pump that pressurises the LNG before vaporisation. Gas is injected directly into the cylinder near top dead centre, after compression of pure air, through a dedicated high-pressure gas valve in the cylinder cover. Pilot diesel (typically 3 to 5 percent of total energy at full load) is injected through the standard fuel injector and provides the high-temperature ignition kernel.

The combustion is a non-premixed diffusion flame, identical in character to a heavy-fuel diesel combustion event with natural gas substituted for the residual hydrocarbon. Crevice slip is essentially zero because the cylinder contains hot air, not premixed mixture, during the compression stroke. The only meaningful methane slip pathway is incomplete combustion in the diffusion flame edges, which manufacturers report at 0.2 to 0.5 percent of fuel mass under design conditions.

The penalty of the Diesel-cycle architecture is fuel-supply system complexity and capex. The high-pressure pump alone adds approximately USD 3 to 5 million to a deep-sea newbuilding fuel-system bill, and the inert-gas safety architecture demands more piping, more valves, and more crew training than a low-pressure system. Tier III NOx compliance requires either exhaust gas recirculation (EGR) or selective catalytic reduction (SCR) because the high-temperature diffusion flame produces NOx in quantities similar to heavy-fuel combustion; the lean-burn NOx benefit of the Otto cycle is not available.

The MEPC.391(82) Annex 1 default for the LNG-Diesel pathway reflects this combustion physics directly: the methane slip term is small enough that the WtW intensity is dominated by combustion CO2 plus upstream WtT, and the headline value sits near 76 gCO2eq/MJ. The Otto-cycle equivalent, with slip an order of magnitude higher and weighted by GWP100 of 28, lands near 85 gCO2eq/MJ. The 9 gCO2eq/MJ gap is almost entirely the slip-times-GWP product.

Methane slip mechanism and per-cycle magnitude

Methane slip in marine engines has three principal sources, with relative magnitudes very different between the two cycles.

Crevice slip is unburned gas trapped in cylinder geometry (top-land clearance, piston ring pack, head-gasket recess, cylinder liner irregularities) during compression and expelled into the exhaust during the exhaust stroke. This pathway exists only when the cylinder contains a premixed charge during compression, which means it is an Otto-cycle phenomenon and not a Diesel-cycle one. Crevice slip is the largest single contributor to LP engine slip and is the primary mechanism that makes Otto-cycle slip an order of magnitude larger than Diesel-cycle slip.

Flame-quench slip is gas in the laminar flame layer adjacent to the cylinder wall that is quenched by heat loss to the wall before combustion completes. This pathway exists in both cycles but is larger in Otto-cycle premixed combustion because the entire cylinder volume burns near-simultaneously (deflagration); in a Diesel-cycle diffusion flame, only the flame envelope itself reaches wall surfaces, and only briefly.

Scavenging slip is gas that passes through the cylinder during the scavenge phase of a two-stroke engine without being captured for combustion. In two-stroke gas engines with port-admitted gas (some early ME-GA generations and certain four-stroke layouts), a fraction of the admitted gas escapes directly into the exhaust manifold. Modern ME-GA designs have substantially reduced this contribution by closing inlet valves before exhaust valves, but a residual scavenging slip term of 0.2 to 0.5 percent remains in the LP two-stroke envelope.

Quantitatively, the per-cycle slip magnitudes in 2026 vintage equipment are:

Otto cycle, four-stroke (Wartsila DF, Caterpillar M46DF): 1.5 to 3.0 percent of fuel mass at full load, rising to 3.0 to 5.0 percent at part load below 50 percent MCR.
Otto cycle, two-stroke (MAN ME-GA): 1.0 to 1.5 percent at full load, rising to 2.0 to 2.5 percent at part load.
Diesel cycle, two-stroke (MAN ME-GI): 0.2 to 0.5 percent at full load, rising to 0.5 to 0.7 percent at part load.

The MEPC.391(82) Annex 1 default values consolidate these ranges into representative figures used in WtW calculations for operators that do not bring forward verified engine-specific test bed data:

LNG-Otto four-stroke slip default: approximately 3.0 percent of fuel mass.
LNG-Otto two-stroke slip default: approximately 1.7 percent of fuel mass.
LNG-Diesel two-stroke slip default: approximately 0.2 percent of fuel mass.

Each percentage point of slip, multiplied by the GWP100 of methane (28) and converted to gCO2eq per MJ of fuel energy, contributes approximately 5.6 gCO2eq/MJ. The Otto-Diesel gap of 1.5 to 2.5 percentage points therefore produces a WtW gap of 8 to 14 gCO2eq/MJ, consistent with the MEPC.391(82) and FuelEU default differentials.

MEPC.391(82) Annex 1 LNG defaults: per-cycle WtW values

The Annex 1 default WtW emission factor table in MEPC.391(82) provides separate rows for LNG by engine technology, recognising that combustion physics determines the bulk of the WtW intensity once fuel is in the bunker tank.

LNG-Otto four-stroke (Wartsila DF, Caterpillar M46DF, smaller MAN four-strokes):

WtT: 18 to 22 gCO2eq/MJ (extraction 9, liquefaction 6, shipping 1, regasification 0.5, distribution 0.5)
TtW CO2: 56 gCO2eq/MJ
TtW CH4 slip (3.0 percent at GWP100 of 28): 8.4 gCO2eq/MJ
TtW N2O: 0.1 gCO2eq/MJ
WtW total: approximately 87 gCO2eq/MJ

LNG-Otto two-stroke (MAN ME-GA):

WtT: 18 to 22 gCO2eq/MJ
TtW CO2: 56 gCO2eq/MJ
TtW CH4 slip (1.7 percent at GWP100 of 28): 4.8 gCO2eq/MJ
TtW N2O: 0.1 gCO2eq/MJ
WtW total: approximately 83 gCO2eq/MJ

LNG-Diesel two-stroke (MAN ME-GI):

WtT: 18 to 22 gCO2eq/MJ
TtW CO2: 56 gCO2eq/MJ
TtW CH4 slip (0.2 percent at GWP100 of 28): 0.6 gCO2eq/MJ
TtW N2O: 0.1 gCO2eq/MJ
WtW total: approximately 76 gCO2eq/MJ

The headline figures of approximately 85 gCO2eq/MJ for LNG-Otto (a representative blend of four-stroke and two-stroke in proportion to the global LNG-fuelled fleet) and 76 gCO2eq/MJ for LNG-Diesel are the values that most operators encounter in /calculators/fuel-wtw-lng and the marine GFS methodology calculations.

The 9 gCO2eq/MJ gap is structural, not a measurement artefact, and it determines whether a given LNG-fuelled vessel sits above or below the Tier 1 required GFI trajectory in any given compliance year.

FuelEU Annex II treatment of LNG

FuelEU Maritime Annex II provides default WtW factors that closely track the MEPC.391(82) values but with EU-specific calibration of the upstream WtT component reflecting the European supply mix.

The 2026 Annex II defaults distinguish four LNG pathways by engine technology:

LNG Otto medium-speed dual-fuel (LNG-Otto MS-DF): 76.08 gCO2eq/MJ default; with engine-specific slip reported, can rise above 90 gCO2eq/MJ for high-slip auxiliary engines.
LNG Otto slow-speed dual-fuel (LNG-Otto SS-DF): 76.08 gCO2eq/MJ default; ME-GA at typical slip lands near 80 to 84 gCO2eq/MJ in actual reporting.
LNG Diesel slow-speed dual-fuel (LNG-Diesel SS-DF): 76.08 gCO2eq/MJ default; ME-GI at typical slip lands near 75 to 77 gCO2eq/MJ in actual reporting.
LNG lean-burn spark-ignited (LNG LBSI): 76.08 gCO2eq/MJ default; rare in deep-sea but used in some inland and short-sea applications.

The default of 76.08 gCO2eq/MJ in the published Annex II table is a single starting figure that operators must then adjust by reporting actual engine-specific methane slip, calibrated against the FuelEU monitoring, reporting and verification (MRV) regime. The technology-specific slip factors that the verifier applies are:

LNG-Otto MS-DF: 3.1 percent slip (default)
LNG-Otto SS-DF: 1.7 percent slip (default)
LNG-Diesel SS-DF: 0.2 percent slip (default)
LNG LBSI: 2.5 percent slip (default)

Applying GWP100 of 28 to each slip factor and adding to the 76.08 baseline yields the actual reported intensity, which is then compared against the FuelEU Maritime intensity cap for the year. A vessel running ME-GI ends up near 76 gCO2eq/MJ, well below the 91.16 baseline minus the year’s percentage reduction; a vessel running Wartsila 50DF ends up near 90 gCO2eq/MJ, leaving very little headroom against the cap.

Operators therefore have a strong commercial incentive under FuelEU to certify lower-than-default slip values through engine test bed reports recognised by the verifier. The 2025 and 2026 reporting cycles have seen widespread submission of slip data 20 to 40 percent below default for the latest-generation Wartsila 31DF and ME-GA engines, lowering the effective FuelEU WtW intensity by 1 to 3 gCO2eq/MJ.

Upstream WtT component: extraction, liquefaction, regas, transport

The WtT segment of LNG WtW intensity is materially larger than for VLSFO or MGO because LNG carries a liquefaction step that is energy-intensive and a regasification step that is a small additional charge. The MEPC.391(82) and FuelEU defaults consolidate this into approximately 18 to 22 gCO2eq/MJ for a global-average LNG supply chain, decomposing as follows.

Upstream extraction (gas production) contributes 8 to 12 gCO2eq/MJ depending on the field. The figure includes wellhead gas processing (acid gas removal, dehydration, NGL recovery), associated venting and flaring, and field-level methane fugitive emissions. Methane fugitives are the dominant variable: a 1 percent fugitive rate at the wellhead, pipeline, and processing chain adds approximately 8 gCO2eq/MJ at GWP100, swamping the CO2 component of upstream emissions. The IEA Methane Tracker reports global average upstream methane intensity at approximately 1.7 percent (range 0.5 to 4 percent across producers), and the MEPC.391(82) default uses a global-weighted figure near 1.0 percent.

Liquefaction adds 5 to 8 gCO2eq/MJ. The energy required to cool natural gas to minus 162 degrees Celsius and store it as a liquid at ambient pressure is approximately 8 to 12 percent of the gas’s energy content, supplied by either gas-driven turbines (most common in stand-alone liquefaction trains) or electricity (occasionally). Both routes produce CO2 emissions that scale with the liquefaction energy load. Modern coil-wound heat exchanger trains (Air Products C3MR, ConocoPhillips Optimised Cascade, Shell DMR) achieve liquefaction at the lower end; older single-mixed-refrigerant trains run at the upper end.

LNG shipping contributes 0.5 to 1.5 gCO2eq/MJ, depending on voyage distance, vessel propulsion (steam turbine, dual-fuel diesel-electric, two-stroke LP-DF, two-stroke ME-GI), and BOG management strategy. A modern Q-Flex carrier on the Qatar to Europe route at full BOG-fuelled propulsion runs near 0.7 gCO2eq/MJ; an older steam turbine vessel on the Qatar to Japan route can reach 1.5 gCO2eq/MJ.

Regasification at the import terminal adds 0.2 to 0.6 gCO2eq/MJ. The regas energy is small (1 to 2 percent of the LNG energy content) and is typically supplied by seawater heat exchange, vaporiser-fired natural gas, or open-rack vaporisers using ambient air. Floating storage and regasification units (FSRUs) carry slightly higher regas penalties because of the on-board boil-off and propulsion overhead.

Distribution and bunkering add 0.3 to 0.8 gCO2eq/MJ. Bunker barge transport, ship-to-ship transfer losses, and small storage venting all contribute. The modern STS LNG bunkering operation with vapour return and inerted lines runs at the lower end; a truck-to-ship bunkering at a small-scale terminal runs at the upper end.

The total WtT bill of approximately 18 to 22 gCO2eq/MJ is therefore roughly double the WtT for petroleum-derived marine fuels (8 to 11 gCO2eq/MJ for VLSFO or MGO), because the liquefaction step and the upstream methane fugitives have no analogue in the petroleum chain at comparable magnitude.

Sourcing variability: US shale, Qatar, Australia NW Shelf, Russia Yamal

The 18 to 22 gCO2eq/MJ default WtT figure is a global-average composite. Actual WtT depends substantially on the sourcing basin, and operators with full chain-of-custody disclosure can certify lower or higher values for individual cargoes.

US shale gas (Henry Hub feedstock). US export LNG drawn from Permian, Marcellus, Haynesville, and Eagle Ford basin gas runs through liquefaction trains on the US Gulf Coast (Sabine Pass, Cameron, Corpus Christi, Freeport, Plaquemines) and the East Coast (Cove Point) at modern coil-wound efficiency. Upstream methane intensity is variable: Permian basin gas carries fugitive rates of 2 to 4 percent (above the MEPC.391(82) default), Marcellus gas runs near 0.5 to 1.5 percent. Cargo-level WtT therefore spans roughly 18 to 26 gCO2eq/MJ, with the high end driven by Permian fugitive-leak performance. EPA methane regulations under the Inflation Reduction Act and Methane Emissions Reduction Programme are tightening this range from 2025 onward.

Qatar (North Field gas). Qatari LNG draws from the offshore North Field, the world’s largest single non-associated gas reservoir. Field-level CO2 venting from gas processing is significant (the North Field gas contains 4 to 5 percent CO2 that must be removed before liquefaction), but methane fugitive performance is low (under 0.5 percent) because the integrated Qatargas and RasGas operations run modern leak detection and repair. Liquefaction at Ras Laffan uses Air Products C3MR trains at high efficiency. Cargo-level WtT for Qatari LNG runs 15 to 18 gCO2eq/MJ, below the global default. With the North Field Expansion (NFE) bringing 32 mtpa of new capacity online by 2027 and integrated CCS at the new trains, Qatari WtT is projected to fall further.

Australia North West Shelf and Gorgon (Australian LNG). Australian export LNG comes from the North West Shelf (Karratha), Pluto, Gorgon, Wheatstone, Ichthys, and Prelude FLNG. Upstream methane performance is generally good (Australian regulations and the use of associated gas from oil developments minimise field-level fugitives). Liquefaction CO2 varies: Gorgon runs CO2 injection (CCS) on the high-CO2 reservoir gas, partially offsetting the elevated raw gas CO2 content. North West Shelf and Pluto run conventional CO2 venting. Cargo-level WtT spans 17 to 22 gCO2eq/MJ, in line with the global default.

Russia Yamal (Yamal LNG, Arctic LNG 2). Yamal LNG and Arctic LNG 2 export cargoes from the Russian Arctic via ice-class LNG carriers and trans-shipment terminals. Upstream methane performance from Russian gas fields has historically been poor (fugitive rates 2 to 5 percent at field level on independent satellite measurements), and the long shipping leg from Sabetta or Murmansk to Asian or European markets adds 1.0 to 1.5 gCO2eq/MJ to the transport term. Cargo-level WtT runs 22 to 30 gCO2eq/MJ, above the global default. Russian LNG flows have been reduced significantly into European markets since 2022 sanctions but continue to East Asian buyers.

A modern CharterParty or Bunker Delivery Note increasingly references the GIIGNL disclosure framework or the MIQ certification standard for upstream methane intensity. Operators that want to claim a lower WtW intensity than the Annex 1 or Annex II default must produce a chain-of-custody trace and a third-party verifier audit, which is the same architecture that applies to bio-LNG and renewable LNG pathways.

AR5 GWP100 vs GWP20 sensitivity

The choice of global warming potential metric is the single largest sensitivity in any LNG WtW calculation, because LNG’s TtW intensity is dominated by methane slip and methane’s climate forcing depends strongly on the integration time horizon.

The IPCC Fifth Assessment Report (AR5) provides:

GWP100 = 28 for fossil CH4 (the time-integrated radiative forcing of 1 kg CH4 over 100 years, normalised to 1 kg CO2).
GWP20 = 84 for fossil CH4 (the same metric over 20 years).

The MEPC.391(82) Annex 1 and FuelEU Maritime Annex II both mandate AR5 GWP100 as the reporting metric. The choice was made by the IMO LCA Guidelines Working Group at MEPC 80 and confirmed in the final resolution at MEPC 82, with the rationale that 100-year horizons align with national greenhouse gas inventories under UNFCCC and with cumulative emissions targets in the Paris Agreement. Some climate scientists and NGOs argue that GWP20 better reflects near-term decarbonisation imperatives and the urgency of methane reductions; the regulatory choice nonetheless rests at GWP100.

The sensitivity is large. Recasting the LNG-Otto four-stroke default at GWP20 instead of GWP100:

TtW CH4 slip (3.0 percent at GWP20 of 84): 25.2 gCO2eq/MJ (versus 8.4 at GWP100)
WtW total at GWP20: approximately 105 gCO2eq/MJ (versus 87 at GWP100)

For LNG-Diesel, the recast is:

TtW CH4 slip (0.2 percent at GWP20 of 84): 1.7 gCO2eq/MJ (versus 0.6 at GWP100)
WtW total at GWP20: approximately 78 gCO2eq/MJ (versus 76 at GWP100)

At GWP20, an LNG-Otto four-stroke vessel reads as higher WtW intensity than VLSFO (which sits near 91 at GWP100 and changes only marginally at GWP20 because VLSFO has negligible methane content). At GWP100, the same vessel reads as lower WtW intensity than VLSFO. The conclusion the regulator and the operator draw from the LNG investment depends entirely on the time horizon embedded in the metric.

LNG-Diesel performance is robust under both GWP100 and GWP20 because the slip term is small enough that the multiplier choice does not move the answer materially. This is the structural reason why ME-GI is increasingly preferred in newbuilding orders where long-asset-life FuelEU and IMO GFS exposure is being underwritten.

The IMO LCA Guidelines retain GWP100 as the binding metric, and this page treats GWP100 as the reporting basis throughout, with GWP20 disclosed only for sensitivity. Operators that want to publish dual-metric intensity figures for sustainability reporting (TCFD, CSRD, GRI) often disclose both, but the regulatory compliance number is the GWP100 figure.

Bio-LNG and renewable LNG (brief, with cross-link)

Bio-LNG (biomethane liquefied) and synthetic LNG (e-LNG, methane synthesised from green hydrogen and captured CO2) share the same downstream combustion physics as fossil LNG. The methane slip percentages, the engine technology distinction, and the WtW math at the funnel are identical. What differs is the WtT component.

Bio-LNG WtT depends on the feedstock pathway. Anaerobic digestion of agricultural residues, food waste, or landfill gas, followed by upgrading to biomethane and liquefaction, can deliver WtW intensities of 15 to 35 gCO2eq/MJ under MEPC.391(82) and FuelEU pathways, depending on feedstock and methane leakage during digestion. Some pathways achieve negative WtW intensity once avoided emissions credits from manure management are included, although the IMO LCA Guidelines and FuelEU treat negative-intensity claims with caution and require RED III sustainability certification.

Synthetic LNG WtT depends on the source of hydrogen and CO2. A green hydrogen route (renewable electrolysis) coupled with biogenic CO2 captured from a biomass facility delivers WtW intensities of 5 to 15 gCO2eq/MJ. A blue hydrogen route (steam methane reforming with CCS) coupled with industrial CO2 capture delivers 30 to 50 gCO2eq/MJ, depending on capture efficiency and upstream gas methane leakage.

Both bio-LNG and synthetic LNG pathways are addressed in detail on dedicated wiki pages; the /calculators/fuel-wtw-bio-lng calculator parameterises feedstock and digestion pathway. Drop-in substitution into existing LNG-fuelled vessels is technically straightforward (the molecule is identical to fossil LNG once delivered to the bunker manifold), and operators increasingly blend bio-LNG into fossil LNG bunkers to lower the effective FuelEU intensity. The /calculators/fuel-wtw-blend calculator handles the energy-weighted intensity arithmetic for any fossil and bio-LNG blend.

The methane-slip story remains unchanged: a vessel burning 100 percent bio-LNG in a Wartsila 50DF still slips 3 percent of its methane to the funnel. The slip is fossil-equivalent in its climate forcing because the bio-LNG molecule is chemically identical to the fossil LNG molecule. The bio-LNG WtW advantage comes entirely from the WtT side.

Commercial bunkering and Type C tanks; BOG management

LNG bunkering infrastructure has expanded substantially since 2018, driven by the growth of the LNG-fuelled merchant fleet and the supporting investment by major bunker suppliers and gas majors.

Bunkering pathways. Three principal bunkering modes exist: ship-to-ship (STS), truck-to-ship (TTS), and pipeline-to-ship (PTS). STS using purpose-built LNG bunker vessels is dominant at major hubs. Operators such as Shell, TotalEnergies, Hoegh LNG (HoeghLNG), Pavilion Energy, Avenir LNG, and Titan LNG operate fleets of small-scale LNG bunker vessels at Rotterdam, Antwerp, Singapore, Yokohama, Marseille, Algeciras, Gibraltar, and the US Gulf and East Coasts. Q-Flex carriers (the 210,000 to 217,000 cbm Qatari class) and Q-Max vessels (the 263,000 to 266,000 cbm class) are the upstream supply backbone, although they do not bunker directly into smaller vessels.

A typical bunker delivery note for marine LNG specifies the methane number (a knock-resistance metric for Otto-cycle engines, typically 65 to 90 depending on heavy-end hydrocarbon content), heating value (LHV near 49.0 MJ/kg, HHV near 54.5 MJ/kg), density at delivery temperature, and trace species (nitrogen, ethane, propane, butane, heavier hydrocarbons). The nitrogen content is significant because nitrogen in the BOG returns ballasts the methane number and can deliver knock-prone bunker into Otto-cycle engines.

Type C tanks are pressure-rated cylindrical or bilobe tanks fabricated from austenitic stainless steel, 9 percent nickel steel, or aluminium alloys, with a design pressure typically between 5 and 10 bar. They are the dominant containment system for LNG-fuelled vessel fuel storage on container ships, dry bulk, tankers, and ferries, because they accommodate boil-off through pressure rise during a holding period without requiring active reliquefaction. The IGC and IGF Codes set the regulatory framework. Tank capacities range from a few hundred cubic metres on inland vessels to 18,000 cbm on the largest dual-fuel container ships.

BOG management is the operational discipline of handling the natural evaporation of LNG inside fuel tanks during voyage and port stays. The pressure rise inside a Type C tank during a 7- to 14-day holding period is the primary mechanism that defines how often a vessel must consume gas to manage tank pressure. Three principal management strategies are deployed:

Gas consumption mode: the engine burns gas at sufficient rate to manage tank pressure. This is the design-condition operation for most LNG-fuelled propulsion vessels.
Forced boil-off: at low engine load (port approach, manoeuvring), tank pressure can be reduced by forced vaporisation through a heater and gas consumption by auxiliary engines or gas combustion units (GCUs).
Reliquefaction: on LNG carriers and some large LNG-fuelled vessels, on-board reliquefaction plant returns BOG to the cargo or fuel tank as liquid.

Holding-period planning matters for WtW accounting because BOG that is vented overboard (rare on modern vessels but occasionally seen at long port stays without a GCU) is a direct methane emission and is counted at the GWP100 of 28 in any WtW disclosure. Operators are increasingly required to log and verify BOG handling under FuelEU MRV.

Formula, assumptions, and limits

Formula

The WtW intensity for LNG in marine dual-fuel engines under MEPC.391(82) and FuelEU Annex II is constructed as the sum of upstream WtT plus combustion CO2 plus methane slip times GWP100 plus N2O times GWP100:

\text{EF}_{\text{WtW,LNG-Otto}} = \text{EF}_{\text{WtT,LNG}} + \text{EF}_{\text{TtW,CO}_2} + \text{GWP}_{100,\text{CH}_4} \cdot \text{slip}_{\text{Otto}} \cdot \text{factor}

The cycle-pair gap is the slip-difference times GWP100:

\Delta\text{EF}_{\text{cycle}} = \text{GWP}_{100,\text{CH}_4} \cdot (\text{slip}_{\text{Otto}} - \text{slip}_{\text{Diesel}})

Substituting representative values:

\Delta\text{EF} = 28 \cdot (0.030 - 0.002) \cdot 49.0 / 1.0 \approx 38 \text{ g CO}_2\text{eq/kg fuel} \approx 7.7 \text{ gCO}_2\text{eq/MJ}

(where 49.0 MJ/kg is the LCV of LNG and the factor handles unit conversion from g CH4 per g fuel to g CO2eq per MJ of fuel energy).

Derivation

The TtW carbon dioxide intensity for LNG comes from the molar mass ratio. Methane (CH4) has a carbon mass fraction of $C_f = 12 / 16 = 0.75$ . Combusting 1 kg of methane produces $0.75 \cdot 44/12 = 2.75$ kg of CO2. Dividing by an LCV of 49.0 MJ/kg gives a TtW CO2 intensity of $2750 / 49 = 56.1$ gCO2/MJ. This is the pure CO2 combustion term in the MEPC.391(82) defaults.

The TtW methane slip intensity is constructed as:

\text{EF}_{\text{slip}} = \text{slip} \cdot \text{LCV}^{-1} \cdot 1000 \cdot \text{GWP}_{100}

where slip is the fraction of fuel mass passing unburned, LCV is in MJ/kg, and the factor 1000 converts g to mg per MJ. For Otto-cycle slip of 3.0 percent and GWP100 of 28:

\text{EF}_{\text{slip}} = 0.030 \cdot (1/49.0) \cdot 1000 \cdot 28 = 17.1 \text{ mgCH}_4\text{eq/MJ} \cdot 28 / 1000 = 8.4 \text{ gCO}_2\text{eq/MJ}

The WtT intensity is constructed as the sum of upstream extraction, processing, liquefaction, shipping, regasification, and distribution emissions per MJ of delivered LNG energy. Each stage has an empirical default in MEPC.391(82) Annex 1 and FuelEU Annex II, calibrated against the GIIGNL, IEA, and JEC datasets.

Assumptions

The published defaults rest on five principal assumptions:

Global-average upstream methane intensity of approximately 1.0 percent across the LNG supply chain. Actual values vary from 0.3 percent (Qatar with integrated LDAR) to 4 percent (older Russian and some US shale fields).
Modern liquefaction efficiency with energy consumption near 8 to 10 percent of LNG energy content. Older trains run higher.
Average shipping distance consistent with the global LNG trade pattern; Atlantic and Pacific basin trade legs and Q-Flex Q-Max vessel mix. Specific cargo voyages can be shorter or longer.
AR5 GWP100 of 28 for fossil CH4 as the binding climate metric. GWP20 of 84 doubles to triples the slip-driven WtW intensity component.
Engine slip at design condition. Part-load slip can be 50 to 100 percent higher than design slip, particularly for four-stroke Otto-cycle engines below 50 percent MCR.

Worked example

Consider a 23,000 TEU container ship with twin MAN ME-GA two-stroke main engines, total fuel consumption 240 t LNG per day at typical service speed.

Energy consumed per day:

E = 240 \times 10^3 \text{ kg} \cdot 49.0 \text{ MJ/kg} = 11.76 \text{ TJ}

WtW intensity at MEPC.391(82) Annex 1 default for LNG-Otto two-stroke:

\text{EF}_{\text{WtW}} = 20 \text{ (WtT)} + 56.1 \text{ (CO}_2\text{)} + 4.8 \text{ (slip)} + 0.1 \text{ (N}_2\text{O)} = 81.0 \text{ gCO}_2\text{eq/MJ}

Daily WtW emissions:

M = 11.76 \times 10^{12} \text{ J} \cdot 81.0 \times 10^{-9} \text{ tCO}_2\text{eq/J} = 952 \text{ tCO}_2\text{eq/day}

Replacing the same vessel with an MAN ME-GI Diesel-cycle engine at the same fuel consumption (LNG-Diesel WtW 76.4 gCO2eq/MJ):

M = 11.76 \times 10^{12} \cdot 76.4 \times 10^{-9} = 898 \text{ tCO}_2\text{eq/day}

The daily WtW gap is 54 tCO2eq, or approximately 20,000 tCO2eq per year at 365 days operating profile. At a 2026 EU ETS price of approximately EUR 90 per tCO2eq and a FuelEU non-compliance penalty in the range of EUR 200 to 300 per tonne of avoided CO2eq for over-cap operation, the annual GHG cost differential between an Otto-cycle and a Diesel-cycle engine on the same vessel can exceed EUR 4 to 6 million per ship.

Edge cases and limits

The defaults break down in four operational regimes.

Sustained part-load operation. Slow steaming below 50 percent MCR for extended periods raises Otto-cycle slip by 50 to 100 percent above design. The MEPC.391(82) Annex 1 defaults, calibrated at approximately 75 percent MCR, understate the actual WtW intensity for vessels operating in sustained slow-steaming mode. Operators with verifier-certified actual slip data can capture or expose this effect; the default does not.

Extreme upstream methane events. Satellite-detected methane super-emitter events in supply basins (Permian flaring incidents, Arctic compressor station leaks) can raise the cargo-level WtT by 5 to 15 gCO2eq/MJ for affected cargoes. The Annex 1 default does not adjust for cargo-specific deviations.

Boil-off venting. A vessel that vents BOG at port (rare under modern operating practice but occasionally seen on older ferries or short-call vessels without GCU) effectively adds a methane emission term not captured in the default. FuelEU MRV is increasingly tightening reporting on BOG venting.

LNG-Diesel low-load CH4 slip. Even ME-GI engines at sub-25 percent load can show slip rising to 0.7 to 1.0 percent because the diffusion flame thinning at low fuel injection rates exposes more flame surface to wall quench. The default of 0.2 percent slip is calibrated at design load and underrepresents low-load behaviour.

Regulatory basis

MEPC.391(82): IMO LCA Guidelines and Annex 1 default emission factor table for LNG by engine technology, mandatory under the IMO Net-Zero Framework from 2027 onward.
MEPC 83 outcomes: GHG Fuel Standard adoption with engine-technology-resolved LNG defaults in scope from 2027.
Regulation (EU) 2023/1805 (FuelEU Maritime): Annex II default factors for LNG by engine technology, in force since 2025.
Regulation (EU) 2015/757 (EU MRV): monitoring and reporting basis for ship-level emission data, including BOG and bunker delivery records.
MARPOL Annex VI Regulation 13 (NOx Tier III): Otto-cycle lean-burn engines achieve Tier III without aftertreatment; Diesel-cycle engines need EGR or SCR.
MARPOL Annex VI Regulation 14: SOx is essentially zero from LNG; sulphur compliance is automatic.
IGF Code: safety standard for low-flashpoint fuels including LNG, governing tank, piping, ventilation, and bunkering arrangements.
IGC Code: safety standard for gas carriers, applicable to LNG carriers using cargo as fuel.

Common errors

Five recurring errors appear in operator and consultant LNG WtW arithmetic:

Treating LNG as a single pathway. Otto and Diesel cycles produce materially different WtW intensities and must be reported separately; a single “LNG” line in a fleet report misrepresents the carbon position.
Using GWP20 alongside MEPC.391(82) defaults. The defaults are calibrated at GWP100; mixing GWP20 sensitivity numbers with GWP100 defaults produces double-counted slip terms.
Ignoring cargo-level WtT variability. Quoting the global default of 20 gCO2eq/MJ for a Russian Yamal cargo or a Qatar cargo without sourcing-specific adjustment misrepresents the true intensity by 5 to 10 gCO2eq/MJ either way.
Conflating methane number with methane slip. Methane number is a fuel knock metric (Otto-cycle only); methane slip is an engine combustion outcome. They are distinct concepts.
Reporting funnel CO2 as WtW. Funnel CO2 is the TtW component only; quoting it as WtW omits 18 to 22 gCO2eq/MJ of upstream burden.