Ask whether LNG as a fuel helps the climate and the honest answer is: it depends on the engine, and the difference between engines is not a rounding error.
Why methane slip exists at all
Not every gas engine burns its fuel the same way, and the difference in combustion strategy is exactly what determines how much escapes unburned.
Diffusion combustion, used in ME-GI-type engines, injects gas directly into already-compressed, already-hot air late in the cycle. The gas burns as it mixes, at the flame front, with very little opportunity to sit unburned anywhere.
Premixed lean-burn combustion, used in most dual-fuel medium-speed engines and, to a lesser degree, in X-DF two-stroke engines, mixes gas and air before ignition. Some of that premixed charge finds its way into crevices around piston rings and combustion chamber corners where the flame cannot reach, and it survives the cycle unburned.
The physical difference — when the gas meets the flame — is the entire explanation for why slip rates vary so much between engine families using ostensibly the same fuel.
Why a small fraction matters disproportionately
Methane is a far more effective greenhouse gas than carbon dioxide, molecule for molecule, over the time horizons usually used in climate accounting — commonly cited at around 80 times more potent over a twenty-year window, falling to a lower but still substantial multiple over a hundred years.
That means a slip rate of only a few per cent can erase a large part of the carbon dioxide advantage LNG has over heavier fuels. The combustion arithmetic — burning methane produces less CO2 per unit of energy than burning fuel oil — is not in question. Whether the full-cycle climate outcome actually improves depends on how much methane escaped along the way, and that is a question about engines and equipment, not about the fuel itself.
Why there is no single answer
“Does LNG reduce emissions” is not a question with one number, because the answer depends on at least three choices that vary by ship, by route and by who is doing the accounting.
Which engine. A diffusion-cycle engine and a lean-burn engine on the same route, burning the same fuel, can produce meaningfully different full-cycle climate outcomes.
Which time horizon. Methane’s warming effect is front-loaded — it is far more potent in the first two decades after release than a hundred years later, because it breaks down in the atmosphere faster than carbon dioxide does. A twenty-year accounting makes slip look far worse than a hundred-year accounting does. Neither horizon is more “correct”; they answer different questions about urgency.
Where the boundary is drawn. An analysis stopping at the ship’s exhaust ignores everything upstream. One that includes production and transmission leakage, discussed further in lifecycle emissions, can produce a very different picture.
Anyone quoting a confident single figure for “LNG’s emissions benefit” has made all three choices, usually without saying so.
Slip is not the only leak
Methane escapes the chain in places that have nothing to do with an engine. Valves, flanges, seals and vents at production sites, processing plants and along transmission pipelines all leak, continuously, in quantities that are notoriously difficult to measure directly and are frequently found to be higher than official inventories assume when independently monitored — for instance by satellite.
That upstream leakage is a separate topic from engine slip, covered in fugitive emissions and lifecycle accounting, and it is frequently the larger of the two effects. Engine slip is easier to discuss precisely because it can be measured on a test bed for a specific, known machine; upstream leakage resists that kind of precision, which does not make it smaller.
What this means for reading a claim
A statement that “LNG-fuelled shipping cuts emissions by X per cent” is only as good as its assumptions about engine type, time horizon and system boundary. The honest version of the claim names all three. Where this site records propulsion type for a vessel, that field is exactly the piece of information needed to ask the first of those three questions about a specific ship — though it does not answer the other two, and this site does not attempt an emissions estimate of its own for any vessel or route.