Car vs Train vs Plane: How to Compare Travel Emissions Fairly

Compare passenger journeys, not vehicle labels

The common functional unit is a passenger-kilometre: one passenger transported one kilometre. Calculate each realistic route from door to door. A flight comparison should include surface travel to and from airports; rail should use the actual network route rather than straight-line distance; a car should use the driven route. If one option requires a hotel or a major detour, record that separately rather than burying it in a generic factor.

Then define the boundary. Tank-to-wheel covers emissions from operating a vehicle. Well-to-tank adds fuel or electricity supply. A lifecycle comparison can add vehicle and infrastructure manufacturing, maintenance, and end of life. These boundaries answer different questions. Use one consistently across modes or display layers separately—never compare a lifecycle rail figure with a flight's direct CO2 and call the result complete.

Cars: occupancy can change the personal result dramatically

For a combustion car, measured fuel over the trip or annual fuel economy is stronger than a vehicle-class average. Multiply fuel by the relevant combustion factor and, if the chosen boundary includes it, add upstream fuel supply. Divide shared trip emissions by occupants only when they are genuinely making the same journey; an extra passenger does not halve manufacturing emissions in an annual ownership study by itself.

For an electric car, use metered charging if available, include charging losses consistently, and apply an electricity factor matched to place and time. Tailpipe CO2 is zero, but power generation and vehicle production are not. Battery size, vehicle mass, lifetime kilometres, and grid mix influence lifecycle comparisons. An electric car is not automatically worse on a high-carbon grid or automatically impact-free on a clean one—the calculation needs the actual case.

Trains: electricity, load, and service type matter

Electric rail emissions depend on the power system and the operator's energy use per passenger. Diesel rail has direct fuel emissions. High-speed, intercity, regional, metro, and sleeper services have different speed, stopping pattern, equipment, and occupancy. Official operator or national factors are preferable to importing a UK value into another country without a proxy label.

Average factors allocate a service's energy across typical passenger load. Your boarding a scheduled train does not immediately change how much electricity it uses, but passenger-kilometre averages remain useful for planning and comparing demand at scale. Infrastructure construction can matter in a full lifecycle study, especially for new lines, yet it should be annualised over use and included for competing road and aviation infrastructure too.

Flights: airport pair and cabin class are not optional details

Aircraft fuel is used for take-off, climb, cruise, descent, and ground operations. The relationship with distance is not linear, and route, aircraft, passenger load, and freight allocation matter. ICAO's Carbon Emissions Calculator uses airport-pair and operational data to estimate passenger CO2. Its 2026 version distinguishes economy, premium economy, business, and first class.

Cabin class matters because premium seats generally use more floor area and reduce the number of passengers among whom the flight's allocated fuel is shared. Multi-leg itineraries also repeat take-off and routing overheads. Calculate each segment rather than entering only the distance between the first and final city. Do not compare a one-way flight with a round-trip rail journey by accident.

Aviation CO2 and non-CO2 effects should be shown separately

ICAO ICEC estimates CO2. Aviation also affects climate through nitrogen oxides, water vapour, contrails, and induced cloudiness, with effects that depend on altitude, location, weather, and time. A carbon calculator may add a non-CO2 scenario or uplift based on a declared scientific method, but labelling the combined number simply as measured flight CO2 is misleading.

Keep at least two lines: route-specific CO2 and an optional wider climate scenario. State whether upstream fuel supply is included and whether the non-CO2 estimate uses a multiplier, effective radiative forcing, or another basis. Because uncertainty is larger than for fuel CO2, a range is more honest than a precise surcharge. Carbon offsets should not be subtracted from gross trip emissions; any credit claim belongs in a separate field with its own quality assessment.

Build the comparison in seven steps

Write the exact origin, destination, date range, passengers, and luggage needs. Map realistic door-to-door itineraries. Record driven kilometres and occupancy, rail service and route, and every flight segment with cabin. Choose direct-only, direct plus energy supply, or lifecycle boundaries. Apply factors from the appropriate geography and year. Divide only genuinely shared emissions. Finally, display CO2, full CO2e, and aviation non-CO2 scenarios with their own labels.

Test sensitivity before deciding. Change car occupancy from one to three, compare an airport connection with a direct train, or test economy rather than business class. If travel time changes the number of hotel nights or meals, show those optional additions rather than hard-coding them into transport factors. Cost and accessibility belong beside emissions; a lower number that is unaffordable or physically inaccessible is not a usable recommendation.

• Use actual route kilometres, not one straight-line distance for every mode.
• Use either fuel or vehicle-km for a car calculation unless one validates the other.
• Match country, year, and lifecycle boundary across factors.
• Keep route CO2 separate from uncertain aviation non-CO2 effects.
• Save occupancy and cabin class with the result so it can be reproduced.

From travel carbon to an ecological footprint

A travel carbon result in tCO2e is not converted wholesale into gha. In a top-down Ecological Footprint model, mobility answers adjust relevant country-baseline cells. Fossil CO2 can inform the carbon-uptake component under the accounts method, while built-up land, infrastructure, services, and other land demands remain separately allocated. Full CO2e and non-CO2 aviation effects remain parallel climate indicators.

EcoSi Footprint is an independent beta estimate and is not affiliated with Global Footprint Network. Travel results should expose factor source, route, occupancy, cabin, boundary, and scenario range. UK DESNZ conversion factors are suitable for UK activities and useful examples of a documented factor set, but any use elsewhere must be identified as a proxy. Transparency is more valuable than a global ranking built from mismatched inputs.