Why measurement is estimation
Nobody can put a meter on a shipment and read off its carbon footprint. A parcel or pallet shares a vehicle with other cargo, travels through several handling points and may change modes along the way. So freight emissions are almost always estimated from other data using agreed methods. The results are useful for comparison and tracking, but they carry uncertainty, and it is worth remembering that when reading any single figure.
Emissions are usually reported as carbon dioxide equivalent, often written CO2e. This combines carbon dioxide with other greenhouse gases, converted to a common scale, so that different gases can be added together.
The basic building blocks
Most methods combine two ingredients: activity data and emission factors.
- Activity data describes what happened: how much fuel was burned, or how much cargo moved and over what distance.
- Emission factors convert that activity into emissions, for example the kilograms of CO2e produced per liter of a given fuel or per tonne of cargo moved one kilometer.
Emission factors come from published databases and industry methodologies. They differ by fuel, vehicle type, load and region, so choosing one that matches the real situation matters.
Fuel-based method
When a carrier or fleet knows how much fuel it used, the simplest approach is to multiply the fuel by a fuel-specific factor. This is generally considered more accurate for a whole fleet, because it starts from what was actually consumed. The harder step is dividing the total among individual shipments. That usually means allocating by weight, volume, pallet spaces or distance, and the allocation rule can change the answer noticeably.
Distance-based method
When fuel data is not available, many shippers estimate with the tonne-kilometer approach. A tonne-kilometer is one tonne of cargo moved one kilometer. The calculation is:
emissions = weight (tonnes) x distance (km) x emission factor (per tonne-km)
Here is a hypothetical example, with an invented factor used only to show the arithmetic. Suppose a shipment weighs 10 tonnes and travels 500 km. That is 10 x 500 = 5,000 tonne-km. If the assumed factor were 0.1 kg CO2e per tonne-km, the estimate would be 5,000 x 0.1 = 500 kg CO2e. Real factors depend on mode, vehicle and load, and should come from a recognized source rather than from this example.
For volume-heavy but light cargo, the weight used is often a chargeable or volumetric weight, or the estimate is based on pallet positions. Which one is appropriate depends on the method adopted.
Mode matters
The same cargo and distance can produce quite different results depending on the mode. Ocean shipping and rail freight typically have lower emissions per tonne-kilometer than road, and air freight typically has much higher ones. Because actual distances also differ by mode, with road routes often more direct than rail and ships taking indirect sea lanes, estimates generally use the real route distance rather than a straight line. Many methods apply a distance adjustment to account for this.
Load factor also has a large effect. A truck running half full or returning empty spreads the same fuel over less cargo, so each tonne carried carries more emissions. Good methods account for empty running and typical utilization, or at least state their assumptions.
System boundaries
Another key choice is what the estimate includes. Some methods count only the fuel burned or energy used by the vehicle, often called tank-to-wheel. Others also include the emissions from producing and delivering the fuel or electricity, often called well-to-wheel. A figure that includes upstream emissions will be higher than one that does not, even for the same trip. Terminals, warehouses and transshipment points can also be included or left out.
In corporate reporting, transport bought from carriers generally falls under indirect emissions in the value chain, while a company’s own vehicles count as direct emissions. Published frameworks, including industry guidance and international standards for transport chain emissions, exist to make methods consistent. Following one consistently matters more than which one is picked, because it keeps year-to-year comparisons meaningful.
Data quality and its limits
Estimates range from rough to detailed. At the rough end, a shipper applies average factors to shipment weight and distance. At the detailed end, carriers supply measured fuel use and actual load data per trip. The detailed approach takes more effort, and better supply chain visibility and data analytics make it more practical, but it is not always available, especially with many small carriers.
Common problems include missing distances, unknown vehicle types, estimated weights, and different carriers using different methods. When comparing two quotes or two years of results, check that the same boundary, factors and allocation rules were used. Otherwise a gap in numbers may reflect method differences rather than real change.
Using the numbers
Once a baseline exists, emissions figures can be tracked alongside logistics KPIs such as cost and delivery performance. They can show which lanes, customers or modes contribute most, and they help test whether a change, such as switching modes or consolidating loads, is likely to help. Treat the output as a guide for decisions, not a precise ledger, and revisit methods as better data becomes available.