Air freight carbon emissions: how to measure and interpret a shipment
The original synthetic editorial scene shows an unbranded cargo aircraft at dawn beside wrapped pallets and containers, with an open loading area. It provides airport-operation context only; it does not identify the fictional AF-204 shipment, verify a carrier or establish measured emissions.
Diagram text description
Air freight carbon emissions for a shipment are usually an allocated estimate of the flight activity associated with moving that cargo. The number depends on the shipment, route, aircraft service, allocation method and emissions boundary. It is not a sensor reading taken from the package.
Start by identifying the shipment and setting the boundary. Collect its activity records, choose a suitable calculation method, allocate the relevant emissions, then review the result and update it when better evidence arrives. Keep what is known separate from what the method models: a declared origin and destination do not establish the actual routing or fuel burned.
The practical aim is a reconstructable result that helps a shipping decision. This guide explains the evidence choices and works through a fictional two-leg shipment. Its arithmetic demonstrates compatible units and acceptance checks; the invented coefficient is neither an official default nor a measured carrier factor. A result can be useful for learning while remaining unsuitable for an operational or public claim.
First define the shipment and emissions boundary
Define the movement and the result before collecting factors. Is the question about an airport-to-airport flight, or the complete service from collection to delivery? Does the result describe CO2 from fuel combustion, tank-to-wheel CO2e, or well-to-wheel CO2e including upstream fuel production? These choices affect what may be added and compared.
A flight-only CO2 result and a wider door-to-door CO2e result cover different activities and potentially different gases. A wider service may include collection trucks, flights, storage and delivery. However, CO2e does not automatically mean door to door, and a door-to-door result does not by itself establish a lifecycle fuel boundary. State the logistics coverage and the gas/fuel basis separately.
The left group represents a flight-only CO2 result for the air movement. The right group represents an explicitly wider service CO2e result that includes road, air and storage under its selected lifecycle method. These are examples of different choices, not definitions that every CO2e result is door to door. Align gases, fuel boundary and logistics coverage before comparing totals.
Diagram text description
ICAO’s Air Freighter methodology, Version 2 dated July 2024, describes a predictive flight CO2 estimate for dedicated freighters. It does not calculate the shipment’s other transport modes. Its modelled answer should not be renamed a measured flight result or a complete logistics CO2e footprint.
Write the boundary beside the number, including the origin and destination, included legs and excluded activities. Where non-CO2 aviation effects are relevant to the decision, identify whether the chosen method addresses them. Do not assume a generic uplift is included or append one without a documented basis. Matching unit symbols alone is insufficient when the underlying coverage differs.
Collect the activity data that changes the answer
Build the activity record around one shipment ID and distinct leg IDs. Keep the original documents and distinguish actual operational evidence from planned routing or modelling assumptions. A forwarder’s invoice and an air waybill can describe the same movement; two documents do not necessarily represent two activities.
| Data field | Evidence to request | Gap or compatibility check |
|---|---|---|
| Shipment and mass | Shipment ID, packing or weight record, shipped kilograms and date | Distinguish physical shipped mass from chargeable weight; use the basis required by the method |
| Route and legs | Origin, hub, destination, flight or routing records where known | Mark planned, modelled and actual distance bases separately |
| Aircraft service | Dedicated freighter or passenger belly cargo; aircraft type where available | Hold a service-specific coefficient if the service is unknown |
| Operator and timing | Carrier, service date, operational evidence and document identity | Confirm that the data apply to this shipment and period |
| Other logistics | Collection, delivery and warehouse activity if included | Retain separate records; missing data do not mean zero emissions |
Table 1, shown one row at a time. Every field is retained.
Row 1
- Data field
- Shipment and mass
- Evidence to request
- Shipment ID, packing or weight record, shipped kilograms and date
- Gap or compatibility check
- Distinguish physical shipped mass from chargeable weight; use the basis required by the method
Row 2
- Data field
- Route and legs
- Evidence to request
- Origin, hub, destination, flight or routing records where known
- Gap or compatibility check
- Mark planned, modelled and actual distance bases separately
Row 3
- Data field
- Aircraft service
- Evidence to request
- Dedicated freighter or passenger belly cargo; aircraft type where available
- Gap or compatibility check
- Hold a service-specific coefficient if the service is unknown
Row 4
- Data field
- Operator and timing
- Evidence to request
- Carrier, service date, operational evidence and document identity
- Gap or compatibility check
- Confirm that the data apply to this shipment and period
Row 5
- Data field
- Other logistics
- Evidence to request
- Collection, delivery and warehouse activity if included
- Gap or compatibility check
- Retain separate records; missing data do not mean zero emissions
A handling unit can include packaging, pallets or containers whose treatment depends on the method. Record the mass definition rather than changing it silently to match a convenient coefficient. Preserve units when converting a carrier file into an internal table.
For a transfer shipment, record each actual or assumed air leg instead of substituting the direct origin-to-destination distance without explanation. If the carrier later supplies a different route, retain the original estimate and identify what changed. This makes a revised result explainable rather than merely different.
Choose the most defensible calculation route
Choose the calculation route that best matches the available evidence and intended use. Suitable carrier fuel and operational data can support a more specific allocation than a generic estimate. They still need a defined boundary, method and shipment relationship; a carrier logo is not sufficient evidence of those choices.
| Route | What supports the calculation | Main limitation |
|---|---|---|
| Carrier operational evidence | Applicable fuel, flight activity, cargo loads and documented allocation | Confirm coverage and allocation; company-wide averages are not automatically shipment-specific |
| Route or aircraft model | Shipment mass, route assumptions, aircraft/service information and model version | Predictive routing, fleet and load assumptions may differ from actual operation |
| Mass-distance estimate | Tonnes shipped, leg kilometres and a compatible emissions factor | Factor service, geography, period and boundary must match the activity |
| Spend screening | Freight expenditure and a documented monetary factor | Price changes and service mix limit physical interpretation; label it screening |
Table 2, shown one row at a time. Every field is retained.
Row 1
- Route
- Carrier operational evidence
- What supports the calculation
- Applicable fuel, flight activity, cargo loads and documented allocation
- Main limitation
- Confirm coverage and allocation; company-wide averages are not automatically shipment-specific
Row 2
- Route
- Route or aircraft model
- What supports the calculation
- Shipment mass, route assumptions, aircraft/service information and model version
- Main limitation
- Predictive routing, fleet and load assumptions may differ from actual operation
Row 3
- Route
- Mass-distance estimate
- What supports the calculation
- Tonnes shipped, leg kilometres and a compatible emissions factor
- Main limitation
- Factor service, geography, period and boundary must match the activity
Row 4
- Route
- Spend screening
- What supports the calculation
- Freight expenditure and a documented monetary factor
- Main limitation
- Price changes and service mix limit physical interpretation; label it screening
IATA’s cargo-emissions publisher page identifies Recommended Practice 1678 as a shipment-level methodology and describes updates adopted in 2022. That publisher description establishes the method’s identity; it is not a factor table or evidence that a particular carrier followed it for your shipment.
ICAO’s reviewed freighter edition uses model inputs, including routing, aircraft and load information, to predict emissions. The simple mass-distance exercise below is a separate teaching calculation, not a reproduction of that calculator. Keep the method version and the actual data source together so a reader can distinguish an operational estimate, a modelled estimate and a spend-based screen.
Freighter and passenger belly cargo are not the same
A dedicated freighter and passenger belly cargo allocate flight activity over different transported loads. A freighter carries cargo, while a passenger flight carries passengers, baggage and cargo under an applicable allocation rule. The same endpoints and distance cannot make those relationships interchangeable.
Belly cargo is not automatically assigned zero because the passenger flight was scheduled anyway. Equally, a dedicated-freighter coefficient should not be transferred to belly cargo without evidence that its method applies. Request the aircraft service and allocation basis rather than inferring them from an airline name.
The freighter side contains cargo and its load allocation. The passenger-flight side contains passengers, baggage and cargo sharing flight activity under a different rule. Belly cargo is not assigned zero simply because the passenger flight exists, and the same distance does not justify reusing a freighter factor. Check the applicable service and allocation before accepting a shipment result.
Diagram text description
Check the coefficient denominator. It may represent cargo tonne-kilometres after allocation, emissions per kilogram carried, or another defined output. If allocation is already embedded in the factor, adding a second arbitrary cargo share can distort the result. If you start from flight fuel instead, you need the method’s load and allocation evidence.
The fictional calculation uses a coefficient defined only for its invented dedicated-freighter service. No belly-cargo estimate follows from it. When the actual service is unknown, hold acceptance of that service-specific result until the evidence resolves the mismatch.
Work through a transparent shipment example
Explicitly fictional teaching example: shipment AF-204 carries 200 kg on two invented dedicated-freighter legs during May 2025. Physical mass is 0.2 tonnes. Distances are assumed air-leg distances, not verified flight tracks. Coefficient T-1 is invented as 0.75 kg CO2 per tonne-kilometre, already allocated to cargo, covering combustion CO2 only. It is not an IATA, ICAO or measured carrier factor.
| Leg identity | Assumed air distance | Mass-distance activity | Fictional calculation | Result |
|---|---|---|---|---|
| AF-204-L1: North Quay to Mesa Hub | 2,400 km | 0.2 t × 2,400 km = 480 t·km | 480 × 0.75 kg CO2/(t·km) | 360 kg CO2 |
| AF-204-L2: Mesa Hub to Lake Gate | 1,600 km | 0.2 t × 1,600 km = 320 t·km | 320 × 0.75 kg CO2/(t·km) | 240 kg CO2 |
| AF-204: both unique air legs | 4,000 km total | 480 + 320 = 800 t·km | 800 × 0.75 kg CO2/(t·km) | 600 kg CO2 |
| Factor acceptance control | Distances still assumed | 800 t·km | Real estimate is 800F until a sourced, route-appropriate F is justified | No real-shipment result established |
Table 3, shown one row at a time. Every field is retained.
Row 1
- Leg identity
- AF-204-L1: North Quay to Mesa Hub
- Assumed air distance
- 2,400 km
- Mass-distance activity
- 0.2 t × 2,400 km = 480 t·km
- Fictional calculation
- 480 × 0.75 kg CO2/(t·km)
- Result
- 360 kg CO2
Row 2
- Leg identity
- AF-204-L2: Mesa Hub to Lake Gate
- Assumed air distance
- 1,600 km
- Mass-distance activity
- 0.2 t × 1,600 km = 320 t·km
- Fictional calculation
- 320 × 0.75 kg CO2/(t·km)
- Result
- 240 kg CO2
Row 3
- Leg identity
- AF-204: both unique air legs
- Assumed air distance
- 4,000 km total
- Mass-distance activity
- 480 + 320 = 800 t·km
- Fictional calculation
- 800 × 0.75 kg CO2/(t·km)
- Result
- 600 kg CO2
Row 4
- Leg identity
- Factor acceptance control
- Assumed air distance
- Distances still assumed
- Mass-distance activity
- 800 t·km
- Fictional calculation
- Real estimate is 800F until a sourced, route-appropriate F is justified
- Result
- No real-shipment result established
The symbolic structure is 200 kg ÷ 1,000 = 0.2 t; multiplied by 4,000 km, this gives 800 tonne-kilometres. For a real shipment the result would be 800F, where F must be a justified, route-appropriate factor with matching units and scope. The 600 kg CO2 teaching answer exists only because this exercise defines an invented coefficient. It is not a defensible real-shipment result.
Apply two acceptance controls. First, an invoice and waybill both referring to AF-204-L1 must link to the same leg. Adding its 360 kg again would create a false 960 kg total. Second, entering 200 as tonnes instead of converting kilograms would produce 600,000 kg CO2, a thousandfold error; reject that unit mismatch.
For a separate fictional sensitivity, change only T-1 to 0.90 kg CO2/(t·km), with activity and boundary fixed. The answer becomes 720 kg CO2, 20% higher. No road legs, warehousing, upstream fuel emissions, other greenhouse gases or non-CO2 aviation effects are included. The coefficient does not apply to belly cargo. Replace the assumptions with compatible carrier evidence before relying on the estimate.
Compare transport choices on a like-for-like basis
Compare actual transport proposals after aligning the cargo and service being assessed. A lower-looking number may describe a shorter boundary, a different gas basis or an option that cannot meet the delivery requirement. Universal air-to-sea or air-to-rail multipliers conceal these differences.
| Comparison test | Evidence needed before deciding |
|---|---|
| Same cargo and destination | Equivalent quantity, handling needs and final delivery point |
| Complete route | Collection, transfers, main transport, storage and delivery included consistently |
| Same emissions basis | Matching gases, fuel lifecycle scope and reported unit |
| Comparable factor evidence | Applicable geography, service, period, load and allocation basis |
| Service feasibility | Delivery time, reliability, storage needs and product condition |
| Unknown field | Resolve the gap or hold the comparison; an unknown boundary is not a pass |
Table 4, shown one row at a time. Every field is retained.
Row 1
- Comparison test
- Same cargo and destination
- Evidence needed before deciding
- Equivalent quantity, handling needs and final delivery point
Row 2
- Comparison test
- Complete route
- Evidence needed before deciding
- Collection, transfers, main transport, storage and delivery included consistently
Row 3
- Comparison test
- Same emissions basis
- Evidence needed before deciding
- Matching gases, fuel lifecycle scope and reported unit
Row 4
- Comparison test
- Comparable factor evidence
- Evidence needed before deciding
- Applicable geography, service, period, load and allocation basis
Row 5
- Comparison test
- Service feasibility
- Evidence needed before deciding
- Delivery time, reliability, storage needs and product condition
Row 6
- Comparison test
- Unknown field
- Evidence needed before deciding
- Resolve the gap or hold the comparison; an unknown boundary is not a pass
Ask whether an alternative changes more than the main transport mode. A slower route may need additional storage, or the proposed service may still contain an air leg. Retain those activities instead of comparing an airport-to-airport air result with an incomplete port-to-port alternative.
If the evidence is incomplete, present the unresolved difference and the next request. Do not calculate a percentage saving from incompatible totals. The fictional AF-204 combustion-only result cannot establish a lifecycle CO2e saving against a sea or rail quotation. A useful comparison explains both the estimated emissions and whether the proposed service can actually deliver the same goods under the required conditions.
Turn a calculation into a repeatable shipment record
Filled fictional reconstruction record: this record describes AF-204 only. Every shipment detail, distance and coefficient is invented. It shows how a reviewer could reconstruct the teaching answer and identify the evidence still needed for real acceptance.
| Record field | Filled fictional value |
|---|---|
| Shipment and period | AF-204; May 2025; invented North Quay–Mesa Hub–Lake Gate routing |
| Unique legs and documents | AF-204-L1 and AF-204-L2; invoice and waybill references for L1 link to one leg |
| Declared mass | 200 kg physical shipped mass = 0.2 t on each leg; no chargeable-weight substitution |
| Distance basis | Assumed 2,400 km and 1,600 km air-leg distances; 4,000 km combined; no verified flight tracks |
| Mode and allocation | Invented dedicated freighter; cargo allocation embedded in T-1; no extra allocation multiplier |
| Method and coefficient | Mass-distance teaching method; T-1 version 1; invented 0.75 kg CO2/(t·km); not an official published factor |
| Calculation and result | 480 t·km gives 360 kg CO2; 320 t·km gives 240 kg CO2; unique-leg total 600 kg CO2 |
| Emissions boundary | Air-leg fuel-combustion CO2 only; not lifecycle CO2e |
| Exclusions | Collection/delivery road legs, storage, upstream fuel, other gases and non-CO2 aviation effects |
| Evidence and reviewer | Teaching status only; illustrative reviewer Shipping Analyst; carrier/service, mass, route and applicable factor evidence unresolved |
| Acceptance and update trigger | Hold operational acceptance; revise when carrier evidence confirms routing, allocation, service and emissions basis |
Table 5, shown one row at a time. Every field is retained.
Row 1
- Record field
- Shipment and period
- Filled fictional value
- AF-204; May 2025; invented North Quay–Mesa Hub–Lake Gate routing
Row 2
- Record field
- Unique legs and documents
- Filled fictional value
- AF-204-L1 and AF-204-L2; invoice and waybill references for L1 link to one leg
Row 3
- Record field
- Declared mass
- Filled fictional value
- 200 kg physical shipped mass = 0.2 t on each leg; no chargeable-weight substitution
Row 4
- Record field
- Distance basis
- Filled fictional value
- Assumed 2,400 km and 1,600 km air-leg distances; 4,000 km combined; no verified flight tracks
Row 5
- Record field
- Mode and allocation
- Filled fictional value
- Invented dedicated freighter; cargo allocation embedded in T-1; no extra allocation multiplier
Row 6
- Record field
- Method and coefficient
- Filled fictional value
- Mass-distance teaching method; T-1 version 1; invented 0.75 kg CO2/(t·km); not an official published factor
Row 7
- Record field
- Calculation and result
- Filled fictional value
- 480 t·km gives 360 kg CO2; 320 t·km gives 240 kg CO2; unique-leg total 600 kg CO2
Row 8
- Record field
- Emissions boundary
- Filled fictional value
- Air-leg fuel-combustion CO2 only; not lifecycle CO2e
Row 9
- Record field
- Exclusions
- Filled fictional value
- Collection/delivery road legs, storage, upstream fuel, other gases and non-CO2 aviation effects
Row 10
- Record field
- Evidence and reviewer
- Filled fictional value
- Teaching status only; illustrative reviewer Shipping Analyst; carrier/service, mass, route and applicable factor evidence unresolved
Row 11
- Record field
- Acceptance and update trigger
- Filled fictional value
- Hold operational acceptance; revise when carrier evidence confirms routing, allocation, service and emissions basis
Keep the source documents attached to the relevant leg identities, alongside the coefficient version and calculation. An updated invoice should not create a new physical leg unless it documents another movement.
When evidence arrives, record whether the change concerns activity, method, allocation or boundary. Retain the previous result and explain any difference. A reviewer can then distinguish a corrected mass from a new route or a wider lifecycle method. The filled table supports reconstruction; it does not turn fictional inputs into measured data or provide independent assurance.
Where this fits in an organization’s inventory
The calculation describes a movement; inventory classification describes its relationship to a particular reporting company. Keep those decisions separate. The same third-party shipment can be relevant to different value chains without one shipment total automatically identifying the correct reporting category.
Check whether the company owns or controls the transport asset, or purchases a third-party transport service. Retain the company boundary, contract and payment evidence. Purchased third-party freight may be relevant to upstream transportation and distribution, but the company’s actual relationship and the applicable inventory rules need review.
Retain the calculated shipment result. For inventory mapping, examine whether the transport asset is owned or controlled by the reporting company or whether the movement is a purchased third-party service. Those relationships guide the subsequent reporting treatment. The diagram does not assign every air movement to Category 4; the company boundary, contract and payment evidence still need review.
Diagram text description
Do not classify every air shipment as Scope 3 Category 4 merely because it is freight. Owned or controlled transport follows a different assessment, and the responsibility for purchasing the service can matter. The dedicated Category 4 explanation addresses that mapping in more detail; this article stays focused on reconstructing the movement calculation. Preserve the shipment result and its boundary so the inventory reviewer can use it without recreating the arithmetic.
What to do after the baseline
Use the baseline as a starting point for better evidence and a practical service decision. Segment recurring lanes by cargo needs, route, service urgency and method quality. Ask the carrier for applicable operational data, then test a feasible alternative where timing and handling permit.
Observe the service outcome as well as the emissions estimate. Record delivery time, condition, reliability and additional logistics activity. Feed those findings into the next procurement decision instead of assuming that a different transport label guarantees a saving.
Separate physical change from a calculation change. A lower revised estimate might reflect a better load assumption or a different factor rather than a change in the shipment. Keep a consistent comparison basis, and explain why the number moved. This is particularly important when replacing a broad baseline with more specific carrier evidence.
For AF-204, the immediate task would be to replace invented distances, service details and T-1 with compatible evidence, resolve the excluded activities needed by the intended boundary, and review the result. No generic percentage reduction follows from the exercise. The useful next action is the evidence request or service test that can change an actual shipping decision.
Maintain shipment evidence for review
Maintain the shipment evidence in a form that another person can follow: product or order identity, leg IDs, original mass and route records, coefficient version, calculation, exclusions and reviewer disposition. Structured product and supplier records can provide context, but they do not establish the emissions of a particular flight.
Before choosing support, check what the proposed system or service actually does. Ask whether it retains the applicable method, exposes allocation assumptions, links duplicate documents to one movement and preserves unresolved evidence. Confirm any carrier-data connection or assurance scope separately rather than inferring it from a carbon-related product description.
An organized record helps a reviewer reproduce the result and request missing information. Keep the operational estimate, inventory classification and permission to communicate a claim as distinct decisions. For the next shipment, start with its identity and boundary, use the best compatible evidence available, and retain the limits alongside the number.
Freight & distribution
