Scope 3 Cat 4 Air Freight Calculator (Bellyhold / Freighter / RFI)

What air freight covers in Scope 3 Category 4

GHG Protocol Scope 3 Category 4 — upstream transportation and distribution — covers the movement of purchased goods between a company’s tier-1 suppliers and its own operations, plus third-party logistics the reporting company pays for. Air freight is the mode used where speed or perishability justifies the cost and the carbon: high-value electronics, pharmaceuticals, time-critical spare parts, fashion replenishment, and fresh produce.

Category 4 vs Category 9 — same factors, different direction

The line between Category 4 and Category 9 is the direction of the goods flow, not the arithmetic. Category 4 is upstream — transport the reporting company pays for, bringing goods in. Category 9 is downstream — transport of sold products after they leave the company’s control. The GLEC air intensities are identical across both; the only difference is which Scope 3 category line the result is booked to. An airline reporting its own jet-fuel burn books it to Scope 1; a shipper buying cargo space books the same tonne-km to Scope 3 Category 4.

Mode boundary — what’s in, what’s elsewhere

The radiative-forcing question — why the headline excludes RF

This is the question that most often trips up an air-freight inventory, and the one an aviation-literate assurer probes first. Burning jet fuel at altitude does more to the climate than the carbon dioxide alone: contrails, contrail-induced cirrus, and high-altitude nitrogen-oxide chemistry add a warming effect on top of the CO₂. The shorthand for that uplift is radiative forcing (RF), and the multiplier applied to capture it is sometimes called the radiative forcing index (RFI).

What the GLEC freight factor includes — and what it does not

The GLEC Framework v3.2 air-freight WTW intensity is a CO₂e figure without a radiative-forcing uplift. It captures the full fuel lifecycle — upstream production plus combustion — but not the non-CO₂ altitude effects. Radiative forcing is a live, separately published adjustment in the passenger-aviation dataset, where a with-RF and a without-RF value sit side by side as distinct figures. It is not carried on the freight rows at all. This is the GLEC and GHG Protocol convention for Category 4 freight: report the without-RF figure as the headline, and treat RF as a documented sensitivity.

How to present an RF sensitivity

The defensible pattern is a two-line disclosure: the GLEC without-RF figure as the reported Category 4 emissions, and an RF-inclusive sensitivity computed by applying a stated multiplier to the combustion component, with the multiplier and its source named. The science on the exact multiplier is still unsettled, so the value you choose is a judgement you must document rather than a fixed constant. What matters for assurance is that the headline traces cleanly to the GLEC factor and the sensitivity is visibly separate and explained.

How the calculation works — activity × WTW intensity

Each leg reduces to one multiplication, run per leg and summed into the absolute emissions total at the end:

kg CO₂e = Mass (tonnes) × Distance (km) × WTW intensity (kg CO₂e per tonne-km)

The activity spine

The activity is cargo mass in tonnes multiplied by the flight distance in kilometres. Distance for air freight is the great-circle distance between airports plus the calculator’s fixed 95 km uplift for indirect routing and holding — see the distance section below. Air uses a single activity unit throughout: there is no TEU-km complication as there is in sea container freight. Every air row is per tonne-km.

Well-to-wheel, well-to-tank, tank-to-wheel

Deck and haul — no regions, no fuel choice

Unlike road freight, the air set has no regional dimension and no fuel options — Jet A/A-1 is assumed throughout. Select the deck that carried the cargo (bellyhold, freighter, or the unknown blend where the carrier does not disclose) and the haul length (short-haul or long-haul). Where the carrier reports neither deck nor haul, the unknown deck combined with the haul that matches the route distance is the defensible modelled default.

Bellyhold vs freighter — the long-haul crossover

The most counterintuitive feature of the air set is that bellyhold is not uniformly cleaner than freighter. The ranking flips with haul length, and getting it backwards is a common modelling error.

On short-haul, bellyhold (1.239) beats freighter (1.516): belly cargo rides on a passenger flight that was operating anyway, so it carries only a share of the fuel burn the passengers do not. On long-haul, the order reverses — a dedicated long-haul freighter (0.608) beats belly (0.936) because purpose-built freighters fly dense, high-utilisation payloads that spread the fuel burn much further per tonne-km, while belly capacity on a long passenger flight is constrained by passenger and baggage weight. The lesson: deck and haul interact, and you cannot assume belly is always the lower-carbon choice.

Why short-haul is roughly double long-haul

Across every deck, short-haul intensity runs roughly double the long-haul figure. The driver is the take-off and climb phase, which burns a disproportionate share of a flight’s fuel. On a short flight that fixed high-burn segment is spread over few kilometres, inflating the per-tonne-km intensity; on a long flight the same take-off burn is amortised over thousands of cruise kilometres at a far lower burn rate.

The reporting consequence is that haul classification materially changes the result, so the route distance that determines short versus long must be applied consistently. A leg sitting near the short/long boundary is worth checking against the carrier’s own classification rather than guessing, because the intensity step between the two bands is large.

The great-circle distance and the 95 km uplift

Aircraft do not fly straight lines from gate to gate. They follow air-traffic-control routings, hold and stack near congested airports, and take indirect departure and approach paths. Great-circle distance — the shortest path over the globe’s surface between two airports — understates the distance actually flown, and so understates tonne-km and emissions.

GLEC and the IATA RP 1678 method address this with a fixed distance correction. The calculator takes the great-circle distance between origin and destination airports and adds a flat 95 km uplift to approximate the extra distance flown:

• Great-circle distance — enter the shortest-path distance between the two airports; the calculator applies the 95 km uplift automatically. This is the standard input.
• Actual flown distance — if the carrier supplies the true distance flown, that already captures the routing, so the fixed uplift would double-count; use the carrier figure as-is and note that the standard uplift was not applied.

Worked example — a two-leg air shipment

This example reproduces against the live calculator. It combines a long-haul freighter leg and a short-haul bellyhold leg in one consignment — the realistic shape of a time-critical shipment that flies the long ocean crossing on a dedicated freighter and the final regional hop in passenger belly hold. All factors are GLEC Framework v3.2 well-to-wheel intensities, without radiative forcing. Distances shown are great-circle; the calculator adds the 95 km uplift before multiplying, and the figures below reflect that uplift.

Leg A — long-haul freighter

Distance = 9,500 great-circle + 95 uplift = 9,595 km. Tonne-km = 12 × 9,595 = 115,140. Emissions = 115,140 × 0.608 = 70,005 kg CO₂e (70.0 tCO₂e). Headline is without radiative forcing.

Leg B — short-haul bellyhold

Distance = 600 great-circle + 95 uplift = 695 km. Tonne-km = 12 × 695 = 8,340. Emissions = 8,340 × 1.239 = 10,333 kg CO₂e (10.3 tCO₂e). The short-haul belly intensity of 1.239 is roughly double the long-haul freighter’s 0.608, which is why a leg one-fourteenth the distance still contributes a meaningful share.

Shipment rollup

Both legs share the same tonne-km unit, so unlike a mixed sea shipment the activity figures add cleanly: 115,140 + 8,340 = 123,480 tonne-km, and the emissions sum to 80.3 tCO₂e. The long-haul leg dominates the footprint at 87% — moving 12 tonnes 9,595 km is far more transport activity than the regional hop, even though the short-haul belly intensity per tonne-km is the higher of the two. If a disclosure policy required a radiative-forcing sensitivity, the 80.3 tCO₂e without-RF figure would be reported as the headline and the RF-uplifted figure shown separately beneath it, with the multiplier named.

Data quality and the GLEC default hierarchy

GLEC frames freight data quality as a hierarchy from primary measured energy down to modelled defaults. The reporting expectation is to climb the hierarchy where the emissions are material — and for most companies air freight is material relative to its tonnage, so it rewards better data more than the surface modes do.

The honest disclosure is the mix. An air-freight inventory that uses carrier-reported figures where the forwarder provides them and the unknown-deck blend on the spot bookings that do not is a well-constructed inventory. Claiming primary-energy precision across every consignment when most are modelled deck-haul defaults is the disclosure failure, not the use of defaults.

Audit checklist — what gets flagged in air-freight assurance

Air-freight assurance under ISO 14083 and ISAE 3000 traces each leg from activity data to the rolled-up total. The findings below are the most common substantive issues raised on an air-freight inventory specifically.

Reporting context — Scope 3, GLEC/ISO 14083, CORSIA, IFRS S2, CSRD

Air-freight emissions feed several disclosure regimes through the same GLEC calculation. The four framework rows below cover the disclosure surface a typical shipper or forwarder navigates; the CORSIA layer sits alongside as the aviation-sector regulatory context.

The CORSIA layer — aviation’s sectoral scheme

Distinct from the corporate-disclosure frameworks above, CORSIA — the Carbon Offsetting and Reduction Scheme for International Aviation — regulates the carbon of international flights directly, requiring operators to monitor emissions and offset growth above a baseline. CORSIA applies to the aircraft operator, not to the shipper booking cargo. A shipper does not file under CORSIA; a carrier does. But CORSIA shapes the fuel and efficiency choices that feed back into the GLEC intensities a shipper reports, and a carrier may cite its CORSIA position when responding to a shipper’s emissions-data request — so it is worth understanding the distinction. CORSIA is a sector-offsetting scheme for the operator, not a per-shipment Category 4 figure for the shipper.

GLEC, IATA RP 1678, and ISO 14083

GLEC v3.2 allocates air-cargo emissions on the IATA RP 1678 method and is structured to align with ISO 14083, the international transport-chain emissions methodology. A calculation built on GLEC air intensities is therefore broadly ISO 14083-conformant — an internationally standardised basis rather than a proprietary one, which assurers and disclosure regimes increasingly expect for transport emissions.

Data sources, factor versioning, and update transparency

Air intensities — source and structure

The air-freight intensities are drawn from GLEC Framework v3.2, retrieved for the MasterBrain factor set in May 2026. The air set is compact: six rows across three decks (bellyhold, freighter, and the unknown blend) and two haul lengths (short-haul and long-haul), all on Jet A/A-1 and all on a per-tonne-km basis. Each row’s headline value is the well-to-wheel intensity on an AR6 GWP-100 basis, without a radiative-forcing uplift; the well-to-tank and tank-to-wheel components are published alongside and are reported as fixed figures in worked examples rather than user-variable inputs. See the DEFRA emission factors reference for the jet-fuel factors that complement the GLEC set on carrier-reported primary-energy legs.

GWP basis

The GLEC air set uses AR6 GWP-100 — the IPCC Sixth Assessment Report’s 100-year global warming potentials, the current GLEC convention, aligned with the GHG Protocol’s expected basis. Where air-freight legs are combined with DEFRA-sourced own-operation legs, note that DEFRA factors are published on an AR5 GWP-100 basis; the calculator does not silently reconcile the two, and a mixed-basis inventory should disclose the mix rather than implying a uniform basis. See the global warming potential definition for the AR5-to-AR6 distinction.

Versioning and update cadence

GLEC publishes factor refreshes annually, and the MasterBrain factor set tracks each publication. The calculator stamps each result with the MasterBrain version against which it was computed, so a leg computed against one factor vintage and the same leg recomputed against a later vintage are distinguishable in restatement work. The GLEC Framework version (v3.2) is independent of the underlying factor vintage — a methodology version applies across successive annual refreshes until the next framework revision ships.

What’s next — completing your Category 4 transport inventory

Air freight is the highest-intensity mode but rarely the largest by tonnage; a complete Category 4 footprint spans every mode the chain uses. Each mode follows its own factor set and, in the case of sea, its own activity units. The road and sea calculators are live; the remaining modes are on the GreenCalculus build roadmap.

Separately from the upstream Category 4 modes above, downstream transport and distribution — the movement of sold products after they leave the company’s control — is accounted under Scope 3 Category 9, not Category 4. It uses the same GLEC air intensities but books to a different category line. A dedicated downstream transport calculator is on the roadmap and is not yet published.

Frequently asked questions

Methodology notes and limitations

Methodology version. Calculator implements the GLEC Framework v3.2 air-freight intensities (retrieved May 2026), allocated per IATA RP 1678 and aligned with ISO 14083. Results are computed on an AR6 GWP-100 basis. Inventories computed against earlier factor vintages remain auditable against the corresponding MasterBrain version stamp.

Mode boundary. Air freight only — three decks (bellyhold, freighter, and the unknown blend) across short-haul and long-haul, all on Jet A/A-1, all per tonne-km. Road, sea, rail, and last-mile freight use separate factor sets and are out of scope.

Radiative forcing excluded from the headline. The well-to-wheel intensity is a CO₂e figure without a radiative-forcing uplift, per the GLEC and GHG Protocol convention for Category 4 freight. Where a disclosure requires the altitude effects, report the without-RF figure as the headline and an RF-inclusive sensitivity separately, with the chosen multiplier and its source named. The calculator does not fold RF into the reported number.

WTW headline; WTT/TTW fixed components. The calculator computes the well-to-wheel total. The well-to-tank and tank-to-wheel components are published by GLEC alongside each row and are reported as fixed figures; they do not vary with user inputs and are not separate live fields.

Distance basis. The calculator applies a flat 95 km uplift to the great-circle distance entered, approximating holding and indirect routing per IATA RP 1678. Where the carrier supplies actual flown distance, enter it without the uplift to avoid double-counting. Document the basis per lane.

Deck and haul matter most. Intensity roughly doubles from long-haul to short-haul, and the bellyhold/freighter ranking flips with haul length. Where the deck is unknown, use the unknown blend rather than guessing the lower deck; where a leg sits near the short/long boundary, check the carrier’s classification.

SAF excluded. Factors assume conventional Jet A/A-1. Any sustainable-aviation-fuel reduction must be substantiated and applied separately, not assumed into the headline.

Activity data taken as entered. The calculator uses the mass and distance the user enters; it does not independently verify them. The user is responsible for the activity-data basis and the distance basis, and for documenting both per leg.

No assurance opinion. Results are estimates and do not constitute an assurance opinion. They should be reviewed by a qualified practitioner before use in IFRS S2 disclosures, CSRD ESRS E1 datapoints, or other regulatory submissions. The full methodological deep-dive — deck and haul selection, the bellyhold/freighter crossover, the distance uplift derivation, and the radiative-forcing treatment — is published on the paired GLEC air-freight methodology page.