Coffee — Roasted Beans

Food & Beverage
Medium Confidence

Carbon Cost Index Score

6.5 kgCO₂e / per kg roasted beans

Per kg

6.5 kgCO₂e / kg

Methodology v1.0 · Last reviewed 2026-04-07

Scope Breakdown

Scope kgCO₂e % of Total Distribution
Scope 1 0.8 12%
Scope 2 0.5 8%
Scope 3 5.2 80%
Total 6.5 100%

Emission Hotspots

Emission Hotspot Scope Est. % of Total
Coffee cultivation (land use, fertiliser application, farm energy) S3 45%
Processing at origin (wet/dry milling, drying) S3 18%
International freight (air, sea, road — green coffee beans) S3 14%
Roasting (natural gas or electric drum roasters) S1 12%
Packaging and retail distribution S3 11%

Manufacturing Geography

Region
Global (Brazil, Colombia, Vietnam, Ethiopia primary)
Grid Intensity
Mixed — Brazil ~220 gCO2e/kWh, Vietnam ~560 gCO2e/kWh, EU roasting ~300 gCO2e/kWh

Material Composition Assumptions

A representative kilogram of roasted coffee beans begins as approximately 1.19 kg of green (unroasted) coffee beans — the difference represents moisture and volatile compounds lost during roasting (the “roast loss”, typically 15–20% by mass). The full upstream supply chain includes:

The CCI score of 6.5 kgCO2e/kg represents a blended average across Arabica and Robusta origins and includes cultivation, on-farm processing, international transport, roasting, and primary packaging.

Why the Score Is What It Is

Coffee is a moderately carbon-intensive beverage ingredient primarily because of the agricultural footprint at origin, not the roasting process that consumers associate most directly with the product. The 6.5 kgCO2e/kg score is driven predominantly by Scope 3 upstream emissions — over 77% of the total.

Farm-level emissions account for the largest single share (~45%). Synthetic nitrogen fertiliser applied to coffee plants releases nitrous oxide (N2O) from soil at rates that are highly variable but climatically significant (N2O is approximately 265 times more potent than CO2 over a 100-year horizon). Land-use change is a structural risk in coffee, particularly in areas of forest frontier expansion in Brazil’s Cerrado, parts of Ethiopia, and Vietnam’s Central Highlands. Where primary forest is converted to coffee production, lifecycle emissions can be three to five times higher than the baseline estimate shown here.

Processing at origin (wet or dry milling, fermentation, drying) contributes roughly 18% of total emissions. Wet processing requires substantial water and produces organic-rich effluent; dry processing relies on sun-drying or mechanical dryers. The energy mix at origin — predominantly diesel generators and biomass — drives the Scope 1 fraction of processing emissions.

International freight contributes ~14%. Green coffee typically moves by sea container (low carbon per tonne-km), but the long distances from Brazil or Vietnam to European or North American roasters accumulate. Specialty coffees sourced from smaller lots may use airfreight at a premium, which can add 0.5–1.5 kgCO2e/kg to the total.

Roasting itself (12% of total) is surprisingly modest — a well-run gas drum roaster adds approximately 0.6–1.0 kgCO2e per kg roasted, including direct combustion and electricity for airflow and automation. Electric roasters powered by renewable energy can reduce this to under 0.1 kgCO2e/kg.

What Drives Variation

Origin and production system are the largest sources of variability. Brazilian sun-dried Robusta from intensively managed farms can have footprints as low as 4.5 kgCO2e/kg because of low transportation costs within Brazil and efficient mechanised harvesting. Ethiopian washed Arabica from smallholder farms typically sits around 6–8 kgCO2e/kg, with variation driven by fertiliser use and drying method. Vietnamese Robusta tends toward the upper end of the range (7–10 kgCO2e/kg) due to high fertiliser application rates and a carbon-intensive grid for processing.

Deforestation risk is the most extreme source of variation. Coffee grown on recently cleared forest land can carry a land-use change (LUC) penalty of 10–30 kgCO2e/kg — potentially multiplying the baseline score several times over. Supply chain traceability tools and sustainability certifications (Rainforest Alliance, UTZ, Fairtrade) aim to reduce this risk but do not eliminate it.

Roasting geography and energy source matter for the Scope 2 portion. A roaster operating in Norway on near-zero-carbon hydroelectricity versus one in Poland on a coal-heavy grid will see scope 2 emissions differ by a factor of five or more, though this shifts the overall score by less than 0.5 kgCO2e/kg in most cases.

Packaging format affects the score modestly. Single-serve capsules (Nespresso-style) add 2–4 g of aluminium or plastic per serving, which, when normalised per kg of coffee used, can add 0.3–0.8 kgCO2e/kg. Bulk whole-bean supply with minimal packaging sits at the low end.

Organic certification typically removes synthetic fertiliser emissions but does not necessarily reduce overall footprint — yield reductions on organic farms can mean more land is required per unit of output, sometimes resulting in a similar or slightly higher per-kg emission factor depending on the yield gap in a specific origin.

Manufacturing Geography

The default manufacturing region reflects the global coffee trade: origins concentrated in tropical belt (Brazil ~40% of global production, Vietnam ~20%, Colombia ~8%, Ethiopia ~5%), with roasting primarily in consuming markets (EU, USA, Japan).

Green bean transport distances are typically 10,000–18,000 km by sea from origin to consuming-country port. Roasting occurs close to the end market in most cases — large-scale industrial roasters (Nestlé, JDE, Lavazza) operate in the EU and USA using grid electricity and natural gas.

The blended grid intensity for roasting is approximately 300–400 gCO2e/kWh in EU and USA facilities, falling as renewable energy penetration increases. Origin-country grids matter primarily for mechanical processing; Brazil’s predominantly hydroelectric grid (~220 gCO2e/kWh average) is a relative advantage for origin-country processing versus Vietnam (~560 gCO2e/kWh).

Provenance Override Guidance

Manufacturers and importers may override the default CCI score by providing:

  1. Farm-level LCA data or verified origin certification demonstrating no deforestation risk (e.g., Rainforest Alliance 2020 standard with verified deforestation monitoring).
  2. Fertiliser application records per farm or cooperative, allowing direct substitution of the default N2O emission factor.
  3. Processing method documentation specifying wet vs. dry processing, energy source (grid, diesel, biomass, solar), and wastewater treatment method.
  4. Freight mode and distance records for green bean shipment — sea vs. air, port of origin to roasting facility.
  5. Roasting facility energy data including gas consumption per kg roasted and renewable energy certificates (RECs) covering facility electricity.

Verified supply chain data can reduce the score by 20–40% relative to the conservative mid-range default.

Methodology Notes

Related Concepts

Related Categories

Sources

  1. Killian et al. — Lifecycle Assessment of Coffee, 2013. Cradle-to-grave analysis across four origin countries; farm-level emissions dominate at 60–70% of total footprint.
  2. Ecoinvent v3.9 — Coffee production datasets covering Arabica and Robusta cultivation, wet and dry processing, and industrial roasting. Regional variants for Brazil, Colombia, Vietnam, and Ethiopia.
  3. International Coffee Organization (ICO) — Sustainability Report 2022. Covers deforestation risk, fertiliser intensity, and water use across major producing regions.
  4. Wiedemann et al. — Carbon footprint of coffee supply chains, 2020. Scope 3 disaggregation for green bean trade and roasting stages.
Scan a product in this category →