Land and ecosystem transitionsOngoing

The full climate cost of food: emissions and the carbon value of land, product by product

The climate cost of about 270 food products once the land they occupy is valued at the carbon it could hold. A published global model, rebuilt so it can be verified, then extended toward 2050.

Project facts

ProgrammeLand and ecosystem transitions · also shown in Systemic transitions
StatusOngoing (rebuild and 2020 benchmark complete; 2050 extension planned) What the status labels mean
MethodsBiophysical modelling · Environmental accounting and LCA · Model reproduction and verification · Uncertainty analysis
ScaleLand system
GeographyGlobal: 10 world regions and 4 countries reported separately

The question

What does each food product cost the climate once the land it occupies is valued at the carbon that land could hold? And which diets, productivity gains and trade shifts lower that cost by 2050 without leaving the space of what is biophysically and economically feasible?

What the figures leave out

The usual footprint of a food counts the emissions of farming and processing. It does not count the carbon the land would hold under natural vegetation: the carbon opportunity cost of land. For land-hungry products this omitted cost can be large, and it is the cost that decides whether sparing or restoring land pays off.

Approach

We took a published global agri-food model and rebuilt it as software that can be checked. Two independent implementations were tested against each other until they matched, then the model was corrected and a 2020 benchmark was frozen. Uncertainty runs through Monte Carlo simulation. The 2050 extension will link the model with equilibrium and spatial land-use models.

PUBLISHED GLOBAL AGRI-FOOD MODELPublished model~270 FOOD PRODUCTSRebuild ARebuild BCROSS-CHECKEDUNTIL THEY MATCHTWO INDEPENDENT IMPLEMENTATIONS2020 benchmarkCORRECTED · FROZEN2050 foresightPLANNED · NOT STARTEDLAND VALUED AT THE CARBON IT COULD HOLD · UNCERTAINTY BY MONTE CARLO SIMULATION2050: LINKS TO EQUILIBRIUM AND SPATIAL LAND-USE MODELS Published model~270 FOOD PRODUCTSRebuild ARebuild BCROSS-CHECKED UNTIL THEY MATCH2020 benchmarkCORRECTED · FROZEN2050 foresightPLANNED · NOT STARTEDLAND VALUED AT THE CARBON ITCOULD HOLD · MONTE CARLOUNCERTAINTY
FIG. 1 · METHOD SCHEMATIC · NO VALUES · Diagram of the verification method: a published model rebuilt twice, cross-checked, frozen as a 2020 benchmark, then extended to 2050.STATUS · ONGOING

Scope

~270food products
10 + 4world regions, plus countries reported separately
2020reference year of the benchmark
2050horizon of the planned foresight

Where it stands

Ongoing

The rebuild and the corrected 2020 benchmark are complete. The 2050 foresight extension is planned and has not started. No 2050 result exists yet.

The decision it informs

Diet, waste and productivity policy; land sparing and restoration; the effect of EU policies on land use abroad; companies' land-related targets and sourcing.

INSTRUMENTS · Product-level climate costs; food and land scenarios to 2050.

What the work required

  • Rebuilding published scientific models so they can be verified
  • Benchmark curation and scenario engineering
  • Uncertainty quantification
  • Research pipelines that are AI-assisted but deterministic

Working on diets, land use or the climate cost of food?

Discuss this work