Emissions from farming are a headache for any company reliant on agricultural supply chains. For CPG giants like Nestlé and Mondelez International, agricultural ingredients make up more than two-thirds of their carbon footprints.
Reducing farm-level emissions is a challenge but just working out what those emissions are and tracking them over time is another problem entirely.
Go deeper with GlobalData
Access deeper industry intelligence
Experience unmatched clarity with a single platform that combines unique data, AI, and human expertise.
Food and drinks companies typically use a standard number for converting the amount of an ingredient to CO2. One kilogram of carrots, strawberries, or potatoes can be converted to an associated mass of CO2 based on a standardised “emissions factor”. But, if a supplier reduces emissions, it is not reflected in emissions factors, so it is not recorded by the manufacturer.
To solve this problem, companies are trying to incorporate more supplier-level data into their emissions calculations. Some parts of this are straightforward. A farmer will know how much fuel they have used, which can be accurately translated to an emissions number.
The more difficult part is measuring “biogenic emissions”, which are from soil, crops and livestock. For these, companies are reliant on estimates and extrapolation, rather than real measurement.
New tools are bearing fruit
Technology can improve the quality of biogenic emissions data. Carbon project developers, which sell carbon offsets, use satellite imagery to monitor farming practices like crop rotations, cover cropping and ploughing, which all impact soil carbon, and enable more accurate farm-level emissions estimates.
US Tariffs are shifting - will you react or anticipate?
Don’t let policy changes catch you off guard. Stay proactive with real-time data and expert analysis.
By GlobalDataHowever, the true emissions figure will still vary based on a range of local factors like climate and the type of soil.
For better data, companies need to move to direct measurement at the field level. One way to do this is with a “flux tower”, which is a tower fitted with sensors that can measure the flow of greenhouse gases just above a field, giving a much more accurate picture of the amount of carbon being stored or released.
A trial at Buck Island Ranch in Florida flux towers were used to show that, when cattle graze freely, they help store carbon in soil and reduce methane emissions. The towers strengthened root systems and reduced the number of dead plants being decomposed by microbes, which were emitting more methane than the cattle.
Better measurement could mean farmers can reduce their emissions through changing farming practices, rather than using emission-reducing feeds, which may be off-putting to consumers. It could also help industry bodies to defend livestock farmers from excessive regulation aimed at tackling methane emissions.
From research experiments to corporate strategy
The adoption of more advanced soil monitoring technologies has so far been limited to non-profit researchers and carbon markets. Some corporate use has been reported by Suzano, the paper and pulp giant, but most companies are put off by the cost; a network of flux towers being built by the Irish government is reported to cost between €80,000 to €250,000 ($93,000 to $290,000) per tower.
However, smaller gizmos and gadgets are becoming available. A company called Eosense produces smaller “flux chambers” (which are about the size of a beachball) while a start-up called YardStick sells carbon sensors farmers can bury in the ground.
Even with these smaller devices, installing sensors in every field would be expensive and impractical. Instead, companies can use data from a small number of devices at test farms to enhance estimates across their suppliers and understand the impact of different farming practices.
If food and drinks companies want to properly understand their carbon footprint and defend their industries against stricter regulations, investment in this type of soil carbon monitoring is worth considering.
