Where does the carbon footprint fall?

Developing a carbon map of food production

by Katharina Plassman, Gareth Edward-Jones, International Institute for Climate Change (IIED), octobre 2009

Keywords : environment - production

Language : French

The concept of local food is appealing to many consumers. However, difficulties remain with defining what actually constitutes local food. Given the globalised nature of agricultural markets, bread which is baked in a small village bakery in England may be made from grain grown in Canada. Similarly many of the inputs (e.g. tractors, fertilisers, diesel and concentrate feed) to a West Country dairy farm selling local ice cream may come from outside the UK.

One of the purported advantages of local food relates to reduced emissions of greenhouse gases from the food chain. This concept was initially encapsulated by measuring food miles, however more recently this simple concept has been replaced by the development of more comprehensive life cycle assessments and carbon footprints.

Carbon footprints report the total levels of greenhouse gas (GHG) emissions from food production, but do not document the actual geographic location of the emissions. If advocates of local food really wanted to differentiate local from non-local food in a quantifiable way, then one way to do this would be to utilise local stocks of carbon and to make GHG emissions locally.

This report advances the discussion about defining the local by examining the geographical location of GHG emissions along the supply chains upstream of two case study farms. The resulting carbon map illustrates the amount and location of the GHG emissions related to the provision of inputs and on-farm processes, and enables characterisation of the ‘localness’ of the two farm systems. Inputs to the two case study dairy farms were documented and the origin of the constituent raw materials was identified. On-farm emissions were also estimated and through combining these sets of data the carbon footprint was calculated for each farm. Results are expressed per hectare and per litre of milk. Through combining the origin of the inputs with details of relative GHG emissions it was possible to develop a ‘carbon map’ which shows the spatial location of emissions at a global scale.

Both case study farms had very similar carbon maps. Less than 5 per cent of GHG emissions related to the provision and use of inputs are considered local (i.e. occur within 50 km of the farm). As the emissions of GHGs from soils and livestock occur on farm they are defined as being local. As a result their inclusion in the carbon footprint changed the carbon map considerably, and greater than 50 per cent of all total GHG emissions then occurred locally. Further analysis considered the emissions derived from soya in livestock meal, which may be grown in South America on land recently cleared from forest. Specific inclusion of emissions resulting from land use change for soya production in the carbon map increased the amount of non-local emissions for both case study farms.This work was conducted over 10 days. Only two case study farms were considered, and there are considerable gaps in the analysis of the relevant supply chains. The knowledge-base on emissions for each input and process are incomplete. As a result the report does not represent a comprehensive analysis of the data and should be viewed more as proof of concept rather than a definitive analysis.

Read the document (270 KiB)