Product carbon footprinting

This paper (Plassmann K, Norton A, Attazadeh N, Jensen M P, Brenton P and Edwards-Jones G (2010). Methodological complexities of product carbon footprinting:a sensitivity analysis of key variables in a developing country context, Environmental Science and Policy, 13, 5. 393-404) looks at how differences in the methodological approaches and boundaries adopted by different methods of carbon footprinting, in combination with problems of data scarcity and the subsequent need to use default data, together give rise to very large differences in the carbon footprint of the same product, in this case sugar. 

It finds that the carbon footprint ranghes from between 0.03 to  0.2 kg CO₂e/kg of sugar cane delivered to the refinery depending on what is included and excluded in the methodology and what data are used.
 

Abstract

Product carbon footprinting schemes adopt different analytical methodologies. The calculations can also be affected by limited data availability and uncertainty surrounding the value of key variables. The combination of these factors reduces the validity of comparing carbon footprints between products and countries.
 
We used data from sugar production in Zambia and Mauritius to test how variations in methodology affected the product carbon footprint (PCF). We calculated a PCF according to PAS 2050 and explored the sensitivity of the results to the variation of key variables.
 
Results showed that land use change emissions can dominate PCFs. The largest potential impact came from assuming global worst case data for land use change emissions where a product’s origin is unknown (+1900%). The issue of land use change can lead to high carbon footprints for products from developing countries where more natural vegetation is still being converted and data are most lacking. When land use change is not important, variables such as electricity emission factors, capital inputs and loss of soil carbon had significant impacts on the PCF.
 
This analysis highlights the large effect of methodology on PCFs. These results are of particular concern for developing countries where data are scarce and the use of global worst case data may be prescribed. We recommend the development of more precise emission factors for tropical countries and bio-regions, and encourage the transparent use of PCF methodologies, where data sources, uncertainties and variability are explicitly noted.
 
NB: this is a diagram from the showing the sorts of things that can be included in or excluded from a PCF, and that affected the results.
 
The paper can be downloaded here.
(This is a pay service from ScienceDirect.)