KLIMAT - a Norwegian research project
Report from the banana case study
The “KLIMAT” research project was a three year Norwegian research project financed by the Norwegian Research Council and partners. The project lasted from 1st of January 2009 to the 31st of August 2012. The partners were retailers Coop and Norgesgruppen, food producers Tine and Nortura and distributor and wholesaler BAMA. The research was carried out by Ostfold Research with advice from experts at UMB (University of environment and life sciences, Norway), food research institutes SIK (Sweden), MTT (Finland), and Aalborg University (Denmark).
The aim of the project was to develop a standardized methodology for calculating Carbon Footprint of food products. 6 case products have been chosen, among them bananas. In the case of bananas BAMA and Dole Fresh Fruit were cooperating partners. This report contains a summary of the research carried out relating to the banana case and methodology development as well as results, discussions and conclusions. The results were planned to be published in an article in a peer-reviewed scientific magazine, other magazines and publications and in scientific conferences.
Three scenarios were investigated, one German and one Norwegian scenario using “pallet boat”, i.e. boats where the product is transported on pallets, and one Norwegian scenario using container boats. All scenarios were Cradle to retail. In addition one cradle to grave scenario was calculated using pallet boats and end consumption in Norway. The total Carbon Footprint (CFP) of the product was found to be 0,78 kg CO2-eq/kg product using container transport, and 1,37 and 1,27 using pallet boats for Norway and Germany respectively.
The main impact came from overseas transport, with the primary production and packaging coming next in importance. Primary production is dominated by direct emissions from landfill and soils. The production of ammonium nitrate fertilizer is also important. The largest contribution from overseas transport was emissions from combustion of fuel to propel ships forward and cool the product. Refrigerant leakage plays a very minor part. For the remaining parts of the value chain the product wastage plays the most important role. Many studies have pointed out that CFPs of food products are associated with high uncertainties because of uncertainty in determining soil emissions such as N2O but in this case soil emissions were found to be of low importance, approx 4 % of the total CFP. Hence the importance of the uncertainty in calculating N2O emissions is small. The effect of waste has a significant impact on the CFP. Landfilled waste in the plantations gives a high negative impact, the return of crop residue waste to the soil is probably positive for the soil carbon content. The waste treatment in Norway is efficient thus limiting the negative impact of the packaging and product waste.
The LCA methodology used is standardised but the standards leave a lot of freedom for the user. The investigation shows that the methodological choices were vital for the end result. System boundaries in particular was very important, for instance the CFP from Cradle to Grave was significantly higher than the Cradle to retail CFP. The inclusion of infrastructure (Capital Goods) was also found to be of importance.
Allocation was of less importance but was nevertheless important. Data quality and choice of data source was very important for the end result. If data for only one plantation had been used, or if another base year had been chosen, the results would have looked significantly different. The emissions from production of fossil fuels and fertilizer differed very much from source to source, having important implications for the end result. The impact of waste was important. The lack of primary data meant that the packaging impact had to be calculated using average numbers which is unfortunate because it is a major impact.
The calculation of CFPs can be done using only activity data and emission factors but this case showed the importance of using a comprehensive LCA approach, use LCA software and thoroughly evaluates data sources. Once this is done a more simplified approach can be used in future calculations.