MEXALCA: a modular method for the extrapolation of crop LCA

Life cycle assessment (LCA) is widely used for the environmental assessment of food products, but difficulties arise when evaluating large portfolios of food products or when faced with a diversity of sources of ingredients and/or frequent changes of suppliers. In such situations, a specific, in-depth assessment of each ingredient is not feasible, and screening approaches using a few LCIA (Life Cycle Impact Assessment) results are not recommended. The goal of this paper is to propose an intermediate solution between a screening assessment using limited data and specific LCA for all products considering all sources of ingredients.

The extrapolation method presented here [Modular EXtrapolation of Agricultural LCA (MEXALCA)] allows deriving LCIA results for a crop in a specific (target) country using the LCIA data of the same crop in another (original) country. The existing crop inventory (LCI) is split into nine modules corresponding to the main on-field and post-harvest activities, each associated to its key farming input. This approach reduces the complexity of the inventories, and data collection is focused on nine inputs in the target country. Additionally, data can be approximated by means of statistical estimators if necessary. Impacts per unit of farming input are calculated for each module in the original country and combined with the quantity of farming inputs in the target countries in order to determine the impacts in the latter.

Combining MEXALCA with available national statistics provides the means for a rapid evaluation of environmental impacts of a given crop for all producing countries globally and the determination of their statistical distribution. The range of the impacts for a number of crops was determined, and a sensitivity analysis with values derived from national statistics was conducted. Validation of the method showed a good performance for the evaluation of energy demand, ozone formation, and global warming potential. The method was inaccurate for eutrophication and acidification and not suitable for toxicity impact categories.

Extrapolation of agricultural LCIAs using statistical data and existing inventories is feasible. Compared with many streamlined LCA approaches, it does not discard processes nor jeopardise the understanding of the production system. MEXALCA shows benefits in terms of the amount of data and time required and allows for data generation on a multi-country scale. It appears promising for an assessment of products from various origins and may also be useful for the extrapolation of one product to another. The method is suitable for environmental assessments at the regional or global scale and to fill data gaps in traditional LCAs. However, extrapolation should not replace a conventional LCA for the assessment of a specific product, and particular caution should be exercised for spatially dependent impact categories such as eutrophication or acidification.