Implications of uncertainty and variability in the Life Cycle Assessment of pig production systems

Goal, Scope and Background. Calculating LCA outcomes implies the use of parameters, models, choices and scenarios which introduce uncertainty, as they imperfectly account for the variability of both human and environmental systems. The analysis of the uncertainty of LCA results, and its reduction by an improved estimation of key parameters and through the improvement of the models used to convert emissions into regional impacts, such as eutrophication, are major issues for LCA. Methods. In a case study of pig production systems, we propose a simple quantification of the uncertainty of LCA results (intra-system variability) and we explore the inter-system variability to produce more robust LCA outcomes. The quantification of the intrasystem uncertainty takes into account the variability of the technical performance (crop yield, feed efficiency) and of emission factors (for NH3, N2O and NO3) and the influence of the functional unit (FU) (kg of pig versus hectare used). For farming systems, the inter-system variability is investigated through differentiating the production mode (conventional, quality label, organic (OA)), and the farmer practices (Good Agricultural Practice (GAP) versus Over Fertilised (OF)), while for natural systems, variability due to physical and climatic characteristics of catchments expected to modify nitrate fate is explored. Results and Conclusion. For the eutrophication and climate change impact categories, the uncertainty associated with field emissions contributes more to the overall uncertainty than the uncertainty associated with emissions from livestock buildings, with crop yield and with feed efficiency. For acidification, the uncertainty of emissions from livestock buildings is the single most important contributor to the overall uncertainty. The influence of the FU on eutrophication results is very important when comparing systems with different degrees of intensification such as GAP and OA. Concerning the inter-system variability, differences in farmer practices have a larger effect on eutrophication than differences between production modes. Finally, the physical characteristics of the catchment and the climate strongly affect the results for eutrophication. In conclusion, in this case study, the main sources of uncertainty are in the estimation of emission factors, due both to the variability of environmental conditions and to lack of knowledge (emissions of N2O at the field level), but also in the model used for assessing regional impacts such as eutrophication. Recommendation and Perspective. Suitable deterministic simulation models integrating the main controlling variables (environmental conditions, farmer practices, technology used) should be used to predict the emissions of a given system as well as their probabilistic distribution allowing the use of stochastic modelling. Finally, our simulations on eutrophication illustrate the necessity of integrating the fate of pollutants in models of impact assessment and highlight the important margin of improvement existing for the eutrophication impact assessment model.