Spatial differentiation? Yes, but, not only for impact characterization!

International audience; Life Cycle Assessment (LCA) faces with spatial differentiation issues that have been discussed for years in LCA community. A 77 papers review reveals that the most commonly studied one is the lack of spatial consideration in the environmental impact characterization step and especially for local and regional impacts as eutrophication. This issue was the first to be identified by Potting and Hauschild (1997). More recently number of developments aim to include spatial and temporal information during the Life Cycle Inventory (LCI) carrying out. If theorical methodologies are available for LCI and characterization steps, we only found practical ones for the two others LCA steps (Goal and scope definition and Results interpretation) despite identified issues (to overcome multifunctionality). This fact can be seen as a paradox as for the life cycle thinking and the iterative, holistic and global aspects of LCA. That is why we propose to develop a conceptual framework to consider the spatial differentiation all over the four steps of LCA in a continous and heterogeneous way. This framework is called “the spatialization continuum”. The first steps of the building of such a conceptual framework consist in the identification of the existing theorical and practical considerations of spatial differentiation in LCA and the potentiality to combine them towards a continous methodology. Two questions are explored: -For each step of LCA, what is necessary to ensure the spatial differentiation consideration? -What does a development on a precise step involve to the other ones? The conceptual framework also set the objective to examine different tools that appear to be indispensible when dealing with spatial information and their potential coupling with LCA methodology and practices. Among these tools, Geographic Information Systems (GIS) are the most commonly used. However GIS appear not sufficient to overcome all the spatial differentiation related issues. In fact the spatialization continuum should ask for multiple external tools, methodologies or concepts. Therefore, the spatialization continuum materializes as a “multi-plugin” that will call for different solutions depending on the concerned issue and the required level of robustness. In this context, it differs from the classification of strategies to overcome LCA limitations proposed by Udo de Haes (2004): LCA extension, Hybrid LCA and Toolbox. We decided to call this new strategy the multi-plugin LCA.; Besoin de différenciation spatiale tout au long des 4 étapes de l'ACV.