Evaluation of Process- and Input-Output-Based Inventories for Environmental Assessment of Production and Consumption Activities: Evaluering av prosess- og kryssløpsbaserte inventardata for miljøanalyser av våre produksjons og konsum aktiviteter

Abstract

Life cycle assessment (LCA) and environmentally extended input-output analysis (IO) both make it possible to account for the direct and indirect environmental impacts of whole life cycles of commodities and services. While LCA is generally recognized as being more vulnerable to truncation error and inconsistent system boundary definitions, IO is criticized as being too aggregated and insufficiently specific for product level assessments. Overcoming partly these shortcomings, hybrid LCA‑IO methodologies have been developed in recent years, though they yet have to penetrate mainstream practice. Little quantitative knowledge is available concerning the actual magnitude and characteristics of truncation errors or aggregation uncertainties in LCA and IO. Furthermore, few studies have compared the content and structures of LCA and IO datasets to evaluate their compatibility and complementarity under hybridization. In this study, an LCA and an IO database were linked and jointly analysed. It was found that the importance of the representation of the different economic sectors within the LCA dataset did not generally match their environmental importance, and that some sectors of the economy were totally ignored, leading to the concept of "database truncation bias". The levels of precision and specificity of the descriptions of the different economic sectors were found to be complementary between LCA and IO databases. The use of process‑based LCA in assessing the aggregation uncertainties of IO datasets was also demonstrated. Our results point to definite shortcomings and inefficiencies in conventional LCA and IO databases, but also highlight hindrances and limitations of their ad hoc hybridization. This leads us to argue in favour of a more integrated hybrid framework for the elaboration of inventories and databases. Key words : Life cycle assessment, LCA, input-output, IO, life cycle inventory, LCI, industrial ecology, uncertainty analysis, aggregation uncertainty, truncation error, system boundary, hybridization, key sector analysis, linkage, database