SETAC Globe - Environmental Quality Through Science
 
  6 October 2011
Volume 12 Issue 10
 

Return to the Globe

Life Cycle Inventory (LCI) Modelling and Attributional/Consequential Issues

Marc-Andree Wolf European Commission, Ispra, Italy and Tomas Ekvall, IVL Swedish Environmental Research Institute, Göteborg, Sweden

The Milan session Life Cycle Inventory (LCI) Modelling and Attributional/Consequential Issues tackled one of the most discussed methodological issues in life cycle assessment (LCA): how to model the inventory of the life cycle. Many LCA practitioners recognise two archetypal approaches to modelling the inventory of the life cycle. Attributional LCA (ALCA) aims to describe the environmentally relevant physical flows to and from a life cycle and its subsystems, and consequential LCA (CLCA) aims to describe the environmental consequences of an analysed decision (e.g., on alternative raw material sources). CLCA is broader in scope in the sense that a CLCA should include all processes that they are expected to be affected by the decision, regardless of whether or not they are part of the existing supply-chain (i.e. also effects that occur via market mechanisms). This includes the identification of affected production technologies, but can also include constraints, rebound effects and positive feedback mechanisms. Modelling of different types of consequences is essential to CLCA, which means that a CLCA could potentially involve a host of tools such as equilibrium models and experience curves.

The choice of specific approaches based on ALCA, CLCA, or a combination of the two should depend on the decision-context of the study and on the knowledge needs of the target audience. This session was concerned with how the LCA methodology can be systematically adapted to the context and knowledge needs. Particular focus was on the choice between ALCA- and CLCA-based approaches, and on the development and application of specific approaches for CLCA. Different modeling approaches were presented as well as proposals regarding how to identify the approach that is most applicable to the decision context. Scenario analysis of modeling assumptions was suggested to increase robustness of decision support. The need for further development, particularly of CLCA was highlighted, since only few of the different types of consequences are supported with modeling approaches so far and the higher structural and data uncertainty are challenges.

The complexity of real world market systems, especially of goods with international markets, was illustrated using specific case study examples. The need for expanding the classical process based modeling of LCA with simulation models such as general equilibrium models was highlighted. The step from modelling single marginal technologies to a mix of marginal technologies was found necessary to better capture real world situations. One presentation addressed the translation of general guidance into sector or product group specific guides. That was seen as a step to make LCA more straightforward to apply, for better reproducibility and acceptance. Across the speakers, while the need for goal and scope dependent modeling was agreed upon, the question of which particular approach would be best suited for which situation was answered in different ways. The “scientific” challenge will be to overcome disagreements on methodological preferences – at least regarding current good practice – to effectively support incorporation of LCA in a policy and wider business context. Key criteria for recommendations on good practice in LCA are: availability and applicability of data and models, acceptance and effectiveness of communicating results, and robustness of decision support.

Reinout Heijungs’ lead-off presentation “Consequential LCA, Attributional LCA and Scenarios” highlighted the important role of scenarios in both ALCA and CLCA, and introduced the new concept of backcasting LCA. The question of attributional and consequential was seen as a continuum, with substitution being part of both principles, instead of an “either-or” divide. Predictive, explorative, normative questions may favor which approaches and kinds of scenarios are best employed, favoring modeling approaches more from the kind of question to be answered over the method applied.

Rolf Frischknecht proposed a scope-dependent choice between ALCA and CLCA models. It was argued that the decision support question may not necessarily require CLCA modeling. For example, supply-chain contracts should not be altered based on CLCA when supporting medium scale-decisions (e.g., on company level) because the consequences of the decisions would be too small to cause actual market effects. Differentiation between small, medium and large scale decisions should be made according to the “size” of the decision analysed, i.e., the relative economic value of the affected share of a market.

Paolo Masoni presented work on operational LCA guidance for fuel cells under the European Commission’s Joint Technology Undertaking (JTI), describing methodological challenges that have been encountered. Specific guidance is being developed for fuel cell-related LCA data development and studies, starting from the need for more consistent, straightforward and efficient sector-specific guides and in line with the International Reference Life Cycle Data System (ILCD) Handbook. This will steer future EU research investments in this area. Data collection templates, training material etc. will ease application. The LCI modeling approach is an attributional approach that includes substitution of avoided burdens, as specified for micro-level decisions in the ILCD Handbook.

Ben Amor presented a study on electricity trade analysis and marginal technologies in consequential LCA, in the case of Quebec’s hydropower in the northeastern American market. The question was how the trade of electricity affects actual greenhouse gas (GHG) emissions related to electricity production in neighboring regions. One element found particularly relevant was the timing of electricity trade (diurnal and across the seasons), due to time-dependent price differences, which affected short-term marginal technologies for power production. Integrated over an entire year, a combination of two main marginal technologies (natural gas and hard coal combustion) was identified. It was shown that combined analysis of the two marginal technologies substantially better captures the actual GHG emissions, compared to analyses of single marginal technologies. Scenario analysis should at least be used, in cases where it is not possible or feasible to determine the specific marginal mix.

Thomas Dandres presented the outcome of his dissertation research on European energy policy analysis. The research used the general equilibrium model GTAP in a consequential and prospective life cycle assessment framework. T he prospective study analysed potential consequences of a bioenergy policy over a 20-year period, with expected benefits for human health but higher impact on natural ecosystems. A number of sensitive macroeconomic parameters were identified for the GTAP model (gross domestic product, population, capital, labor force, and technology innovation). The research again highlighted the need for scenario analysis and better uncertainty assessment due to the inherently high uncertainties in these kinds of studies.

In the poster spotlights, further methodological challenges and solutions were presented, as well as applications to specific cases. Yalda Shayeghi informed us about parameterising land use change patterns and applying demand forecast scenarios to achieve robust bioenergy decision support. Antonino Marvuglia presented a poster on partial and general equilibrium modeling to integrate consequential effects of indirect land use changes (ILUC) in a biogas LCA. Richard Wallace focused on the role of constraints, in a constraint-based linear programming model for performing LCI analysis. This has been applied to a case study on the energy valorization (price-setting) of grape marc (a biomass waste of the vinting industry). Samuel Vionnet presented a poster on the current lack and needs for LCI data for water footprinting.

Author contact information: marc-andree.wolf@jrc.ec.europa.eu, tomas.ekvall@ivl.se

Return to the Globe

 
SETAC mission statement Contact SETAC Globe
Contact the SETAC North America office
Contact the SETAC Europe office