How to take time into account in the inventory step: a selective introduction based on sensitivity analysis
Résumé
Life cycle assessment is usually an assessment tool, which only considers steady-state processes, as the temporal and spatial dimensions are lost during the life cycle inventory (LCI). This approach therefore reduces the environmental relevance of certain results, as it has been underlined in the case of climate change studies. Given that the development of dynamic impact methods is based on dynamic inventory data, it seems essential to develop a general methodology to achieve a temporal LCI. This study presents a method for selecting the steps, within the whole process network, for which dynamics need to be considered while others can be approximated by steady-state representation. The selection procedure is based on the sensitivity of the impacts on the variation of environmental and economic flows. Once these flows have been identified, their respective timescales are compared to the inherent timescales of the impact categories affected by the flows. The timescales of the impacts are divided into three categories (days, months, years) based on a literature review of the ReCiPe method. The introduction of a temporal dynamic depends on the relationship between the timescale of the environmental and economic flows on the one hand and that of the concerned impact on the other hand. This approach is illustrated by the life cycle assessment of palm methyl ester and ethanol from sugarcane. In both cases, the introduction of a temporal dynamic is limited to a small proportion of the total number of flows: 0.1 % in the sugarcane ethanol production and 0.01 % in the palm methyl ester production. Future developments of time integration in the LCI and in the life cycle impact assessment (LCIA) are also discussed in order to deal with the need of characterization functions and the recurrent problem of waiting times. This work provides a method to select specific flows where the introduction of temporal dynamics is most relevant. It is based on sensitivity analyses and on the relationship between the timescales of the flows and the timescale of the involved impact. The time-distributed LCI generated by using this approach could then be coupled with a dynamic LCIA proposed in the literature.