Teixeira, R.F.M., Morais, T.G., Barão, L. & Domingos, T. (2016) Improving regionalized life cycle inventories with mass-balance models and scale-consistent data. Proceeding of LCA Food 2016 Conference, Dublin, Ireland, 1228-1235.NON-cE3c affiliated
Objective: Among all Life Cycle Assessment (LCA) stages, life cycle inventories (LCI) commonly demand the most time and effort. LCI datasets are rarely fully regionalized, particularly for agri-food materials. Agricultural products inputs and outputs crucially depend on location and technology. The level of regional representability of materials in standard databases is thus indeterminate. Here we propose a new mass-balance model to build inventories that are fully scale-consistent.
Method: We present the model framework that can be used to obtain the inputs from nature and the technosphere and the outputs to air, water and soil. We implemented the framework for carbon (C) and nitrogen (N) and present results for the soil N sub-model. The N cycle is determined for the main agricultural exports in Portugal (e.g. maize, vine). We used activity data from Portuguese agricultural fact sheets and quantified the parameters in the balance using process-based models for each variable (e.g. regional productivity, soil organic matter accumulation and loss, nitrate leaching) that use spatial
environmental data. The model was applied at regional scales, and to assure scale consistency, in the sense introduced by Morais et al (2016). We inserted as a restriction that the aggregated sum of all inventory flows is equal to country -level totals.
Results: We provide regional results that include all soil N-related inputs and outputs in the country. The model provides direct inventory flows that can be used in highly spatialized LCI. We found that the most important advantage is the spatial dimension of results, which due to the scale-consistency of data can be aggregate at any level of detail from the local to the country scale. Nitrate leaching is the main source of uncertainty.
Conclusion: LCI is lagging behind the geospatial detail of most recent impact assessment models. The approach presented here enables the construction of more powerful inventories relating field data with regional or country aggregates.