Assessment of an LCIA method for evaluating biodiversity impact from food production An LCA case study of pork

Abstract

The global loss of biodiversity is mainly driven by land use and land use change due to e.g., agriculture and food production. This study presents a case study of pork that assesses the applicability of the biodiversity life cycle impact assessment method (LCIA) developed by Chaudhary & Brooks (2018). The purpose of the study was to increase the understanding of the method in three aspects: (1) aspects of biodiversity captured in relation to food production, (2) the potential inclusion of land transformational impact, and (3) the spatial resolution. The outcome of the study was intended to provide valuable insights to the development of a database for biodiversity impact from food carried out by RISE (Research Institutes of Sweden), a database to be used for consumer communication purposes. The case study consists of a life cycle assessment (LCA) of three pork production systems with different feed compositions. The functional unit was 1 kg of edible pork meat and covered production phases from cradle to farm-gate. Considerable weight was put on the inventory data collection for land occupation and land transformation flows, where modelling the feed composition and the feed crop cultivation was an essential part. Several methods to assess biodiversity impact in LCA are being developed, in this study one of the most promising methods was selected. What aspects of biodiversity the method capture in relation to food production, was assessed through mapping drivers of biodiversity loss and characteristics of biodiversity to illustrate what parts of these were covered. The evaluation of which spatial resolution of the method to be preferred and how to allocate the impacts from land transformation was assessed through applying different characterization factors to demonstrate the variation in result depending on methodological choices. In some respects, the studied method fails to cover the complexity of biodiversity loss, as it only assesses species loss and how it is impacted by land use interventions, but no other indicator or driver. Yet, the method was found relatively easy to apply and could be used when comparing products of similar character for indicating land use impacts on species. For cropland, three intensity levels are included. However, the method is not developed enough to show differences in a comparative LCA on crops cultivated under different agricultural systems (e.g., differences between organic and conventional). When applying the higher spatial resolution provided by the method, ecoregional approach, the results differ from when applying a higher resolution using the country approach, although not by much. Since applying high spatial resolution was found to be more complicated and time consuming, one should consider focusing on other parameters that might have a larger influence, such as assessing a detailed feed composition. The method is not yet developed enough to include several agricultural land use types nor taxa groups (e.g., insects and microorganisms), two important factors potentially providing a larger difference in result between products of different character. Altogether, the lower spatial resolution might in many cases be good enough. A harmonized way of applying the characterization factors for land transformation was not found. However, when included using the approach of this study, the impact from land transformation had a large contribution to the total biodiversity damage. Whether to include land transformation is suggested to be decided by if the food products under study contributes or have contributed to land use change, and not through system delimitations. Altogether, despite highlighted drawbacks of the method, it serves a purpose by capturing potential differences in biodiversity damage between food products. One should however keep in mind that the method accounts for one only driver of biodiversity decline and one biodiversity indicator.