Assessing the environmental impact of diet – Influence of using different databases of foods' environmental footprints

Current food systems compromise many environmental impact boundaries, and a dietary transition towards sustainable dietary patterns is needed. The definition of policies to promote the dietary transition implies accurate knowledge of dietary choices' environmental impact. For its assessment, life cycle assessment (LCA) has been employed, but methodological heterogeneity and limited access to LCA data hinder comprehensive comparisons. Several publicly accessible standardized foods’ environmental impact databases have been developed to address these challenges. However, variations in indicators included and data sources raise questions about their impact on assessing dietary environmental footprints and the correlation amid various indicators within and between databases. This study aims to evaluate the effect of using different public-access food LCA databases on estimating the individual dietary environmental impact in a nationally representative survey through multiple indicators as well as the correlations between them. Food-specific environmental impact indicators data from three databases (Poore&Nemececk, SHARP-ID, SU-EATABLE LIFE) were merged with individual food consumption data from the Portuguese Food, Nutrition, and Physical Activity Survey (IAN-AF, 2015–2016) (n = 5811) to estimate the usual environmental impact of diet for each indicator. Food groups' percentual contribution (%) to the environmental footprints and the Pearson correlation between indicators were also estimated. Our results showed that different databases of foods’ environmental impacts led to diverse estimates for common indicators when linked to the same food consumption data (e.g., GHGE – P50(P25–P75) SHARP-ID: 4.42(3.44–5.64)kgCO2eq, Poore&Nemececk: 6.17(4.46–8.41)kgCO2eq, SU-EATABLE LIFE: 5.64(4.26–7.36)kgCO2eq). However, except for water footprints, most indicators from all databases were highly correlated with each other, with meat and other animal-based foods as the top contributors. In conclusion, the absolute differences observed can compromise the validity of the findings and their comparability with other countries' estimates when different LCA data are used. A standardized, consolidated database covering a judicious selection of indicators across Europe along with official guidelines for assessing dietary environmental impacts would facilitate its assessment benefiting the establishment of food sustainability policies and recommendations at national and pan-European l