Biodiversity footprints of 151 popular dishes from around the world

requirements and only mammals, birds and amphibians were considered. Similarly, the ranges of species could be refined to consider only area of suitable habitat [25]. Our analyses were standardised based on calorie content but this does not necessarily reflect nutritional needs. This standardization contributes to the result by which low biodiver- sity footprint dishes identified are mostly carbohydrates with low nutritional content. Future analyses standardised on nutritional needs such as protein, essential lipids, minerals, and vita- mins is needed to better reveal nutritious dishes with low biodiversity footprint for a step towards sustainable diets. Our list of dishes is far from exhaustive and not likely to be representative of the dishes of the world. One factor is that we focused on dishes from countries that are top 25 in terms of GDP. This led to no dishes from Africa. In addition, due to constraints resulting in the com- plexity of intermediate products used in recipes, we did not exhaustively analyse all available dishes in a given country. Our dish sample needs thus to be regarded as an illustration of the potential of the methods to be applied to a variety of dishes. Our limited sample of dishes high- lights the need to develop datasets of biodiversity footprint of main ingredients and products used in dishes worldwide. While this study used the world biodiversity footprint in globally produced scenarios to consider the possibility of global trade, future research could consider using the biodiversity

footprints of specific countries that the ingredients were sourced from considering actual trade links with their embedded biodiversity impacts [8]. Finally, dishes recipes can present high variability which could lead to different results. Conclusion Our study attempts to provide a higher level of aggregation of units of consumption, i.e. dishes, with regards to their biodiversity footprint across multiple countries. Overall, our results point to the importance of certain ingredients and regions of production in the overall biodiversity footprint of a dish, namely dishes with high biodiversity footprints include beef dishes such as fraldinha (beef cut) originating from Brazil and legume dishes such as chana masala (chickpea curry) from India. On the other hand, vegan and vegetarian dishes tend to have low biodiver- sity footprints. Comparisons between locally and globally produced scenarios suggests the importance of dish’s country of origin to determine the source of production with a lower bio- diversity footprint where India was observed to be involved with the production of mostly high biodiversity footprint dishes with biodiversity impacts driven by ingredients (e.g. rice, legumes, chicken) that are not commonly flagged as having a high environmental footprint. Moving towards the estimation of dishes’ biodiversity footprints across multiple countries can help to fill consumers’ knowledge gap in an accessible manner. A combination of consumer awareness to facilitate a transition towards sustainable diets is imperative to mitigate the large

impacts of food production on biodiversity. Supporting information S1 Table. Outliers and their replacement crops datasets used in the calculation of biodiver- sity footprint of dishes. (DOCX) S2 Table. Correspondence between ingredients and crop and pasture categories in the bio- diversity footprint dataset. (XLSX) PLOS ONE Biodiversity footprint of popular dishes PLOS ONE | https://doi.org/10.1371/journal.pone.0296492 February 21, 2024 11 / 15