Biodiversity footprints of 151 popular dishes from around the world

impacts of pesto are calculated from its ingredients such as olive oil, basil, pine nuts, facilitating the purchasing decision of consumers that buy pesto directly) [27]. Following the analogy with composite products, a fundamental higher level of aggregation for consumption decisions are dishes. Individuals decide what to consume at a restaurant or cook at home at the dish level, rather that at the primary ingredients level, following cultural and social factors, e.g. an Indian household may consider chicken chaat (dish made of chicken, vegetables and spices) versus chana masala (chickpea curry), or, in Brazil a common choice could be fraldinha (beef cut popular for BBQ, “churrasco”) versus arroz carreteiro (rice with grilled meat dish). Our knowledge of the biodiversity footprint at the dish level across multiple countries is limited, even though it could be very useful to simplify the choices of environmen- tally conscious consumers. In the context of low biodiversity footprint information at the dish level across multiple countries, here we aim to calculate the biodiversity footprints of 151 popular dishes from around the world as an illustration. We do this assuming both locally and globally produced scenarios and under assumptions of feedlot-grown and pasture-grown livestock. We use recent biodiversity footprint databases of crops that are based on the concept of species ranges encroached by specific crops [26]. Methods Overview We considered popular dishes from countries that were in the top 25 in terms of gross domes-

tic product in 2019. Dishes were selected from Cable News Network and when a list of dishes was not available for a given country, Taste Atlas was used instead [28, 29]. A total of 151 pop- ular dishes were selected. The dishes were grouped into three categories: vegan, vegetarian, and dishes containing meat. Vegan dishes do not contain any animal-derived products, vege- tarian dishes do not contain meat but include animal products such as milk and egg. Dishes containing meat include dishes with meat, poultry or fish. The biodiversity footprint was calculated using three biodiversity indicators, namely species richness, threatened species richness, and range rarity affected by converting natural habitat to cropland or pastureland. Additionally, we considered four scenarios, feedlot-grown and locally produced, feedlot-grown and globally produced, pasture-grown and locally produced, and pasture-grown and globally produced. In the globally produced scenarios, biodiversity foot- print was calculated based on the global distribution of species and crops, while the locally pro- duced scenarios were calculated at the country level. The biodiversity footprint of each of the ingredients composing the popular dishes after a calorical contribution normalization was assessed. This was done using biodiversity indexes [29] which analysed the agricultural footprint of individual crops and pasture on three taxa (mammals, birds, and amphibians). For livestock ingredients, they were translated to the crop

and pasture inputs necessary to produce the livestock ingredient before using the biodiversity indexes. All data analysis and figures were produced in R version 4.1.1 [30]. Biodiversity footprint of ingredients The biodiversity footprint per unit of local crop yield (biodiversity footprint Mg-1) was calcu- lated using the abovementioned biodiversity indicators [26]. The biodiversity footprint related to a crop or pasture area was the number of species, number of threatened species, or range rarity affected by converting natural habitat to cropland or pastureland. Biodiversity footprint per: (i) unit of local crop yield; and (ii) across agricultural areas datasets were used to calculate the biodiversity footprint of each dish ingredient. Dataset (i) was used for calculating the PLOS ONE Biodiversity footprint of popular dishes PLOS ONE | https://doi.org/10.1371/journal.pone.0296492 February 21, 2024 3 / 15