Towards a more comprehensive life cycle assessment framework for assessing toxicity-related impacts for livestock products: The case of Danish pork

In life cycle assessments of livestock systems, toxicity-related impacts are not commonly considered or only specific aspects (such as pesticides, manufacturing of inputs) are assessed. In this context, the aim of this study was to define a framework for assessing toxicity-related impacts and to characterize human toxicity and freshwater ecotoxicity for a livestock product based on applying the state-of-the-art models PestLCI Consensus and USEtox. Furthermore, methodological gaps were discussed and ways forward were suggested. The case study focused on Danish pork production and the toxicity results were reported per kg 'meat' (the parts of pig used for human consumption) leaving the slaughterhouse. The assessment framework included the use of pesticides and heavy metals in feed production, the use of veterinary pharmaceuticals in pig production, and the manufacturing of inputs. The use of cleaning agents could not be assessed with the currently available methods. New characterization factors were calculated for 35 chemicals not available in USEtox. For Danish pork production, feed production was the main contributor to the analyzed toxicity impacts. The use of pesticides was the main driver for organic substances while heavy metal emissions related to the application of pig manure to fields were the hotspot for metal-based substances. The use of veterinary pharmaceuticals contributed only to freshwater ecotoxicity by 3%. PestLCI Consensus estimates were compared with different approaches. The impact of metabolites of pesticides and veterinary pharmaceuticals was assessed and discussed. Methodological gaps and research needs were identified regarding the assessment of pesticides, veterinary pharmaceuticals, metal-based substances, inorganic substances, and combined exposure to multiple chemicals. Better data related to the use and chemical properties of substances are needed to reduce uncertainty in toxicity modeling.