Comparison of carbon footprint and water scarcity footprint of milk protein produced by cellular agriculture and the dairy industry

Purpose This paper studies the carbon footprint and water scarcity footprint (WSF) of a milk protein, beta-lactoglobulin, produced by cellular agriculture and compares this to extracted dairy protein from milk. The calculations of the microbially produced proteins were based on a model of a hypothetical industrial-scale facility. The purpose of the study is to examine the role relative to dairy of microbially produced milk proteins in meeting future demand for more sustainably produced protein of high nutritional quality. Methods The evaluated process considers beta-lactoglobulin production in bioreactor cultivation with filamentous fungi T. reesei and downstream processing for product purification. The model considers four production scenarios in four different locations (New Zealand, Germany, US, and Australia) with a cradle-to-gate system boundary. The scenarios consider different sources of carbon (glucose and sucrose), different options for the fungal biomass treatment (waste or animal feed) and for the purification of the product. Allocation to biomass was avoided by considering it substituting the production of general protein feed. The carbon footprint and WSF (based on AWaRe factors) modelling is compared to calculations and actual data on extracted dairy protein production in NZ. The uncertainties of modelled process were addressed with a sensitivity analysis. Results and discussion The carbon footprint of microbially produced protein varied depending on the location (energy profile) and source of carbon used. The lowest carbon footprint (5.5 t CO(2)e/t protein) was found with sucrose-based production in NZ and the highest (17.6 t CO(2)e/t protein) in Australia with the glucose and chromatography step. The WSF results varied between 88-5030 m(3) world eq./t protein, depending on the location, type of sugar and purification method used. The avoided feed production had a bigger impact on the WSF than on the carbon footprint. Both footprints were sensitive to process parameters of final titre and protein yield from sugar. The results for milk protein were of similar magnitude, c.10 t CO(2)e/t protein and 290-11,300 m(3) world eq./t protein. Conclusions The environmental impacts of microbially produced milk protein were of the same magnitude as for extracted dairy protein. The main contributions were sugar and electricity production. The carbon footprints of proteins produced by cellular agriculture have potential for significant reduction when renewable energy and more sustainable carbon sources are used and combined with evolving knowledge and technology in microbial production. Similarly, the carbon footprint of milk proteins can potentially be reduced through methane reduction technologies.