Carbon footprint of raw milk for direct human consumption produced in Po Valley: Data input uncertainties and effect on output variance

A survey on 61 dairy farms in the Po Valley was carried out using an LCA approach to identify milk production's carbon footprint (CF) at the farm gate, with 1 kg of fat and protein corrected milk (FPCM) as a functional unit. Data were collected and grouped into five main categories: feed purchase, in farm feed production, energy & resources, enteric fermentation, and livestock manure. The CF calculation was conducted according to IPCC guidelines 2006. Dairy herds were classified into three groups according to quartiles for the number of lactating dairy cows: small (< 87), medium (between 87 and 270), and large herds (> 270). The overall CF was 1.19 kg CO2-eq kg−1 FPCM, and the major contribution was from enteric fermentation (43.3%), feed purchase (29.1%), and manure handling (14.6%). Small farms have shown a higher CF (1.35 kg CO2-eq kg−1 FPCM) than medium and large herds (1.13 kg CO2-eq kg−1 FPCM, on average). A Monte Carlo simulation analysis showed a high contribution to the variance of CF due to feed purchase, mainly from proteic extra EU and cereals. Results could help plan mitigation initiatives in less performing herds.