Fat- and protein-corrected milk formulation to be used in the life-cycle assessment of Mediterranean dairy goat systems

Abstract The aim of this paper is to develop a new specific algorithm (fat- and protein-corrected milk calculation) for dairy goat to be used in life-cycle assessment (LCA) studies. Though the contribution of goat milk to world milk production ranks third, the literature does not report any specific correction for goat's milk. Using the available bibliographic data, a multiple regression was performed that allowed obtaining the relationship between the energy content (EC) and the fat (FC) and protein content (PC) of dairy goat milk. The multiple regression resulting from the 3 variables analyzed through the data drawn from the literature was significant (R2= 0.99; p ≤ 0.001). The equation resulting from the correlation was used to develop algorithms for the calculation of fat- and protein-corrected milk (FPCM) at specific FC and PC. Since FC and PC are very different depending on the goat breed two different groups have been defined: i) goat breeds with FC in milk below 4% (FCB4), and ii) goat breeds with FC in milk above 4% (FCA4). The EC found for the FCB4 group (FC = 3.70 and PC = 3.27) was 728.11 kcal kg−1, while that of the FCA4 group (FC = 4.92 and PC = 3.61) was 860.69 kcal kg−1. After substituting the EC values obtained before, the resulting FPCM equations to calculate specific FC and PC by unit of mass (kg of milk = M), according to each group, are shown below: FCB4 group = FPCM (FC = 3.70, PC = 3.27) = M * [(0.12 * FC + 0.10 * PC + 0.23)]; FCA4 group = FPCM (FC = 4.92, PC = 3.61) = M * [(0.10 * FC + 0.08 * PC + 0.20)]. Finally, the variation between the FPCM values calculated using the specific equation obtained in this study for goats and those previously published (for sheep or modifications made from dairy cattle) was evaluated; in FCB4 group the differences varied between (-) 32% and (+) 14% and in FCA4 group between (-) 21% and (+) 35%. Values are overestimated if sheep's FPCM calculations are used (because sheep milk has higher fat content than goat milk) and underestimated if dairy cattle's FPCM calculations are used (because cow milk has lower fat content than goat milk). In conclusion, the results found in the present study show the need to use a specific FPCM formulation in the LCA equation for goat's milk.