Life cycle assessment of milk production in Italian intensive dairy farms

Environmental concerns are having increasing priority upon political, social, and economic agendas, in particular when related to agriculture. Food production has an environmental impact, and as the global populations continue to increase, it is critical to produce sufficient high-quality food from a finite resource supply in order to mitigate the effects upon the environment (Capper et al., 2009). In the North of Italy favorable climatic and infrastructural conditions promoted a great concentration of livestock farms with intensive utilisation of natural resources (i.e. land, air, water). The objective of this study was to assess the environmental impact of milk production in intensive dairy systems, in order to identify farm characteristics that guarantee at the same time low environmental impact and economic sustainability. A cradle to farm gate Life Cycle Assessment (LCA) was performed on 41 intensive dairy farms in Northern Italy. In addition to the evaluation of greenhouse potential, impact categories as acidification, eutrophication, land use and energy use were considered. The functional unit was 1 kg Fat and Protein Corrected Milk (FPCM). Farm key parameters about crop production for feeding, livestock, manure management, purchased feed, fertilisers, pesticides, electricity and fuels, milk and meat sold were collected by personal interview to the farmers. LCA was carried out using LCA software package, SimaPro 7.3.2 (Pr\ue9 Consultants, 2011). Gross margin, i.e revenues minus the direct production costs, excluding labour costs (expressed in \u20ac/t FPCM), was used as economic indicator. Database was analysed using the CLUSTER procedure (SAS, 2000). In order to identify different farming systems the following variables were considered: gross margin, feed self-sufficiency, dairy efficiency and stocking density. Two main clusters of farms were identified (A and B); moreover in each of the two clusters two subgroups of farms were defined (Table 1). Farms from cluster B were slightly less intensive than farms from cluster A: they had significantly larger farm land (ha), lower stocking density (LU/ha) and higher feed self-sufficiency (%). Economic results were similar between the two main clusters but ecological performances were better for farms from cluster B: nitrogen and phosphorus balances at farm gate (kg/ha) and the off-farm components of total climate change, acidification, eutrophication, energy use and land use per kg FPCM were significantly lower in cluster B than in A. In fact farms from cluster B had higher feed self-sufficiency and purchased less feed, reducing off-farms fraction of all impacts but increasing on-farm component of eutrophication, acidification and land use. Farms from cluster B impacted more in term of total eutrophication (on- and off-farm components) in comparison with farms from cluster A. Onfarm crop production weighted for the 50% on eutrophication, because of the use of fertiliser which could determine nitrate leaching in the water and ammonia emission in the air. Considering the subgroups, farms from cluster 4 had better economic performances than cluster 3; they were characterised by low stocking density, high feed self-sufficiency and balanced partition of farm land among different crop production (lucerne: 15.0% ; grass: 14.9%; maize for silage 21.2% of farm land). Farm included in cluster 4 had lower nitrogen and phosphorus balances than cluster 3; they probably paid more attention in using fertilisers (134 vs 178 kg of N input from artificial fertilisers) and sold feed (446 vs 0 kg of N output from sold feed). The energy use of cluster 4 was lower than cluster 3 (P=0.10) as a consequence of reduced use of off-farm products, especially feeds. In the context of intensive dairy farming of Northern Italy cluster 4 identifies a type of farming system that can produce good economic performances without increasing environmental impact