Life cycle inventory of 23 dairy farms in south-western Sweden.

The importance of obtaining greenhouse gas (GHG) and corresponding resource use data on dairy farms cannot be emphasised enough, as baseline data is required to evaluate where to put emphasis in mitigation or change. The publication from South-Western Sweden cited provides such an opportunity. The authors collected data and use Life Cycle Assessment (LCA) methodology in analysis. The purpose of the study was to gain knowledge of the environmental impact of contemporary milk production and how farms differ in resource use and emissions.

Data were collected from 23 farms. The farms were divided into three groups: (1) nine conventional high producing farms producing more than 7500 kg energy corrected milk (ECM)/ha (Conv High); (2) eight conventional medium producing farms producing less than 7500 kg ECM/ha (Conv Med), and (3) six organic farms (Org). All farms participated in the official advisory programme and had detailed information on feed and fertiliser consumption. In addition, data were collected on use of diesel, electricity, pesticides and plastics. For ammonia, nitrate, nitrous oxide and methane, different emission models were used in order to infer the losses. The functional (reference) unit was one kg ECM at the farm gate. Emissions of ammonia and nitrate, as well as pesticide use, were also evaluated per ha of arable land. The systems analysed included all phases in the life cycle of fertilisers, feed products, diesel, pesticides and plastics. Transport was also taken into account, but not buildings, machinery and medicines. Allocation between milk and meat was done on an economic basis, distributing 90% of the impact to milk and 10% to meat. Economic allocation was also used in the Life Cycle Inventory (LCI) of the concentrate feed. In the LCIs of the 23 farms, the results were presented as use of resources (energy, nutrients and land), use of pesticides, emissions of GHG, and emissions of nitrifying and acidifying substances. The LCI results were statistically analysed by ANOVA.

The results showed the following: (1) Resources: There was no significant difference in the use of energy resources between the two conventional groups. The overall energy use was significantly lower for the organic farm group in comparison to the conventional groups, and it was the total lower use of fossil fuels that explained this difference. The use of non-renewable resources for nutrient supply (P and K) was also significantly lower for the organic group. (2) Land use: The yearly land use for producing one kg of milk was significantly higher for the organic group, for the arable land, as well as the natural grazing areas. (3) Pesticides: The pesticide use was 71-81 mg active substance per kg milk for the conventional groups, with no significant difference between them. In the organic group the average pesticide use was 7.8 mg active substance per kg milk, which was due to the conventional feed ingredients in the purchased feed. (4) GHG emissions: The variation between farms was large, ranging from 760 to 1260 g CO₂ eq/kg milk for the conventional farms and 730 to 1110 g CO₂ eq/kg milk for the organic farms, and consequently no significant differences between the three farm groups. (5) Ammonia and nitrous oxide emissions: The calculated emissions/ha of arable land were highest in the Conv High group due to the higher livestock density in this group, but the difference with the other groups was not significant when expressed per kg milk. (6) Farm N surplus: The numbers were respectively 166, 122 and 66 kg N/ha for Conv High, Conv Med and Org, the latter being significantly different from the other groups.

In summary: Although LCI data were collected from 23 dairy farms in an area with uniform and favourable conditions for milk production, the results showed large variations between farms. Better knowledge of these variations and their cause can assist benchmarking and environmental improvement on farms. The differences in parameters between the two conventional groups were mostly not significant, but the organic farms had significantly lower use of fossil energy, P and pesticides, and larger land use. Greenhouse gas emissions did not differ between production system. The N-surplus was lower at the organic farms, but there was no significant difference between farm groups in the area-based nitrate leaching. The area-based ammonia emissions were primarily due to livestock density and not production system.