Temporal, spatial and management variability in the carbon footprint of New Zealand milk

The carbon footprint of milk from year-round grazed-pasture dairy systems and its variability has had limited research. The objective of this study was to determine temporal, regional, and farm system variability in the carbon footprint of milk from New Zealand (NZ) average dairy production. Farm production and input data were collected from a national database for 2010/11 to 2017/18 across regions of NZ and weighted on relative production supplied to the major dairy cooperative Fonterra to produce an NZ-average. Total greenhouse gas emissions were calculated using a life cycle assessment methodology for the cradle-to-farm gate, covering all on- and off-farm contributing sources. The NZ-average carbon footprint of milk varied from 0.81 kg of CO2 equivalent (CO2eq)/kg of fat- and protein-corrected milk (FPCM) in 2010/11 (with widespread drought) to 0.75 to 0.78 kg of CO2eq/kg of FPCM in 2013/14 to 2017/18, with a trend for a small decrease over time. Regional variation occurred with highest carbon footprint values for the Northland region due to greatest climatic and soil limitations on pasture production. Dairy cattle diet was approximately 85% from grazed pasture with up to 15% from brought-in feeds (mainly forages and by-products). The CO2 emissions from direct fuel and electricity use constituted <2% of total CO2eq emissions, whereas enteric methane was near 70% of the total. An estimate of potential contribution from direct land use change (plantation forest to pasture) was 0.13 kg of CO2eq/kg of FPCM. This was not included because nationally there has been a net increase in forest land and a decrease in pasture land over the last 20 yr. Data used were highly representative, as evident by the same estimated carbon footprint from 368 farms (in 2017/18) from the national database compared with that from a direct survey of 7,146 farms. New Zealand-specific nitrous oxide emission factors were used, based on many validated field trials and as used in the NZ greenhouse gas inventory, resulting in an 18% lower carbon footprint than if default Intergovernmental Panel on Climate Change factors had been used. Evaluation of the upper and lower quartiles of farms based on per-cow milk production (6,044 vs. 3,542 kg of FPCM/cow) showed a 15% lower carbon footprint for the upper quartile of farms, illustrating the potential for further decrease in carbon footprint with improved farm management practices.