A farm-scale framework to assess potential farm- and regional-scale implications of removing palm-kernel expeller as a supplementary feed for dairy cows

Farm-scale models were integrated with spatially discrete estimates of pasture production to examine the potential farm and regional implications of removing palm-kernel expeller (PKE) as a supplementary feed from dairy farms in Southland, New Zealand. The following two farm-production systems representing the majority of dairy farms in the region were modelled: a System 3 farm (D3; mid-intensification, with 10–20% of imported feed) and a System 4 farm (D4; mid- to high intensification, with 20–30% of imported feed). Within each system, the impact of the following four PKE options was explored: (1) a control with PKE (Baseline); (2) no PKE, with fewer cows producing the same amount of milk per cow as in Baseline; (3) no PKE, with the same number of cows producing less milk per cow than in Baseline; and (4) PKE replaced with barley grain. Barley grain provides for similar flexibility (timing of purchase and feeding), and can be sourced locally. Faced with the need to remove PKE as a dietary ingredient, farmers would benefit from adopting the second PKE option (no PKE, with fewer cows producing the same amount of milk per cow as in Baseline); farm-operating profits were reduced by only 3% (compared with 30% of System 4 farms adopting the third PKE option, i.e. no PKE, with the same number of cows producing less milk per cow than in Baseline) relative to the Baseline farms. The narrow range of mean annual nitrate-nitrogen (nitrate-N) leaching losses (estimates ranged from 30 to 33 kg N/ha) reflects similar estimates of N intake and N excreted in urine across the modelled options. Substantial amounts of barley grain would need to be transported into the region or produced locally to replace PKE.