Jump to Main Content
Evaluation of a feeding strategy to reduce greenhouse gas emissions from dairy farming: The level of analysis matters
- Van Middelaar, C.E., Berentsen, P.B.M., Dijkstra, J., De Boer, I.J.M.
- Agricultural systems 2013
- European Union, carbon dioxide, climate change, compliance, corn, corn silage, dairy cows, dairy farming, diet, farms, fermentation, grass silage, grasses, grasslands, greenhouse gas emissions, greenhouse gases, income, labor, land use change, life cycle assessment, linear programming, mechanistic models, methane, nitrous oxide, plowing, sandy soils, summer, winter
- The dairy sector contributes to climate change through emission of greenhouse gases (GHGs), via mainly carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O). Replacing grass silage with maize silage is a feeding strategy to reduce enteric CH4 emission. The effect of this strategy on GHG emissions can be analyzed at three different levels: animal, farm, and chain level. The level of analysis might affect results and conclusions, because the strategy affects not only enteric CH4 emissions at animal level, but also other GHG emissions at farm and chain levels. The objective of this study was to determine if the level of analysis influences conclusions about the GHG reduction potential of increasing maize silage at the expense of grass and grass silage in a dairy cow’s diet.First, we used a linear programming (LP, maximizing labor income) dairy farm model to define a typical Dutch dairy farm on sandy soils without a predefined feeding strategy (i.e. reference situation). Second, we combined mechanistic modeling of enteric fermentation and life cycle assessment to quantify GHG emissions at all three levels. Third, continuing from the diet derived in the reference situation, maize silage was increased by 1kg DM per cow per day at the expense of grass (summer), or grass silage (winter). Next, the dairy farm model was used again to determine a new optimal farm plan including the feeding strategy, and GHGs were quantified again at the three levels. Finally, we compared GHG emissions at the different levels between the reference situation and the situation including the feeding strategy. We performed this analysis for a farm with an average intensity (13,430kgmilk/ha) and for a more intensive farm (14,788kgmilk/ha).Results show that the level of analysis strongly influences results and conclusions. At animal level, the strategy reduced annual emissions by 12.8kg CO2e per ton of fat-and-protein-corrected-milk (FPCM). Analysis at farm and chain level revealed first of all that the strategy is not feasible on the farm with an average intensity because this farm cannot reduce its grassland area because of compliance with the EU derogation regulation (a minimum of 70% grassland). This is reality for many Dutch dairy farms with an intensity up to the average. For the more intensive farm, that can reduce its area of grassland, annual emissions reduced by 17.8kg CO2e per ton FPCM at farm level, and 20.9kg CO2e per ton FPCM at chain level. Ploughing grassland into maize land, however, resulted in non-recurrent emissions of 913kg CO2e per ton FPCM. At farm and chain levels, therefore, the strategy does not immediately reduce GHG emissions as opposed to what results at animal level may suggest; at chain level it takes 44years before annual emission reduction has paid off emissions from land use change.