Van Middelaar C E, Berentsen P B M, Dijkstra J, Van Arendonk J A M, De Boer I J M
Animal Production Systems group, Wageningen University, PO Box 338, 6700 AH Wageningen, the Netherlands.
Business Economics group, Wageningen University, PO Box 8130, 6700 EW Wageningen, the Netherlands.
J Dairy Sci. 2015 Jul;98(7):4889-903. doi: 10.3168/jds.2014-8310. Epub 2015 Apr 23.
Breeding has the potential to reduce greenhouse gas (GHG) emissions from dairy farming. Evaluating the effect of a 1-unit change (i.e., 1 genetic standard deviation improvement) in genetic traits on GHG emissions along the chain provides insight into the relative importance of genetic traits to reduce GHG emissions. Relative GHG values of genetic traits, however, might depend on feed-related farm characteristics. The objective of this study was to evaluate the effect of feed-related farm characteristics on GHG values by comparing the values of milk yield and longevity for an efficient farm and a less efficient farm. The less efficient farm did not apply precision feeding and had lower feed production per hectare than the efficient farm. Greenhouse gas values of milk yield and longevity were calculated by using a whole-farm model and 2 different optimization methods. Method 1 optimized farm management before and after a change in genetic trait by maximizing labor income; the effect on GHG emissions (i.e., from production of farm inputs up to the farm gate) was considered a side effect. Method 2 optimized farm management after a change in genetic trait by minimizing GHG emissions per kilogram of milk while maintaining labor income and milk production at least at the level before the change in trait; the effect on labor income was considered a side effect. Based on maximizing labor income (method 1), GHG values of milk yield and longevity were, respectively, 279 and 143kg of CO2 equivalents (CO2e)/unit change per cow per year on the less efficient farm, and 247 and 210kg of CO2e/unit change per cow per year on the efficient farm. Based on minimizing GHG emissions (method 2), GHG values of milk yield and longevity were, respectively, 538 and 563kg of CO2e/unit change per cow per year on the less efficient farm, and 453 and 441kg of CO2e/unit change per cow per year on the efficient farm. Sensitivity analysis showed that, for both methods, the absolute effect of a change in genetic trait depends on model inputs, including prices and emission factors. Substantial changes in relative importance between traits due to a change in model inputs occurred only in case of maximizing labor income. We concluded that assumptions regarding feed-related farm characteristics affect the absolute level of GHG values, as well as the relative importance of traits to reduce emissions when using a method based on maximizing labor income. This is because optimizing farm management based on maximizing labor income does not give any incentive for lowering GHG emissions. When using a method based on minimizing GHG emissions, feed-related farm characteristics affected the absolute level of the GHG values, but the relative importance of the traits scarcely changed: at each level of efficiency, milk yield and longevity were equally important.
育种有潜力减少奶牛养殖中的温室气体（GHG）排放。评估遗传性状发生1个单位变化（即遗传标准差提高1个单位）对整个产业链温室气体排放的影响，有助于深入了解遗传性状在减少温室气体排放方面的相对重要性。然而，遗传性状的相对温室气体值可能取决于与饲料相关的农场特征。本研究的目的是通过比较一个高效农场和一个低效农场的产奶量和寿命值，评估与饲料相关的农场特征对温室气体值的影响。低效农场未采用精准饲喂，每公顷饲料产量低于高效农场。产奶量和寿命的温室气体值通过使用全农场模型和两种不同的优化方法来计算。方法1通过最大化劳动收入来优化遗传性状变化前后的农场管理；对温室气体排放的影响（即从农场投入生产到农场大门）被视为一种副作用。方法2通过在保持劳动收入和牛奶产量至少不低于性状变化前水平的同时，使每千克牛奶的温室气体排放量最小化，来优化遗传性状变化后的农场管理；对劳动收入的影响被视为一种副作用。基于最大化劳动收入（方法1），低效农场产奶量和寿命的温室气体值分别为每头奶牛每年每单位变化279千克和143千克二氧化碳当量（CO2e），高效农场为每头奶牛每年每单位变化247千克和210千克CO2e。基于最小化温室气体排放（方法2），低效农场产奶量和寿命的温室气体值分别为每头奶牛每年每单位变化538千克和563千克CO2e，高效农场为每头奶牛每年每单位变化453千克和441千克CO2e。敏感性分析表明，对于这两种方法，遗传性状变化的绝对影响取决于模型输入，包括价格和排放因子。仅在最大化劳动收入的情况下，由于模型输入变化，性状之间的相对重要性才会发生显著变化。我们得出结论，与饲料相关的农场特征假设会影响温室气体值的绝对水平，以及在使用基于最大化劳动收入的方法时性状在减少排放方面的相对重要性。这是因为基于最大化劳动收入来优化农场管理不会对降低温室气体排放产生任何激励。当使用基于最小化温室气体排放的方法时，与饲料相关的农场特征会影响温室气体值的绝对水平，但性状的相对重要性几乎没有变化：在每个效率水平上，产奶量和寿命同样重要。
