The global warming potential of milk production is a key aspect in the assessment of its eco-efficiency. The carbon footprint summarises the climate-relevant emissions of the production cycle, with this study focussing on the emissions from cradle to farm gate as boundaries. Numerous life cycle assessments were published in recent years, presenting the carbon footprint for different regions, production systems and management scenarios. However, despite the obvious high scientific interest in this topic, general conclusions on the climate-friendliness of contrasting production systems can hardly be drawn and there is no clear answer to the question of whether or not grazing systems provide an overall more climate-friendly alternative to confinement systems.

To resolve this discussion, a meta-analysis was applied to a dataset, created with data from a selection of 30 published life cycle assessments, comprising in total 87 carbon footprint values from 15 different countries. After a standardisation process, three production system categories (pasture-based, mixed and confinement) were defined based on feeding parameters. Aside from the comparison of the production systems, the effects of various production variables (pasture and concentrate intake, milk yield per cow, milk yield per kg of metabolic live weight, mineral nitrogen fertilisation, feed efficiency and replacement rate) were analysed. Therefore, linear and level-log mixed models were developed and analyses of covariance performed. As the dataset used for the analysis covered a large range of different intensities per production system, the results are based on a robust analysis and can be extrapolated to any milk production system with known in- and outputs.

The results show that increases in milk yield, pasture intake and feed efficiency decreased the carbon footprint of milk significantly, independent of the production system. However, the mitigation potential is limited across production systems with regard to their productivity thresholds. The comparison of the systems without consideration of other production parameters did not reveal any significant differences. When controlling for milk yield, however, the pasture-based system achieved lower carbon footprints compared to the other production systems. Thus, irrespective of the importance of milk yield for emissions and the generally lower milk yields of the pasture-based production system, this system still offers a competitive climate impact.

牛奶生产的全球变暖潜力是评估其生态效率的关键方面。碳足迹总结了生产周期中与气候相关的排放，而本研究则侧重于从摇篮到农场大门的排放。近年来发布了许多生命周期评估，介绍了不同地区，生产系统和管理方案的碳足迹。然而，尽管该主题显然具有很高的科学兴趣，但几乎无法得出关于对比性生产系统的气候友好性的一般性结论，并且对于放牧系统是否提供总体上更有利于气候的替代方案这一问题尚无明确答案。限制系统。

为了解决这一讨论，对数据集进行了荟萃分析，该数据集是从30种已发布的生命周期评估中选择的数据创建的，其中包括来自15个不同国家的总计87个碳足迹值。经过标准化过程后，根据进料参数定义了三个生产系统类别（基于牧草的，混合的和封闭的）。除了比较生产系统外，还分析了各种生产变量（牧草和精饲料的摄入量，每头牛的产奶量，每公斤代谢活重的产奶量，矿质氮肥，饲料效率和替代率）的影响。因此，开发了线性和水平对数混合模型，并进行了协方差分析。由于用于分析的数据集涵盖了每个生产系统很大范围的不同强度，

结果表明，与生产系统无关，牛奶产量，牧草摄入量和饲料效率的提高显着降低了牛奶的碳足迹。但是，就生产力阈值而言，整个生产系统的缓解潜力有限。在不考虑其他生产参数的情况下进行的系统比较没有发现任何显着差异。但是，在控制牛奶产量时，与其他生产系统相比，基于牧场的系统的碳足迹更低。因此，尽管基于牧场的生产系统的牛奶产量对于排放的重要性和通常较低的牛奶产量，该系统仍然具有竞争性的气候影响。