Environmental and animal welfare impacts of different dairy production systems: a holistic approach.

传统的英国乳制品生产，牛平均有6个月的放牧时间，并在冬季饲养几个月，已被证明对温室气体排放、生物多样性、土壤健康和水污染有很高的环境影响。研究和农业社区已经确定了两种潜在的更可持续的方法：再生放牧系统，它将保护和粮食生产整合在同一块土地上(“土地共享”)；以及可持续强化系统，将高产粮食生产农场与受保护的栖息地分开进行保护(“土地节约”)。由于这些乳品系统使用的管理方法不同，它们产生了不同的环境、经济和动物福利结果。此外，社会对乳制品生产环境问题的认识不断提高，很可能已经影响了公众对乳制品生产系统及其生产的牛奶的态度。在农场层面评估动物福利是耗时的，因此，已提议使用人工智能(AI)监视工具(在摄像机的帮助下)来识别饲养系统中的关键动物福利特征，但尚未在放牧条件下推广。该项目旨在评估传统、再生和可持续强化奶制品系统的环境、经济和动物福利方面，包括通过新的基于视觉的机器学习技术，以及社会对它们的看法。成功的候选人将有机会参加一个多学科小组，在这个小组中，可持续动物生产、经济和环境足迹评估、动物行为和福利、数据科学机器视觉和人工智能以及公众感知评估方面的学者专家将共同对乳品生产系统进行全面评估。学生将受益于能够使用布里斯托尔大学和罗萨姆斯特德大学的研究设施，并将接受培训并获得对动物生产系统的不同方面进行深入评估所需的技能。此外，学生还将有机会与我们的合作伙伴牧场饲养牲畜协会(PFLA)的农民、研究人员和顾问直接接触。

Conventional UK dairy production, where cattle have on average 6 months grazing and spend the winter months housed has been shown to have high environmental impacts on greenhouse gas emissions, biodiversity, soil health and water pollution. The research and farming communities have identified two potentially more sustainable approaches: regenerative grazing systems, which integrate conservation and food production on the same land ("land sharing"); and sustainably intensified systems, which separate high-yield food production farms from protected habitats for conservation ("land sparing"). Since these dairy systems differ in the management practices they use, they result in different environmental, economic, and animal welfare outcomes. Moreover, it is likely that the rising societal awareness of dairy production environmental concerns has affected public attitudes regarding dairy production systems and the milk they produce. Assessing animal welfare at the farm level is time consuming and thus, the use of artificial intelligence (AI) surveillance tools (aided by video cameras) has been proposed to identify key animal welfare traits in housed systems but has not been extended under grazing conditions yet. This project aims to evaluate the environmental, economic and animal welfare aspects of conventional, regenerative and sustainably intensified dairy systems, including through novel vision-based machine learning technologies, as well as societal perceptions towards them. The successful candidate will have the opportunity to participate of a multidisciplinary group where academics experts in sustainable animal production, economic and environmental footprint assessment, animal behaviour and welfare, data science machine vision and artificial intelligence and public perception evaluation have joined together to undertake a comprehensive evaluation of the dairy production systems. The student will benefit from having access to the University of Bristol and Rothamsted Research facilities and will receive training and obtain the skills needed to undertake an in deep evaluation of the different aspects of the animal production systems. Moreover, the student will have the opportunity to directly engage with farmers, researchers and advisors from our partner, Pasture-Fed Livestock Association (PFLA).