From Carbon Footprint Accounting to Decision Support Systems in Dairy Sheep: Harmonised LCA Evidence, Digital Infrastructure, and Benchmarking

This PhD thesis targets carbon footprint (CF) assessment as a decision-support and governance lever for dairy sheep systems, under the premise that its usefulness depends on three enabling conditions: high-quality and structured farm data, methodological robustness and transparency, and the translation of Carbon Footprint (CF) results into actionable priorities and monitorable KPIs from farm to supply-chain level. Chapter 1 delivers an integrative PRISMA based systematic review (2010–2025) to clarify why Life Cycle Assessment (LCA) results for sheep products (milk, meat, wool) are often difficult to compare. The review confirms that variability across studies is strongly influenced by methodological settings system boundaries, functional unit, allocation, Global Warming Potential (GWP) metric choices, and the inclusion/exclusion of soil carbon sequestration besides real differences among production systems. Chapter 2, developed within the project LIFE Green Sheep framework, translates CF from a comparative indicator into an operational tool by applying an intra-national profiling approach. Beyond comparing national calculators, the chapter focuses on identifying measurable differences between low- and high-emission farms and on prioritising drivers that explain emission intensity within the same production system. This approach operationalises four complementary research questions to answer, “what to mitigate” and “how”, distinguishing generalisable levers from those that must be adapted to the specific farm profile (e.g., productivity per head, purchased feed dependence, grazing management, direct energy use, nitrogen pressure, reproduction and milk composition). Chapter 3 addresses a key bottleneck, without a reliable data infrastructure, CF remains episodic and not auditable. The thesis contributes through APPàre, an integrated digital platform for the Italian sheep sector, embedding a GHG calculator designed to reuse existing farm data, formalise calculation rules, and ensure traceability and repeatability of the assessment. Validation on 10 Sardinian dairy sheep farms provides CF results expressed in kg CO₂ eq/kg FPCM, with an emissions profile dominated by enteric sources and a negative relationship between emission intensity and production level, supporting productivity as a guiding KPI and purchased inputs/energy as secondary levers to be managed according to yield. Chapter 4 closes the accounting to decision support cycle through the CAO cooperative case study (18 farms, Sardinia), applying standardised LCA. The chapter frames CF as a diagnostic-strategic device to identify hotspots and convert farm variability into a streamlined KPI dashboard supporting cooperative benchmarking, capacity building and continuous annual monitoring. Overall, the thesis systematises three levels that are often treated separately, evidence synthesis, profiling methods, and digital/cooperative governance, showing how their integration makes CF a management indicator rather than a communication-only metric, particularly in sheep milk supply chains where climate impact is strongly farm-driven.