Developing a predictive model for the energy content of goat milk as the basis for a functional unit formulation to be used in the life cycle assessment of dairy goat production systems. (27th July 2017)

• Record Type: Journal Article
• Authors: Danieli, P. P.; Ronchi, B.
• Is Part Of: Animal. Volume 12:Number 2(2018)
• Page Start: 408
• Page End: 416
• Publication Date: 2017-07-27
• Subjects: gross energy, -- environmental sustainability, -- regression model, -- Monte Carlo permutation test, -- model selection; Animal breeding -- Periodicals; Animal genetics -- Periodicals; Animal nutrition -- Periodicals; Animal physiology -- Periodicals; Environmental sciences -- Periodicals; 636.005
• DOI: 10.1017/S1751731117001720
• Languages: English
• ISSNs: 1751-7311

Abstract

Abstract : Recent reports on livestock environmental impact based on life cycle assessment (LCA) did not fully consider the case of the dairy goat. Assignment of an environmental impact (e.g. global warming potential) to a specific product needs to be related to the appropriate 'unitary amount' or functional unit (FU). For milk, the energy content may provide a common basis for a definition of the FU. To date, no ad hoc formulations for the FU of goat milk have been proposed. For these reasons, this study aimed to develop and test one or more predictive models (DPMs) for the gross energy (GE) content of goat milk, based on published compositional data, such as fat (F), protein, total solids (TS), solid non-fat matter (SNF), lactose (Lac) and ash. The DPMs were developed, selected and tested using a linear regression approach, as a meta-analysis (i.e. meta-regression) was not applicable. However, in the final stage, a control procedure for spurious findings was carried out using a Monte Carlo permutation test. Because several published predictive models (PPMs) for GE in cow milk and goat milk were found in the literature, they were tested on the same data set with which the DPMs were developed. The best-performing DPMs and PPMs were compared directly with a subset of the individual data retrieved from the literature. Overall, the paucity of direct measurements of the GE in goat milk was a limiting factor in collecting data from the literature; thus, only a small data set ( n =26) was established, even though it was considered sufficiently representative of milks from different goat breeds. The three best PPMs based on F alone gave more biased estimates of the GE content of the goat milk than the three new DPMs based on F, F and SNF and F and TS, respectively. Accordingly, three different formulations of FU are proposed, depending on the availability of data including both F and TS (or F and SNF) or F alone. Even though several metrics can be used in defining the FU for milk to be used in LCAs of goat farming systems, the proposed FU formulations should be adopted in place of the similar energy-based ones developed for other dairy species.