Electronic Nose Analysis to Detect Milk Obtained from Pasture-Raised Goats

Abstract

The nutritional peculiarities of dairy products made with milk from pasture-fed ruminants would require a rapid control to be authenticated and limit the risk of fraud. In the current study, ninety milk samples from two groups of goats were analysed by electronic nose, quantitative descriptive sensory (QDA) and gas chromatography-mass spectrometry analysis with the aim of discriminating between milk produced on grazing and on a confinement feeding system. The raw milk samples were taken at five different times over a period of three months (April, May and June 2021) from eighteen individual Saanen goats divided into two groups, one of which was fed outdoors on a highly biodiverse pasture. Linear discriminant analysis (LDA), carried out on electronic nose data, was able to classify the two types of milk in terms of an animal feeding system (88% correct classification). Pasture milk scored higher for sensory descriptors such as “Grassy” and “Sweet aromatic” odours. Terpene compounds were the chemical class that qualitatively differentiates the pasture milk while volatile fatty acids were the most present quantitatively. Electronic nose has proven to be a rapid, reproducible and simple method for authenticating pasture raw milk in routine control analyses.

Keywords

• E-nose
• pattern recognition techniques
• MOS sensors
• small ruminants
• raw milk
• milk volatile organic compounds