Browsing by Author "Vermeulen, Marguerite Antoinette"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Vinnige identifikasie van botaniese- en geografiese oorsprong van Suid-Afrikaanse heuning
    (Stellenbosch : Stellenbosch University, 2020-12) Vermeulen, Marguerite Antoinette; Manley, Marena; Williams, Paul James; Allsopp, Mike; Stellenbosch University. Faculty of AgriSciences. Dept. of Food Science.
    ENGLISH ABSTRACT: The botanical and geographic origin has the greatest influence on the properties of honey, and so it affects the price of honey as well. Therefore, it is important to determine the origin of the honey. Mellisopalionology analysis is the traditional method of determining the botanical and geographical origins of honey, but it is time consuming and requires an expert. This study aimed to develop a quick and easy method to determine the botanical and geographical origin of South African honey. NIR spectroscopy was used in conjunction with partial least squares discriminant analysis (PLSDA) for classification. Two different instruments were used for scanning honey, namely the benchtop instrument and the handheld instrument. The benchtop instrument in transflection and the handheld in diffusion reflection. The handheld instrument was used with two different presentation modes, a Teflon cup and a glass vial. The average spectra of the benchtop instrument indicated four main absorption bands at 1460 (sugar and moisture related), 1940 (moisture related), 2090 and 2280 nm (both sugar related) while the handheld intrument only indicated two main absorption bands for each presentation mode at 995 (aromatic related) ) and 1200 nm (sugar related), and 1200 and 1450 nm (sugar and moisture related) for the Teflon cup and glass vial respectively. The main component analysis (PCA) score plots for the botanical classification showed poor separation between the classes [fynbos (Sandveld and Strandveld) and ‘others’ (lucerne, macadamia, red river gum, saligna, citrus)]. This is due to the complexity of honey. For the geographic classification, the PCA score plots showed better separation especially for the Hopefield and Stanford classes. The Stellenbosch class overlapped with the other two classes. The reason for this was that Stellenbosch's vegetation is a mixture of mostly fynbos and Eucalyptus, while Hopefield has mostly fynbos and Stanford mostly has Eucalyptus. The hand model (Teflon cup) showed the best classification botanical classification with a prediction accuracy of 79.63%. This accuracy shows that fynbos could not be successfully distinguished from the ‘other’ honey, but does have the potential if a larger sample set can be collected. The geographical classification achieved an overall good predictive accuracy of 84.21% with the hand model (glass vial). This therefore appears to have been classified between the geographical origins, but by going back to the PCA score plots it can be seen that the classification is rather based on the botanical origin of the samples which is therefore a distinction between fynbos and Eucalyptus honey. With this finding, it shows that a distinction can be made between the botanical origin of South African honey, but a larger sample set is needed.