Authentication of the geographical origin of Australian Cabernet Sauvignon wines using spectrofluorometric and multi-element analyses with multivariate statistical modelling. (15th January 2021)

• Record Type: Journal Article
• Authors: Ranaweera, Ranaweera K.R.; Gilmore, Adam M.; Capone, Dimitra L.; Bastian, Susan E.P.; Jeffery, David W.
• Is Part Of: Food chemistry. Volume 335(2021)
• Publication Date: 2021-01-15
• Subjects: A-TEEM absorbance-transmission and fluorescence excitation emission matrix -- CV cross-validation -- DA discriminant analysis -- EEM excitation-emission matrix -- GI geographical indication -- ICP-MS inductively coupled plasma-mass spectrometry -- IFE inner filter effects -- IRMS isotope ratio mass spectrometry -- MIR mid-infrared -- NIR near-infrared -- PARAFAC parallel factor analysis -- PCA principal component analysis -- PDO Protected Designation of Origin -- SIMCA soft independent modelling of class analogy -- SVM support vector machine -- TA titratable acidity -- TAn total anthocyanins -- TP total phenolics -- XGB extreme gradient boosting; Fluorescence spectroscopy -- Phenolics -- Colour -- Chemometrics -- Excitation-emission matrix -- Authenticity; Food -- Analysis -- Periodicals; Food -- Composition -- Periodicals; 664
• DOI: 10.1016/j.foodchem.2020.127592
• Languages: English
• ISSNs: 0308-8146

Abstract

Highlights: Authenticity of Cabernet Sauvignon wine modelled using elements and spectroscopy. Excitation-emission matrices provided a unique molecular fingerprint for each wine. Novel use of multivariate algorithm XGBoost yielded 100% correct classification. The approach could be used in the supply chain to verify the provenance of wine. Abstract: With the increased risk of wine fraud, a rapid and simple method for wine authentication has become a necessity for the global wine industry. The use of fluorescence data from an absorbance and transmission excitation-emission matrix (A-TEEM) technique for discrimination of wines according to geographical origin was investigated in comparison to inductively coupled plasma-mass spectrometry (ICP-MS). The two approaches were applied to commercial Cabernet Sauvignon wines from vintage 2015 originating from three wine regions of Australia, along with Bordeaux, France. Extreme gradient boosting discriminant analysis (XGBDA) was examined among other multivariate algorithms for classification of wines. Models were cross-validated and performance was described in terms of sensitivity, specificity, and accuracy. XGBDA classification afforded 100% correct class assignment for all tested regions using the EEM of each sample, and overall 97.7% for ICP-MS. The novel combination of A-TEEM and XGBDA was found to have great potential for accurate authentication of wines.