Quality Assessment of Red Wine Grapes through NIR Spectroscopy

• Autor(a) principal: Rouxinol, Maria Inês
• Data de Publicação: 2022
• Outros Autores: Martins, Maria Rosário, Murta, Gabriela Carneiro, Barroso, João Mota, Rato, Ana Elisa
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10174/32681; https://doi.org/Rouxinol, M.I.; Martins, M.R.; Murta, G.C.; Mota Barroso, J.; Rato, A.E. Quality Assessment of Red Wine Grapes through NIR Spectroscopy. Agronomy 2022, 12, 637. https://doi.org/10.3390/agronomy12030637; https://doi.org/10.3390/agronomy12030637
• topic: NIR-spectroscopy; phenolic; flavonoids; anthocyanins; tannins; SSC; wine grapes

Red wine grapes require a constant follow-up through analytical chemistry to assure the greatest wine quality. Wet chemical procedures are time-consuming and produce residues that are hard to eliminate. NIR (near infrared radiation) spectroscopy has been referred as an accurate, rapid, and cost-efficient technique to evaluate quality in many fruit species, both in field and in industry. The main objective of this study was to develop predictive models using NIR spectroscopy to quantify important quality attributes in wine grapes. Soluble solids content (SSC), titratable acidity (TA), total phenolic content, total flavonoids, total anthocyanins, and total tannins were quantified in four red wine grape varieties, ‘Aragonês’, ‘Trincadeira’, ‘Touriga Nacional’, and ‘Syrah’. Samples were collected during 2017 and 2018 along véraison. Prediction models were developed using a near-infrared portable device (Brimrose, Luminar 5030), and spectra were collected from entire grapes under near field conditions. Models were built using a partial least square regression (PLSR) algorithm and SSC, TA, total anthocyanins, and total tannins exhibited a determination coefficient of 0.89, 0.90, 0.87, and 0.88, respectively. The Residual Prediction Deviation (RPD) values of these models were higher than 2.3. The prediction models for SSC, TA, total anthocyanins, and total tannins have considerable potential to quantify these attributes in wine grapes. Total flavonoids and total phenolic content were predicted with a slightly lower capacity, with R2 = 0.72 and 0.71, respectively, and both with a RPD of 1.6, indicating a very low to borderline potential for quantitative predictions in flavonoids and phenols models