European Food Research and Technology

, Volume 243, Issue 6, pp 941–953

Assessing wine sensory attributes using Vis/NIR

  • José Antonio Cayuela
  • Belén Puertas
  • Emma Cantos-Villar
Original Paper

DOI: 10.1007/s00217-016-2807-9

Cite this article as:
Cayuela, J.A., Puertas, B. & Cantos-Villar, E. Eur Food Res Technol (2017) 243: 941. doi:10.1007/s00217-016-2807-9

Abstract

The quality of wine involves the presence and intensity of many flavors and nuances. The assessment by tasting panels of the sensory quality of wine and by measuring the intensity of their attributes or defects is more difficult than assessing the global score of wine quality, which is a single value for each wine. This research focused on the feasibility of the visible and near-infrared spectroscopy for assessing wine sensory attributes. Predictive models of sensory attributes such as positive and negative for red and white wines have been developed, based on their spectra and reference values from a specialized tasting panel. The results indicate that the technique is feasible for predicting some of the most characteristic sensory attributes of red and white wines such as Flavor intensity, Astringency, Color intensity, Length, Persistency, Pleasantness and Balance. Their correlation coefficients of validations were about 0.9 in most cases. The technique was also suitable for predicting some important defects such as Oxidation, Unclean, Ethyl acetate and Acetic acid. The main potential use of this technique is for contrasting or confirming the assessment of wine for the positive and negative sensory attributes, determined by the sensory analysis of tasting panels. This can be particularly useful in cases where there is discrepancy in the assessments of a sensory panel. It is reasonable to consider the possibility of tuning the technique for detection in routine analysis of some negative attributes.

Keywords

Attributes Sensory analysis Tasting panel Vis/NIR Wine 

Abbreviations

CSIC

Spanish Council for Scientific Research

IFAPA

Institute Research and Training Agricultural and Fisheries

MN

Mean normalization

NIRS

Near-infrared spectroscopy

PC

Principal components

RC

Coefficient of calibration

r

Correlation coefficient between the analyzed and predicted values in the external validation exercises

RCV

Coefficient of cross-validation

RPD

Residual predictive deviation

SD

Standard deviation

SEC

Standard error of calibration

SEP

Standard error of performance

Vis/NIR

Visible and near infrared

Vis/NIRS

Visible and near-infrared spectroscopy

Copyright information

© Springer-Verlag Berlin Heidelberg 2016

Authors and Affiliations

  • José Antonio Cayuela
    • 1
  • Belén Puertas
    • 2
  • Emma Cantos-Villar
    • 2
  1. 1.CSIC-IG, Fisilogía y Tecnología de Productos VegetalesSevilleSpain
  2. 2.Instituto de Investigación y Formación Agraria y Pesquera (IFAPA) Rancho de la MercedJerez de la FronteraSpain

Personalised recommendations