Evaluation of chiavennasca grape technological and phenolic ripening indexes by a portable vis-nir device

Grape composition at harvest is one of the most important factors to determine the future quality of wine. The conventional methods for the determination of grape quality parameters are time consuming, require preparation of samples, are expensive, and generally highlight only one or few aspects of grape quality. Visible - Near Infrared Spectroscopy (Vis/NIR) is a rapid and non-destructive technique, requiring minimal sample processing before analysis, and coupled with chemometric methods appears to be one of the most convenient and straightforward analytical tools for studying food products. The aim of this study is the design, realization and experimentation of a portable optical device (Vis-NIR spectrophotometer) for single sample, non-destructive and quick prediction, in the wavelength range 450-980 nm of quality parameters of Nebbiolo grapes, ecotype Chiavennasca, harvested in the Valtellina viticultural area. PLS regression models were built both for technological ripening parameters (soluble solids content, titrable acidity, pH value) and for phenolic ripening indexes (polyphenols content, total and extractable anthocyanins content). The analysis was performed both on fresh berries and on homogenized samples. Good results were obtained especially for homogenized samples with correlation coefficient of cross-validation about 0.8 for all the indexes. Finally, a classification analysis was performed both for fresh berries and for homogenized samples, using PLS-discriminant technique, for TSS and titrable acidity. Good results were obtained for homogenized samples, with about 85% of correctly classified in validation for both the parameters. Thanks to the obtained results, a simple, inexpensive and quick experimental system has been built.