Estimation of grape Isabel's quality in pre-harvest based on spectral signature

Abstract

Near Infrared Spectroscopy is a valid non-destructive technique to quantify / estimate grape quality parameters. The objective of this study was to evaluate the logistics of using a motorized vehicle and a portable infrared spectrophotometer (NIR) to quantify the field phenolic compounds (anthocyanin, polyphenols and tannins) of the Isabel grape by spectral data. The spectra were taken in the field in a portable NIR (Brimrose Corp, MD, USA), attached to a Notebook and positioned in a Gator (John Deere) motor vehicle, which has a suitable dimension (1.52 m) to lines without damaging the plants. The phenolic compounds were estimated using calibration models previously determined. Seventy clusters of grape were evaluated in the 2017 and 2018 harvests, after pigmentation of the bark. It was observed that the models tested to quantify the phenolic compounds had an excellent predictive capacity, presenting low standard deviation. The instrumentation and software used were well integrated, making it possible to quickly visualize the variability of quality indicators in the field before harvest.