Quality evaluation of grapes for mechanical harvest using vis NIR spectroscopy

Abstract

Mechanical harvest of grapes is one of the operations that mostly influence the quality of the future wine. The shaking frequency of the harvesting machine is usually adjusted on the basis of the grape berry characteristics in order to limit grape juice production that is a potential cause of uncontrolled fermentations. These evaluations usually require time, personnel and laboratory analyses. The introduction of a vis NIR system to rapidly and reliably evaluate the berry properties in field before mechanical harvest could be a good alternative. The aim of this study was to evaluate the feasibility of applying vis NIR spectroscopy as a non-destructive technique on grapes cv. Syrah and Chardonnay to predict pedicel detachment force, pH and total soluble solids before mechanical harvest. The spectral acquisitions were performed using a portable vis NIR device (600 - 1000 nm). An Ordinary Least Square evaluation was applied to assess vis NIR prediction ability on grapes. The system gave excellent performance in predicting pH for both varieties (R² = 0.99), also confirmed by the indicators SECV/M and Bias/M respectively equal to 0.024 and 0.014 for cv. Syrah and 0.002 and -0.009 for Chardonnay. The vis NIR device showed satisfactory prediction ability even regarding total soluble solids (R² = 0.99) with SECV/M = 0.090,  Bias/M = 0.071 for cv. Syrah and SECV/M = 0.00,  Bias/M = -0.002 for Chardonnay. However, the results showed the low vis NIR ability to predict detachment force for Chardonnay grapes (R² = 0.85, SECV/M = 1.008; Bias/M  = -0.834), and an acceptable one for Syrah grapes (R² = 0.87; SECV/M = 0.362; Bias/M = -0.109). Since detachment force has an enormous importance in grapes mechanical harvest, the possibility of applying vis NIR spectroscopy in field before harvest is very encouraging for cv. Syrah (red grapes) and need to be improved for cv. Chardonnay (white grapes).