For wineries processing hand harvested grapes, a visual inspection of microbial spoilage is expedient and cost effective. However, due to the increasing adoption of machine harvesters, which frequently rupture berry skins making visual inspection less precise, and the high degree of error of visual inspections between individuals, a quantitative approach to assess spoilage is needed. Fourier Transform Mid-Infrared Spectroscopy (FT-MIR), combined with multivariate analysis, is being investigated as an approach to predict grape health as a sample that can be analyzed in less than one minute. Calibration data was obtained from grape samples of Chardonnay, Riesling, Petite Sirah, and Zinfandel, which were sorted into fractions of 0, 5, 10, 15, and 20% microbially impacted clusters in healthy grape material, and analyzed using the spectrophotometer. The spectral data was analyzed using the Partial Least Squares (PLS) regression. Predicting factors included volatile acidity (VA), gluconic acid, ethanol, lactic acid, glucose-fructose content, Brix, titratable acidity, tartaric acid, malic acid, pH, alpha amino nitrogen, ammonia, and yeast assimilable nitrogen (YAN). A model was selected which optimized for a high coefficient of determination (R2) and a low root mean squared error (RMSE). Additional selection criteria included the extent to which predicting factors have been observed to correlate with microbial spoilage in other studies. The model selected had a R2 of 0.620 and a RMSE of 4.596, making it suitable for identifying grapes marked by spoilage. The output of the model was converted to Grape Health Index (GHI) Scores for better usability by operators at the test stand. The GHI was implemented at the test stands of two large-scale wineries during the 2022 and 2023 seasons. Additional wineries will be added in 2024. The testing of the impact of microbial spoilage on wine aging is currently underway. Wines made in 2018, 2019 and 2020 are evaluated for color degradation and browning, anthocyanins, tannins, pigments, and acetaldehyde. The results will be compared with the original harvest results for comparative analysis concluding the project.
