Inter-regional characterization of New Zealand pinot noir wines: Assessing geographical origin through mid-FTIR and...

Pinot Noir wines from New Zealand, celebrated for their high-quality and regional distinctiveness, are ideal for origin differentiation studies due to their sensitivity to environmental variations. This study aims to identify the characteristics of wine origins and establish a model for tracing these origins, with the goal of protecting the economic position of New Zealand Pinot Noir wine. Exploratory and multivariate analyses were conducted on phenolic and Fourier-transform infrared (FTIR) spectroscopic data using 116 Pinot Noir wine samples from three regions. The Mid-FTIR spectra, particularly between 3050 and 1950 cm^-1 wavelengths, were successfully correlated with the major phenolic compounds in wines through Partial Least Squares Discriminant Regression (PLSR), achieving R²_Pre values ranging from 0.75 to 0.99. Multivariate analyses, including Linear Discriminant Analysis (LDA), Partial Least Squares Discriminant Analysis (PLS-DA), Artificial Neural Network (ANN) and Random Forest (RF) based on either phenolic data alone, spectral data alone, or a fusion of both, achieved high classification acccuracy. Among these methods, the LDA and ANN models demonstrated superior classification performance compared to PLS-DA and RF. The phenolic profile provides detailed information on key compositional markers, while FTIR spectra offer a holistic chemical representation. This fusion enables a more comprehensive differentiation of wine origins, reducing potential misclassification caused by the limitations of individual datasets. In conclusion, exploratory and multivariate analyses proved to be rapid and efficient methods for tracing the origin of pinot noir wines produced in New Zealand.