Complementary spectral and volatile fingerprints enable discrimination of post-harvest processed arabica coffee using NIR spectroscopy and HS-GC-IMS.

Post-harvest processing strongly influences the chemical composition and aroma development of coffee, yet rapid analytical approaches for differentiating processing methods remain limited. This study evaluates near-infrared (NIR) spectroscopy and headspace gas chromatography-ion mobility spectrometry (HS-GC-IMS) as complementary tools to discriminate Arabica coffee processed by honey, natural, and washed methods. Roasted Arabica Gayo coffees were analyzed using NIR spectroscopy and HS-GC-IMS, followed by chemometric evaluation with principal component analysis and supervised machine learning. Support vector machine models provided the most robust performance, with ion mobility sum spectra (IMSS) achieving 100% test accuracy without preprocessing, while first-derivative NIR spectra also reached perfect classification. Model interpretation highlighted class-discriminant IMSS relative drift-time variables and derivative-enhanced NIR wavelength regions, although these fingerprint variables were not assigned to individual compounds. These results demonstrate that NIR spectroscopy and HS-GC-IMS provide complementary chemical fingerprints enabling rapid and reliable discrimination of coffee post-harvest processing.