Chemical and sensory characterization of the conilon coffee terroir in the State of Espírito Santo
Name: ANA CAROLINA DOS SANTOS BATISTA
Publication date: 23/04/2026
Examining board:
Name |
Role |
|---|---|
| ALDEMAR POLONINI MORELI | Examinador Externo |
| EMANUELE CATARINA DA SILVA OLIVEIRA | Coorientador |
| LUCAS LOUZADA PEREIRA | Presidente |
| RICARDO MACHADO KUSTER | Examinador Interno |
Summary: The study aimed to characterize the terroir of conilon coffee from Espírito Santo through an integrated evaluation of chemical and sensory profiles, using infrared spectroscopy (FTIR) combined with chemometric and machine learning methods to discriminate both geographic origin and variability across harvest seasons. Samples of Coffea canephora from different municipalities and consecutive harvests were analyzed, undergoing various post-harvest processes, including fermentation, followed by drying, standardized roasting, and sensory analysis according to the UCDA protocol. Chemical characterization was performed using FTIR, and the spectral data were organized into multivariate matrices and analyzed using exploratory (PCA) and supervised (Random Forest and PLS-DA) techniques, with cross validation and evaluation of metrics such as accuracy, sensitivity, and specificity. The results showed that, although the spectra presented overall similarity, subtle variations were associated with terroir and harvest year. PCA revealed partial grouping among regions, while supervised models showed good performance in
