Abstract

Two approaches were examined for the preparation of near-infrared (NIR) calibrations for the prediction of α-acids content, β-acids content, and hop storage index (HSI) in baled hop samples. In the first approach, NIR calibrations were developed using a hop sample calibration set that was representative of the full analytical range of values for each constituent (α-acids: 1.86–19.57%; β-acids: 1.45–11.58%; and HSI values: 0.220–0.480). In the second approach, hop samples were segregated by α-acids content before calibration development. NIR calibrations were produced from calibration sets with high (9–20%), medium (6–10%), and low (1–7%) α-acids contents. For α-acids predictions, the overall performance of the full-analytical-range α-acids calibration (first approach) was comparable to calibrations prepared using either the high- or medium-α-acids-range calibration sets. The standard error of cross validation (SECV) and coefficient of determination (r2) values for the full-analytical-range α-acids calibration were 0.31% and 0.99, respectively. The α-acids calibration prepared from the low-α-acids-range set had the best overall performance statistics of all the α-acids calibrations developed. The SECV and r2 values for that calibration were 0.22% and 0.97, respectively. NIR calibrations for β-acids and HSI were not substantially improved when calibrations were produced using hop samples segregated into high, medium, and low α-acids ranges. The SECV and r2 values for the full-analytical-range β-acids and HSI calibrations were 0.20% and 0.99, and 0.010 and 0.89, respectively.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call