|Authors: ||K. Ncama, L.S. Magwaza, S.Z. Tesfay, A. Mditshwa, N. Mbili|
|Keywords: ||oil content, chemometrics, fruit quality, dry matter|
Harvest maturity of avocado fruit is currently estimated based on dry matter content (DM). The DM is analyzed destructively from few samples representing the entire orchard.
However, oil content (OC) is the only maturity parameter with a direct link to eating quality of avocados.
Successful studies indicating non-destructive assessment of avocado OC are limited.
Consequently, this study was conducted to develop models for non-destructive assessment of maturity stages of 'Hass' avocado using portable visible to near infrared spectrometer (Felix, F-750, Vis-NIR system) equipped with xenon-tungsten lamp as light source, and lead-sulphide detector.
Partial least square (PLS) regression models for assessing OC, DM and moisture content (MC) of avocado fruit were developed from spectral and biochemical reference data using The Unscrambler® X chemometric software.
Cross-validation was used to develop models.
Final models were tested on independent test set.
Previous studies have indicated a change in the operation of spectrometers at different temperatures.
As a result, the robustness of models developed in this study was increased by acquiring three spectra from each sample at 10, 20 and 30°C from morning to early evening.
A model for assessing OC was successfully developed as indicated by the high regression coefficient (R2=0.979) and the ratio of performance deviation (RPD=6.817), and low root mean square error of prediction (RMSEP=1.115%). Successful models were developed for MC (R2=0.955; RPD=4.221; RMSEP=0.405%) as well as DM (R2=0.984; RPD=7.789; RMSEP=0.220%). Commercial application of portal spectrometers with the developed models can increase the number of assessed harvest maturity parameters and the number of reference samples because the system can estimate the maturity parameters without harvesting of sample fruit.
Download Adobe Acrobat Reader (free software to read PDF files)
URL www.actahort.org Hosted by KU Leuven