Non-destructive evaluation of avocado fruit maturity using near infrared spectroscopy and PLS regression models

• NIRS was evaluated for prediction of avocado maturity status.
• Avocado mesocarp dry matter, moisture contents were predicted accurately.
• Results demonstrated importance of integrating multi-seasonal dataset in PLS models.
• Prediction model robustness across orchards was successfully evaluated.

Harvest maturity of avocado fruit is currently determined using indices such as mesocarp oil content, dry matter (DM) or moisture content (MC), both measured destructively using representative samples of a batch in an orchard. Although useful, destructive techniques are time-consuming and results reflect properties of specific produce evaluated. In this study, the feasibility of near-infrared spectroscopy (NIRS) as a rapid non-destructive method for predicting maturity parameters of individual avocado fruit was evaluated. NIRS prediction results showed that MC and DM were predicted accurately, with residual predictive deviation (RPD) of 2.00 and 2.13 respectively. However, models for predicting oil content, though promising, were not accurate, with RPD value of less than 1.0. The good prediction statistics between NIRS predicted MC and DM content demonstrated the potential of this system for non-destructive evaluation of avocado fruit maturity status. The stability and accuracy of models developed over two seasons, 2013 and 2014, to predict maturity parameters of avocado fruit demonstrated model robustness. NIRS combined with PLS models developed in this study are recommended for non-destructive evaluation of individual avocado fruit maturity status.

Figure optionsDownload as PowerPoint slide