NIRS technology for fast authentication of green asparagus grown under organic and conventional production systems

• MEMS-NIRS technology for the authentication of organic origin of intact green asparagus at industrial level.
• The tip and the middle portion of the spear proved to be the most suitable for authentication of organic asparagus.
• NIRS technology provides reliable information about asparagus quality, i.e. harvest month and postharvest storage time.
• Differences caused by water loss and fiber profiles might contribute to variations in quality of spears as a function of storage time.

This study sought to evaluate the ability of near-infrared reflectance spectroscopy (NIRS) to classify intact green asparagus as a function of growing method (organic vs. conventional) during postharvest refrigerated storage, and as a function of harvest month and postharvest cold storage duration. It also sought to identify the portion of the spear best suited for this purpose. A total of 300 green asparagus spears (Asparagus officinalis L., cv ‘Grande’), were sampled after 7, 14, 21 and 28 d of refrigerated storage (2 °C, 95% RH) and at commercial harvest time. Three commercially available spectrophotometers were evaluated for this purpose: a scanning monochromator (scanning range 400–2500 nm), a diode-array Vis/NIR spectrophotometer (range 400–1700 nm) and a handheld MEMS spectrophotometer (range 1600–2400 nm). Models constructed using partial least squares 2-discriminant analysis (PLS2-DA) correctly classified 91% of samples by growing method using the diode array instrument, between 86% and 91% using the scanning monochromator and between 82% and 84% using the handheld spectrometer. The tip and the middle portion of the spear proved to be the most suitable for this purpose employing the MEMS instrument. Using similar models, the diode array instrument correctly classified 100% of samples by harvest month, compared with between 97% and 98% using the scanning monochromator and between 87% and 96% using the handheld instrument. Models also correctly classified between 66% and 97% of samples by postharvest storage time, depending on the instrument used. The results indicate good performance of the prediction models, particularly for predicting harvest month and growing method, determination of the latter being of considerable importance for the authentication of organic asparagus at industrial level.