| Abstract: | There is a strong economic interest in routine measurements of cotton quality as production processes and final products depend on it. An important cotton property is “micronaire,” a parameter that is indicative of the fibers' maturity and its fineness. Currently, micronaire is normally measured in laboratories with equipment that prohibits routine field analyses. The goal of this study is a proof-of-principle demonstrating that cotton quality as determined via fiber micronaire is correlated to fiber properties and that these properties can be determined by near-infrared (NIR) reflection spectroscopy using portable instrumentation in conjunction with Principal Component Regression for micronaire prediction. A set of 191 cotton samples was acquired from over 100 different upland cotton varieties, and initial spectroscopic studies confirmed the feasibility of NIR spectroscopy to measure cotton micronaire in the laboratory with portable NIR instrumentation. Sample reproducibility was an issue which has been resolved with two approaches, that is, model spectral artifacts, mainly baseline shifts, by means of chemometric calibrations or application of second derivative spectroscopy to suppress baseline drifts. Results from this study demonstrated in up to 90% of the test samples that the micronaire values fall into the acceptance range. Thus, a promising new approach for field analyses is on the horizon and has been assessed in this study. Further, the acceptance range could be reduced to ±0.2 m.u. and still ≥70% of the samples fell inside the restricted acceptance range. Up to 60% of the samples fell inside an acceptance range of ±0.1 m.u. |
