Prediction and Classification of Energy Content in Growing Cane Stalks for Breeding Programmes Using Visible and Shortwave Near Infrared

The efficient selection of sugarcane varieties for energy conversion requires rapid and non-destructive methods for the evaluation of the energy content. The objective of this study was to develop spectroscopic models for direct measurement of lower heating values (LHV) of sugarcane in fields and to examine the precision of the classification based on fibre content values and LHV accumulation. A portable visible-shortwave near-infrared (Vis/SWNIR) spectrometer across a wavelength range of 700–1000 nm was used for spectra collection under normal field conditions. The LHV models were developed by partial least squares (PLS) regression. The model that was calibrated using the Savitzky–Golay first derivative spectra had the best performance with a coefficient of determination (r²) of 0.75 and root mean square error of predictions (RMSEP) of 179.7 J g⁻¹. The PLS prediction was also used to classify the characteristics of LHV and fibre content of cane samples compared to reference values. The results showed a precision of 0.89 for classifying the 90–100th percentile of the total samples having high LHV with low fibre content and obtaining high LHV and high fibre content. These results illustrated that the spectroscopic technique could be applied for rapid and non-destructive measurement of the LHV of cane stalks during their period of growth under field conditions. In addition, it could be an alternative screening tool for helping breeders in selecting the 90–100th percentile of the total clones, having high LHV and low fibre content and obtaining high LHV and high fibre content. Picking the 90th to 100th percentile of the entire clones in the first step of breeding programmes is crucial because only these clones will be employed to explore the study in the following stages: the investigation of disease resistance, ratooning performance, etc. Consequently, clone screening with the promising method of LHV estimation instead of visualizing the physiology can increase the success rate of obtaining a promising clone, leading to a high percentage of success in breeding programmes.