Using hyperspectral measurements to estimate ratio of leaf carbon to nitrogen in winter wheat

Ratio of carbon to nitrogen (C/N) in leaves, defined as the ratio of LCC (leaf carbon concentration) to LNC (leaf nitrogen concentration), is a good indicator for synthetically diagnosing the balance of carbon and nitrogen, nutrient status, and growth vigor in crop plants. So it is very significant for effective diagnosis and dynamic regulation of crop growth in field to monitor changes of leaf C/N quickly and accurately. Considering the close relationships between chlorophyll, nitrogen (N) and C/N, five typical indices aimed at N estimation were tested to estimate C/N in winter wheat as well as five indices aimed at chlorophyll evaluation in this study. The multi-temporal hyperspectral data from the four stages (flag-leaf, anthesis, filling, and milk-ripe) of winter wheat were obtained to calculate these selected spectral indices for evaluating C/N in winter wheat. The results showed that some tested indices were able to estimate leaf C/N in winter wheat, especially the spectral indices, MCARI/OSAVI2 and MTCI had the better performance of estimating C/N with R² of 0.51 and 0.49, RMSE of 1.35 and 1.39, respectively. In order to improve the accuracy of C/N estimates, PLS (Partial Least Squares) was used to estimate C/N in winter wheat, and the analyses showed that a better accuracy with R² of 0.60 and RMSE of 1.23 was obtained when using PLS. It indicates that applying hyperspectral reflectance measurements for monitoring leaf C/N in winter wheat appears very potential.