Fig. 2 | Scientific Reports

Fig. 2

From: Improving nitrogen use efficiency in rice by estimating leaf nitrogen content with near-infrared spectroscopy and chemometric modeling

Fig. 2

Evaluation of NIR spectra acquired for fresh leaves from five rice genotypes under different nitrogen treatments (n = 2700), including spectra collected on three different days (D1, D2, and D3). (a) Raw averaged NIR spectra (350–2500 nm) with baseline correction across nitrogen treatments, showing the characteristic spectral patterns of fresh rice leaves. Nitrogen-associated absorption regions are highlighted in blue, indicating key bands linked to proteins, amino acids, and chlorophyll-related compounds involved in nitrogen metabolism. (b) Transformed averaged spectra (350–2500 nm) showed no significant differences between nitrogen treatments. (c) PCA scores plot displaying the distribution of spectra based on the first two principal components (PC-1 and PC-2), explaining the majority of the variance in the dataset (97.7%). Spectra from D1 are clustered on the left side, while spectra from D2 and D3 are grouped on the right. (d) PCA loadings illustrating the dominant wavelengths contributing to the observed variance in the PCA scores plot.

Back to article page