Applications of Derivative Spectroscopy on determination of nitrogen status of tea plant (Camellia sinensis)

¹ School of Mathematics & Computer Science, Wuyi University Wuyishan, Fujian, 354300, China
² Fujian Key Laboratory of Big Data Application and Intellectualization for Tea Industry (Wuyi University), China
³ College of Computer Science and Technology, Zhejiang University of Technology, Zhejiang, China.
⁴ Information Technology and Laboratory Management Center, Wuyi University, Wuyishan, Fujian, 354300, China

^* Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

Nitrogen (N) fertilizer plays an important role in tea plantation with significant impacts on the formation of vital compounds, pigments and flavor-related substances. Excess and deficiency of N can reduce crop yield and quality yet impact environmental concerns. Therefore, it is crucial to tailor fertilization levels based on the N status of tea plants to avoid both inadequate and excessive N application. This study has evaluated if it is possible to improve the effectiveness of spectral indices to estimate the nitrogen concentration using derivative spectroscopy and dimension reduction techniques to process the spectral signatures. The results demonstrated that this study involved spectrum smoothing, derivative implementation, and measurements of Normalized Difference Spectral Index (NDSI) and Ratio Spectral Index (RSI) indices. Analysis identified the top 25 Pearson correlation coefficient (PCC) for both indices and employed Principal Component Analysis (PCA) to determine ten principal component scores. A ten-component linear regression model was developed, and its effectiveness was assessed. This model was then used to predict N levels in tea gardens. Derivative spectroscopy techniques can improve the effectiveness of indices to estimate the nitrogen concentration as compared to non-derivative indices. The statistics values are always higher for derivative indices. This model was then used to predict N levels in tea gardens. To increase Signal to Noise Ratio (SNR), a scheme added three spectra and corresponding N values together then find an average. This procession made very good results on statistics for multiple linear regression. The results are promising, suggesting that further research in this field is needed.