Estimation of the Relative Chlorophyll Content in Spring Wheat Based on an Optimized Spectral Index

• Published in: Photogrammetric engineering and remote sensing 2018-12, Vol.84 (12), p.801-811
• Main Authors: Kasim, Nijat, Sawut, Rukeya, Abliz, Abdugheni, Qingdong, Shi, Maihmuti, Balati, Yalkun, Ahunaji, Kahaer, Yasenjiang
• Format: Article
• Language: English
• ISSN: 0099-1112
• DOI: 10.14358/PERS.84.12.801

The relative chlorophyll content is one of the essential factors that affect crop growth and yield, and chlorophyll is an important parameter that reflects the stress and health of vegetation. The spectral feature parameter method is widely applied to estimate the relative chlorophyll content of wheat. To provide a scientific basis for wheat growth monitoring and agronomic decision-making, we estimated the relative chlorophyll content using hyperspectral technology. During the summer of 2017, we collected canopy reflectance spectra using field spectroscopy along with the relative chlorophyll content of wheat. To comprehensively analyze the field-collected hyperspectral data, various band combinations were used to calculate a simple spectral index (ratio spectral index, RSI), normalized difference spectral index (NDSI) and chlorophyll index (CI). We compared simple spectral indices with 17 different indices from the literature. The relationships between the indices and relative chlorophyll content were then examined, and the strongest relationships were demonstrated. The partial least squares regression (PLSR) method was utilized to develop a predictive model of the relative chlorophyll content. The newly identified NDSIs, RSIs, and CIs always performed better than the spectral indices from previous studies, and the relative chlorophyll content exhibited the highest correlations with RSI (R849 nm, R850 nm), CI (R849 nm, R850 nm), and NDSI (R849 nm, R850 nm), calculated using leaf reflectance spectra (|r| ≥ 0.7). The -model revealed that the highest R2Pre (0.74) and lowest RMSEPre (2.72 SPAD) were identified with four optimized chlorophyll indices (CI (R849 nm, R850 nm), CI (R539 nm, R553 nm), CI (R540 nm, R553 nm), and CI (R536 nm, R553 nm)). The spatial information from these parameters will aid the proper nutrient management of optimal spring wheat crop growth and forecasting models for a precision wheat agriculture system.