Using hyperspectral indices to estimate foliar chlorophyll a concentrations of winter wheat under yellow rust stress

Abstract The canopy hyperspectral reflectance of winter wheat infected with yellow rust at different levels of severity were measured by an ASD FieldSpec Pro FR™ spectrometer in the field and the concentrations of chlorophyll a (Chl a) in the leaves corresponding to the spectra were determined by biochemical methods in the laboratory. Correlation analyses were made between Chl a concentrations and canopy hyperspectral data of diseased wheat. Results show that foliar Chl a concentrations are strongly correlated with canopy spectrum in the visible region and the first‐order derivative spectrum in blue edge, green edge, and red edge. Linear and nonlinear models for estimating Chl a concentrations of the diseased wheat were built based on several spectral indices. Results indicate that SDr/SDg , in which SDr and SDg are the sums of the first derivative within red and green edges, outperformed the other indices in predicting Chl a concentrations. The relative estimation errors for Chl a for 12 unseen samples are 17.5%. It is concluded that derivative spectra in red edge and green edge have strong prediction power for foliar Chl a concentrations of diseased winter wheat. Using hyperspectral remote sensing data to monitor crop disease and nutrition status is very promising.