Detecting ozone effects in four wheat cultivars using hyperspectral measurements under fully open-air field conditions

The screening of sensitive spectral indictors is essential for quantitative diagnosis of ozone (O3)-induced stress in plants. Four wheat (Triticum aestivum L.) cultivars with different degrees of O3-tolerance were grown under an elevated O3 concentration (E-O3) in fully open-air field conditions for two consecutive growth seasons from 2012 to 2013. The aim was to find sensitive hyperspectral indictors for real-time detection of O3 effects. The results showed that E-O3 caused a significant decrease in leaf thickness and pigment concentrations, resulting in a change in leaf reflectance. The effects of E-O3 on both physiological variables and reflectance characteristics were wheat cultivar-specific, with a greater and earlier O3 effect found in the O3-sensitive wheat cultivars than in the O3-tolerant cultivars. Spectral indices that were previously developed to detect leaf biological variables were examined, and highly correlated relations were found between the chlorophyll content and the three spectral parameters ND705, mND705 and R550, with Pearson r values of 0.896, 0.892 and − 0.872, respectively. When independently determined data from 2013 were used to test the derived equations, the coefficients (R2) of correlation between the measured and estimated chlorophyll were 0.817 (ND705), 0.833 (mND705) and 0.776 (R550); the root mean square errors (RMSE) were 0.227 (ND705), 0.228 (mND705) and 0.254 (R550) (mg kg− 1); and the mean relative errors (RE) were 8.7% (ND705), 8.1% (mND705) and 9.1% (R550). Furthermore, O3-induced changes in the three optical parameters were in accordance with the leaf chlorophyll responses in wheat. Our study suggested that the reflectance indices mND705, ND705 and R550, especially the former two spectral indices that contained information from several bands, could help to support the diagnosis and real-time monitoring of O3-induced damage in wheat. To estimate the wheat yield accurately using the selected spectral indices, the filling stage was found to be the best time for measuring canopy reflectance.