Spectral Identification of Stress Types for Maize Seedlings Under Single and Combined Stresses

Plants frequently suffer from different types of stress and their combination. Timely and effective monitoring of plant stresses is necessary for the precision management of crops and environmental protection. Hyperspectral remote sensing may help monitoring demand based on spectral feature analysis. However, existing studies are still insufficient for the spectral identification of plant stress types, including the combined stress type. In this paper, drought, copper, and their combined stresses on maize seedlings were designed to analyze differences of plant parameters and spectral indices by comparing with a control group. The experimental results indicate that: 1) chlorophyll content, leaf area, and relative water content could be used as key parameters to express the inter-type stress differences, and in particular, chlorophyll content was the most important bio-parameter due to its unique characteristic to distinguish a combined stress from drought stress; 2) red-edge position, the first derivative at the red edge and shortwave infrared water stress index was found to be effective for characterizing the three plant parameters under plant stresses because they could minimize the effect of variations of stress types on the prediction of these parameters; and 3) the three spectral indices might be used to identify the three stress types of maize seedlings by a decision tree analysis. The results may be useful for the precision management of crops and for environmental protection and monitoring as well.