General Spectral Characteristics of Leaf Reflectance Responses to Plant Stress and Their Manifestation at the Landscape Scale

Remarkably similar results have been reported in a number of studies that evaluated patterns of change in leaf reflectance spectra within the 400–850 nm wavelength range that occur with plant physiological stress. A variety of stressors have been imposed on species ranging from grasses to conifers and deciduous trees. In all cases, the maximum difference between control and stressed states occurred as a reflectance increase near 700 nm. This common response near 700 nm, as well as correspondingly increased reflectance in the green-yellow spectrum, are explained by the tendency of stress to reduce leaf chlorophyll concentration and by the in vivo absorption properties of chlorophyll. To determine the extent to which stress-induced changes in the reflectance of stressed vegetation at the landscape scale may be similar to those observed commonly for individual leaves, a row crop of corn was exposed to various levels of N fertilization, and canopy reflectances were measured using AVIRIS imagery. Changes in corn canopy reflectance with N deficiency were spectrally similar to the commonly observed leaf reflectance responses to stress, with maximum reflectance differences between N-deficient and control plots at 730 nm. Only far-red reflectance increased significantly (P=0.05) with relatively mild N deficiency, but reflectance in the green and far-red spectra correlated equally well with field estimates of leaf chlorophyll and laboratory measurements of leaf N concentration. A complete lack of N fertilizer increased reflectance significantly in both the green and far-red spectra and decreased reflectance in the near-infrared. Additionally, short-term water stress caused changes in corn canopy reflectance that differed from the responses to N deficiency, altering reflectance substantially only in the near-infrared, where it increased by as much as 2.5 percent. Consequently, remote sensing may be used not only to detect plant stress in monoculture stands but also to predict its cause.