Detection of nutrient stress in Douglas-fir seedlings using spectroradiometer data

Narrow-band spectral reflectance measurements in the visible and near-infrared region of the spectrum were investigated for the detection of nutrient deficiencies, and estimating the foliar concentrations of nitrogen, phosphorus, sulphur, and total chlorophyll. One-year-old Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) seedlings were treated with 24 nutrient solutions containing nitrogen, phosphorus and sulphur levels ranging from 1 to 400 mg/L. After one growing season, newly matured needles were harvested, spectral reflectance measured from 400 to 1100 nm, and foliar samples analyzed for nutrient and chlorophyll levels. Several nutrient deficiencies were diagnosed. There were no unique changes in spectral reflectance which could be attributed to a specific nutrient deficiency; rather changes in reflectance were non-specific responses influenced by how varying nutrient levels affected total chlorophyll concentration. All deficiencies caused decreases in total chlorophyll, thus demonstrating the value of total chlorophyll as an indicator of nutrient stress. Correlation coefficients were calculated describing the degree of association between foliar nitrogen, phosphorus, sulphur and total chlorophyll, and the following spectral parameters: the wavelength of the red edge, percent spectral reflectance at 554 and 630 nm (red rise), and 15 vegetation indices (VIs). VIs were first determined using combinations of spectral reflectances at 480, 554, 800 nm and the wavelength of the red well (674 nm; called red well VIs), and then recalculated using red rise (called red rise VIs) rather than red well measurements to see if greater correlation with foliar constituents could be obtained. Although the use of red rise measurements in the calculation of VIs 1, 2, 3, 12 and 14 resulted in higher correlation coefficients, differences between coefficients were seldom statistically significant. Red well and red rise VI15 were most correlated with the needle nitrogen content of individual Douglas-fir, red rise VI10 with phosphorus, and spectral reflectance at 554 nm and red well VI15 with total chlorophyll. These parameters proved useful indicators of relative nitrogen, phosphorus and total chlorophyll content. None of the VIs or other spectral parameters followed the same relationship with foliar sulphur for the sulphur, nitrogen and phosphorus treatments; consequently, none were deemed suitable for sulphur estimation.Forestry, Faculty ofGraduat