Assessment of Vegetation Indices for the Early Detection of Nitrogen Stress in Corn.

The derivation of vegetation indices (VIs) from remotely sensed spectral reflectance data for the assessment of crop biophysical status has remained a subject of interest within the remote sensing and agronomic communities.  Determination of the spatial variability of a crop’s nitrogen status through remote sensing techniques can be used to prescribe the variable rate application of fertilizers, focusing inputs on areas exhibiting signs of N stress while reducing applications on regions not under stress, thereby improving nitrogen use efficiency.  Because there is a short time period between the first evidence of N stress and the last feasible opportunity to apply supplementary N, the effectiveness of VIs for detecting N stress at an early growth stage is an important consideration.  A multi-year investigation of the efficacy of VIs for the determination of N status, as exhibited by leaf chlorophyll content, was conducted at the Beltsville Agricultural Research Center (BARC) in Beltsville, Maryland.  Biophysical and remotely sensed data were collected every three to seven days at growth stages from V5 to R1.  A Cropscan multi-spectral radiometer was used to collect spectral reflectance measurements in 13 bands centered at wavelengths ranging from 530 nm to 880 nm.  A Minolta SPAD meter was used to measure leaf level chlorophyll.  VIs calculated from spectral reflectance data were evaluated for correlations with measurements of leaf level chlorophyll.  VIs which incorporated a spectral band from the “red edge” region (720 nm) exhibited higher correlation coefficients with leaf chlorophyll levels at early growth stages (V6 -V8) than indices that did not utilize this region of the spectrum.  At later growth stages correlations between VIs and leaf level chlorophyll increased, while differences between VIs that utilized the “red edge” region and VIs that did not decreased.