Exploring Factors Influencing the Effectiveness of Vegetation Indices for Corn Nitrogen Management.

Improving the efficiency of Nitrogen (N) fertilizer use is a primary issue for the agricultural community. The environmental impact of excessive N application and economic considerations of optimizing input to yield ratios warrant research in topics including the optimal time to apply N fertilizer, effectiveness of variable rate applications to improve Nitrogen Use Efficiency (NUE), and capabilities of remote sensing techniques to detect the onset and spatial variability of N stress. A multi-year assessment of the efficacy of spectral remote sensing derived vegetation indices (VIs) to discern N stress in corn has been conducted in Maryland from 2004 to 2012. Satellite multispectral imagery, aerial hyperspectral imagery, ground based passive sensor reflectance data, and ground based active sensor reflectance data were collected. Biophysical parameters including plant height, leaf area index (LAI), leaf chlorophyll content, and development stage were acquired weekly from growth stages V6 to R1. Canopy reflectance simulation models were run and the output data analyzed to explore the impacts of individual canopy components on the overall reflectance seen at the sensor. Vegetation indices derived from spectral remote sensing can be grouped into two broad classes, biomass responsive indices and plant pigment responsive indices. It was found that VIs effectiveness for discerning N stress in corn plants varies the most at growth stages prior to canopy closure, when the effects of background components and variations in plant biomass confound the detection of the leaf chlorophyll levels. The performance of various plant pigment responsive VIs converge as LAI increases and the plant canopy closes. Normalizing a VI appropriate for detecting leaf chlorophyll levels with a VI well correlated to plant biomass was shown to improve the ability to extract information about leaf chlorophyll variation from canopy level remote sensing spectra collected at growth stages prior to canopy closure.