|Reese, Cheryl -|
|Clay, Sharon -|
|Clay, David -|
|Beck, Dwayne -|
Submitted to: Journal of Terrestrial Observation
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: April 20, 2010
Publication Date: March 1, 2010
Citation: Reece, C.L., Long, D.S., Clay, S.A., Clay, D.E., Beck, D. 2010. Nitrogen and Water Stress Impacts Hard Red Spring Wheat (Triticum aestivum) Canopy Reflectance. Journal of Terrestrial Observation. 2(1):Article 7. Available:http://docs.lib.purdue.edu/jto/vol2/iss1/art7. Interpretive Summary: Applying nitrogen fertilizer during the growing season to match crop demand is seen as a way to improve farm efficiency. This approach relies upon sensing the greenness in the crop, which is linked to the amount of chlorophyll or nitrogen in plant leaves. This study was conducted to see how the sensing techniques can accommodate water stress, which can be confused with nitrogen stress in the crop. A replicated experiment with wheat was conducted over three years in Aurora, SD. Light reflectance from the crop was measured in the green, red, red edge, and near infrared wavelengths using a hand held radiometer, and the chlorophyll and nitrogen status of the crop was quantified through laboratory analysis of leaf samples. Grain samples were analyzed for concentrations of carbon isotope 12, which is an indicator of water stress in the plant. Certain bands of light together with vegetation indices derived from ratios of these bands, were excellent predictors of grain yield, grain protein, and yield loss from nitrogen stress. This information is potentially useful for determination of the nitrogen status of wheat in early season and applying nitrogen topdressing at this time to improve yield.
Technical Abstract: Remote sensing offers a simple, time efficient method for making in-season nitrogen (N) recommendations for spring wheat (Triticum aestivum). However, spectral crop reflectance can be confounded by water and N stress that simultaneously impact protein content and yields. The objective of this study was to determine the impact of water and N stress on crop reflectance and the sensitivity of structural indices (NDVI, GNDVI, and BNDVI), and chlorophyll indices (CGreen and CRedEdge) to water and N stress. A split block experiment containing 4 blocks was conducted in 2002, 2003, and 2005 at Aurora South Dakota. The treatments were 2 soil moisture regimes and 4 N rates. Reflectance was measured with a handheld multispectral radiometer at Haun 2, 4-4.5, 6, and 10-10.2. Remote sensing-based prediction models for yield, yield loss due to N stress, yield loss for water stress, and protein were developed. Yield loss due to N stress decreased with increasing N, while yield loss to water stress had the opposite relationship. Protein concentration generally increased with N. The remote sensing models for protein and yield loss due to N stress explained more of the variability than the yield model at Haun 4-4.5 and Haun 6. Bands used in developing the prediction models were broadband reflectance in green (520-660 nm) and NIR (760-900 nm) and narrow band reflectance in the red (654-666 nm), green (563-573), red edge (704-716 nm) and NIR (755-765 and 804-816). These data suggest that spectral bands together with chlorophyll indices, which do not saturate as readily as structural indices, may be good regression estimators of yield, grain protein, and yield losses due to N stress.