Skip to main content
ARS Home » Pacific West Area » Pendleton, Oregon » Columbia Plateau Conservation Research Center » Research » Publications at this Location » Publication #216332

Title: Combined Spectral Index to Improve Ground-Based Estimates of Nitrogen Status in Dryland Wheat

Author
item EITEL, JAN - GRADUATE STUDENT
item Long, Daniel
item GESSLER, PAUL - UNIVERSITY OF IDAHO
item Hunt Jr, Earle

Submitted to: Agronomy Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/9/2007
Publication Date: 11/1/2008
Citation: Eitel, J.U.H., D.S. Long, P.E. Gessler, E.R. Hunt. 2008. Combined Spectral Index to Improve Ground-Based Estimates of Nitrogen Status in Dryland Wheat. Agronomy Journal. 100:1694-1702.

Interpretive Summary: New optical sensing technology now exists to simultaneously detect the nitrogen nutrition status of the crop and apply fertilizer to meet crop needs within each 1-m^2 area of a field. This ground-based remote sensing technique has relied upon the NDVI: a widely used spectral index of the amount of green vegetation derived from the ratio of spectral reflectance in the chlorophyll absorbing red and leaf internal scattering near infrared bands. Unfortunately, the NDVI is inappropriate in dryland fields where crop variations are most often linked to variations in plant available soil moisture and not soil nitrogen. This research proposes use of a new combined spectral index, termed: MCARI/MTVI2 formed from the ratio of the Modified Chlorophyll Absorption Reflectance Index and second Modified Triangular Vegetation Index. Data simulations from radiative transfer models showed MCARI/MTVI2 to be both sensitive to variations in chlorophyll and resistant to variations in crop cover, which is desired for successful remote prediction of crop nitrogen. Regression analyses confirmed that MCARI/MTVI2 was superior in performance to NDVI. The results suggest that the use of new index MCARI/MTVI2 may improve ground-based sensing of crop N status under dryland conditions where spectral variability is dominated by soil moisture rather than soil nitrogen fertility.

Technical Abstract: Recent studies have demonstrated the usefulness of the single ratio Normalized Difference Vegetation Index (NDVI) and ground-based remote sensing for estimating crop yield potential and basing in-season nitrogen (N) fertilizer application. The NDVI is positively related to crop N status and leaf area index (LAI) primarily under N limiting conditions. However, under water limiting conditions, variations in LAI are often driven by soil moisture rather than plant available N, and as a consequence, may confound spectral estimates of crop N status. The objective of this study was to evaluate the performance of single ratio and combined spectral indices for ground sensing of crop chlorophyll (C_ab) and N status in dryland wheat production fields. Sensitivity of selected spectral indices to varying C_ab and LAI values was assessed using canopy reflectance spectra simulated by the PROSPECT+SAIL radiative transfer model. Simulated data showed the proposed index Modified Chlorophyll Absorption Ratio Index/second Modified Triangular Vegetation Index (MCARI/MTVI2) to be both sensitive to C_ab and resistant to variations in LAI. Relationships between spectral indices (including NDVI and MCARI/MTVI2), and LAI, flag leaf N, or C_ab were evaluated relative to crop canopy reflectance measured with a multispectral radiometer in three dryland wheat (Triticum aestivum L.) fields. Overall, the evaluation showed NDVI and other single ratio indices to be highly correlated with LAI (r^2 ' 0.72) and poorly correlated with Cab (r^2 ' 0.35) and flag leaf N (r^2 ' 0.27) versus combined index MCARI/MTVI2, which was poorly correlated with LAI (r^2 = 0.01) and more highly correlated with Cab (r^2 = 0.63) and flag leaf N (r^2 = 0.50). The results suggest that the use of the combined index MCARI/MTVI2 may improve ground multispectral estimates of crop N status under dryland conditions where spectral variability is dominated by moisture versus N-induced variations in LAI.