Abstract
The objective of the research was to compare multispectral images of the Ikonos high resolution satellite with optical data collected with a continuously moving ground‐based platform. A Cropscan, Inc. multispectral radiometer was used to automatically take a measurement every 1–1.5 m. Field images of both devices were compared in their capability for detecting nitrogen variability in a wheat field that had three seeding densities and five nitrogen application rates. Every treatment had four repetitions, resulting in 60 plots of 12 m by 16 m. Due to frequent cloud cover, only one Ikonos image could be taken during the growing season. At harvest, yield variables of the plots were collected to relate to the images of both systems. Calculated normalized difference vegetation index (NDVI) from both measurement systems was a good indicator of applied nitrogen in the field: correlation was 0.76 and 0.85 for Ikonos and the platform, respectively. NDVI was also well correlated to yield variables at harvest. The best correlation found for both systems was with grain nitrogen content (kg[N] ha−1): 0.86 and 0.91 for Ikonos and the platform, respectively. The ground‐based system provided a better estimate of nitrogen in the wheat crop and was better related to dry matter content. Ikonos images were higher correlated to straw yield and biomass. When estimating yield variables with NDVI measurements, the average error percentage was lower for the platform than for Ikonos. In conclusion, both measurement devices could be used to indicate wheat nitrogen stress by measuring canopy reflectance, but measurements of the platform were more accurate.
Acknowledgments
The authors gratefully thank the Soil Service of Belgium (SSB), CNH n.v. and the Flemish Institute for the Promotion of the Scientific–Technological Research in Industry (IWT). The research was party funded by EC commissions DG‐XII, Framework IV, Environment and Climate Programme.