Previous research indicated
that a linear-plateau function using relative green difference vegetation
index (RGDVI) from aerial color-infrared (CIR) photography could be used
to predict optimum N rates in corn (Zea mays L.) at tasseling (VT).
The objective of this research was to validate this RGDVI-based remote
sensing technique for determining in-season N requirements for corn at
the VT growth stage, and to test the robustness of the model across years.
A two-way factorial experimental design was implemented as a split-plot
in randomized complete blocks with N at planting (NPL) as main
plot factor and sidedress N at VT (NVT) as sub-plot factor at
10 irrigated and non-irrigated sites in North Carolina during 2003. Results
indicate that the linear-plateau model describing the relationship between
economic optimum NVT rates and RGDVI was the best predictor
and as observed in previous research appears robust over a variety of moisture
regimes and years. The difference between predicted and observed optimum
NVT rates ranged from -30 to 90 kg N ha-1. A greater
difference between predicted and observed N rates was observed when N requirement
was high and was attributed to lower yield potential observed in this study
compared to model development years. Overall, the remote sensing technique
was successful in predicting optimum NVT rates (r2
= 0.85) given the inherent constraints of predicting yield potential in
a particular year. Although the model tended to over-predict N rates, it
was able to capture changes in N requirements across the range of conditions
tested. The results indicate that the RGDVI-based remote sensing technique
can be used to adjust late in-season N rates. Further research is underway
to validate the model across years.
See more of Nitrogen
See more of S04 Soil Fertility & Plant Nutrition
See more of The ASA-CSSA-SSSA International Annual Meetings (November 6-10, 2005)