Mid-season Prediction of Wheat Grain Yield Potential and Nitrogen Response

Abstract

Soil nutrient management has made significant advances in efficiency, especially with nitrogen (N) fertilization. Nonetheless, there is still room for improvement surrounding mid-season prediction of grain yield and ensuing fertilizer nitrogen (N) rates. Sequential normalized difference vegetation index (NDVI) measurements from two long-term nutrient management experiments (Experiment 222 and Experiment 502) were used to improve the prediction of yield potential, and to decipher situations where added N would be unlikely to increase winter wheat (Triticum aestivum L.) grain yields in the southern Great Plains. These sequential readings were used by-date, and over dates to evaluate grain-yield-prediction collected from the same plots at harvest. Additional climatological data was also employed by site to improve yield prediction indices, including cumulative growing degree days from planting to sensing greater than zero (GDD>0). The coefficient of determination (r2) for each NDVI/yield relationship was then plotted as a function of corresponding GDD>0. A linear plateau model was applied to these relationships for Experiment 222 and Experiment 502, resulting in an r2 of 0.98 and 0.47, respectively. Utilizing the number of days where GDD>0 is more refined than growth stage because it embeds climatological estimates of growth that can be used in another year and/or environment. Knowing this value can serve as a guide as to exactly when the NDVI reading should be collected.