Response Analysis of In-Season Rice Biomass and Grain Yield to Nitrogen Fertilization.

Recent remote sensing technologies are gradually employing nitrogen (N) fertilizer management tools aimed at increasing crop fertilizer use efficiency. This study was initiated in 2008 to (1) determine the relationship of rice biomass and grain yield in response to different pre-flood N rates application, and (2) determine if the optimal rate predicted at mideason based on biomass response to N can be used to estimate N rate requirement to maximize grain yield. This study was superimposed on existing variety by N trials located at the LSU AgCenter Rice Research Station in Crowley, Colvin Farm in Rayville, LA and Mississippi State University Delta Research and Extension Center. Above ground plant samples were collected at panicle differentiation (PD) and 50% heading from a 0.9 m section of the middle drill row. Grains from each plot were harvested using a small combine. Relationship of biomass and grain yield, and estimated optimal N rates using biomass and grain yield were determined using regression analysis. Biomass collected at PD and 50% heading explained 65% and 63% of variability in grain yield, respectively implying that information on biomass production as early as PD can be used to project in-season rice grain yield. Linear-plateau (LP) model suggested that biomass at PD predicted an optimal N rate of 132 kg ha^-1 while at 50% heading the predicted optimal N rate was  126 kg ha^-1. At harvest, a higher N requirement at a rate of 151 kg ha^-1 was determined for optimal grain production. Biomass is a good in-season predictor of rice yield potential, thus remote sensing can be a powerful tool to non-destructively acquire this information instantly.