## Factors Affecting Internal Nitrogen Efficiency of Corn

Internal N efficiency
(IE) is defined as the amount of grain dry matter (GDM) produced per unit of N
in the above ground plant at physiological maturity (PMN). Currently, a static
value of IE (48 kg GDM kg

^{-1 }N) is used to define the optimal PMN in yield goal-based N recommendations used in 30 U.S. states and several N recommendation models. To evaluate the accuracy and variability of this value of IE at the economic optimum N rate (IE_{E}), experiments were conducted at 47 sites located in eight states over a three year period (2014-2016). To establish IE_{E}, N treatments ranged from 0 to 315 kg N ha^{-1}in 45 kg N ha^{-1}increments, applied either at-planting or split with 45 kg N ha^{-1}at-planting and the remainder at the V9±1 V-stage. Average IE_{E}across all site-years was 53 kg GDM kg^{-1}N with 79% of the observations between 46 and 60 kg GDM kg^{-1}N, higher than the currently accepted value of IE. Half of the time the timing of N application affected IE_{E}, with greater IE_{E}with split N in 70% of these instances due to lower PMN arising from reduced stover dry matter. In most cases the timing of N did not affect IE_{E}. Across all site-years, GDM at the EONR or EONR were unrelated to IE_{E}. Plant N content at VT of the non-fertilized and 45 kg N ha^{-1}at planting treatments were single variables most highly correlated with IE_{E}(*p*≤ 0.10, r = -0.42 and -0.50, respectively). These variables reflected the amount of residual or available N retained in the plant and/or SDM at the optimal N rate. Other factors such as plant available water content at various depths and crop reflectance at the V9 leaf stage (sufficiency and simple ratio indices for both NDVI and NDRE at 0 and 45 kg N ha^{-1}) were negatively related to IE_{E}across all site-years, but only weakly. Predictive models for IE_{E}at planting and prior to sidedressing accounted for < 50% of the variation in IE_{E}. Internal N efficiency varied considerably, but was difficult to predict, thus contributing to the inaccuracy of the yield-goal based N recommendations.