Abstract

Crop nitrogen (N) fertilization diagnoses via the N nutrition index (NNI)-yield relationship have been tested for several crop species, but a cross-species comparison of that relationship has not been performed yet. This study aimed to perform a cross-species comparison of the relationship between NNI and yield with emphasis on the yield sensitivity to N deficiency, slope of the models. Additionally, we conducted an evaluation to determine the best NNI sampling moment to predict relative yield, with focus on major grain crops. Based on a recently published global dataset to parametrize critical dilution curves, we calculated integrated NNI, instantaneous NNI, relative yield, and relative shoot biomass for annual ryegrass, tall fescue, maize, potato, rice, and wheat. We obtained 238 observations to fit integrated NNI-relative yield linear mixed-effects models and 1606 observations to fit instantaneous NNI-relative yield models. Subsequently, we performed a sensitivity analysis to determine the best NNI sampling moment to predict relative yield, with focus on major grain crops (maize, rice, and wheat). Our results show that there was low inter-species variation of sensitivity to N deficiency, i.e., the slope of the relationship between relative yield and integrated NNI. For grain crops, instantaneous NNI around anthesis demonstrated a better predictive capability for relative yield, outperforming other vegetative stages. This finding contributed to improving the understanding of the association between relative yield and NNI with implications for breeding programs, nutrient management practices, and crop modelling. Most importantly, this study is a contribution to improving the N nutrition diagnosis for several crop species, by using an integral, comparative approach.