Does water availability affect the critical N dilution curves in crops? A case study for maize, wheat, and tall fescue crops

Previous studies reported an effect of water deficit on crop nitrogen (N) demand but the statistical significance of this effect is unclear and the physiological mechanism explaining this effect remains poorly understood. The specific aims of this review are (i) assess whether the parameters of the N dilution curve are significantly influenced by the level of water availability and (ii) provide a synthesis of the plausible physiological mechanisms involved on changes in N dilution. To address the first goal, several datasets assessing the effects of water availability on N status of maize (Zea mays L.), wheat (Triticum aestivum L.), and tall fescue (Lolium arundinaceum (Schreb.) Darbysh.) were combined and then used to fit N dilution curves with a Bayesian statistical approach. Concerning the second objective, we investigate two possible mechanisms, namely a modification in water soluble carbohydrates (WSCs) and an effect on plant architecture and morphology impacting the allometry between structural and metabolic compartment. Our re-analysis of published data for tall fescue, wheat, and maize reveals no conclusive evidence supporting an effect of water availability effect on the N dilution curve. In addition, we show that WSCs accumulation under water deficit condition is unlikely to be accompanied by N-rich soluble compounds, and that a reduction of water availability has limited effect on the allometry of plant mass allocation. Lastly, our conclusion provides new insights on the use of a common critical N dilution curve (one model across water levels) for estimating the crop N status where both water (under no extreme deficiency) and N stresses are operating in order to separate direct effect of water availability from its indirect effect on crop N nutrition status.