More aeration is not always better: Bayesian construction of critical nitrogen dilution curves and the equivalence of diagnostic indicators for tomatoes under aerated drip irrigation

This study explores how aerated drip irrigation affects nitrogen nutrition diagnosis in tomatoes and systematically evaluates the applicability of various variables in developing critical nitrogen dilution curves. Three levels of aeration (5, 15, 40 mg L⁻¹) and three nitrogen levels (120, 180, 240 kg ha⁻¹) were tested. A Bayesian hierarchical model was used to construct critical nitrogen dilution curves based on aboveground biomass (AGB), leaf area index (LAI), and leaf dry matter (LDM). Results showed that aerated drip irrigation has a nonlinear threshold effect on tomato growth and nitrogen uptake. The 15 mg L⁻¹ treatment significantly increased all indicators—LDM by 28.2–49.3% and nitrogen uptake by 14.7–32.5%. The beneficial effect decreased at 40 mg L⁻¹, indicating that 15 mg L⁻¹ is the optimal aeration level. The Bayesian approach effectively quantified uncertainty in curve parameters; LDM was the most sensitive indicator of aeration, with its parameter significantly higher at 15 mg L⁻¹ than at 5 mg L⁻¹. The accuracy of nitrogen nutrition index (NNI) simulations (n-RMSE < 10%) and yield predictions ( R ² > 0.80) for AGB, LAI, and LDM were similar, with minimal differences. This suggests that all three can serve as effective diagnostic indicators, with LDM showing potential for remote sensing-based diagnosis. Aerated drip irrigation significantly lowered the NNI threshold required to achieve maximum relative yield, indicating improved nitrogen use efficiency. In conclusion, this research recommends an aeration intensity of 15 mg L⁻¹ and an NNI diagnostic threshold of 0.97–0.98 for tomatoes under aerated drip irrigation, providing a scientific basis for precise nitrogen management to reduce inputs while maintaining yield.