Winter wheat nitrogen status monitoring using bayesian critical nitrogen dilution curves and satellite imagery

The Nitrogen Nutrition Index (NNI) serves as an agronomic indicator for assessing crop nitrogen status and guiding precision fertilization strategies. Its calculation relies on the critical nitrogen dilution curve (Nc), which in conventional approaches is commonly derived from gradient nitrogen (G–N) treatments. This requirement limits their applicability in practical agricultural scenarios for large-scale monitoring with remote sensing data. Here, we first estimated above-ground biomass (AGB) and plant nitrogen concentration (PNC) by integrating remote sensing data with phenological information. Then we adopted a Bayesian-based Nc model specifically for winter wheat under non-gradient nitrogen (NG-N) conditions, yielding the fitted curve: Nc=6.14 × AGB -0.53 . Independent datasets from regions with G-N treatments were used to validate this curve, confirming its robustness with an R 2 of 0.73 and an RMSE of 0.52%. The remote sensing-derived AGB and PNC values were subsequently combined with the Nc curve to calculate NNI, thereby establishing an indirect NNI estimation model. Then the optimal NNI model NNI pred =Na -NDI×GDD / Nc -TNDVI×GDD achieved R² of 0.62 and RMSE of 0.14, was subsequently applied to Sentinel-2 imagery in Mingji Town to generate nitrogen diagnostic maps at key phenological stages. This approach enables spatially explicit and temporally continuous monitoring of crop nitrogen status, providing a practical and scalable solution for precision nitrogen management in agricultural production.