Estimation of canopy nitrogen content in winter wheat from Sentinel-2 images for operational agricultural monitoring

Abstract Canopy nitrogen content (CNC, kg/ha) provides crucial information for site-specific crop fertilization and the usability of Sentinel-2 (S2) satellite data CNC monitoring at high levels in managed agricultural fields is still underexplored. Winter wheat samples were collected France Belgium 2017 (n = 126) 2018 18), analysed S2-spectra extracted sample locations. A comparison three established remote sensing methods to retrieve was carried out: (1) look-up-table (LUT) inversion canopy reflectance model PROSAIL, (2) Partial Least Square Regression (PLSR) (3) nitrogen-sensitive vegetation indices (VI). The spatial temporal transferability new rigorously assessed. PROSAIL-LUT approach predicted with a root mean squared error 33.9 kg/ha on dataset slightly larger value 36.8 dataset. Contrary, PLSR showed an 27.9 kg N/ha (R2 0.52) calibration (2017) but substantially 38.4 independent (2018). VIs revealed errors than results much higher validation (> 50 kg/ha). PROSAIL more stable robust VI when applied data. obtained maps may support farmers adapting their management according actual status.