Nitrogen status of durum wheat derived from Sentinel-2 satellite data in central Italy

• Published in: Remote sensing applications 2024-11, Vol.36, p.101323, Article 101323
• Main Authors: Nino, Pasquale, D'Urso, Guido, Vanino, Silvia, Di Bene, Claudia, Farina, Roberta, Falanga Bolognesi, Salvatore, De Michele, Carlo, Napoli, Rosario
• Format: Article
• Language: English
• Subjects: Canopy chlorophyll content; Chlorophyll indices; Leaf area index; Nitrogen nutrition index; Remote sensing
• ISSN: 2352-9385; 2352-9385
• DOI: 10.1016/j.rsase.2024.101323

In agriculture, nitrogen (N) is a key element in plant nutrition that affects, both positively and negatively, the productive and qualitative results of the crop. Accurate quantification of nitrogen levels is crucial for devising effective plant nutrition strategies. The objective of this study was to validate a novel method to estimate the N content at different phenological stages of durum wheat (Triticum durum Desf. cv. Iride) under different N management strategies (chemical synthetic fertilizer - SYN and organic fertilizer - ORG) in Italy, using the Nitrogen Nutrition Index (NNI)as a diagnostic tool for improving nitrogen fertilization timing and doses. The NNI is the ratio between the actual crop nitrogen content (Na) and the optimal level (Nc) required for ideal growth conditions. On ground level, Leaf Area Index (LAI), Canopy reflectance, leaf chlorophyll content (LCC) and N concentration in leaves were measured. At landscape level, LAI and Canopy Chlorophyll Indices (CIs) were derived from Sentinel 2 (S2) multispectral images captured on the same days as the ground measurements: Chlorophyll Indexes were used for estimating the canopy chlorophyll content, CCC. Na in leaves and canopy were calculated from LCC and CCC respectively. In the study area, Nc is Nc = 4.65LAI−0.35, R2 = 0.92. Among the tested Chlorophyll Indices (CIs) regression models, the linear regression was the more accurate to predict Na content, even though most of the tested Chlorophyll Indices (CIs) showed an R2 > 0.8,a. The best-performing spectral index in both calibration and validation steps resulted from the IRECI, with R2 = 0.90 and RMSE = 0.31. The developed NNI well-captured the seasonal N dynamic for durum wheat, under different N management and meteorological conditions. The NNI calculated from S2 data for crop N status assessment, showed to be an accurate estimation of the Nitrogen Nutrition Index and can be used for the fertilization plans without costly on ground measurements.