Estimation of nitrogen content in sugarcane based on vegetation indices derived from Sentinel-2 data

• Autores: Filho, Jose Neto Soares, Pereira, Douglas Endrigo Perez, Noronha, Amanda Soares Regis
• Formato: artículo
• Fecha de Publicación: 2025
• Institución: Universidad Nacional de Trujillo
• Repositorio: Revistas - Universidad Nacional de Trujillo
• Lenguaje: portugués
• OAI Identifier: oai:ojs.revistas.unitru.edu.pe:article/6200
• Enlace del recurso: https://revistas.unitru.edu.pe/index.php/scientiaagrop/article/view/6200
• Nivel de acceso: acceso abierto
• Materia: agro-modelo; cultivares; dossel; sensoriamento remoto; inteligencia artificial; agro-model; canopy; remote sensing; artificial intelligence

Sugarcane occupies a large territorial scale in the world and is constantly searching for mechanisms to monitor nutrients in the crop production cycle, using non-destructive methods. The study aimed to estimate the nitrogen content in the sugarcane leaf was developed in the 2021/2022 harvest on two commercial fields of dryland cultivars (RB867515 = 50.75 ha) and (CVSP7870 = 48.56 ha) at the Serranópolis-Goiás mill, evaluating the efficiency of the biochemical vegetation indices Fraction of Absorbed Photosynthetically Active Radiation (fAPAR) and Canopy Chlorophyll Content (CCC) processed using the radiation transfer model RTM PROSAIL, compared to the Normalized Difference Vegetation Index (NDVI) and Green Normalized Difference Vegetation Index (GNDVI), processed using mathematical band ratio models. Both were based on a time series of Sentinel-2 data as input variables. The validation of the Agro-Model occurred through analysis of leaf tissue collected in seven interspersed evaluations during the period the crop remained in the field. The functionality of the four indexes was evidenced, highlighting the biochemical index fAPAR from the perspective of descriptive statistics (R² = 0.970 and RMSE = 0.46) for the cultivar RB867515 and (R² = 0.940 and RMSE = 0.69) for the cultivar CVSP7870.