Variabilidade espacial da qualidade fisiológica de sementes de soja e sua relação com o índice de vegetação "N-Sensor"

• Ano de defesa: 2025
• Autor(a) principal: Boelter, Jessica Hoch
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Santa Maria; Brasil; UFSM; Programa de Pós-Graduação em Agronomia; Centro de Ciências Rurais
• Palavras-chave em Português: Agricultura de precisão e digital; Imagens de satélite; Produção de sementes; Precision and digital agriculture; Satellite imagery; Seed production; CNPQ::CIENCIAS AGRARIAS::AGRONOMIA
• Link de acesso: http://repositorio.ufsm.br/handle/1/37018

The soybean crop plays an evident and prominent role in Brazil and worldwide. However, there is a growing need to adopt strategies that sustainably maximize the productive potential of the areas available for cultivating this oilseed. To meet this demand, the use of Precision and Digital Agriculture tools has become increasingly frequent, one of their applications being the use of vegetation indices as a predictive tool for the physiological quality of soybean seeds. In this context, the objective was to evaluate the relationship between the N-Sensor vegetation index, determined at different times, and soil fertility attributes, with the spatial variability of qualitative attributes and soybean seed yield, in order to improve harvest logistics efficiency. The experiment was conducted in two agricultural fields intended for soybean seed production, located in the Central Depression region of Rio Grande do Sul, during the 2020/2021 and 2021/2022 growing seasons. The N-Sensor vegetation index was determined from Sentinel-2 satellite images, with a spatial resolution of 10 meters, at three times that coincided with the soybean phenological stages R1, R4, and R6. Georeferenced sampling of seeds, soil, and plants was carried out using a sampling grid of one point per hectare. The seed samples were taken to the laboratory for the analysis of physiological quality attributes; the soil samples were analyzed for fertility attributes; and the plant samples were used to measure the crop yield components. The physiological quality data of the seeds were interpolated using the kriging method to generate spatial variability maps. Analyses were performed to categorize the relationship between the physiological quality attributes of seeds and the vegetation index at the three sampling times, as well as principal component analysis considering soil fertility data, yield components, and seed physiological quality attributes, according to the different vegetation index levels and at each evaluation time. The categorization levels between the vegetation index and the physiological quality attributes of the seeds varied according to the sampling time and the physiological attribute evaluated. The principal component analysis revealed the existence of contrasting fertility conditions, productive performance, and seed quality among the different vegetation index zones and evaluation times within the production fields.