ABSTRACT
The management of nitrogen fertilization in wheat has proven to be a limiting factor for grain yield in the southern region of Brazil, and new techniques like the normalized difference vegetation index (NDVI) can be useful for estimating crop yield potential and, consequently, for determining N fertilization requirements. The objective of this study was to evaluate the use of NDVI, obtained through active and passive sensors, as a tool for estimating the grain yield of wheat cultivars under different N application rates. The experiment was conducted in Guarapuava, PR, Brazil, using a randomized block design and a 4 × 5 factorial structure, with four cultivars and five N rates. Evaluations consisted of NDVI readings; shoot dry matter (DM) at the phenological stages of tillering, booting, and flowering; leaf N concentration; and grain yield. For both sensors, significant differences in NDVI readings were found among cultivars in all phenological stages and at all N rates, as well as among N rates and phenological stages. The results revealed DM and leaf N effects on NDVI readings, and the possibility of generating models to estimate these parameters based on NDVI readings and also for estimating grain yield at an early crop development stage.
nitrogen fertilization; remote sensing; Triticum aestivum
