Nitrogen (N) fertilization in spring wheat in southern Brazil is based on grain yield potential, soil organic matter content and previous crop (soybean or corn). However, these variables are not precise and subjected to errors, resulting on N losses and yield potential reduction. Moreover, grain yield potential definition is difficult, since it is affected by weather conditions that are variable between years. For nitrogen management, shoot biomass and N uptake are important components and should be used for estimating optimal topdressing nitrogen rates. Such evaluations can be done by characterizing canopy reflectance signatures using active remote sensing. The development of proximal active sensors based on NDVI (normalized difference vegetation index) evaluations as the Greenseeker sensor is an approach for site-specific N fertilization in field crops and one of the tools available to estimate yield potential for nitrogen management in wheat. In this sense, the objective of the present study was to study the variation of NDVI, measured by an active canopy sensor, during crop development as affected by N availability and to evaluate the use of NDVI as a tool for estimating shoot dry biomass and N uptake, as well as to evaluate the use of NDVI for determining optimum topdressing N rates in different wheat genotypes. Experiments were carried out at field conditions in Eldorado do Sul (State of Rio Grande do Sul, southern Brazil) between 2014 and 2016. The experimental site has been cultivated in no-tillage system for 22 years and the soil is a typical red dystrophic argisol with low organic matter content. Climate in the region is subtropical (Cfa type) with wet and humid summers. Treatments consisted of different wheat genotypes and different N rates applied at plant emergence (no N up to 60 kg N/ha) and at the 6th leaf stage (no N up to 80 kg N/ha). NDVI in different growth stages (from 4-leaf-stage up to grain filling) using the Greenseeker sensor, shoot biomass and N uptake at 6th leaf stage and grain yield were evaluated. NDVI values increased during crop ontogeny and were affected by nitrogen availability. This index was efficient to detect growth variability generated by N availability and correlated well with wheat grain yield for all cultivars tested. Models for the relationship between NDVI, shoot biomass and N uptake were developed. The models proposed in this study presented good results, allowing its use for real-time estimation of shoot biomass and N uptake in different wheat genotypes. The N rate of maximum technical efficiency was negatively correlated with NDVI values measured at the moment of N topdressing. Our results indicated that NDVI measurement by an active sensor is an efficient tool to evaluate N nutritional status and grain yield potential. This information can be used for site-specific variable N fertilization according to spatial variability within a field.