Mapeamento da responsividade ao uso do nitrogênio por meio de imagens na cultura do milho

Abstract

High-throughput phenotyping (HTP) is a strategic tool for maize breeding, enabling rapid and accurate evaluation of agronomic traits. This study aimed to select RGB image–based vegetation indices for identifying productive maize genotypes under distinct nitrogen fertilization conditions at the plot level. Thirty-five maize genotypes were evaluated in a randomized complete block design arranged in a split-plot scheme with three replications, subjected to two nitrogen levels (20 kg ha⁻¹ and 160 kg ha⁻¹). Four UAV flights were conducted at two altitudes (60 m and 80 m), and mixed models (REML/BLUP) were used for genetic and spatial analyses. The results showed that the flight performed at 61 days after planting (V7/V8) at an altitude of 60 m presented the highest repeatability and accuracy. The BGI (Blue Green Pigment Index) vegetation index showed high sensitivity in discriminating against nitrogen levels and in the indirect selection of productive genotypes. Genotypes G42, G4, G34, G31, G20, G16, and G1 were identified as superior for grain yield. It is concluded that blue-band–based indices, such as BGI, are efficient for selecting genotypes under nitrogen stress before flowering.