Cultivo de milho sob restrição hídrica, relações entre adubação nitrogenada, índices de vegetação e produtividade em ambiente irrigado e não irrigado

• Ano de defesa: 2023
• Autor(a) principal: Leite, Felipe Guadagnini
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Santa Maria; Brasil; Engenharia Agrícola; UFSM; Programa de Pós-Graduação em Engenharia Agrícola; Centro de Ciências Rurais
• Palavras-chave em Português: Sensoriamento remoto suborbital; Irrigação; Disponibilidade hídrica; Agricultura de precisão; Suborbital remote sensing; Irrigation; Water availability; Precision agriculture; CNPQ::CIENCIAS AGRARIAS::ENGENHARIA AGRICOLA
• Link de acesso: http://repositorio.ufsm.br/handle/1/29645

Corn cultivation is an important activity in Brazilian agribusiness and contributes in several ways to the country's economy. The aim of this research was to use Precision Agriculture (PA) technologies in corn cultivation, such as remote sensing with remotely piloted aircraft (RPA), generation of vegetation indices (VI), varying nitrogen (N) doses, and irrigation to understand their relationships with each other and with productivity. The research was carried out on a commercial corn plantation at UFSM Campus in Santa Maria-RS, using the hybrid Brevant B2688PWU (80,000 plants ha-1 ), in both irrigated and non-irrigated environments. An experimental block of 20m x 45m was installed in each environment, with four micro-weather stations installed in each block to monitor water inputs (precipitation + irrigation). Irrigation was recommended through the Irrigation System platform, with only 36.25 mm of water applied in three irrigations between V5 and V12 (11.56% of the recommended amount), resulting in a -140 mm water deficit during this period. Different doses of N were applied using NPK at sowing and urea (45% N) broadcasted at V4, V7, and for the highest N dose also at VT, with doses of 30 (only at sowing), 122.6, 147, 180, 250 kg N ha-1 . Throughout the productive cycle (six phenological stages), the VI NDVI, NDRE, NDWI, PSRI, and MPRI were obtained using the multiespectral RedEdge-MX Micasense sensor on the Phantom 4 DJI RPA. The five treatments were randomly distributed in the blocks, with an unbalanced number of repetitions. The micro-weather station data were interpolated in a GIS, the treatment repetitions were delimited, and the VI data were obtained through zonal statistics. Productivity, precipitation, irrigation, and total water were determined, along with the N doses and coverage fertilizations, which were verified by factorial multivariate analysis (FMA) for the environments and experiment. Correlation, regression, and ANOVA analyses were also performed for the most expressive variables in the experiment. In the FMA, the effect of low water availability was observed, mainly in the non-irrigated environment. For the experiment, MPRI and NDRE VI presented high positive correlations with water inputs and productivity, whereas the correlations with N doses were low. PSRI and NDWI VI presented moderate negative correlations with water inputs and productivity, whereas with N doses, the correlations were low. ANOVA, for productivity, showed that water inputs were the significant factor (F = 123.30; p < 0.001) compared to N doses (F = 1.108; p = 0.362). The average grain productivity in the irrigated environment (7.26 Mg ha-1 ) was higher than in the non-irrigated environment (3.51 Mg ha-1 ) according to the Tukey test (p = 0.05). Regression analyses between N doses and productivity in BIR indicated (r²=0.41), in BNIR (r²=0.591), and for both (r²=0.508). It is concluded that water deficit affected N utilization, VI reflected stress, and productivity was reduced. Irrigation attenuated its effect, and there was higher productivity in the irrigated environment, being 106.84% higher than in the non-irrigated environment.