Doses de nitrogênio em tabaco virgínia cultivado na agricultura familiar monitorado por plataforma multiespectral embarcada em aeronave remotamente pilotada

Abstract

Tobacco production is one of the main activities of family farming in southern Brazil, thanks to its high added value and guarantee of purchase by large companies in the sector, which places the country as the second largest producer and largest exporter in the world in this segment. . The objective of this work was to apply Precision Agriculture (PA) techniques, good management practices and nitrogen fertilization variation in Tobacco Virginia, making the production system more profitable and sustainable, for the reality of family farming. The study was carried out in a commercial area of tobacco cultivation in Virgínia, in the municipality of Agudo/RS. Good management practices, precision agriculture principles and different doses of nitrogen were used in the cultivation. A factorial experimental design in randomized blocks was used and two management practices were compared: principle of precision agriculture and good management practices (partial correction of soil acidity in the soil and localized fertilization) and implantation of tobacco in the traditional way, used by the tobacco grower in family farming – Factor 1, with N dosages (Chilean saltpeter) of 60, 100, 170, 200, 250 kg ha ha-1 – Factor 2. Flights were carried out to obtain the vegetation index (IV) MPRI, NDRE, NDVI and NDWI through multispectral images from the RedEdge-Mx sensor on board and adapted in a Remotely Piloted Aircraft, with the IV also generated with data from the FieldSpec spectroradiometer - HH2, in different phenological stages of tobacco. In addition, measurements of the relative chlorophyll content were carried out with a SPAD Chlorophyllometer, measuring the length, width and height of the plants. Data analysis was performed using multivariate statistics with the Factor Analysis (FA) method. Of the 143 variables measured in the experiment, 29 eigenvalues were detected and among these, factors 1 and 2 were the most important, as they carry the greatest variability in the original data set. The variables under study were initially analyzed by three-way ANOVA, to detect whether there was an environmental effect in the blocks, and the influence of this factor was not detected, two-way ANOVA was used and comparison of means by Tukey's test (p=0.05), using Sigmaplot software version 15.0. Graphs for ANOVA and regression analysis were performed using the Statistica 12 program. The IV evolution graphs were performed using an electronic spreadsheet. The results showed that the applications of AP in tobacco cultivation, made it possible to reduce the doses of Nem in relation to that applied by the farmer, has the potential to bring returns in productivity, leaf quality and profitability. The doses of N used in the experiment were adequate to obtain variability in the responses of N content in leaves, productivity, profitability and vegetation indices. The managements and blocks did not obtain significant differences when evaluating the factors that contributed most in the experiment. A longer study period is recommended to detect effects of the application of PA techniques in tobacco cultivation.