Sensores de vegetação na predição da necessidade na fertilização nitrogenada em cobertura complementar no arroz irrigado

Abstract

Among the factors that determine the productivity of irrigated rice, nitrogen fertilization stands out in the amount demanded by the plant at different phenological stages. Thus, evaluating the nutritional status during different phenological stages of the crop, through sensors, represents an approach capable of increasing the efficiency of fertilization. The objective of this study was to evaluate the use of remote sensing to prescribe nitrogen fertilization in irrigated rice and the efficiency of the use of NDVI, measured by two optical vegetation sensors. The evaluations were carried out during two agricultural years, in stage R1, according to the scale of Caunce et al. (2000) of the selected crop as the main phenological monitoring phase. Sowing of irrigated rice was carried out in a commercial production area and when reaching the phenological stage V2/V3, doses of 0, 69, 138 and 207 kg ha-1 of N were applied, in a case-by-case block design, with eight replicates for each treatment. The vegetation index used to indirectly monitor the nutritional status of rice was the NDVI for both collection platforms, ARP (remotely piloted aircraft) and Proximal, with Mapir Survey 3 and GreenSeeker sensors, respectively. Rice nutritional status was directly evaluated by dry mass yield, shoot N content and absorbed N determined by micro-Kjedahl digestion at R1 stage. The productive stability and gradual positive response to nitrogen fertilization, both in pre-irrigation fertilization and in the combined pre-irrigation fertilization + coverage of 37 kg ha-1 of N in the R1 stage, demonstrated the feasibility of using it in a commercial crop area. the “Rich Strip of N” strategy to maximize the use of the input. Nitrogen doses influenced rice dry mass yield and N content. The highest grain yields were obtained with N content of 2.3% of dry mass in the R1 stage. Both monitoring platforms showed efficiency in detecting N doses through the use of the NDVI index in irrigated rice, but this efficiency varied between sensors and agricultural year, where the vegetation sensor embedded in ARP obtained better performance.