Relação da resposta multiespectral da cultura da soja com parâmetros químicos do solo

Abstract

The development of technology in agriculture is transforming farming practices, with the soybean crop being one of the most benefited. Precision agriculture, driven by tools such as geoprocessing, remote sensing and the use of Unmanned Aerial Vehicles (UAVs), has enabled a deeper understanding of the parameters that permeate agricultural production. This study sought to analyze the relationship between the multispectral response of the soybean crop, using remote sensing with UAVs, and the chemical parameters of the soil. The experiment was carried out in a plot located in the municipality of Marmeleiro - PR. The soybean crop was planted in mid-November, and the cultivar used was “Brasmax Cromo TF”. The chemical parameters were obtained by sampling 40 points; the samples were taken from 0 to 20 cm deep. The Organic Matter, Phosphorus, Potassium, Calcium, Magnesium and Aluminum contents were analyzed, as well as the soil pH, Base Saturation and Cation Exchange Capacity. A DJI Phantom 4 Multispectral drone was used for the aerial survey. A total of three flights were carried out, the first at the close of the rows (V6), the second at the start of flowering (R1), and the third at the end of grain filling (R5.5). DJI Ground Station PRO, Agisoft Metashape PRO and Qgis software were used, while the indices used were NDVI and GNDVI. For both indices, the highest coefficients of determination were between phosphorus and the first flight indices, with R² = 0.3962 and 0.2344 for NDVI and GNDVI, respectively. The lowest correlations were obtained for aluminum, with the best indices being the NDVI from the second flight and the GNDVI from the third flight, with R² = 0.1244 and 0.0105, respectively. Furthermore, some of the correlations obtained were negative, contrary to the initial expectations of the work.