Spectral detection of nematodes in soybean at flowering growth stage using unmanned aerial vehicles

ABSTRACT: Soybean is one of the main crop species grown in the world. However, there is a decline in productivity due to the various types of stress, including the nematodes Heterodera glycines and Pratylenchus brachyurus. The objectives were to determine the best spectral band for detecting H. glycines and P. brachyurus at the beginning of flowering (R1). Soil and root sampling was conducted at nine sampling sites in each of the five nematode-infested regions, totaling 45 sampling points. Flights were made at all regions using Phantom 4 Advanced, Sequoia and 14-band customized Sentera. For H. glycines, the red spectral band best explained the variability on soil and root nematode counts as well as the second stage of juveniles in soil. For P. brachyurus, Sentera RedEdge best explained the variability in root nematode counts and Sequoia NIR best explained soil juveniles. A multiple linear regression model using spectral data for detecting P. brachyurus and H. glycines improved R² compared to simple linear regressions. At flowering growth stage (R1), soybean spectral reflectance was associated with the number of H. glycines and P. brachyurus on soil and roots using low-cost and multispectral sensors.