INSETOS HERBÍVOROS EM PLANTIO DE SOJA NO SUDESTE GOIANO

Abstract

Integrated Pest Management (IPM) helps to reduce production costs in agricultural crops of economic importance. Sampling, one of the pillars of the MIP, can reveal which insects use a specific plant species as a host. And this helps the farmer to make a decision for control purposes. The objective was to evaluate the occurrence of insect pests in the soybean crop during the 2018/2019 harvest in different soybean genotypes with IPRO technology. The study was conducted in the municipality of Campo Alegre de Goiás, GO, Brazil. The experimental design used was in randomized blocks consisting of 10 soybean genotypes: Bônus, Desafio, Flecha, Foco, ICS7019, M5917, M7110, Power, ST721 and ST797, with four replications. During the cycle of each genotype, weekly samplings were carried out according to the phenological stages using the vertical batting cloth. Survey data and insect identification were noted in identification worksheets called MIP Soybean. The variables were subjected to linear correlation, after the phylogenetic tree was built, then the biplot canonical variables method was used, in addition to unsupervised computational learning through Artificial Neural Networks, using the K-means and Kohonen Map algorithms. The genotypes differed in terms of the severity of the pest attack, showing a strong correlation between the phenological state and the incidence of caterpillars and bedbugs. The 8473RR genotype hosted more lepidopterans, however, less heteropterans in our field observations.