Sensoriamento remoto para detecção de seringais

• Ano de defesa: 2017
• Autor(a) principal: Lemes, Ernane Miranda
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Agronomia
• Palavras-chave em Português: Agronomia; Nematóide de galha; Seringueira - Doenças e pragas; Sensoriamento remoto - Agricultura; Nematoide de galha; Hevea brasiliensis; NDVI; Índice razão simples; RapidEye; Root-knot nematode; Simple ratio index; CNPQ::CIENCIAS AGRARIAS::AGRONOMIA
• Link de acesso: https://repositorio.ufu.br/handle/123456789/19716; http://dx.doi.org/10.14393/ufu.te.2017.46

The production of natural rubber from rubber trees (Hevea brasiliensis) is a strategic activity for Brazilian economy. The main factor reducing productivity of this activity is the occurrence of diseases in the rubber plantation. The root-knot nematode (Meloidogyne exigua) is the main root disease of this culture in Brazil and its early identification enables better management of its outbreak in orchards and the reduction of their losses. This study identified areas with the occurrence of M. exigua in rubber plantations through the use of spectral vegetation indices - Normalized Difference Vegetation Index (NDVI) and Simple Ratio Index (SR) - generated from images of the constellation RapidEye® satellites (5 meters resolution). Productive rubber plantations, at least 7 years old, in Minas Gerais (irrigated) and Goiás (non-irrigated) states were sampled and georeferenced in the second half of the rainny season and the quantities of nematodes in soil, roots and total were estimated. The rubber plantation in Minas Gerais was sampled initially following the abnormalities detected by the NDVI index generated through the images of Landsat 8 satellite (30 meters resolution). The rubber plantation in Goiás was sampled randomly to validate the results observed in the first rubber plantation (Minas Gerais). In spite of the identification of significant regressions with the use of high resolution images (RapidEye®) between the variables quantities of nematodes and spectral vegetation indices, none showed a coefficient of determination (R2) greater than 0.31. The descriptive statistics of the data set from both surveys did not identify differences in the quantities of nematodes found in the roots or the total quantity between the areas, however, the amount of nematodes in the soil was approximately 236% greater in Goiás rubber plantation, while both spectral vegetation indices (NDVI and SR) in this rubber plantation were lower than the indices estimated for the rubber plantation in Minas Gerais. This difference was a result of irrigation in the area in Minas Gerais. The correlations of Pearson, Spearman and Kendall were calculated between the densities of nematodes and the spectral vegetation indexes. The correlations between the amount of nematodes in rubber trees plantation and NDVI index showed the best significant coefficients for any of the correlations evaluated. Through the NDVI index it is possible to differentiate an infested rubber tree plantation from a non-infested rubber tree plantation by M. exigua root-knot nematode.