Estimativa da desfolha e o índice de área foliar no patossistema soja-ferrugem asiática utilizando o programa QUANT

• Ano de defesa: 2007
• Autor(a) principal: Oliveira, Max Lenine Rezende de
• Orientador(a): Vale, Francisco Xavier Ribeiro do
• Banca de defesa: Costa, Luiz Cláudio , Jesus Júnior, Waldir Cintra de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Viçosa
• Programa de Pós-Graduação: Mestrado em Fitopatologia
• Departamento: Etiologia; Epidemiologia; Controle
• País: BR
• Palavras-chave em Português: Soja; Desfolhamento; Ferrugem da soja; Imagens digitais; Phakopsora pachyrhizi
• Palavras-chave em Inglês: Soybean; Defoliation; Soybean rust; Digital images; Phakopsora pachyrhizi
• Área do conhecimento CNPq: CNPQ::CIENCIAS AGRARIAS::AGRONOMIA::FITOSSANIDADE::FITOPATOLOGIA
• Link de acesso: http://locus.ufv.br/handle/123456789/4443

The objective of this work was propose a new methodology to quantify the Leaf Area Index (LAI) and defoliation caused by soybean rust, using digital images processing by Quant software. Two experiments were accomplished in areas located in Federal University of Viçosa, Viçosa County - MG, in agricultural year of 2006/2007. To establish the different treatments, seeking a gradient of disease, varying rust intensity and defoliation, the experimental units were powdered with fungicide tebuconazole in different stages of host growing. One area of approximately 2 m2 in the superior part of plants canopy, of each experimental units, was delimited to be taken the digital images to be processed with the software Quant. The digital images were later processed and analyzed by Quant using the color reduction by palette. Colors of the original images were reduced to only two, green (leaf area) and black (soil), which correspond to plant covering and to the area without plant covering, respectively. The green area values in cm2 and the black area (%) were used to quantify the LAI and defoliation caused by rust, respectively. Defoliation data obtained with Quant were correlated with defoliation obtained through destructive sampling. The area values in cm2 were used to estimate a model to quantify LAI, which was after correlated with other quantification methodologies, like destructive sampling and ceptometer. When compared between them, real defoliation and estimated defoliation by software Quant showed high association degree. The model to estimate LAI was significant and showed to be efficient when they were used with the data evaluated until the total closing of canopy or, when the rust started to defoliate the plants and the soil to be exposed. However, it showed to be inefficient to accomplish quantification in the stages that plants started to reach a maximum vegetative point, covering all soil surface. It can be said so, that the program Quant can be used and adopted as a new method to estimate defoliation and Leaf Area Index in soybean with great speed and effectiveness, becoming a useful tool in epidemiologic studies to quantify host growing.