Comparação entre métodos de elaboração e validação de escala diagramática de bacteriose foliar em diferentes clones de Eucalyptus urophylla X Eucalyptus grandis
Data
2016-09-08
Autores
Damasceno, Vítor Flávio Furtado [UNESP]
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
A avaliação da severidade - porcentagem da área lesionada em relação à área total - pode ser feita com auxílio de escalas diagramáticas, que são elaboradas com base na Lei WeberFechner e utilizam regressão linear simples em seu processo de validação, porém apresentam erros na metodologia estatística. O objetivo desta pesquisa foi propor um método de elaboração e validação de escalas de bacteriose em um híbrido Eucalyptus urophylla x Eucalyptus grandis, avaliar a da influência de concentrações de diferentes espécies de bactérias na validação de escalas, verificar a possibilidade de avaliações utilizando tons de cinza, e a influência da dispersão de lesões nas estimativas dos avaliadores no uso desta ferramenta. Foram conduzidos quatro experimentos com amostras das cidades de Bofete-SP e Botucatu-SP. No primeiro experimento, foram utilizadas 100 folhas de três clones (H13, I144 e AEC 1528) de Eucalyptus urophylla x Eucalyptus grandis infectadas naturalmente com bacteriose, que tiveram sua severidade medida para elaboração, por quartis e Weber-Fechner, e validação das seis escalas diagramáticas. No segundo, mudas de eucalipto com três concentrações diferentes (105, 107 e 109 indivíduos/mL) de inóculo de três bactérias (Pseudomonas sp., Xanthomonas sp. e mix) tiveram folhas coletadas para construção de três escalas diagramáticas. No terceiro, imagens digitais das amostras de folhas dos três clones de eucalipto infectados com bacteriose foram analisadas em tons de cinza. Houve estudo de correlação entre as variáveis considerando os três clones como um único grupo e separadamente. No quarto, a escala diagramática construída para o clone AEC 1528 foi repetida em diferentes níveis de dispersão, com a mesma severidade das folhas representativas dos níveis da escala, seguidas de avaliação. No primeiro experimento, o método de quartis apresentou melhoras nas leituras quando comparado ao método Weber-Fechner, houve diferença entre os clones segundo as avaliações e o uso de escalas foi favorável. No segundo, houve melhoras nas estimativas com o uso das escalas. No terceiro, os tons de cinza podem diferenciar a severidade dos clones como um único grupo e há separação de clones por severidade. No quarto, foi detectada influência direta da dispersão das lesões da doença nas estimativas dos avaliadores. Conclui-se que o método de quartis obteve melhores resultados que o método Weber-Fechner; escalas levando em consideração a concentração de inóculo podem resultar em diferenças nas avaliações; tons de cinza podem avaliar severidade e diferenciar clones; variações na dispersão de uma mesma severidade influenciam diretamente as estimativas de um avaliador.
The severity evaluation – percentage of the injured leaf area relative to its total – could be done with the assistance of diagrammatic scales, which are elaborated based on the WeberFechner law and uses simple linear regression for its validation process. However, errors can be found in statistical methodology. The objective of this research was to propose methods of elaboration and validation for diagrammatic scales of bacterial leaf blight on the hybrid Eucalyptus urophylla x Eucalyptus grandis, evaluate the influence of concentrations and different species of bacteria on scales validation, verify the possibilities of evaluations using gray scale, and the influence of leaf injuries dispersion on the estimates done by evaluators using this tool. Four experiments were conducted with samples from Bofete-SP and Botucatu-SP. In the first experiment, 100 leaves were used from three different clones (H13, I144 e AEC 1528) of Eucalyptus urophylla x Eucalyptus grandis naturally infected with bacterial leaf blight, which had the severity measured for the elaboration, by quartiles and Weber-Fechner law, and the validation of the six resultant diagrammatic scales. In the second one, eucalypt leaves with three different concentrations (105, 107 e 109 individuals/mL) of three bacterial inoculums (Pseudomonas sp., Xanthomonas sp. and mix), had their leaves collected for the construction of three diagrammatic scales. In the third, the digital images of the sampled leaves of the three eucalypt clones were analyzed using gray scale values. There was made a correlation among the variables considering the three clones as one single group and separately. In the fourth, the diagrammatic scale built for the clone AEC 1528 was replicated into different levels of dispersion with the same severity of the levels present in the scale, followed by evaluation. In the first experiment, the quartiles method presented reading improvements in comparison with the Weber-Fechner method. There was difference among the clones and the use of the scale was favorable. In the second, there were estimates improvements using the scales. In the third, the gray scale can detect the difference of severity of the clones as a single group and there is clone differentiation by severity. In the fourth, it was detected direct influence of the injuries dispersion in the evaluations. It concludes that the quartiles method had better results than the Weber-Fechner method; scales considering the concentration of the inoculum could result into evaluation differences; gray scale can be used for severity evaluation and clone differentiation; variations of the same dispersion of severity can influence directly the estimates of and evaluator.
The severity evaluation – percentage of the injured leaf area relative to its total – could be done with the assistance of diagrammatic scales, which are elaborated based on the WeberFechner law and uses simple linear regression for its validation process. However, errors can be found in statistical methodology. The objective of this research was to propose methods of elaboration and validation for diagrammatic scales of bacterial leaf blight on the hybrid Eucalyptus urophylla x Eucalyptus grandis, evaluate the influence of concentrations and different species of bacteria on scales validation, verify the possibilities of evaluations using gray scale, and the influence of leaf injuries dispersion on the estimates done by evaluators using this tool. Four experiments were conducted with samples from Bofete-SP and Botucatu-SP. In the first experiment, 100 leaves were used from three different clones (H13, I144 e AEC 1528) of Eucalyptus urophylla x Eucalyptus grandis naturally infected with bacterial leaf blight, which had the severity measured for the elaboration, by quartiles and Weber-Fechner law, and the validation of the six resultant diagrammatic scales. In the second one, eucalypt leaves with three different concentrations (105, 107 e 109 individuals/mL) of three bacterial inoculums (Pseudomonas sp., Xanthomonas sp. and mix), had their leaves collected for the construction of three diagrammatic scales. In the third, the digital images of the sampled leaves of the three eucalypt clones were analyzed using gray scale values. There was made a correlation among the variables considering the three clones as one single group and separately. In the fourth, the diagrammatic scale built for the clone AEC 1528 was replicated into different levels of dispersion with the same severity of the levels present in the scale, followed by evaluation. In the first experiment, the quartiles method presented reading improvements in comparison with the Weber-Fechner method. There was difference among the clones and the use of the scale was favorable. In the second, there were estimates improvements using the scales. In the third, the gray scale can detect the difference of severity of the clones as a single group and there is clone differentiation by severity. In the fourth, it was detected direct influence of the injuries dispersion in the evaluations. It concludes that the quartiles method had better results than the Weber-Fechner method; scales considering the concentration of the inoculum could result into evaluation differences; gray scale can be used for severity evaluation and clone differentiation; variations of the same dispersion of severity can influence directly the estimates of and evaluator.
Descrição
Palavras-chave
Pseudomonas sp., Xanthomonas sp., Weber-Fechner, Quartis, Concentração de inóculo, Tons de cinza, Dispersão de lesões, Doença, Viveiros