Modelagem de dados de cana-de-açúcar: uma abordagem multivariada
Carregando...
Arquivos
Data
2016-11-29
Autores
Araújo, Elton Gean [UNESP]
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
Neste trabalho, foram estudadas as interações entre os componentes de produção e tecnológicos da cana-de-açúcar, bem como da dinâmica de carbono no solo e os fatores latentes formados pelos atributos físicos e químicos do solo, por meio de modelos de análise de equações estruturais, com auxílio das técnicas estatísticas multivariadas análise dos componentes principais e análise fatorial. Ensaios experimentais foram conduzidos em diferentes localidades dos estados de São Paulo (Áreas com Argissolo Vermelho Distrófico abrúptico textura média/argilosa, Latossolo Vermelho Eutroférrico textura muito argilosa e Latossolo Vermelho Eutroférrico textura argilosa) e Mato Grosso do Sul (Área com Latossolo Vermelho-Amarelo distrófico textura franco argilo arenosa). Na camada de 0 a 0,20 m do solo, os componentes de produção e tecnológicos da cana-de-açúcar foram influenciados de forma negativa e significativa apenas pelo fator latente atributos físicos (densidade do solo, porosidade total e macroporos do solo) na ordem de 2,6 (em módulo) mais do que o fator latente atributos químicos associados a acidez do solo (pHCa, pHK e pHag), evidenciando compactação causada pelo tráfego de máquinas (cana crua em 3º corte). No entanto, na camada de 0,20 a 0,40 m, apenas o fator latente atributos químicos influenciou de maneira positiva e significativa os componentes de produção e tecnológicos da cultura, na ordem de aproximadamente 3 vezes mais que os atributos físicos, evidenciando o efeito benéfico da aplicação de gesso após o primeiro corte. Em relação à dinâmica de carbono, expressa neste trabalho pelas variações da emissão de CO2 do solo e do estoque de carbono no solo (calculado para a camada de 0-0,10 m), estas foram principalmente influenciadas pelo fator latente atributos físicos (porosidade livre de água, densidade do solo e macroporos), na ordem de pelo menos 4 vezes (em módulo) mais do que o fator latente atributos químicos (teor de cálcio, teor de magnésio e capacidade de troca de cátions do solo), para as áreas de estudos localizadas no estado de São Paulo (municípios de Pradópolis e Guariba). Na área de Aparecida do Taboado-MS, constatou-se que as variações de emissão de CO2 não sofreram influência significativa dos fatores latentes atributos químicos e físicos; porém, as variações dos valores de estoque carbono, foram explicadas de forma estatisticamente significativas por esses dois fatores, sendo os coeficientes estruturais em relação aos fatores latentes atributos físicos e atributos químicos, iguais a -0,34 e 0,28, respectivamente.
In this work, were studied the interactions between production components and of technology of sugarcane as well as the carbon dynamics in soil and latent factors formed by the physical and chemical properties of the soil, through of structural equations analysis models, with the help of the statistical multivariate principal components analysis and factor analysis. Experimental trials were conducted at different locations in the states of São Paulo (Areas with Dystrophic Red Argisol, with medium-clay texture , Eutroferric Red Latosol very clayey texture and Eutroferric Red Latosol with clay texture) and Mato Grosso do Sul (Area with Dystrophic Red-Yellow Latosol, with sandy clay franc texture). In the layer 0 to 0.20 m above the ground, the components of production and technology of sugarcane were influenced negatively and significantly only by the latent factor physical attributes (bulk density, porosity and macropores), in the order of 2.6 (in module), more than latent factor attributes chemical associated with soil acidity (PHCA, PHK and pHag), showing compaction caused by machinery traffic (crude cane in 3rd cut). However, at the layer 0.20 to 0.40 m, only the latent factor chemical attributes it influenced significantly and positively the production components and technological of the culture, on the order of about 3 times more than the physical attributes, highlighting the beneficial effect of plaster application after the first cut. Regarding the carbon dynamics expressed in this work by the variations of soil CO2 emission and the stock of carbon in the soil (calculated for the layer of 0-0.10 m), these were mainly influenced by the latent factor physical attributes (porosity water-free, bulk density and macropores) in the order of at least 4 times (in module) more than the latent factor chemical attributes (calcium, magnesium content and exchange capacity of soil cations), for areas of localized studies in São Paulo (municipalities of Pradópolis and Guariba). In area Aparecida of Taboado-MS, it was found that variations in CO2 emission suffered no significant influence of the latent factors chemical and physical attributes; however, variations in carbon stock values were explained in statistically significant way by these two factors, being the structural coefficients in relation to the latent factors physical attributes and attributes chemical, equal to -0.34 and 0.28, respectively.
In this work, were studied the interactions between production components and of technology of sugarcane as well as the carbon dynamics in soil and latent factors formed by the physical and chemical properties of the soil, through of structural equations analysis models, with the help of the statistical multivariate principal components analysis and factor analysis. Experimental trials were conducted at different locations in the states of São Paulo (Areas with Dystrophic Red Argisol, with medium-clay texture , Eutroferric Red Latosol very clayey texture and Eutroferric Red Latosol with clay texture) and Mato Grosso do Sul (Area with Dystrophic Red-Yellow Latosol, with sandy clay franc texture). In the layer 0 to 0.20 m above the ground, the components of production and technology of sugarcane were influenced negatively and significantly only by the latent factor physical attributes (bulk density, porosity and macropores), in the order of 2.6 (in module), more than latent factor attributes chemical associated with soil acidity (PHCA, PHK and pHag), showing compaction caused by machinery traffic (crude cane in 3rd cut). However, at the layer 0.20 to 0.40 m, only the latent factor chemical attributes it influenced significantly and positively the production components and technological of the culture, on the order of about 3 times more than the physical attributes, highlighting the beneficial effect of plaster application after the first cut. Regarding the carbon dynamics expressed in this work by the variations of soil CO2 emission and the stock of carbon in the soil (calculated for the layer of 0-0.10 m), these were mainly influenced by the latent factor physical attributes (porosity water-free, bulk density and macropores) in the order of at least 4 times (in module) more than the latent factor chemical attributes (calcium, magnesium content and exchange capacity of soil cations), for areas of localized studies in São Paulo (municipalities of Pradópolis and Guariba). In area Aparecida of Taboado-MS, it was found that variations in CO2 emission suffered no significant influence of the latent factors chemical and physical attributes; however, variations in carbon stock values were explained in statistically significant way by these two factors, being the structural coefficients in relation to the latent factors physical attributes and attributes chemical, equal to -0.34 and 0.28, respectively.
Descrição
Palavras-chave
Biomassa, Modelagem, Rendimento, Sustentabilidade agrícola, Biomass, Modeling, Yield, Agricultural sustainability