Análise espacial e temporal de atributos do solo e da produtividade das culturas visando manejo específico em áreas agrícolas
Carregando...
Data
2013-03-21
Autores
Rodrigues, Marcos Sales [UNESP]
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
Os atributos do solo e a produtividade das culturas podem variar no tempo e no espaço, assim como suas relações de causa/efeito. O presente trabalho objetivou analisar atributos do solo e da produtividade das culturas visando o manejo específico de áreas agrícolas. Para isto, foram utilizadas três ferramentas: uso de modelos de regressão espaço-temporal para entendimento da relação entre atributos do solo e produtividade do milho; uso de algoritmo de agrupamento 'fuzzy c-means' para delineamento de zonas homogêneas de manejo; e uso de um sensor de pH do solo em tempo real. Os dois primeiros estudos foram realizados em Jaboticabal, SP. Foram avaliados a produtividade de milho e os atributos químicos e físicos do solo de seis anos agrícolas entre os anos de 2001 a 2010. O primeiro estudo avaliou quatro modelos de regressão os quais foram: mínimos quadrados ordinários; mínimos quadrados generalizados assumindo heteroscedasticidade; modelo espacial e temporal, assumindo homocedasticidade; e modelo espacial e temporal, assumindo heteroscedasticidade. Verificou-se que o modelo espacial e temporal, assumindo heteroscedasticidade foi superior aos demais e que a acidez do solo (pH) foi o fator que mais influenciou a variabilidade da produtividade de milho em todos os anos agrícolas. No segundo estudo, verificou-se que o uso de algoritmo de agrupamento fuzzy c-means' foi eficiente para delinear zonas homogêneas de manejo, quando foi considerada a variabilidade espacial e temporal dos atributos do solo e da produtividade de milho. O terceiro estudo foi conduzido em três campos de produção agrícola nos EUA, e o objetivo foi avaliar o sensor de pH em tempo real Veris (MSP)® em Kentucky, avaliando a acurácia das medidas de pH, 'buffer pH' e a necessidade de calagem usando técnicas analíticas e propondo um...
Soil attributes and crop yield may vary over space and time as well as their cause-effect relationships. Therefore, the objectives of this study were analyze soil attributes and crop yield to apply site-specific management on crop fields, in this regard, it was used three techniques, which were: models that accounted for spatial-temporal autocorrelation to evaluate the relationship between corn yield and soil attributes; identify management zones using fuzzy c-means clustering analysis based on the spatial and temporal variability of soil attributes and corn yield, and use the Veris (MSP)® on-the-go soil sensing system to lime requirement recommendation. The first two studies were carried out in Jaboticabal, SP, Brazil. Soil chemical and physical attributes were measured between 2001 and 2010 year. Ordinary least square, generalized least squares assuming heteroscedasticity, spatial-temporal least squares assuming homoscedasticity, and spatial-temporal model assuming heteroscedasticity analyses were used to estimate corn yield in the first study. The spatial and temporal autocorrelation assuming heteroscedasticity was superior to the other studied model for prediction and soil acidity (pH) was the factor that most influenced corn yield over time in this study. In the second study, which the fuzzy c-means cluster analysis based on the spatial and temporal variability of soil attributes and corn yield was used, it was observed that this technique was efficient to delineate management zones. The third study was carried out in three crop fields in Kentucky, EUA, and it aimed to evaluate the Veris on-the-go pH sensor system for soil pH management by assessing the accuracy of soil pH measurements and associated spatial predictions of buffer pH, and lime requirement proposed a new method for correcting interpolation... (Complete abstract click electronic access below)
Soil attributes and crop yield may vary over space and time as well as their cause-effect relationships. Therefore, the objectives of this study were analyze soil attributes and crop yield to apply site-specific management on crop fields, in this regard, it was used three techniques, which were: models that accounted for spatial-temporal autocorrelation to evaluate the relationship between corn yield and soil attributes; identify management zones using fuzzy c-means clustering analysis based on the spatial and temporal variability of soil attributes and corn yield, and use the Veris (MSP)® on-the-go soil sensing system to lime requirement recommendation. The first two studies were carried out in Jaboticabal, SP, Brazil. Soil chemical and physical attributes were measured between 2001 and 2010 year. Ordinary least square, generalized least squares assuming heteroscedasticity, spatial-temporal least squares assuming homoscedasticity, and spatial-temporal model assuming heteroscedasticity analyses were used to estimate corn yield in the first study. The spatial and temporal autocorrelation assuming heteroscedasticity was superior to the other studied model for prediction and soil acidity (pH) was the factor that most influenced corn yield over time in this study. In the second study, which the fuzzy c-means cluster analysis based on the spatial and temporal variability of soil attributes and corn yield was used, it was observed that this technique was efficient to delineate management zones. The third study was carried out in three crop fields in Kentucky, EUA, and it aimed to evaluate the Veris on-the-go pH sensor system for soil pH management by assessing the accuracy of soil pH measurements and associated spatial predictions of buffer pH, and lime requirement proposed a new method for correcting interpolation... (Complete abstract click electronic access below)
Descrição
Palavras-chave
Solos - Analise, Produtividade agrícola, Milho, Análise espacial (Estatística), Calagem dos solos, Algoritmos, Precision farming
Como citar
RODRIGUES, Marcos Sales. Análise espacial e temporal de atributos do solo e da produtividade das culturas visando manejo específico em áreas agrícolas. 2013. xii, 90 p. Tese (doutorado) - Universidade Estadual Paulista Júlio de Mesquita Filho, Faculdade de Ciências Agrárias e Veterinárias de Jaboticabal, 2013.