Correlação linear e espacial dos atributos físicos do solo em área sob cultivo de soja no Sistema Plantio Direto
Carregando...
Arquivos
Data
2023-03-03
Autores
Mário, Nilton Eugénio
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
A geoestatística é uma ferramenta extremamente importante e indispensável quando se trata de estudos dedicados à análise ou avaliação de atributos do solo e a consequente geração de semivariogramas e mapas de krigagem. Deste modo, a presente pesquisa objetivou-se em avaliar as correlações entre os atributos físicos do solo em área sob cultivo de soja e sua variabilidade espacial no sistema plantio direto, nas camadas de 0,00 – 0,20 m (1) e 0,20 – 0,40 m (2) de profundidade. A pesquisa foi conduzida em uma área localizada no município de Selvíria – MS, onde foram realizadas coletas de dados de solo utilizando-se uma malha experimental de 0,5 ha, totalizando-se 60 pontos amostrais georreferenciados para análise dos seguintes atributos físicos: a granulometria (Are, Arg e Sil), macroporosidade (Ma), microporosidade (Mi), a porosidade total determinada (PTd), densidade do solo (DS), resistência a penetração com penetrômetro de impacto (RPi), resistência a penetração com o penetrômetro de bancada (RPpb), umidade gravimétrica (UG), densidade de partícula (DP), porosidade total calculada (PTc) e DNAt, com base nas técnicas estatísticas e geoestatísticas feitas por meio de semivariogramas experimentais e mapas de krigagem com auxílio do software estatístico SAS e GS+ 7.0, respectivamente. Foram encontradas correlações que apresentaram melhores resultados na primeira camada do solo, para os seguintes pares de atributos: SIL1 x RPi1, ARE1 x UG1, SIL1 x DP1, SIL1 x ARG1, DNAt x ARG1, ARE1 x Mi1, DNA x SIL1, PTc1 x RPb1, PTc1 x Ma1, DNAt x RPb1, e para a segunda camada, os seguintes pares: ARG2 x RPi2, ARE2 x RPi2, ARE2 x DP2, Mi2 x ARE2, Mi2 x ARG2, PT2 x ARG2, DS2 x Mi2 e PTc2 x Mi2. Além disso, verificou-se que a camada de 0,00-0,20 m de profundidade, foi a que apresentou maiores variabilidades dos atributos físicos com base no coeficiente de variação. O DNAt do solo quando correlacionado com atributos em estudo, mostrou-se positiva para SIL1 e RPb1 e negativo para ARG1, ambos na primeira camada do solo. Entretanto, dos atributos pesquisados, a maior parte deles não apresentaram dependência espacial, isto é, demostraram-se ser aleatórios, heterogêneos e com muita variabilidade espacial, podendo-se modelar seus semivariogramas do tipo efeito pepita puro.
Geostatistics is an extremely important and indispensable tool when it comes to studies dedicated to the analysis or evaluation of soil attributes and the consequent generation of semivariograms and kriging maps. Thus, the present research aimed to evaluate the correlations between the physical attributes of the soil in an area under soybean cultivation and its spatial variability in the no-tillage system, in the layers of 0.00 - 0.20 m (1) and 0 .20 – 0.40 m (2) depht. The research was conducted in an area located in the municipality of Selvíria - MS, where soil data were collected using an experimental mesh of 0.5 ha, totaling 60 georeferenced sampling points for analysis of the following physical attributes: granulometry (Are, Arg and Sil), macroporosity (Ma), microporosity (Mi), the total porosity determined (PTd), soil density (DS), penetration resistance with impact penetrometer (RPi), resistance to penetration with the bench top penetrometer (RPpb), gravimetric humidity (UG), particle density (SD), calculated total porosity (PTc) and DNAt, based on statistical and geostatistical techniques made by means of experimental semivariograms and kriging maps with the aid of statistical software SAS and GS+ 7.0, respectively. Correlations were found that presented better results in the first layer, for the following pairs of attributes: SIL1 x RPi1, ARE1 x UG1, SIL1 x DP1, SIL1 x ARG1, DNAt x ARG1, ARE1 x Mi1, DNAt x SIL1, PTc1 x RPb1, PTc1 x Ma1, DNA x RPb1, and for the second layer, the following pairs: ARG2 x RPi2, ARE2 x RPi2, ARE2 x DP2, Mi2 x ARE2, Mi2 x ARG2, PT2 x ARG2, DS2 x Mi2 and PTc2 x Mi2. In addition, it was found that the 0.00-0.20 m depth layer presented the greatest variability of physical attributes based on the coefficient of variation. In addition, it was found that the 0.00-0.20 m depth layer presented the greatest variability of physical attributes based on the coefficient of variation. Soil DNAt, when correlated with the attributes under study, was positive for SIL1 and RPb1 and negative for ARG1, both in the first soil layer. However, of the attributes researched, most of them did not present spatial dependence, that is, they were shown to be random, heterogeneous and with a lot of spatial variability, and their semivariograms of the pure nugget effect type could be modeled.
Geostatistics is an extremely important and indispensable tool when it comes to studies dedicated to the analysis or evaluation of soil attributes and the consequent generation of semivariograms and kriging maps. Thus, the present research aimed to evaluate the correlations between the physical attributes of the soil in an area under soybean cultivation and its spatial variability in the no-tillage system, in the layers of 0.00 - 0.20 m (1) and 0 .20 – 0.40 m (2) depht. The research was conducted in an area located in the municipality of Selvíria - MS, where soil data were collected using an experimental mesh of 0.5 ha, totaling 60 georeferenced sampling points for analysis of the following physical attributes: granulometry (Are, Arg and Sil), macroporosity (Ma), microporosity (Mi), the total porosity determined (PTd), soil density (DS), penetration resistance with impact penetrometer (RPi), resistance to penetration with the bench top penetrometer (RPpb), gravimetric humidity (UG), particle density (SD), calculated total porosity (PTc) and DNAt, based on statistical and geostatistical techniques made by means of experimental semivariograms and kriging maps with the aid of statistical software SAS and GS+ 7.0, respectively. Correlations were found that presented better results in the first layer, for the following pairs of attributes: SIL1 x RPi1, ARE1 x UG1, SIL1 x DP1, SIL1 x ARG1, DNAt x ARG1, ARE1 x Mi1, DNAt x SIL1, PTc1 x RPb1, PTc1 x Ma1, DNA x RPb1, and for the second layer, the following pairs: ARG2 x RPi2, ARE2 x RPi2, ARE2 x DP2, Mi2 x ARE2, Mi2 x ARG2, PT2 x ARG2, DS2 x Mi2 and PTc2 x Mi2. In addition, it was found that the 0.00-0.20 m depth layer presented the greatest variability of physical attributes based on the coefficient of variation. In addition, it was found that the 0.00-0.20 m depth layer presented the greatest variability of physical attributes based on the coefficient of variation. Soil DNAt, when correlated with the attributes under study, was positive for SIL1 and RPb1 and negative for ARG1, both in the first soil layer. However, of the attributes researched, most of them did not present spatial dependence, that is, they were shown to be random, heterogeneous and with a lot of spatial variability, and their semivariograms of the pure nugget effect type could be modeled.
Descrição
Palavras-chave
Camadas do solo, Semivariograms, Geoestatística, Analise do solo, Solos porosidade