ABSTRACT Spatial variability of soil attributes affects crop development. Thus, information on its variability assists in soil and plant integrated management systems. The objective of this study was to assess the spatial variability of the soil apparent electrical conductivity (ECa), electrical conductivity of the saturation extract (ECse), water content in the soil (θ) and soil texture (clay, silt and sand) of a sugarcane crop area in the State of Pernambuco, Brazil. The study area had about 6.5 ha and its soil was classified as orthic Humiluvic Spodosol. Ninety soil samples were randomly collected and evaluated. The attributes assessed were soil apparent electrical conductivity (ECa) measured by electromagnetic induction with vertical dipole (ECa-V) in the soil layer 0.0.4 and horizontal dipole (ECa-H) in the soil layer 0.0-1.5 m; and ECse, θ and texture in the soil layers 0.0-0.2 m and 0.2-0.4 m. Spatial variability of the ECa was affected by the area relief, and had no direct correlation with the electrical conductivity of the saturation extract (ECse). The results showed overestimated mean frequency distribution, with means distant from the mode and median. The area relief affected the spatial variability maps of ECa-V, ECa-H, ECse and θ, however, the correlation matrix did not show a well-defined cause-and-effect relationship. Spatial variability of texture attributes (clay, site and sand) was high, presenting pure nugget effect. | RESUMO A variabilidade espacial dos atributos do solo, interferem sobre o desenvolvimento dos cultivos. Assim, o conhecimento dessa variabilidade permite o manejo integrado de solo e planta. Objetivou-se determinar a variabilidade espacial da condutividade elétrica aparente (CEa), condutividade elétrica do extrato de saturação (CEes), conteúdo de água (θ) e textura (argila, silte e areia) do solo em uma área cultivada com cana-de-açúcar, no Estado de Pernambuco. A área de estudo possui cerca de 6,5 ha e o solo da área é um Espodossolo Humilúvico órtico. As amostras de solo foram avaliadas em 90 pontos de amostragem distribuídos aleatoriamente. Foram amostrados os seguintes atributos: condutividade elétrica aparente (CEa) medida por indução eletromagnética (dipolo vertical: CEa-V e dipolo horizontal: CEa-H) nas camadas de 0,0-0,4 m e 0,0-1,5 m de profundidade respectivamente. Os demais atributos foram medidos nas camadas de 0,0-0,2 m e 0,2-0,4 m de profundidade (CEes, θ, argila, silte e areia). A variabilidade espacial da condutividade elétrica aparente do solo medida por indução eletromagnética (CEa-V e CEa-H) foi influenciada pelo relevo, não apresentando relação direta com a condutividade elétrica do extrato de saturação do solo (CEes). Os atributos em estudo apresentaram distribuição de frequência com média superestimada, com valores de média se distanciando dos valores de moda e mediana. O relevo influenciou os mapas de variabilidade espacial da CEa-V, CEa-H, CEes e θ, apesar da matriz de correlação não demonstrar relação de causa e efeito bem definida. Os atributos texturais (argila, site e areia) apresentaram elevada variabilidade espacial, apresentando efeito pepita puro (EPP).

Spatial variability of soil attributes affects crop development. Thus, information on its variability assists in soil and plant integrated management systems. The objective of this study was to assess the spatial variability of the soil apparent electrical conductivity (ECa), electrical conductivity of the saturation extract (ECse), water content in the soil (θ) and soil texture (clay, silt and sand) of a sugarcane crop area in the State of Pernambuco, Brazil. The study area had about 6.5 ha and its soil was classified as orthic Humiluvic Spodosol. Ninety soil samples were randomly collected and evaluated. The attributes assessed were soil apparent electrical conductivity (ECa) measured by electromagnetic induction with vertical dipole (ECa-V) in the soil layer 0.0.4 and horizontal dipole (ECa-H) in the soil layer 0.0-1.5 m; and ECse, θ and texture in the soil layers 0.0-0.2 m and 0.2-0.4 m. Spatial variability of the ECa was affected by the area relief, and had no direct correlation with the electrical conductivity of the saturation extract (ECse). The results showed overestimated mean frequency distribution, with means distant from the mode and median. The area relief affected the spatial variability maps of ECa-V, ECa-H, ECse and θ, however, the correlation matrix did not show a well-defined cause-and-effect relationship. Spatial variability of texture attributes (clay, site and sand) was high, presenting pure nugget effect.

RESUMO O estudo das propriedades físicas e químicas do solo é um procedimento de custo e tempo relativamente elevado. Na busca de alternativas para predizer esses atributos a partir de um número menor de amostras do solo, o uso de Redes Neurais Artificiais (RNA) tem sido apontado como uma técnica computacional com grande capacidade de resolver problemas por meio da experiência, e possuem a capacidade de aquisição e posterior aplicação deste conhecimento. Esse trabalho teve por objetivo utilizar a RNA para estimar os atributos físicos e químicos de solo. Os dados utilizados foram provenientes da análise física e química de solo, coletados em 120 pontos amostrais, os quais foram submetidos à análise descritiva, análise geoestatística, treinamento e análise das RNAs. Na análise geoestatística, para cada atributo do solo, foi verificado o modelo de semivariograma que apresentou melhor ajuste ao modelo experimental, e como método de interpolação foi usada técnica da krigagem ordinária. As RNAs foram treinadas, selecionadas considerando a assertividade no mapeamento dos padrões considerados e utilizadas na estimativa de todos dos atributos de solo. O erro médio de cada estimativa obtida pela técnica da krigagem ordinária foi comparado com o erro médio da estimativa obtida pela RNA e, posteriormente foram comparadas com os valores originais por meio do teste-t de Student. Os resultados mostram que a técnica de RNAs apresenta assertividade compatível à krigagem ordinária. O uso da técnica de RNA apresentou-se promissora para obter estimativas de atributos de solo empregando um número menor de amostras de solo. | ABSTRACT Soil physical and chemical analyses are relatively high-cost and time-consuming procedures. In the search for alternatives to predict these properties from a reduced number of soil samples, the use of Artificial Neural Networks (ANN) has been pointed out as a great computational technique to solve this problem by means of experience. This tool also has the ability to acquire knowledge and then apply it. This study aimed at using ANNs to estimate the physical and chemical properties of soil. The data came from the physical and chemical analysis of 120 sampling points, which were submitted to descriptive analysis, geostatistical analysis, and ANNs training and analysis. In the geostatistical analysis, the semivariogram model that best fitted the experimental variogram was verified for each soil property, and the ordinary kriging was used as an interpolation method. The ANNs were trained and selected based on their assertiveness in the mapping of considered standards, and then used to estimate all soil properties. The mean errors of ordinary kriging estimates were compared to those of ANNs and then compared to the original values using Student's t-Test. The results showed that the ANN had an assertiveness compatible with ordinary kriging. Therefore, such technique is a promising tool to estimate soil properties using a reduced number of soil samples.

Soil physical and chemical analyses are relatively high-cost and time-consuming procedures. In the search for alternatives to predict these properties from a reduced number of soil samples, the use of Artificial Neural Networks (ANN) has been pointed out as a great computational technique to solve this problem by means of experience. This tool also has the ability to acquire knowledge and then apply it. This study aimed at using ANNs to estimate the physical and chemical properties of soil. The data came from the physical and chemical analysis of 120 sampling points, which were submitted to descriptive analysis, geostatistical analysis, and ANNs training and analysis. In the geostatistical analysis, the semivariogram model that best fitted the experimental variogram was verified for each soil property, and the ordinary kriging was used as an interpolation method. The ANNs were trained and selected based on their assertiveness in the mapping of considered standards, and then used to estimate all soil properties. The mean errors of ordinary kriging estimates were compared to those of ANNs and then compared to the original values using Student's t-Test. The results showed that the ANN had an assertiveness compatible with ordinary kriging. Therefore, such technique is a promising tool to estimate soil properties using a reduced number of soil samples.