ABSTRACT The present study aimed at assessing the influence of physical properties associated with soil structure on the Srelative index. Two soil conditions under Ficus carica L. cultivation were studied (with or without liquid bovine biofertilizer in the irrigation water), in the 0-0.1 m and 0-0.3 m layers. Disturbed and undisturbed soil samples were collected from the above-mentioned layers and physical analyzes, pertinent to the study objective, were performed. The response function model was applied to verify how the soil independent physical variables influence on the Srelative index in the 0-0.1 m and 0-0.3 m layers, with 0 and 60% biofertilizer, being combined two to two. Once the response functions were known, the slopes of both functions were compared, being represented by the coefficients β11-β12 and β21-β22. This comparison enabled verifying whether there was an effect of the treatments on the response variable. In addition, the standard error of the difference between coefficients was calculated, and the Student's t-test applied. The method of multiple regression was also used to confirm the effect of the variables on the Srelative index for the 0-0.3 m layer in both treatments. Then, the variables with greater weight were selected by a backward elimination method to estimate the Srelative. The results showed that the Srelative index is strongly influenced by properties of the porous fraction, with total porosity and continuity of pores being of significant influence. Management with liquid bovine biofertilizer results in improvement in the soil structure, with effects measured by the Srelative index. | RESUMO Objetivou-se com esta pesquisa estudar a influência de atributos físicos associados à estrutura do solo no índice Srelativo. Foram contempladas duas situações de solo (sob cultivo de Ficus carica L., sem aplicação e com aplicação de biofertilizante bovino líquido na lâmina de irrigação), nas camadas de 0-0,1 m e 0-0,3 m. Foram coletadas amostras de solo com estrutura preservada e não preservada nas camadas supracitadas e realizadas análises físicas pertinentes ao objetivo do estudo. Foi aplicado o modelo de função de resposta para verificar como as variáveis físicas do solo independentes influenciam no Srelativo, nas camadas de 0-0,1 m e 0-0,3 m tratadas com 0 e 60 % do biofertilizante combinadas duas a duas. Conhecidas as funções de resposta compararam-se as inclinações de ambas as funções, representadas pelos coeficientes β11-β12 e β21-β22, para verificar se houve efeito dos tratamentos na variável resposta. Além disso, calculou-se o erro padrão da diferença entre os coeficientes e aplicou-se o teste t de Student. Também foi aplicado o método de regressão múltipla para verificar o efeito das variáveis sobre o Srelativo para a camada de 0-0,3 m em ambos os tratamentos. Em seguida, por meio do método backward elimination foram selecionadas as variáveis com maior peso para a estimativa do Srelativo. Concluiu-se que o índice Srelativo é fortemente influenciado por atributos da fração porosa, sendo a porosidade total e a continuidade de poros os de influência significativa. O manejo com biofertilizante bovino líquido resulta em melhora na estrutura do solo, com efeitos mensuráveis pelo índice Srelativo.

The present study aimed at assessing the influence of physical properties associated with soil structure on the Srelative index. Two soil conditions under Ficus carica L. cultivation were studied (with or without liquid bovine biofertilizer in the irrigation water), in the 0-0.1 m and 0-0.3 m layers. Disturbed and undisturbed soil samples were collected from the above-mentioned layers and physical analyzes, pertinent to the study objective, were performed. The response function model was applied to verify how the soil independent physical variables influence on the Srelative index in the 0-0.1 m and 0-0.3 m layers, with 0 and 60% biofertilizer, being combined two to two. Once the response functions were known, the slopes of both functions were compared, being represented by the coefficients β11-β12 and β21-β22. This comparison enabled verifying whether there was an effect of the treatments on the response variable. In addition, the standard error of the difference between coefficients was calculated, and the Student's t-test applied. The method of multiple regression was also used to confirm the effect of the variables on the Srelative index for the 0-0.3 m layer in both treatments. Then, the variables with greater weight were selected by a backward elimination method to estimate the Srelative. The results showed that the Srelative index is strongly influenced by properties of the porous fraction, with total porosity and continuity of pores being of significant influence. Management with liquid bovine biofertilizer results in improvement in the soil structure, with effects measured by the Srelative index.