Variabilidade espacial de atributos do solo e correlação com a produtividade da cultura da soja

Abstract

The Precision Agriculture (PA) is a technology based on the management of spatial variability of production's component, where it seeks to increase production efficiency through site-specific crop management. Therefore, this study aimed to determine the spatial variability and their respective correlations between yield of soybeans with the physical and chemical soil, through a systematic sampling grid with regular distance of 70.0 m, totaling 51 sampling points. The experiment was conducted in an experimental area of 24.8 ha located in the town of Maracaju-MS during soybean harvest 2011/2012. The variables were clay, silt, sand, soil density (SD), macroporosity (Ma), microporosity (Mi), total pore volume (TPV), geometric mean diameter (GMD), mean weight diameter (MWD), resistance penetration (RP), pH CaCl2, phosphorus (P), potassium (K), calcium (Ca), magnesium (Mg), aluminum (Al), zinc (Zn), copper (Cu), manganese (Mn), iron (Fe), cation exchange capacity (CEC), sum of bases (SB), in addition to base saturations (V%), K (% K), Ca (Ca%) Mg (% Mg) and Al ( m%). Already the yield was measured at the end of its cycle by mechanical harvesting. The results of the geostatistical analysis indicated that the sampling grid used was capable of identifying the structure of variability in productivity and most soil properties, however can be considered for defining sampling schemes and management practices. The values range varied from 128.0 to 1176.0 m for the physical attributes, from 113.0 to 378.0 m for the chemical attributes, until 304.0 m for the productivity of soybean. As the correlation analysis, there was significant only for sand, silt and RP to a depth of 0.20 m, demonstrating that the individualized study of physical and chemical factors is inefficient to elucidate the variability of crop yield soybeans.