The space variability of the indexes of fertility of the soil increases with the no-till system, being verified differences in the nutrients contents of a superficial layer of the soil in relation to the other lower layer. Therefore, it requires the definition of new sampling procedures to contemplate those alterations. The study was aimed at evaluating the spatial variability of the pH, organic matter and micronutrients (B, Cu, Fe, Mn and Zn) in a Acrustox loamy texture, cultivated since 1995 under no till system. The soil was sampled in regular intervals of 50m, in regular mesh of 121 points, in three depths (0-10cm, 10-20cm and 0-20cm). The data evaluated by descriptive statistics and geostatistics, based on semivariograms adjustment. Larger contents of organic matter, Mn and Zn were verified in the first 10cm of the soil, whereas contrary behaviors were observed for Fe. The highest variabilities were found for B and the lowest for pH in water. For the attributes that presented spatial dependence, this was classified as moderate or strong. Most of the data was adjusted to the semivariograms of exponential model, folloed by the spherical model. Only the variable zinc, in the depth of 0-20cm, did not present spatial dependence. The results obtained results should be considered in the definition of plans of soil sampling and management fertilization of the studied area by the analysis of range and horizontal and vertical variability of the chemical attributes evaluated.
geostatístics; soil fertility; semivariogram; sampling
