The study of the spatial variability of soil attributes can help in the determination of a specific management strategy that enhances agricultural productivity. The present study aimed at the evaluation of a possible influence of the soil management on spatial variability of a Dystroferric Red Latosol (Oxisol) physical properties. The study was carried out in 2002, in three adjacent areas, in Lavras, Minas Gerais state, Brazil, with the following management systems: A1 - fallowing for three years; A2 - recently planted coffee plants and A3 - area with three-year-old coffee plants. One hundred and eighty-eight grid points were sampled at distances of 1, 5 and 10 m. Disturbed soil samples were collected at each sampling point in the 0-20 cm layer to evaluate the clay, silt, and sand contents and particle density. Data were analyzed by classical statistics (Sisvar software) and geo-statistics (GeoR software). It was concluded that management and land position influenced the spatial variability of the studied attributes, and the widest range of semivariograms was found in the areas with greatest soil disturbance. On average, silt and particle density presented the largest and smallest variation coefficients, respectively.
soil management; geostatistics; kriging
