The objective of the research was to evaluate the efficiency of the sampling grids, used in areas managed with precision agriculture, for the characterization of spatial variability levels of phosphorus (P) and potassium (K). The study was conducted in 30 agricultural areas, located in Rio Grande do Sul state, presenting soils classified as Oxisols. The sizes of the sampling grids studied were: 100x100m (10 areas), 142x142m (10 areas) and 173x173m (10 areas).Were analyzed the levels of P and K, in depth of sampling from 0.00-0.10m. The data were subjected to descriptive statistical analysis and geostatistical analysis. In, general, the areas showed average levels of P and K suitable for crops development. However, were observed subareas with low levels of P, justifying fertilization on site-specific. Based on geostatistical parameters, sizes of the sampling grids used are efficient to capture the different scales of spatial variability of P and K, however, it is indicated smaller sampling grids to consider the spatial variability over short distances.
precision agriculture; soil sampling; soil fertility; geostatistical; Oxisols
