Nitrogen is a plant nutrient of great importance. To determine optimum N levels for wheat, a field trial was carried out in a Red Latosol, during the winter of 2005, at Embrapa Cerrados, Planaltina (DF - Brazil). The experiment was set up in a randomized block design, with six replications and the treatments consisted of six N doses (0, 70, 140, 210, 280 and 350 kg ha-1). Soil samples were collected in the 0-20 cm and 20-40 cm layers and three different chemical extractors were used to estimate N availability. Highest determination coefficients were obtained by a quadratic regression between wheat yield and soil N extracted by the different extractors from samples of the 0-20 cm layer, in the following order: KCl 2 mol L-1 a 100 °C (R2 = 0,997**) > H2O2/MnO2 (R² = 0,992**) > buffer pH 11,2 + NO3- (R² = 0,985**) > buffer pH 11,2 (R² = 0,960**). The high determination coefficients suggest that the 0-20 cm layer is the most appropriate for routine N analysis for wheat. The buffer pH 11.2 + NO3- solution can be recommended for routine N analysis, because of its simplicity and high determination coefficient. The economic critical levels for soil N calculated for the different chemical extractors, at a price ratio of 5 to 7 (kg kg-1 of N wheat) were 90 to 68 mg kg-1 (H2O2/MnO2), 87 to 66 mg kg-1 (KCl 2 mol L-1 at 100 °C), 81 to 62 mg kg-1 (buffer pH 11.2 + NO3-) and 64 to 47 mg kg-1 (buffer pH 11.2). The critical levels calculated by the Cate-Nelson mathematical method were 94 mg kg-1 (H2O2/MnO2), 90 mg kg-1 (KCl 2 mol L-1 at 100 °C), 82 mg kg-1 (buffer pH 11.2 + NO3-) and 61 mg kg-1 (buffer pH 11.2).
nitrogen fertilization; Triticum aestivum; chemical extractors
