Growth stagnation on native grassland during the winter season has stimulated the introduction of winter species in order to increase forage offer for animals in the State of Rio Grande do Sul, Brazil. To determine the nutrient losses by erosion as affected by different methods of native grassland improvement, an experiment was carried out at the Agronomic Experimental Station of the Federal University of Rio Grande do Sul, in Eldorado do Sul. A mixture of the winter forage species black oat, Italian ryegrass and arrowleaf clover was introduced on a Paleudult soil under long time native grassland use. Experimental plots of 3,5 by 11,0 m at 0.107 m m-1 of average slope were submitted to simulated rainfall of 64 mm h-1 average intensity and 75 minutes of duration in three different runs: 55 days after soil tillage and sowing; 125 days after soil tillage and sowing (right after the first grazing); and 175 days after soil tillage and sowing (right after the second grazing). The experimental design was completely randomized with five treatments for the introduction of winter species: Zero tillage (Control), No-tillage, Disking; Sub-soiling, and Conventional tillage. During each rainfall, runoff samples were collected every 15 minutes, and the available phosphorus, calcium, magnesium and potassium nutrient concentrations were determined by the extraction method of ionic exchange resin. There was a difference between the rain application runs and the treatments in relation to the nutrient concentrations and losses in the runoff. Highest losses occurred in the first run. In general, highest nutrient losses occurred in the Control treatment and lowest in Conventional tillage. These nutrient losses were not directly related with soil and water losses, but determined by soil surface conditions as well as fertilizer and lime application methods.
soil tillage; winter forrage; runoff; liming; fertilizer supply