Abstract:Soil erosion is an important cause of grassland soil fertility loss and ecosystem degradation. The process of and degree of damage caused by soil and water loss will differ under different grazing systems because of different vegetation coverage rates, and this information is also important to instruct practices and governance polices for ecological and environment protection in the grasslands. The study area, the Hulunbuir grassland, located in the Inner Mongolia region of north China, is dominated by sandy soil and has been subject to grassland degradation and loss of area. In this area, we used different stocking rates to select three grazing system-free grazing, rotation grazing, and no-grazing systems-and then chose one grazing grassland in each system. We collected soil samples, and investigated vegetation growth conditions in all three grazing systems. We also assessed soil total phosphorus and total nitrogen. We uses a rainfall simulator to simulate two rainfall intensities-0.74 mm/min and 1.50 mm/min. Each rainfall intensity in the same grazing pasture two rainfall experiment. In each simulated rainfall experiment, we collected all the runoff in order to test runoff volume. When the rainfall began, we collected runoff samples every 3 min for a total of eight times. We tested runoff, sediments, total phosphorus, and total nitrogen of each runoff sample, and assessed the mechanism of phosphorus and nitrogen loss via rainfall runoff. The result shows that the stocking rates had important effects on reducing surface runoff and sediment between different grazing grassland types. The no-grazing system had the most reclining function and free grazing grassland had the least. With respect to silt content in runoff, free grazing was the least effective system and no-grazing the most effective, suggesting that no-grazing grassland would best reduce soil erosion. Tolerance of rainfall intensity differed under different vegetation coverage. The runoff coefficient was reduced 25.0%-45.7% for the no-grazing grassland under the two rainfall intensities; that of the rotation grazing grassland was reduced 13.2%-20.2%; and the free grazing grassland was reduced 7.5%-12.4%. Runoff nitrate concentrations were influenced significantly by rainfall intensity under the free grazing grassland (nitrogen loss significantly affected) and no-grazing grassland (phosphorus loss significantly affected). Moreover, the nitrogen and phosphorus loss curve followed a power function distribution. Soil and nutrient loss were mainly controlled by runoff volume and silt content in all three grazing systems. The nitrogen and phosphorus loss in sediment were highest under the no-grazing grassland. The rotation grazing system had the lowest enrichment rate. The greatest loss of nitrogen and phosphorus was observed in the free grazing grassland. Nitrogen in sediment first increased and then decreased with increasing grazing intensity. The distribution of phosphorus loss in sediment was contrary to that for nitrogen in the three grazing types. Enrichment was observed in the loss of both nitrogen and phosphorus in runoff. The free grazing grassland had the most significant effects on the main influences. Accordingly, these results could provide vital baseline information for choosing grazing methods and grassland management practices and could facilitate the reduction of nitrate runoff losses and encourage the development of adaptable grazing methods.