氮是地球上所有生物的关键营养物质，在调节陆地生态系统的结构和功能上扮演重要角色。草原生态系统是陆地生态系统重要的组成部分，放牧作为人类管理利用草原生态系统最主要的途径，是自然草原非常重要的一种扰动类型，同时也是生态系统功能关键的调控因子，对草原土壤氮转化过程有着深刻的影响。目前，关于放牧对草原土壤氮转化过程影响的研究结果还不一致。 本研究以内蒙古锡林郭勒盟的典型草原为研究对象，实验设计为4个放牧压梯度，3个空间重复。放牧梯度分别为未放牧（CK），轻度放牧（GI170，小区放养4只家畜，放牧压为170 SSU d hm-2 y-1，170），中度放牧（GI340，小区放养8只家畜，放牧压为340 SSU d hm-2 y-1，340），重度放牧（GI510，小区放养12只家畜，放牧压为510 SSU d hm-2 y-1，510）。经过2016和2017两年的实验，从土壤氮含量、氮矿化量和N2O通量三个方面，探究了不同放牧强度对土壤氮转化过程的影响，得出的主要研究结论如下： 在2016年和2017年，不同放牧强度之间土壤氮储量都没有显著的差异（P > 0.05）。土壤全氮含量随着土壤深度的增加而逐渐降低，且表层土壤（0-10cm）全氮含量最高。2016年，同一土层不同放牧强度之间土壤全氮含量没有显著性差异（P > 0.05）。2017年，不同放牧强度0-10cm、10-20cm土壤全氮有显著性的差异（P < 0.05），且中度放牧样地土壤全氮含量最高。土壤全氮含量与土壤有机碳呈极显著的正相关（P < 0.01），与土壤粘粒含量、地上枯落物含量以及地上总生物量有显著的正相关关系（P < 0.05），与土壤容重存在着显著的负相关关系（P < 0.05）。在整体上，2017年土壤氮储量低于2016年，说明放牧能够降低土壤氮储量。 在2016年和2017年的生长季中，不同放牧强度下土壤净硝化累积量和净矿化累积量变化趋势大体一致，表现出强烈的季节动态变化。2016年，不同放牧强度之间土壤净硝化和矿化速率没有显著差异（P > 0.05）。2017年6-7月，土壤净硝化和净矿化速率最高且为正值，重度放牧样地土壤净硝化和净矿化速率显著高于其它样地（P < 0.05）。在7-9月，土壤净硝化和净矿化速率都为负值。此外，土壤中硝化微生物氨氧化古菌AOA的含量显著高于氨氧化细菌AOB的含量，且AOA在硝化作用过程中的作用大于AOB。在整体上，未放牧样地土壤净硝化累积量和净矿化累积量最高，说明放牧能够降低土壤净矿化累积量。 在2016年和2017年的生长季，不同放牧强度样地土壤N2O通量呈上下波动的变化，且N2O通量平均值之间并不存在显著差异（P > 0.05）。2016年6月，不同放牧强度样地土壤的N2O通量较高；2017年生长季，不同放牧强度样地土壤N2O通量大体呈先升高后降低的趋势，两年生长季的变化范围分别为-2.49 ~ 10.09 μg m-2 h-1和-4.68 ~ 9.79 μg m-2 h-1。2016年，土壤N2O通量与矿化速率、硝化速率呈显著负相关（P < 0.05），与氨化速率呈显著正相关（P < 0.05）；2017年，土壤N2O通量与矿化速率、硝化速率、氨化速率都呈显著负相关（P < 0.05）。在整体上，放牧样地N2O的累积通量都高于未放牧样地，说明放牧能够增加N2O的排放。 DNDC模型能够在点位尺度上模拟土壤氮储量，土壤有机碳、容重是模型模拟土壤氮储量变化的敏感因素，年份之间氮储量模拟值的变化幅度较实测值的变化幅度小。土壤N2O通量的模拟值与实测值的拟合度较高，但是模型模拟N2O通量的变化趋势与实测值有一定差异，土壤有机碳含量、土壤容重、pH值、生物量、植物固氮系数、孔隙度是N2O通量变化的敏感因素。

Nitrogen is a key nutrient for all living things on earth and plays an important role in regulating the structure and function of terrestrial ecosystems. Grassland ecosystem is an important part of terrestrial ecosystem. Grazing is the most important approach for human management and utilization of grassland ecosystem. It is a very important type of disturbance in natural grassland and a key regulator of ecosystem function and has a profound effect on soil nitrogen. At present, the results of the impact of grazing on grassland soil nitrogen transformation are still inconsistent. This research took the typical grassland of Xilin Gol in Inner Mongolia as the research area. The experiment was designed with 4 grazing intensities and 3 spatial replicates. The grazing gradients were non-grazing (CK), light grazing (GI170, with 4 domesticated animals at grazing intensity of 170 SSU d hm-2 y-1), moderate grazing (GI340, with 8 domesticated animals at grazing intensity of 340 SSU d hm-2 y-1), heavy grazing (GI510, with 12 domesticated animals at grazing intensities of 510 SSU d hm-2 y-1). The experiments were carried out in 2016 and 2017, and the effects of different grazing intensities on soil nitrogen transformation were explored in terms of soil nitrogen content, nitrogen mineralization, and N2O flux. The main research conclusions are as follows: There was no significant difference in soil nitrogen storage between different grazing intensities in 2016 and 2017 (P > 0.05). The total nitrogen content of soils with different grazing intensities gradually decreased with the increase of soil depth, and the total nitrogen content in soil surface soil (0-10cm) was the highest. In 2016, there was no significant difference in soil total nitrogen content among different grazing in the same layer (P > 0.05). In 2017, there were significant differences in soil total nitrogen of different grazing intensities in soil layer of 0-10cm and 10-20cm (P < 0.05), and the soil total nitrogen content was the highest in moderately grazing plots. Soil total nitrogen content was significantly positively correlated with soil organic carbon (P < 0.01), and positively correlated with soil clay content, aboveground litter content, and total aboveground biomass (P < 0.05) and significantly negatively correlation with soil bulk density (P < 0.01). On the whole, soil nitrogen storages in 2017 were lower than in 2016, indicating that grazing can reduce soil nitrogen storages. During the growing seasons in 2016 and 2017, the trends of soil cumulative net nitrification and cumulative net mineralization were consistent and showed obvious seasonal changes. In 2016, there was no significant difference in soil net nitrification and mineralization rates between different grazing intensities (P > 0.05). From June to July 2017, the soil net nitrification and net mineralization rates were the highest and positive and the values of the heavily grazing plots were significantly higher than those of grazing plots (P < 0.05). From July to September, the soil net nitrification and net mineralization rates were negative. In addition, the abundance of AOA in soil was significantly higher than that of AOB, and the effect of AOA on nitrification was greater than that of AOB. On the whole, the values were the highest in the non-grazing plots and this shows that grazing can reduce the amount of soil net mineralization. N2O fluxes fluctuated in different grazing intensities, and there were no significant difference between average values. In June 2016, N2O fluxes were higher in different grazing intensities. During the growing season in 2017, N2O fluxes in different grazing intensities first increased and then decreased. The variation ranges of two years were -2.29 ~ 10.09 μg m-2 h-1 and -4.68 ~ 9.79 μg m-2 h-1, respectively. During the growing seasons in 2016 and 2017, there were no significant differences in the average N2O fluxes between different grazing intensities (P > 0.05). In 2016, soil N2O flux was significantly negatively correlated with mineralization rate and nitrification rate (P < 0.05) and positively correlated with ammonification rate (P < 0.05); in 2017, soil N2O flux and mineralization rate, nitrification rate and ammonification rate were all negatively correlated (P < 0.05). On the whole, the cumulative fluxes of N2O in grazing plots were higher than that of non-grazing plots, indicating that grazing can increase N2O emission. The DNDC model can simulate soil nitrogen storage at the point scale. Soil organic carbon and bulk density are sensitive factors for changes in soil nitrogen storage. Through the comparison between simulated values and measured values, the variation range of the simulated value of nitrogen storage between years was smaller than measured values. The fitting degree of simulated and measured values of soil N2O flux was relatively high, but there was a certain difference between the simulated and the measured values of N2O flux. Soil organic carbon content, soil bulk density, pH value, biomass, and plant nitrogen fixation coefficient, porosity are sensitive factors in the change of N2O.