Nitrogen limitation has been well documented in grasslands on the Qinghai-Tibet Plateau (QTP), significantly affecting predictions of plant growth and carbon sequestration potential under future climate change scenario. Beside atmospheric deposition, asymbiotic nitrogen fixation (ANF) may be crucial for nitrogen input in QTP grasslands, due to the lack of artificial fertilization and legume plants. However, little is known about the ANF’s contribution to nitrogen input on the QTP. To fill this knowledge gap, we studied the composition, diversity and activity of ANF diazotrophs across the QTP grasslands by using multiple methods of transect sampling, ¹⁵N-labeling and DNA stable isotope probing (SIP), amplicon sequencing, Random Forest algorithm modelling and digital mapping. We found that Skermanella and Mesorhizobium were the most abundant diazotrophic genera. Soil pH and total phosphorus concentration were the dominant driving factors for their composition and diversity. DNA stable isotope probing with ¹⁵N₂ revealed that Mesorhizobium were the most active nitrogen-fixing microorganisms. The potential N-fixation rates of these diazotrophs ranged from 0 to 18.1 kg N ha^–1 yr^–1, resulting in an estimated annual input of approximately 0.50 Tg N across the entire QTP’s alpine grasslands (i.e., ~25% of annual nitrogen input). The most important factor affecting the ANF rate was soil micronutrient molybdenum, a cofactor in the nitrogen-fixing nitrogenase, accounting for 24.64% of the variance. These findings suggested that ANF diazotrophs play important roles in maintaining nitrogen balance in the QTP grasslands and expand our understanding of Mesorhizobium’s ecological roles beyond traditional symbiotic interactions.

The response of N₂O emissions to nitrogen (N) addition is usually positive, but its response to phosphorus (P) addition varies, and the underlying mechanisms for the changes in N₂O emissions remain unclear. We conducted field studies to examine the response of N₂O emissions to N and P addition over two years in three typical alpine grasslands, alpine meadow (AM), alpine steppe (AS), and alpine cultivated grassland (CG) on the Qinghai-Tibet Plateau (QTP). Our results showed consistent increases in N₂O emissions under N addition alone or with P addition, and insignificant change in N₂O emissions under P addition alone in all three grasslands. N addition increased N₂O emissions directly in AM, by lowering soil pH in AS, and by lowering abundance of denitrification genes in CG. N and P co-addition increased N₂O emissions in AM and AS but only showed an interactive effect in AM. P addition enhanced the increase in N₂O emissions caused by N addition mainly by promoting plant growth in AM. Overall, our results illustrate that short-term P addition cannot alleviate the stimulation of N₂O emissions by N deposition in alpine grassland ecosystems, and may even further stimulate N₂O emissions.

When the dominant species in a plant community are palatable, many believe that large herbivores will reduce the dominant species and promote the proportion of previously suppressed species. However, this view may not always hold true. We conducted a 4-year yak grazing experiment on the Qinghai-Tibet Plateau and tracked the plant compositions of the rotational grazing (RG) and grazing exclusion (GE) grasslands during the four years. The results showed that in the absence of yaks under GE, the plant community was dominated by two palatable species, Kobresia pygmaea and Stipa capillata, due to their small leaf area and rapid growth strategy. The presence of yaks under RG significantly inhibited S. capillata and over half of the forbs, while the proportion of K. pygmaea increased and it became the absolute dominant species, contradicting the view that large herbivores inhibit palatable species. Interannually, the dominance of K. pygmaea under RG decreased in the dry year, leading to an increase in the dominance of the other eight species. Under GE, the dominance of K. pygmaea declined notably in the dry year, while S. capillata and seven other forbs increased substantially. Overall, these results suggest that K. pygmaea is grazing-tolerant but not drought-tolerant, whereas the other eight species are drought-tolerant but not grazing-tolerant. At the community level, community composition shifts resulting from succession after grazing exclusion exceeded those caused by drought, drought tends to induce community species turnover while grazing tends to induce species abundance variations. In summary, our conclusions remind ranch managers that when considering the impact of livestock on plant community composition, they should factor in local conditions and climate change rather than simply assuming that livestock will suppress the palatable species.