Abstract: The objective of this research was to explore the distribution characteristics of soil bacterial communities in alpine meadows with different degrees of degradation in the central Qilian mountains. Illumina HiSeq PE250 high-throughput sequencing technology was used to study the bacterial communities in soils of grasslands with light, moderate and severe degradation. The relationships between soil bacterial communities, soil enzyme activities, and soil physiochemical characteristics were also analyzed. It was found that: with increased intensity of degradation, vegetation cover, height, aboveground biomass and diversity index were all significantly decreased (P<0.05), while the soil enzyme activities and soil physiochemical characteristics generally showed significant decline (P<0.05). The high throughput sequencing identified 257971 effective sequences, 219017 high quality sequences and 2004 operational taxonomic units. The three categories of grassland ranked in order: lightly degraded>moderately degraded>severely degraded, for abundance of soil bacterial communities, while the diversity index ranked lightly degraded>severely degraded>moderately degraded grassland. Analysis of Beta diversity showed that the differences among all plots ranked lightly degraded>severely degraded>moderately degraded grassland. The main feature of change in soil bacterial communities across the three categories of grassland degradation was a change in microbial biomass, and the effect on bacterial community structure was not obvious. Actinobacteria, Firmicutes, Acidobacteria, and Proteobacteria were the dominant phyla at all three levels of degradation in the tested grassland soils, with the proportion of the total bacterial population from these phyla being 77.25%, 84.27% and 78.66%, in mildly, moderately and severely degraded soils, respectively. Lactococcus was found to be a dominant genus in all three soil degradation categories, and the corresponding proportion of this genus in the total population was 14.29%, 38.84% and 7.39%, respectively. Redundancy analysis showed that soil enzyme activities and soil physiochemical characteristics differed based on the distribution of bacterial communities, and that the soil pH is the main driving factor affecting the distribution of soil bacterial communities. The degradation-linked changes in soil bacterial communities in alpine meadow grassland in the central Qilian mountains were strongly linked to the soil enzyme activities and the soil physiochemical characteristics.

Key words: Qilian Mountains, alpine meadow, degraded grassland, high throughput sequencing, bacterial community structure