Soil quality degradation arising from vegetation deterioration in loess-covered hilly regions threatens the ecological environment and the development of sustainable agriculture. Consequently, improving soil quality and speeding up the process of vegetation restoration is an important goal of ecological rehabilitation projects in these regions. Rhizosphere and non-rhizosphere soil under nine typical plant species from a grassland area in the hilly region of southern Ningxia were used to study their microbial functional diversity using the Biolog method. The objective was to provide scientific evidence for preservation or improvement of soil quality, soil fertility and nutrient use efficiency, thus enhancing protection of the ecological environment.
Soil chemical properties (organic C, total N, ammonium N, nitrate N, total P, available P, and available K), soil enzyme activity (invertase, urease, alkaline phosphatase, and dehydrogenase), soil microbial biomass (Cmic, Nmic, and Pmic) and basal respiration differed between the rhizosphere and non-rhizosphere soils from the nine plant species. Enzyme activities and Cmic, Nmic, and basal respiration in the rhizosphere soil of Artemisia frigida were higher than for the other plant species, whereas the main trend of chemical properties, soil enzyme activity, and soil microbial biomass in the rhizosphere and non-rhizosphere soil of Lespedeza bicolor Turcz were lower. The rhizosphere and non- rhizosphere soils of the nine plant species showed significant differences in soil microbial activity, soil microbial diversity index and uniformity index. Except for Agropyron cristatum, soil microbial functional diversity in the rhizosphere soils was higher than in the non-rhizosphere soils. The main carbon sources of microbes in the rhizosphere of the nine plant species were carboxylic acids and amino acids, and in the non-rhizosphere soil were carboxylic acids, amines, and amino acids. Significant correlations were found between organic C, total N, invertase, Cmic, and Nmic in the rhizosphere and non-rhizosphere soil from the different plant species. Significant correlations were found between soil microbial activity, soil microbial diversity index and uniformity index under the different plant species, but correlations of soil microbial activity, soil microbial diversity index and uniformity index with other indicators were not significant. Principal components analysis showed that organic C, total N, invertase, alkaline phosphatase, Cmic, Nmic, basal respiration, the soil microbial diversity index, and the uniformity index can be used as indices to evaluate soil quality under the nine plant species in this typical grassland area.